SAVITRIBAI PHULE PUNE UNIVERSITY

A PROJECT REPORT ON

C2C- PREDICTIVE ANALYSIS OF STUDENT CAMPUS

PLACEMENT

SUBMITTED TOWARDS THE

FULFILLMENT OF THE REQUIREMENTS OF

BACHELOR OF ENGINEERING (Computer Engineering)

BY

Vilas Dasalkar B191054219

Nihar Jachak B191054245
Kajol Pawar B191054311
Prasad Kurale B191054318

Under The Guidance of

Prof.A.M.Chadchankar

DEPARTMENT OF COMPUTER ENGINEERING

Sinhgad Technical Education Society’s
NBN Sinhgad Technical Institutes Campus
NBN Sinhgad School of Engineering, Ambegaon (Bk),Pune-411041
Savitribai Phule Pune University
Academic Year: 2023-24
Sinhgad Technical Education Society’s
NBN Sinhgad Technical Institutes Campus
NBN Sinhgad School of Engineering, Ambegaon (Bk),Pune-411041
DEPARTMENT OF COMPUTER ENGINEERING

CERTIFICATE

This is to certify that the Project Entitled

C2C- PREDICTIVE ANALYSIS OF STUDENT CAMPUS PLACEMENT

Submitted by

Vilas Dasalkar B191054219

Nihar Jachak B191054245
Kajol Pawar B191054311
Prasad Kurale B191054318
is a bonafide work carried out by Students under the supervision of Prof.A.M.Chadchankar
and it is submitted towards the fulfillment of the requirement of Bachelor of Engi-
neering (Computer Engineering).

Prof.A.M.Chadchankar Dr.S.P.Bendale
Internal Guide H.O.D
Dept. of Computer Engg. Dept. of Computer Engg.

Dr. S. P. Patil
Principal
NBN Sinhgad Technical Institute Campus

Signature of Internal Examiner Signature of External Examiner

PROJECT APPROVAL SHEET

A Project Title

C2C- PREDICTIVE ANALYSIS OF STUDENT CAMPUS PLACEMENT

Is successfully completed by

Vilas Dasalkar B191054219

Nihar Jachak B191054245
Kajol Pawar B191054311
Prasad Kurale B191054318

at

DEPARTMENT OF COMPUTER ENGINEERING

Sinhgad Technical Education Society’s

NBN Sinhgad Technical Institutes Campus
NBN Sinhgad School of Engineering, Ambegaon (Bk),Pune-411041

SAVITRIBAI PHULE PUNE UNIVERSITY,PUNE

ACADEMIC YEAR 2023-24

Prof.A.M.Chadchankar Prof.S.P.Bendale
Internal Guide H.O.D
Dept. of Computer Engg. Dept. of Computer Engg.
Abstract

Student placement is a critical factor influencing admissions and a university’s over-

all prestige. This study addresses the paramount importance of enhancing placement
services through advanced predictive algorithms. By analyzing recent student data,
our research aims to predict placement potential, contributing to both student success
and institutional standing.
The methodology involves meticulous preprocessing of institution-specific datasets,
showcasing a substantial increase in predictive accuracy compared to conventional
algorithms. Our campus placement prediction system integrates educational data
mining and key parameters from academic records, utilizing logistic regression for
precise predictions. This innovative model not only outperforms traditional algo-
rithms but also enhances academic performance and soft skills, offering valuable
insights through predictive analytics.
The motivation behind this research lies in the transformative potential it holds for
universities and students alike. By optimizing placement strategies, institutions can
elevate their effectiveness, positively impacting student outcomes. This study en-
courages a paradigm shift in placement approaches, emphasizing the importance of
proactive, data-driven strategies in the ever-evolving landscape of higher education.

NBNSTIC, Department of Computer Engineering 2023-24 I

Acknowledgments

It gives us great pleasure in presenting the preliminary project report on ‘C2C-

PREDICTIVE ANALYSIS OF STUDENT CAMPUS PLACEMENT’.

I would like to take this opportunity to thank my internal guide Prof.A.M.Chadchankar

for giving me all the help and guidance I needed. I am really grateful to them for
their kind support. Their valuable suggestions were very helpful.

I am also grateful to Prof.S.P.Bendale, Head of Computer Engineering Department,

NBNSTIC for his indispensable support, suggestions.

In the end our special thanks to S.C.Sethi for providing various resources such as
laboratory with all needed software platforms, continuous Internet connection, for
Our Project.

Vilas Dasalkar
Nihar Jachak
Kajol Pawar
Prasad Kurale
(B.E. Computer Engg.)

NBNSTIC, Department of Computer Engineering 2023-24 II

INDEX

1 Synopsis 1
1.1 Project Title . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2
1.2 Project Option . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2
1.3 Internal Guide . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2
1.4 Sponsorship and External Guide . . . . . . . . . . . . . . . . . . . 2
1.5 Technical Keywords (As per ACM Keywords) . . . . . . . . . . . . 2
1.6 Problem Statement . . . . . . . . . . . . . . . . . . . . . . . . . . 2
1.7 Abstract . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3
1.8 Goals and Objectives . . . . . . . . . . . . . . . . . . . . . . . . . 3
1.9 Relevant mathematics associated with the Project . . . . . . . . . . 3
1.10 Names of Conferences / Journals where papers can be published . . 5
1.11 Review of Conference/Journal Papers supporting Project idea . . . . 5

2 Technical Keywords 8
2.1 Area of Project . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9
2.2 Technical Keywords . . . . . . . . . . . . . . . . . . . . . . . . . . 9

3 Introduction 10
3.1 Project Idea . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11
3.2 Motivation of the Project . . . . . . . . . . . . . . . . . . . . . . . 11
3.3 Literature Survey . . . . . . . . . . . . . . . . . . . . . . . . . . . 11

4 Problem Definition and scope 14

4.1 Problem Statement . . . . . . . . . . . . . . . . . . . . . . . . . . 15
4.1.1 Goals and objectives . . . . . . . . . . . . . . . . . . . . . 15
4.1.2 Statement of scope . . . . . . . . . . . . . . . . . . . . . . 15
4.2 Major Constraints . . . . . . . . . . . . . . . . . . . . . . . . . . . 16
4.3 Methodologies of Problem solving and efficiency issues . . . . . . . 17
4.4 Outcome . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 18
4.5 Applications . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 20
4.6 Hardware Resources Required . . . . . . . . . . . . . . . . . . . . 21
4.7 Software Resources Required . . . . . . . . . . . . . . . . . . . . . 21

5 Project Plan 22
5.1 Project Estimates . . . . . . . . . . . . . . . . . . . . . . . . . . . 23
5.1.1 Reconciled Estimates . . . . . . . . . . . . . . . . . . . . . 25
5.1.2 Project Resources . . . . . . . . . . . . . . . . . . . . . . . 25
5.2 Risk Management w.r.t. NP Hard analysis . . . . . . . . . . . . . . 26
5.2.1 Risk Identification . . . . . . . . . . . . . . . . . . . . . . 26
5.2.2 Risk Analysis . . . . . . . . . . . . . . . . . . . . . . . . . 27
5.2.3 Overview of Risk Mitigation, Monitoring, Management . . 27
5.3 Project Schedule . . . . . . . . . . . . . . . . . . . . . . . . . . . 30
5.3.1 Project task set . . . . . . . . . . . . . . . . . . . . . . . . 30
5.3.2 Task network . . . . . . . . . . . . . . . . . . . . . . . . . 31
5.3.3 Timeline Chart . . . . . . . . . . . . . . . . . . . . . . . . 33
5.4 Team Organization . . . . . . . . . . . . . . . . . . . . . . . . . . 33
5.4.1 Team structure . . . . . . . . . . . . . . . . . . . . . . . . 33
5.4.2 Management reporting and communication . . . . . . . . . 33

6 Software requirement specification 34

6.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 35
6.1.1 Purpose and Scope of Document . . . . . . . . . . . . . . . 35
6.1.2 Overview of responsibilities of Developer . . . . . . . . . . 35
6.2 Usage Scenario . . . . . . . . . . . . . . . . . . . . . . . . . . . . 36
6.2.1 User profiles . . . . . . . . . . . . . . . . . . . . . . . . . 36
6.2.2 Use-cases . . . . . . . . . . . . . . . . . . . . . . . . . . . 38
6.2.3 Use Case View . . . . . . . . . . . . . . . . . . . . . . . . 38

NBNSTIC, Department of Computer Engineering 2023-24 IV

6.3 Data Model and Description . . . . . . . . . . . . . . . . . . . . . 38
6.3.1 Data Description . . . . . . . . . . . . . . . . . . . . . . . 38
6.4 Functional Model and Description . . . . . . . . . . . . . . . . . . 39
6.4.1 Data Flow Diagram . . . . . . . . . . . . . . . . . . . . . . 40
6.4.2 Description of Functions . . . . . . . . . . . . . . . . . . . 40
6.4.3 Activity Diagram: . . . . . . . . . . . . . . . . . . . . . . 42
6.4.4 Non Functional Requirements: . . . . . . . . . . . . . . . . 44
6.4.5 State Transition Diagram: . . . . . . . . . . . . . . . . . . 45
6.4.6 Design Constraints . . . . . . . . . . . . . . . . . . . . . . 46
6.4.7 Software Interface Description . . . . . . . . . . . . . . . . 46

7 Detailed Design Document using Appendix A and B 48

7.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 49
7.2 Architectural Design . . . . . . . . . . . . . . . . . . . . . . . . . 49
7.3 Data design (using Appendices A and B) . . . . . . . . . . . . . . . 50
7.3.1 Internal Software Data Structure . . . . . . . . . . . . . . . 50
7.3.2 Global Data Structure . . . . . . . . . . . . . . . . . . . . 50
7.3.3 Temporary Data Structure . . . . . . . . . . . . . . . . . . 50
7.3.4 Database Description . . . . . . . . . . . . . . . . . . . . . 51
7.4 Component Design . . . . . . . . . . . . . . . . . . . . . . . . . . 51
7.4.1 Class Diagram . . . . . . . . . . . . . . . . . . . . . . . . 52

8 Project Implementation 53
8.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 54
8.2 Tools and Technologies Used . . . . . . . . . . . . . . . . . . . . . 54
8.3 Methodologies/Algorithm Details . . . . . . . . . . . . . . . . . . 55
8.3.1 Procedural Steps . . . . . . . . . . . . . . . . . . . . . . . 55
8.3.2 Utilized Algorithms . . . . . . . . . . . . . . . . . . . . . 56

9 Software Testing 58
9.1 Type of Testing Used . . . . . . . . . . . . . . . . . . . . . . . . . 59
9.2 Test Cases and Test Results: . . . . . . . . . . . . . . . . . . . . . 59

NBNSTIC, Department of Computer Engineering 2023-24 V

10 Results 61
10.1 Screen shots . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 62

11 Deployment and Maintenance 66

11.1 Installation and Uninstallation . . . . . . . . . . . . . . . . . . . . 67
11.1.1 Installation . . . . . . . . . . . . . . . . . . . . . . . . . . 67
11.1.2 Uninstallation . . . . . . . . . . . . . . . . . . . . . . . . . 68

12 Conclusion and Future Scope 69

References 71

Annexure A Laboratory assignments on Project Analysis of Algorithmic

Design 74
A.1 PROBLEM STATEMENT DEVELOPMENT . . . . . . . . . . . . 75
A.2 JUSTIFYING FEASIBILITY USING IDEA MATRIX . . . . . . . 75
A.3 FEASIBILITY ASSESSMENT USING MATHEMATICAL MOD-
ELS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 76
A.3.1 NP-Hard, NP-Complete, or Satisfiability Issues . . . . . . . 76
A.3.2 Input-Output Relationship (y=f(x)) . . . . . . . . . . . . . 76
A.4 INTEGRATION OF KNOWLEDGE CANVAS . . . . . . . . . . . 76
A.4.1 Identification of Opportunity for Product . . . . . . . . . . 76
A.4.2 Feasibility from a Business Perspective . . . . . . . . . . . 76
A.4.3 Integration of Concepts . . . . . . . . . . . . . . . . . . . . 76

Annexure B Laboratory assignments on Project Quality and Reliability

Testing of Project Design 77

Annexure C Project Planner 81

Annexure D Reviewers Comments of Paper Submitted 83

Annexure E Plagiarism Report 85

Annexure F Information of Project Group Members 87

NBNSTIC, Department of Computer Engineering 2023-24 VI

List of Figures

6.1 Use case diagram . . . . . . . . . . . . . . . . . . . . . . . . . . . 38

6.2 Data Flow Diagram . . . . . . . . . . . . . . . . . . . . . . . . . . 40
6.3 Data Flow Diagram . . . . . . . . . . . . . . . . . . . . . . . . . . 41
6.4 Activity diagram . . . . . . . . . . . . . . . . . . . . . . . . . . . 43
6.5 State transition diagram . . . . . . . . . . . . . . . . . . . . . . . . 45

7.1 Architecture diagram . . . . . . . . . . . . . . . . . . . . . . . . . 50

7.2 Class Diagram . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 52

E.1 Plagiarism Report . . . . . . . . . . . . . . . . . . . . . . . . . . . 86

List of Tables

4.1 Hardware Requirements . . . . . . . . . . . . . . . . . . . . . . . 21

5.1 Risk Table . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 28

5.2 Performance Table . . . . . . . . . . . . . . . . . . . . . . . . . . 28
5.3 Risk Impact . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 29
5.4 Progress Report . . . . . . . . . . . . . . . . . . . . . . . . . . . . 33

6.1 Use Cases . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 38

C.1 Project Timeline . . . . . . . . . . . . . . . . . . . . . . . . . . . . 82

CHAPTER 1

SYNOPSIS
1.1 PROJECT TITLE

C2C- Predictive Analysis of Student Campus Placement

1.2 PROJECT OPTION

Sponsored.

1.3 INTERNAL GUIDE

Prof.A.M.Chadchankar

1.4 SPONSORSHIP AND EXTERNAL GUIDE

Mr.Akhilesh Deo (Automate Engineering)

1.5 TECHNICAL KEYWORDS (AS PER ACM KEYWORDS)

1. Predictive Analysis

2. Data Science

3. Placement

4. Random Forest

5. Logistic Regression

6. Decision Tree

7. Model Evaluation

1.6 PROBLEM STATEMENT

Despite the availability of extensive historical data related to student academic per-
formance, skills, and extracurricular activities, traditional placement processes often
lack the precision necessary to match graduates with suitable employment opportu-
nities. The challenge lies in developing a predictive model that not only considers

NBNSTIC, Department of Computer Engineering 2023-24 2

academic achievements but also integrates soft skills and other relevant factors to
enhance the overall accuracy of placement predictions.

1.7 ABSTRACT

• Student placement’s vital role in academic institutions influences admissions

and university prestige. This study focuses on analyzing recent student data
to predict placement potential, employing an advanced predictive algorithm.
The methodology, based on a meticulously preprocessed dataset from the same
institution, demonstrates a significant accuracy increase over established clas-
sification algorithms. The campus placement prediction system utilizes educa-
tional data mining and parameters from college records and tests, employing
logistic regression for precise predictions. This model not only outperforms
traditional algorithms but also enhances academic performance and soft skills,
providing valuable insights through predictive analytics. In offering a com-
prehensive perspective on predictive analytics in campus placement, this study
contributes to the ongoing discourse on optimizing student outcomes.

1.8 GOALS AND OBJECTIVES

• Enhance Placement Accuracy

• Facilitate Informed Decision-Making

• Improve Employability Outcomes

• Incorporate Soft Skills Assessment

• Adapt to Changing Employment Landscapes

1.9 RELEVANT MATHEMATICS ASSOCIATED WITH THE PROJECT

System Description: An overview of the key components, inputs, outputs, and strate-
gies employed in your project. Here’s a more detailed breakdown for a project on
the prediction of campus placement

NBNSTIC, Department of Computer Engineering 2023-24 3

• Input:

– Student data, including academic records and test scores.

– Institutional data on placement history and campus recruitment activities.

• Output:

– Predicted placement outcomes for individual students.

– Statistical insights into placement trends and success rates.

• Identify data structures, classes, divide and conquer strategies to exploit dis-
tributed/parallel/concurrent processing, constraints.

• Functions :

– Preprocess Data: Cleans and the input data (e.g., handling missing val-
ues, feature scaling).

– Predict Placement: Predicts the likelihood of placement for a given stu-

dent using a trained model.

• Mathematical formulation:

– Precision: Precision is the ratio of true positive predictions to the total

positive predictions made by the model. It measures the accuracy of the
positive predictions.

– Recall (Sensitivity or True Positive Rate): Recall is the ratio of true pos-
itive predictions to the total actual positive instances in the dataset. It
measures the model’s ability to identify all relevant instances.

– F1 Score: The F1 score combines precision and recall into a single met-
ric, providing a balance between them. It is calculated as the harmonic
mean of precision and recall

• Success Conditions:

– Accurate prediction of placement outcomes.

– Improvement in academic performance and soft skills.

NBNSTIC, Department of Computer Engineering 2023-24 4

• Failure Conditions:

– Inaccurate Predictions: System failures resulting in inaccurate placement

predictions.

– Lack of Improvement: Failure to observe improvement in academic per-

formance and soft skills despite system intervention.

1.10 NAMES OF CONFERENCES / JOURNALS WHERE PAPERS CAN

BE PUBLISHED

• IEEE/ACM Conference/Journal 1

• Conferences/workshops in IITs

• Central Universities or SPPU Conferences

• IEEE/ACM Conference/Journal 2

1.11 REVIEW OF CONFERENCE/JOURNAL PAPERS SUPPORTING PROJECT

IDEA

1. Student Placement Prediction Using Supervised Machine Learning:

Authors: M. Siva Surya, Dr. M. Sathish Kumar, Dr. D.Gandhimathi.

Description: This study explores the vital domain of student placement in

academic institutions, emphasizing its impact on university admission and
reputation. With a focus on improving placement services, the research uti-
lizes recent pupil data and integrates a predictive algorithm to forecast current
placement probabilities. Developed with the collaboration of the institution re-
sponsible for placement prediction and rigorous pre-processing, the proposed
model showcases exceptional accuracy, surpassing conventional classification
algorithms.

2. Anticipating Placement Status of Engineering Students using Machine

Learning based Prediction Models - A Case Study of Computer Science
Stream:

NBNSTIC, Department of Computer Engineering 2023-24 5

Authors: Prof. Kajal Rai, Prof. Pawan Kumar, Prof. Sanjay Sood.

Description: This research examines student placements in academic institu-

tions, focusing on placement quality’s impact on university rankings. Machine
learning is used to predict computer science engineering placement status. Lo-
gistic Regression and Decision Tree algorithms show the best predictive per-
formance, identifying factors like ’Xth’ and ’XIIth’ marks, ’Current CGPA,’
and ’Present Attendance’ as significant influencers. Emphasizing the impor-
tance of intelligent computing for anticipating placement trends, the research
offers insights for enhancing overall placement quality.

3. Placement Prediction System using Machine Learning:

Authors: Nitesh Kumar Sharma, Aniket Kumar Singh, Shubham Salvi.

Description: This study on student placement, vital for educational institu-

tion credibility, introduces a K-Nearest Neighbors (k-NN) supervised machine
learning system, achieving 87 percent accuracy in predictions. It provides
placement improvement suggestions and an academic material exchange por-
tal. The system enables skill display, team formation, and material man-
agement. Emphasizing placement’s importance, it explores machine learning
techniques, with a functional block diagram. It discusses the k-NN algorithm
and system performance, concluding with insights on benefits, limitations, and
future enhancements like online courses and coding assessments for compre-
hensive evaluation.

4. Placement Prediction System using Machine Learning:

Authors: Gurivireddy Mary Spandana, L Pallavi (Associate Professor).

Description: This study aims to enhance the student placement system in aca-
demic institutions with a predictive model. Utilizing Logistic Regression,
Random Forest, KNN, and SVM machine learning algorithms, it predicts
placement outcomes based on various parameters. Comparing algorithm per-
formance on a shared database, KNN achieves the highest accuracy at 75.90
percent, making it the most suitable for the dataset. This predictive system is
vital for student career development and institutional reputation.

NBNSTIC, Department of Computer Engineering 2023-24 6

5. Campus Placement Prediction System Using Supervised Machine Learn-
ing Techniques:

Authors: Navuluri Divya, Sravya Namburu, Rajalakshmi Raja.

Description: This research focuses on student placement, crucial for educa-

tional institutions’ reputation and admissions. It predicts current students’
placement chances using past data. The proposed algorithmic model, tailored
with institution-specific data and pre-processing methods, outperforms tradi-
tional algorithms like Decision Tree and Random Forest. Results highlight its
superior accuracy, contributing significantly to improving placement predic-
tions for both students and institutions striving for enhanced placement suc-
cess.

NBNSTIC, Department of Computer Engineering 2023-24 7

CHAPTER 2

TECHNICAL KEYWORDS
2.1 AREA OF PROJECT

1. Data Science and Analytics: Utilizing data analysis techniques to gain insights
from the collected data. Employing statistical models and machine learning
algorithms for predictive analysis.

2. Educational Technology: Applying technology to enhance education-related

processes, in this case, student placement. Leveraging data to improve educa-
tional outcomes and decision-making.

3. Human Resources and Talent Management: Focusing on predicting and man-

aging talent in the context of student placement. Implementing strategies to
optimize the matching of students with job opportunities.

4. Ethical Data Use and Privacy: Addressing ethical considerations related to

handling student data. Ensuring privacy and compliance with relevant regula-
tions.

5. Decision Support Systems: Building a system that aids decision-making in the

context of student placement. Providing insights to educational institutions
and recruiters for informed choices.

2.2 TECHNICAL KEYWORDS

1. Predictive Analysis

2. Data Science

3. Placement

4. Random Forest

5. Logistic Regression

6. Decision Tree

7. Model Evaluation

NBNSTIC, Department of Computer Engineering 2023-24 9

CHAPTER 3

INTRODUCTION
3.1 PROJECT IDEA

• The project endeavors to create an intelligent placement advisor system that

utilizes predictive analysis to elevate students’ employability. This system
aims to offer personalized insights into placement prospects by considering a
comprehensive array of factors, encompassing academic performance, skills,
and soft skills. The goal is to provide precise predictions and actionable rec-
ommendations tailored to individual students, educational institutions, and
employers, thereby enhancing the overall placement experience.

3.2 MOTIVATION OF THE PROJECT

• The main aim of this project is to improve the employability of students by

transcending the traditional focus on academic achievement and by focusing
on students’ employability as a whole. It is the aim of the project to acknowl-
edge and appreciate a wider range of skills, which includes technical profi-
ciency as well as soft skills, as well as a more extensive spectrum of abilities.
Taking a broader view of the job market will provide students with a better
understanding of the ever-evolving dynamics of this market.

3.3 LITERATURE SURVEY

1. Student Placement Prediction Using Supervised Machine Learning:

Authors: M. Siva Surya, Dr. M. Sathish Kumar, Dr. D.Gandhimathi.

Description: Description: This study explores the vital domain of student

placement in academic institutions, emphasizing its impact on university ad-
mission and reputation. With a focus on improving placement services, the
research utilizes recent pupil data and integrates a predictive algorithm to fore-
cast current placement probabilities. Developed with the collaboration of the
institution responsible for placement prediction and rigorous pre-processing,
the proposed model showcases exceptional accuracy, surpassing conventional
classification algorithms.

NBNSTIC, Department of Computer Engineering 2023-24 11

2. Anticipating Placement Status of Engineering Students using Machine
Learning based Prediction Models - A Case Study of Computer Science
Stream:

Authors: Prof. Kajal Rai, Prof. Pawan Kumar, Prof. Sanjay Sood.

Description: This research examines student placements in academic institu-

tions, focusing on placement quality’s impact on university rankings. Machine
learning is used to predict computer science engineering placement status. Lo-
gistic Regression and Decision Tree algorithms show the best predictive per-
formance, identifying factors like ’Xth’ and ’XIIth’ marks, ’Current CGPA,’
and ’Present Attendance’ as significant influencers. Emphasizing the impor-
tance of intelligent computing for anticipating placement trends, the research
offers insights for enhancing overall placement quality.

3. Placement Prediction System using Machine Learning:

Authors: Nitesh Kumar Sharma, Aniket Kumar Singh, Shubham Salvi.

Description: This study on student placement, vital for educational institu-

tion credibility, introduces a K-Nearest Neighbors (k-NN) supervised machine
learning system, achieving 87 percent accuracy in predictions. It provides
placement improvement suggestions and an academic material exchange por-
tal. The system enables skill display, team formation, and material man-
agement. Emphasizing placement’s importance, it explores machine learning
techniques, with a functional block diagram. It discusses the k-NN algorithm
and system performance, concluding with insights on benefits, limitations, and
future enhancements like online courses and coding assessments for compre-
hensive evaluation.

4. Placement Prediction System using Machine Learning:

Authors: Gurivireddy Mary Spandana, L Pallavi (Associate Professor).

Description: This study aims to enhance the student placement system in aca-
demic institutions with a predictive model. Utilizing Logistic Regression,
Random Forest, KNN, and SVM machine learning algorithms, it predicts

NBNSTIC, Department of Computer Engineering 2023-24 12

placement outcomes based on various parameters. Comparing algorithm per-
formance on a shared database, KNN achieves the highest accuracy at 75.90
percent, making it the most suitable for the dataset. This predictive system is
vital for student career development and institutional reputation.

5. Campus Placement Prediction System Using Supervised Machine Learn-

ing Techniques:

Authors: Navuluri Divya, Sravya Namburu, Rajalakshmi Raja.

Description: This research focuses on student placement, crucial for educa-

tional institutions’ reputation and admissions. It predicts current students’
placement chances using past data. The proposed algorithmic model, tailored
with institution-specific data and pre-processing methods, outperforms tradi-
tional algorithms like Decision Tree and Random Forest. Results highlight its
superior accuracy, contributing significantly to improving placement predic-
tions for both students and institutions striving for enhanced placement suc-
cess.

NBNSTIC, Department of Computer Engineering 2023-24 13

CHAPTER 4

PROBLEM DEFINITION AND SCOPE

4.1 PROBLEM STATEMENT

• The existing models often neglect crucial dimensions such as soft skills, ex-
tracurricular achievements, and a holistic evaluation of a student’s capabilities,
resulting in suboptimal placement outcomes. Consequently, there is a pressing
need for a comprehensive and intelligent placement advisor system that goes
beyond conventional methods, leveraging predictive analysis to provide a nu-
anced understanding of a student’s potential and aligning them with suitable
employment opportunities.

4.1.1 Goals and objectives

Goal and Objectives:

• Enhance Placement Accuracy

• Facilitate Informed Decision-Making

• Improve Employability Outcomes

• Incorporate Soft Skills Assessment

• Address Ethical Considerations

• Adapt to Changing Employment Landscapes

4.1.2 Statement of scope

• This project focuses on creating a specialized predictive analysis system for

campus placements. Utilizing advanced algorithms and holistic data, it aims
to predict student placement outcomes, considering academic achievements,
technical skills, and soft skills. The scope includes algorithm development,
real-time predictions, intuitive user interfaces, and strategies for enhancing
student growth. Deployment involves thorough testing, optimization, and doc-
umentation, culminating in a comprehensive project evaluation.

NBNSTIC, Department of Computer Engineering 2023-24 15

4.2 MAJOR CONSTRAINTS

1. Data Privacy and Security:

• Constraint: Adhering to strict data privacy regulations and ensuring the

security of sensitive student information.

• Implications: Implementation of robust encryption, access controls, and

compliance with data protection laws may add complexity and cost to
the project

2. Budgetary Constraints:

• Constraint: Limited budget for development, deployment, and mainte-

nance of the system.

• Implications: Balancing features and functionalities within the available

budget, potentially limiting the scope of the project or requiring phased
development.

3. Integration Challenges:

• Constraint: Difficulties in seamlessly integrating with existing educa-

tional institution systems and external platforms.

• Implications: Delays in implementation, potential data inconsistencies,

and the need for effective communication with diverse systems.

4. Limited Accessibility to External Data Sources:

• Constraint: Restricted access to external data sources for real-time up-

dates on industry trends.

• Implications: Dependence on periodically updated datasets, potentially

leading to a lag in reflecting the latest job market demands.

5. Technology Stack Limitations:

• Constraint: Constraints imposed by the selected technology stack for de-

velopment.

NBNSTIC, Department of Computer Engineering 2023-24 16

• Implications: Limited flexibility in adopting emerging technologies, po-
tential challenges in system scalability, and adaptability to future techno-
logical advancements.

4.3 METHODOLOGIES OF PROBLEM SOLVING AND EFFICIENCY IS-

SUES

1. Define Objectives:

• Clearly define the goals of predictive analysis in student campus place-

ment.

• Understand what specific problems or inefficiencies you aim to address

through predictive modeling.

2. . Data Collection and Preprocessing:

• Gather relevant data on student profiles, academic performance, skills,

and placement outcomes.

• Clean and preprocess the data to handle missing values, outliers, and
ensure data quality.

3. Exploratory Data Analysis (EDA):

• Analyze the distribution of variables and relationships between features.

• Identify patterns or trends that may impact student placement.

4. Define Key Performance Indicators (KPIs):

• Establish KPIs related to placement success and efficiency.

• Metrics may include placement rate, time to placement, and accuracy of

predictions.

5. Select Predictive Modeling Techniques:

• Choose appropriate machine learning algorithms for predictive analysis.

NBNSTIC, Department of Computer Engineering 2023-24 17

• Consider algorithms like Logistic Regression, Decision Trees, Random
Forest, or Neural Networks.

6. Feature Engineering:

• Create new features that might enhance the predictive power of the model.

• Select the most relevant features based on their impact on placement out-
comes.

7. Data Splitting:

• Divide the dataset into training and testing sets to evaluate model perfor-
mance.

8. Model Training:

• Train the selected predictive model on the training dataset.

9. Model Evaluation:

• Evaluate the model’s performance using relevant metrics (accuracy, pre-

cision, recall, F1 score).

• Assess how well the model predicts student placement outcomes.

4.4 OUTCOME

• Improved Placement Accuracy:

Outcome: The predictive model enhances the accuracy of forecasting whether

a student will get placed or not.

Impact: Institutions can better allocate resources, and students receive more
tailored support based on their likelihood of successful placement.

• Efficient Resource Allocation:

Outcome: Identification of key factors influencing placement success helps

optimize resource allocation.

NBNSTIC, Department of Computer Engineering 2023-24 18

Impact: Institutions can focus on areas that contribute most to successful
placements, streamlining efforts and increasing overall efficiency.

• Enhanced Student Counseling:

Outcome: The analysis provides insights into individual student strengths and
weaknesses.

Impact: Guidance counselors can offer targeted advice, helping students im-
prove specific skills or take advantage of particular opportunities to enhance
their employability.

• Personalized Career Development Plans:

Outcome: Tailored recommendations for skill development and career paths.

Impact: Students receive personalized career development plans, increasing

their readiness for the job market.

• Optimized Internship and Training Programs:

Outcome: Identification of skills gaps and areas for improvement.

Impact: Educational institutions can refine internship and training programs to

address specific skill needs highlighted by the predictive analysis.

• Better Recruitment Strategies:

Outcome: Understanding recruiter preferences and requirements.

Impact: Institutions can tailor placement strategies to match student profiles

with recruiter expectations, improving the overall success rate.

• Reduced Time-to-Placement:

Outcome: Streamlined processes and quicker identification of suitable oppor-

tunities.

Impact: Students find placements faster, reducing the time between graduation
and employment.

NBNSTIC, Department of Computer Engineering 2023-24 19

4.5 APPLICATIONS

• Early Intervention and Student Support:

Application: Predicting students at risk of not finding placement.

Benefit: Allows for early intervention and targeted support, such as additional
career counseling, skill development programs, or mentorship.

• Tailored Career Guidance:

Application: Predicting students’ potential career paths based on their profiles.

Benefit: Enables personalized career guidance, helping students make informed

decisions about their academic and professional trajectories.

• Optimizing Curriculum Design:

Application: Analyzing historical placement data to identify skills in high de-

mand.

Benefit: Guides the development of curriculum to align with industry needs,

ensuring graduates are equipped with relevant skills.

• Internship and Placement Program Optimization:

Application: Predicting the success of internship and placement programs.

Benefit: Institutions can optimize program structures, ensuring they provide

the most valuable experiences for students.

• Recruitment Strategy Enhancement:

Application: Analyzing historical data to understand recruiter preferences.

Benefit: Institutions can tailor their recruitment strategies, connecting students

with opportunities that align closely with recruiter expectations.

• Alumni Engagement and Success Tracking:

Application: Tracking the career trajectories of alumni.

Benefit: Institutions can showcase successful alumni stories and maintain con-
nections for mentorship and networking opportunities.

NBNSTIC, Department of Computer Engineering 2023-24 20

• Resource Allocation for Career Services:

Application: Predicting peak periods for career services demand.

Benefit: Helps allocate resources efficiently, ensuring that career services are
adequately staffed during high-demand times.

4.6 HARDWARE RESOURCES REQUIRED

Sr. No. Component Requirement

1 CPU Processor Multi-core processor
2 Memory 8 GB RAM or higher
3 Graphics Card Dedicated GPU for accelerated machine learning tasks
4 Network High-speed internet connection
5 Operating System 64-bit operating system
6 Software Dependencies Python, Jupyter Notebook,Flask

Table 4.1: Hardware Requirements

4.7 SOFTWARE RESOURCES REQUIRED

Platform :

1. Operating System: Windows 10

2. IDE: Anaconda Jupyter

3. Programming Language: Python 3.9

NBNSTIC, Department of Computer Engineering 2023-24 21

CHAPTER 5

PROJECT PLAN
5.1 PROJECT ESTIMATES

1. Scope Definition:

Task: Clearly define the scope of the project, including the objectives, deliver-
ables, and features to be implemented.

2. Resource Planning:

Task: Identify the human resources required for the project, including data
scientists, domain experts, and IT professionals.

Estimate: Consider the expertise and time commitment of each team member.

3. Hardware and Software Requirements:

Task: Specify the hardware and software needed for the predictive analysis.

Estimate: Include the costs of acquiring or upgrading hardware, purchasing

software licenses, and any cloud service subscriptions.

4. Data Collection and Preprocessing:

Task: Estimate the time and resources needed to collect and preprocess the
dataset.

Estimate: Consider the complexity of data cleaning, normalization, and feature

engineering.

5. Model Development:

Task: Develop and train predictive models based on the dataset.

Estimate: Consider the time required for model selection, training, and hyper-
parameter tuning.

6. Validation and Testing:

Task: Validate and test the predictive models to ensure accuracy and reliability.

Estimate: Allocate time for testing different scenarios and fine-tuning the mod-
els.

NBNSTIC, Department of Computer Engineering 2023-24 23

7. Integration with Existing Systems:

Task: If applicable, integrate the predictive analysis system with existing cam-
pus systems.

Estimate: Consider the time and resources required for seamless integration.

8. Deployment:

Task: Deploy the predictive analysis system for practical use.

Estimate: Include time and resources for deploying the system and ensuring
its functionality in a real-world setting.

9. Training and Documentation:

Task: Provide training to end-users and create documentation for system main-
tenance.

Estimate: Consider the time and resources needed to prepare training materials
and documentation.

10. Monitoring and Maintenance:

Task: Establish monitoring mechanisms and plan for ongoing maintenance.

Estimate: Allocate resources for continuous monitoring and periodic updates.

11. Contingency Planning: Task: Identify potential risks and plan for contingen-
cies.

Estimate: Allocate a contingency budget and time buffer to address unforeseen

challenges.

12. Project Management: Task: Manage the project, including communication,

coordination, and progress tracking.

Estimate: Factor in project management time and tools.

13. Ethical and Legal Compliance:

Task: Ensure compliance with ethical and legal standards, especially concern-
ing student data.

Estimate: Allocate resources for legal consultations and compliance measures.

NBNSTIC, Department of Computer Engineering 2023-24 24

14. Timeline:

Task: Develop a timeline for the project from initiation to completion.

Estimate: Break down the project into phases and estimate the time required
for each.

5.1.1 Reconciled Estimates

5.1.1.1 Cost Estimate

• Cost Estimates:

Task: Sum up the estimated costs for human resources, hardware, software,
training, and other expenses.

Estimate: Create a detailed budget that includes all anticipated costs.

5.1.1.2 Time Estimates

The development timeline for a campus placement prediction system can range from
a few months to a year or more, depending on the project’s scope. Here’s a rough
breakdown:

• Data Collection and Annotation: 1-2 months

• Model Development and Training: 2-4 months

• User Interface Development: 2-3 months

• Integration and Testing: 2-3 months

• User Education and Documentation: Ongoing

5.1.2 Project Resources

People:- 4

• Hardware Resources

1. System : Intel i5.

NBNSTIC, Department of Computer Engineering 2023-24 25

2. Hard Disk : 512 GB.

3. RAM : 8 GB

• Software Requirements

1. Operating System - Linux

2. Server-Side Scripting Language - Python

3. Frontend Technologies (HTML, CSS, JavaScript)

4. Framework for Web Development (Flask and React)

5. Postman to test backend APIs

6. Microsoft Excel to store college student’s data.

5.2 RISK MANAGEMENT W.R.T. NP HARD ANALYSIS

This section discusses Project risks and the approach to managing them.

5.2.1 Risk Identification

Analyzing risks in the context of NP-hard problems, such as student campus place-
nent prediction, often involves identifying computational complexity and efficiency-
related concerns.

1. Algorithm Complexity: Developing algorithms for predictive analysis of stu-

dent campus placements may involve complex models, leading to high com-
putational costs.

2. Scalability: As the number of students and companies participating in campus

placements increases, handling large datasets efficiently becomes challenging.

3. Real-time Processing: Achieving real-time predictions for student placements,

especially during peak recruitment seasons, can be difficult due to the compu-
tational burden of complex algorithms.

4. Resource Constraints: Limited computational resources, such as server ca-

pacity or processing power, may hinder the efficiency of predictive analysis
algorithms.

NBNSTIC, Department of Computer Engineering 2023-24 26

5. Parameter Tuning: Manual tuning of algorithm parameters for accurate place-
ment predictions can be time-consuming and requires domain expertise.

6. Data Preprocessing: Preprocessing large volumes of student profiles, company

data, and historical placement records can be computationally intensive and
time-consuming.

7. False Positives/Negatives: Inaccurate predictions may lead to false positives

(students predicted to be placed but aren’t) or false negatives (students not
predicted to be placed but are), impacting the credibility of the placement pro-
cess.

8. Interoperability: Integrating predictive analysis systems with existing campus

databases and management systems may pose challenges in data exchange and
compatibility.

9. Algorithm Maintenance: Regular maintenance and updates of predictive anal-

ysis algorithms are essential to ensure their effectiveness and relevance over
time.

10. Data Storage/Retrieval: Efficient storage and retrieval of student profiles, com-
pany information, and placement data are critical for the smooth functioning
of predictive analysis systems.

By recognizing these risks, stakeholders involved in student campus placements can

devise strategies to mitigate them and enhance the accuracy and efficiency of the
placement process.

5.2.2 Risk Analysis

The risks for the Project can be analyzed within the constraints of time and quality

5.2.3 Overview of Risk Mitigation, Monitoring, Management

Following are the details for each risk.

NBNSTIC, Department of Computer Engineering 2023-24 27

Impact
ID Risk Description Probability
Schedule Quality Overall
1 Inaccurate Predictions Medium Low High High
2 Data Entry Errors Low Low Medium Medium
3 External Market Factors Medium Medium Medium Medium
(e.g., economic conditions)
4 Lack of Maintenance of Pre- Low Low High High
diction Models
5 Integration Challenges with Medium Medium Medium Medium
Existing Systems
6 Data Security Breaches Low Low High High
7 Privacy Violations Low Low High High
8 Inaccurate Data Calibration Medium Medium Medium Medium

Table 5.1: Risk Table

Risk Category Description Impact

Technical Inaccurate prediction models affect placement ac- High
curacy
External Factors Economic fluctuations affect job market predictions Medium
Operational Data entry errors lead to incorrect predictions Medium
Maintenance Lack of maintenance leads to outdated prediction High
models
Integration Difficulty integrating prediction systems with exist- Medium
ing databases
Data Security Breaches in data security compromise student and High
company information
Privacy Violations of privacy regulations lead to legal con- High
sequences
Calibration Inaccurate calibration of prediction models affects Medium
accuracy

Table 5.2: Performance Table

NBNSTIC, Department of Computer Engineering 2023-24 28

Impact Description
High Significant consequences; demands immediate action.
Medium Moderate consequences; requires attention.
Low Minimal consequences; easily manageable

Table 5.3: Risk Impact

Risk ID 1
Risk Description The predictive analysis algorithm may struggle to accurately
predict student placements due to unforeseen patterns or fluc-
tuations in the job market, leading to inaccurate recommenda-
tions.
Category Technical Risk.
Source Historical Placement data inconsistencies.
Probability Moderate
Impact High
Response Monitor and Adjust
Strategy Implement a continous monitoring system to identify and
adapt changing patterns
Risk Status Ongoing

Risk ID 2
Risk Description Incomplete or inaccurate student data in the dataset may im-
pact the accuracy of predictions, resulting in suboptimal place-
ment recommendations.
Category Data Risks
Source Incomplete student profiles.
Probability Moderate
Impact Moderate
Response Validate and Enhance Data
Strategy Implement thorough data validation processes and regularly
update the dataset with accurate information.
Risk Status Identified

NBNSTIC, Department of Computer Engineering 2023-24 29

Risk ID 3
Risk Description Technical issues with the user interface may lead to difficulty
in accessing or interpreting placement predictions, impacting
user experience.
Category Usability Risks
Source User interface design flaws
Probability Low
Impact Moderate
Response User Training
Strategy Conduct user training sessions to familiarize users with the
system interface and troubleshoot potential issues.
Risk Status Mitigated

5.3 PROJECT SCHEDULE

5.3.1 Project task set

Major Tasks in the Project stages are:

• Task 1:Project Planning and Requirements Gathering (4 weeks)

– Define project objectives and scope.

– Identify stakeholders and gather requirements.

– Develop a detailed project plan, including budget and resource allocated.

• Task 2:Technology Research and Prototyping (6 weeks)

– Research and evaluate various machine learning algorithms suitable for

predicting student campus placements based on historical data.

– Consider algorithms such as decision trees, random forests, support vec-

tor machines, or gradient boosting based on the characteristics of the
placement data.

• Task 3: System Development And Testing(6 weeks)

– Develop predictive algorithms for student placements based on academic

records and skills.

NBNSTIC, Department of Computer Engineering 2023-24 30

– Implement a user-friendly interface for data input and result visualiza-
tion.

– Optimize the predictive model based on testing feedback and real-time

data.

• Task 4: Integration and Compatibility (4 weeks)

– Ensure compatibility with various data sources, such as student records

and industry trends databases.

– Validate interoperability with different devices and platforms for acces-

sibility.

• Task 5: : User Acceptance Testing, Refinement, and Deployment (6 weeks)

– Refine the predictive model and user interface based on UAT feedback
and performance evaluations.

– Plan and execute the deployment of the refined system, ensuring a smooth
transition.

5.3.2 Task network

1. Project Initiation:

• Define Project Objectives

• Stakeholder Identification

• Project Kickoff Meeting

2. Planning:

• Scope Definition

• Risk Analysis and Management Plan

• Timeline and Milestones

3. Data Collection and Preparation:

• Data Identification

NBNSTIC, Department of Computer Engineering 2023-24 31

• Data Gathering

• Data Preprocessing

4. Exploratory Data Analysis (EDA):

• Variable Analysis

• Correlation Analysis

5. Predictive Modeling:

• Model Selection

• Feature Engineering

• Model Training

6. Model Evaluation:

• Performance Metrics

• Cross-Validation

7. Deployment:

• System Integration

• User Training

8. Monitoring and Optimization:

• Continuous Monitoring

• Model Optimization

9. Reporting and Documentation:

• Results Reporting

• Documentation

10. Project Closure

• Final Review

• Project Closure Report

NBNSTIC, Department of Computer Engineering 2023-24 32

5.3.3 Timeline Chart

5.4 TEAM ORGANIZATION

5.4.1 Team structure

1. Vilas Dasalkar- Data Scientist.

2. Nihar Jachak - Team Lead, Analyst.

3. Kajol Pawar - Front-End Developer.

4. Prasad Kurale - Back-End Developer.

5.4.2 Management reporting and communication

Meet Date Project Work

1 04/08/23 First meet with guide
2 18/08/23 Discuss various ideas
3 15/09/23 Understanding the concept of Placement Analysis
4 29/09/23 Preparing a presentation
5 06/10/23 Study of various methods to solve the problem
6 13/10/23 Presentation on Methodologies
7 20/10/23 Presentation on Synopsis
8 31/10/23 Submit Rough copy of synopsis

Table 5.4: Progress Report

NBNSTIC, Department of Computer Engineering 2023-24 33

CHAPTER 6

SOFTWARE REQUIREMENT
SPECIFICATION
6.1 INTRODUCTION

6.1.1 Purpose and Scope of Document

The primary goal of this Software Requirements Specification (SRS) document is to

define and outline the functional and non-functional requirements for the develop-
ment of a predictive analysis system focused on student campus placement. It serves
as a comprehensive guide for all stakeholders involved in the project, providing a de-
tailed understanding of the software’s features, functionalities, and constraints. This
document aims to establish a common understanding among development teams,
testers, project managers, and other relevant stakeholders regarding the specific re-
quirements that the predictive analysis system must fulfill.

6.1.2 Overview of responsibilities of Developer

• Requirement Analysis: Review and understand the software requirements spec-

ified in the Software Requirements Specification (SRS) document. Clarify any
ambiguities or uncertainties in the requirements with the project manager or
stakeholders.

• Design and Planning: Plan the overall software architecture, including the
system’s structure, components, and data flow.Create detailed technical design
specifications that outline the software’s structure and modules.

• Coding: Write and implement the code that constitutes the software applica-
tion, adhering to coding standards and best practices. Develop and implement
algorithms and logic to enable the software to perform its intended functions.

• Testing and Debugging: Create unit tests to verify the correctness of individual
code components.Debug and troubleshoot issues, resolving defects and errors
that are identified during testing.

• Integration: Integrate developed components into the broader software system,

ensuring they work harmoniously together.

NBNSTIC, Department of Computer Engineering 2023-24 35

• Documentation: Document the code and its functionality to provide clear,
maintainable code for future reference. Maintain user and developer docu-
mentation for the software application.

• Code Review: Participate in peer code reviews to ensure code quality, adher-
ence to coding standards, and identifying potential improvements.

• Performance Optimization: Optimize code for performance, scalability, and

efficiency, addressing any bottlenecks or resource utilization issues.

• Security Considerations: Implement security best practices, including input

validation, authentication, and protection against common vulnerabilities.

• Version Control: Use version control systems (e.g., Git) to track changes and
collaborate with other team members effectively.

• Collaboration: Collaborate with other team members, such as designers, qual-

ity assurance testers, and project managers, to ensure a cohesive and successful
development process.

• Continuous Learning: Stay updated with the latest technologies, programming

languages, and best practices in software development.

• Troubleshooting and Support: Provide support and troubleshooting assistance

for users and resolve issues reported after deployment.

• Testing Automation: Develop and maintain automated test scripts for contin-
uous integration and regression testing.

• Deployment: Assist in deploying the software to production environments and

address any deployment-related issues.

6.2 USAGE SCENARIO

6.2.1 User profiles

• Actor: Student (end-user)

NBNSTIC, Department of Computer Engineering 2023-24 36

– Goal: The student wants to predict the likelihood of successful placement
based on academic performance and other relevant factors.

– Preconditions: The student has logged into the predictive analysis sys-
tem.The required data, including academic records and other relevant
information, is available in the system

• Steps:

– Initiate Prediction: The student navigates to the ”Placement Prediction”

section of the system. Clicks on the ”Initiate Prediction” button.

– Input Data: The system prompts the student to input academic details,
such as GPA, previous internship experience, and any additional relevant
information.The student enters the required data into the provided fields.

– Review and Confirm: The system displays a summary of the entered data
for the student to review.The student confirms the accuracy of the input
data.

– Run Prediction Model: The student triggers the system to run the predic-
tive analysis model.The system processes the input data using the under-
lying algorithm to predict the likelihood of successful placement.

– View Results: The system presents the prediction results to the stu-
dent.Results may include the likelihood percentage and any additional
insights or recommendations.

– Interpret Results: The student interprets the prediction results to under-

stand the potential outcomes.Considers the insights provided by the sys-
tem in decision-making.

– Save or Share Results:The student has the option to save the prediction
results for future reference.Additionally, the student can choose to share
the results with career counselors or other relevant stakeholders

• Postconditions: The student has successfully used the predictive analysis sys-
tem to assess the likelihood of successful placement. Results and insights are
stored for future reference

NBNSTIC, Department of Computer Engineering 2023-24 37

6.2.2 Use-cases

All use-cases for the software are presented. Description of all main Use cases using
use case template is to be provided.
Sr No. Use Case Description Actors Assumptions
1 Use Case 1 Description Actors Assumption

Table 6.1: Use Cases

6.2.3 Use Case View

Use Case Diagram.

Figure 6.1: Use case diagram

6.3 DATA MODEL AND DESCRIPTION

6.3.1 Data Description

• Student:

– Attributes: StudentID (Primary Key), Name,GPA (Grade Point Aver-

age),Internship Experience,Other relevant attributes (e.g. extracurricular
activities, skills)

• Prediction Result:

NBNSTIC, Department of Computer Engineering 2023-24 38

– Attributes: StudentID (Foreign Key referencing Student), Prediction-
Date, LikelihoodPercentage, AdditionalInsights

• SystemAdministrator:

– Attributes: AdminID (Primary Key), Username, Password, Name, Email

• CareerCounselor:

– Attributes: CounselorID (Primary Key), Name, Email, Phone

6.4 FUNCTIONAL MODEL AND DESCRIPTION

• User Authentication: Authenticate users (students, administrators, career coun-

selors) to ensure secure access to the system.

• Student Data Input: Allow students to input relevant data, including GPA,
internship experience, and other factors required for predictive analysis.

• Initiate Placement Prediction: Enable students to initiate the placement pre-

diction process based on the provided data.

• Run Predictive Analysis: Execute the predictive analysis algorithm using the
student’s input data and the selected predictive model.

• Display Prediction Results: Present the results of the predictive analysis to the
student, including the likelihood percentage and additional insights.

• Save Prediction Results: Allow students to save the prediction results for fu-
ture reference.

• Manage Predictive Model: Allow administrators to manage and update pre-

dictive models, including uploading new model files or adjusting parameters.

• Generate Placement Reports: Trigger the generation of reports summarizing

placement predictions and outcomes for distribution to relevant stakeholders.

• Review Student Predictions: Enable career counselors to review the placement

predictions of individual students and provide guidance.

NBNSTIC, Department of Computer Engineering 2023-24 39

• User Management: Allow administrators to manage user accounts, including
creating, updating, and deactivating accounts.

6.4.1 Data Flow Diagram

6.4.1.1 Level 0 Data Flow Diagram

Figure 6.2: Data Flow Diagram

6.4.1.2 Level 1 Data Flow Diagram

6.4.2 Description of Functions

Software Functions

1. Academic Data Input: This function allows students to input their academic
data (grades, courses taken, extracurriculars, etc.) into the system for predic-
tion analysis.

NBNSTIC, Department of Computer Engineering 2023-24 40

Figure 6.3: Data Flow Diagram
2. Prediction Algorithm: Responsible for processing the student’s academic
data using machine learning or AI algorithms to predict potential job place-
ments.

3. Results Presentation: Once predictions are generated, this function presents

the results to the student, displaying suggested job positions or industries.

4. Feedback Loop: Enables students to provide feedback on predicted results,

which can be used to enhance future predictions.

5. Profile Update: Allows students to update their personal and academic infor-
mation within the system for more accurate predictions in subsequent analy-
ses.

NBNSTIC, Department of Computer Engineering 2023-24 41

Processing Narrative for Function (Academic Data Input)

1. Student Prompt: The system prompts the student to input their academic data
by providing a user-friendly interface or form.

2. Data Input: The student enters their academic information, including grades,
courses taken, project details, internships, and other relevant data.

3. Validation: The system checks the entered data for completeness and validity,
ensuring all required fields are filled and the format is correct.

4. Data Storage: Validated data is stored securely in the system’s database, as-
sociated with the respective student’s profile.

5. Pre-Processing: The stored data undergoes pre-processing steps like normal-

ization or standardization to ensure uniformity and compatibility with the pre-
diction algorithm.

6. Pass to Prediction Algorithm: The pre-processed academic data is passed to

the prediction algorithm module for analysis and prediction generation.

7. Prediction Output: The algorithm processes the data and generates predic-
tions for potential job placements based on historical data and patterns.

8. Results Display: The system presents the predicted placements to the student
through a user interface or report, detailing suggested job positions or indus-
tries.

9. End: The process of inputting academic data and receiving predictions con-
cludes.

6.4.3 Activity Diagram:

Main Flow of the System

1. Start: User initiates the system.

NBNSTIC, Department of Computer Engineering 2023-24 42

Figure 6.4: Activity diagram
2. Activity 1: Academic Data Collection

3. Activity 2: Data Preprocessing

4. Activity 3: Prediction Model Analysis

5. Activity 4: Result Generation

6. Activity 5: Recommendations Display

7. End: Student receives placement predictions.

NBNSTIC, Department of Computer Engineering 2023-24 43

Academic Data Collection (Detailed)

1. Start: System prompts the student for academic data.

2. Activity 1: Student provides academic records.

3. Activity 2: Data is captured and uploaded.

4. Activity 3: Data undergoes validation and preprocessing.

5. Activity 4: Preprocessed data is fed into the prediction model.

6. End: Predicted results are available for display.

6.4.4 Non Functional Requirements:

1. Performance: The system should process predictive analyses for individual

students within seconds to ensure a responsive user experience.

2. Scalability: The system should scale effectively to accommodate a growing

number of students, administrators, and career counselors without significant
performance degradation.

3. Reliability: The system should demonstrate a high level of reliability, mini-

mizing errors or downtime. Predictive analyses must maintain accuracy and
consistency.

4. Security: User authentication and data transmission must be secured through

encryption to protect sensitive information. Access controls should restrict
user privileges based on roles.

5. Usability: The user interface should be intuitive and user-friendly for students,
administrators, and career counselors. Adequate documentation and training
materials should be provided.

6. Availability: The system should maintain 24/7 availability with minimal planned
downtime for maintenance. Users should be notified in advance of scheduled
maintenance.

NBNSTIC, Department of Computer Engineering 2023-24 44

7. Compatibility: The system should be compatible with popular web browsers
and devices to ensure accessibility for a diverse user base.

8. Data Integrity: Data stored in the system should uphold accuracy and integrity
over time. Data backups and recovery mechanisms should prevent data loss.

9. Auditability: The system should maintain audit logs tracking user activities,
especially for critical functions like running predictive analyses and accessing
sensitive data.

6.4.5 State Transition Diagram:

Fig.6.5 shows the state transition diagram. The state of system gets changed when
certain events occur. The transitions from one state to the other are represented by
arrows. The Figure shows important states and events that occur while creating new
project.

Figure 6.5: State transition diagram

NBNSTIC, Department of Computer Engineering 2023-24 45

6.4.6 Design Constraints

1. Data Availability and Quality: Availability of accurate academic and historical

placement data might be limited, impacting prediction accuracy.

2. Regulatory Compliance: Adherence to data privacy laws (e.g., GDPR, CCPA)

influences how student data is collected, stored, and used for predictions.

3. Resource Limitations: Limited computational resources for handling a large

volume of student data can impact prediction processing efficiency.

4. Integration with Educational Systems: Challenges in integrating with exist-

ing educational databases due to varying data formats, access limitations, or
compatibility issues.

5. Security and Privacy Concerns: Ensuring data security against cyber threats
and maintaining user privacy while handling sensitive student information is
critical.

6. Scalability: Designing the system to accommodate a growing number of users

without compromising performance is a significant constraint.

7. Testing and Validation: Limited availability of real-world data for testing and
validating prediction models may affect accuracy and reliability.

8. Financial Limitations: Budget constraints might limit system features, tech-

nology stack choices, and resource acquisition.

9. Technology Compatibility: Institutional policies or preferences might restrict

the choice of technology stack used for the system.

10. Project Timeline: Development cycles affected by academic schedules may

impact feature implementation speed.

6.4.7 Software Interface Description

Application Programming Interface (API)

Description: The system offers APIs for seamless integration with external applica-

NBNSTIC, Department of Computer Engineering 2023-24 46

tions or services. These APIs enable programmatic access to prediction results and
facilitate data exchange with external databases.

Database Interface

Description: The system interacts with a database using various functionalities such
as queries, transactions, and protocols. This interface ensures efficient data storage
and retrieval within the system.

External Predictive Modeling Tools

Description: In cases where the system employs external predictive modeling tools
or libraries, this interface manages the exchange of data between the system and
these tools. It encompasses data formats, model parameters, and retrieving results.

Authentication and Authorization Systems

Description: Integration with authentication and authorization systems is essential

for user login and access control within the system. This interface may involve
interaction with identity providers or Single Sign-On (SSO) services.

Reporting and Analytics Tools

Description: The system interfaces with reporting and analytics tools to generate
comprehensive reports regarding placement predictions. This interface includes data
extraction, transformation, and presentation functionalities.

External Student Databases

Description: If the system depends on external student databases for supplementary

information, this interface specifies the procedures for data retrieval, synchroniza-
tion, and updates between the predictive analysis system and external databases.

NBNSTIC, Department of Computer Engineering 2023-24 47

CHAPTER 7

DETAILED DESIGN DOCUMENT USING

APPENDIX A AND B
7.1 INTRODUCTION

Student placement prediction system marks a significant advancement in educational

technology, offering a proactive approach to streamline the placement process. This
innovative system harnesses predictive analytics, leveraging historical and real-time
data to forecast and optimize student placements within the workforce. At its core,
the placement prediction system aims to revolutionize how educational institutions
facilitate students’ transition from academia to their professional careers. By ana-
lyzing a myriad of factors, including academic performance, skills, extracurricular
activities, and industry trends, this system aims to forecast a student’s likelihood of
success in various job roles or industries. Through the amalgamation of sophisti-
cated algorithms and data-driven methodologies, the system endeavors to provide
tailored recommendations and insights, guiding students towards career paths that
align with their strengths and aspirations. Moreover, it assists academic advisors,
career counselors, and placement cells in making informed decisions to enhance
placement success rates. The predictive system’s introduction signifies a paradigm
shift in optimizing the employment outcomes of students by fostering a data-centric
approach. It seeks to bridge the gap between academic preparation and industry
demands, enabling educational institutions to adapt their curriculum and support
services dynamically. Ultimately, the placement prediction system is poised to rev-
olutionize the educational landscape, offering a proactive and data-driven solution
to optimize student placements, thereby enhancing their career prospects and con-
tributing to the overall success of both students and educational institutions.

7.2 ARCHITECTURAL DESIGN

A description of the program architecture is presented. Subsystem design or Block

diagram,Package Diagram,Deployment diagram with description is to be presented.

NBNSTIC, Department of Computer Engineering 2023-24 49

Figure 7.1: Architecture diagram
7.3 DATA DESIGN (USING APPENDICES A AND B)

7.3.1 Internal Software Data Structure

Data Structures Passed Among Components

• Feature Vectors: Structures representing student profiles, including academic

performance, skills, and extracurricular activities.

• Prediction Output Format: Data structures encapsulating prediction results for

each student.

7.3.2 Global Data Structure

Available Data Structures Across Architecture

• Placement Prediction Configuration: Parameters and settings affecting the pre-

diction algorithm across the system.

• Common Student Profile Format: A standardized format for storing and ac-
cessing student data across various modules.

7.3.3 Temporary Data Structure

Interim Files for Usage

• Temporary Training Data Files: Temporary datasets created for model training
and validation.

NBNSTIC, Department of Computer Engineering 2023-24 50

• Intermediate Prediction Results: Temporary storage for predictions during
processing.

7.3.4 Database Description

Database and Files Used/Created

• Student Information Table: Stores detailed student profiles including academic

records, personal details, and extracurricular activities.

• Prediction Results Log: Log file/table to store prediction outcomes for future
analysis and reference.

• Model Parameters File: Storage for machine learning model weights, hyper-
parameters, or configurations.

7.4 COMPONENT DESIGN

Data Preprocessing Component

• Class Diagram: Illustrates classes for data cleaning, transformation, and fea-
ture extraction.

• Interaction Diagram: Depicts interactions among data preprocessing classes.

• Algorithms: Feature scaling, normalization, and data encoding techniques

used.

Machine Learning Model Component

• Class Diagram: Classes representing various machine learning models (e.g.,

decision trees, neural networks).

• Interaction Diagram: Depicts the flow of data through model training, valida-
tion, and prediction phases.

• Algorithms: Machine learning algorithms like Random Forest, SVM, or Neu-

ral Network training and prediction methodologies.

NBNSTIC, Department of Computer Engineering 2023-24 51

Prediction Output Component

• Class Diagram: Classes handling the format and storage of prediction results.

• Interaction Diagram: Shows the flow of prediction results to different modules.

• Algorithms: Formatting algorithms for presenting prediction outcomes.

7.4.1 Class Diagram

Detailed class diagrams depicting the relationships and interactions between differ-
ent classes within the system:

• User Interface Classes: Representations of classes handling user interactions.

• System Module Classes: Illustrations of classes within each system module.

and their relationships

Figure 7.2: Class Diagram

NBNSTIC, Department of Computer Engineering 2023-24 52

CHAPTER 8

PROJECT IMPLEMENTATION
8.1 INTRODUCTION

The ”C2C- Predictive Analysis of Student Campus Placement” system is designed to

assist students, administrators, and career counselors in making informed decisions
about student placement based on academic performance and relevant factors. The
system follows a layered architecture with various modules and components working
collaboratively to provide a seamless and secure user experience.

8.2 TOOLS AND TECHNOLOGIES USED

1. Reactjs: The primary frontend framework for developing the user interface of
the web application.

2. Material-UI: Provides pre-designed React components following Google’s

Material Design guidelines, ensuring a consistent and visually appealing UI.

3. ApexCharts (React): A React wrapper for ApexCharts, allowing the integra-

tion of interactive and responsive charts into the application for data visualiza-
tion.

4. Flask: A lightweight Python web framework used for building the backend of
the application, handling HTTP requests, and serving data to the frontend.

5. Redhat Openshift: A containerization platform used for deploying, manag-

ing, and scaling the application in a cloud environment.

6. Docker: Utilized for containerization of the application components, ensuring

consistency in development and deployment across different environments.

7. Numpy: A Python library for numerical computing, used for efficient han-
dling of large datasets and mathematical operations in data processing.

8. Pandas: Another Python library used for data manipulation and analysis, pro-
viding data structures and functions to work with structured data.

9. Matplotlib: A Python plotting library used for creating static, interactive, and
animated visualizations of data.

NBNSTIC, Department of Computer Engineering 2023-24 54

10. Seaborn: Built on top of Matplotlib, Seaborn is used for statistical data visual-
ization, providing a high-level interface for drawing attractive and informative
graphics.

11. Scikit-learn: A machine learning library in Python, used for implementing

various machine learning algorithms for data analysis and prediction tasks.

8.3 METHODOLOGIES/ALGORITHM DETAILS

The project begins with formulating objectives and collecting data by the college.
Subsequently, the gathered data undergoes preparation, which involves segmenting
it into training and testing datasets. This process includes separating the data into
dependent and independent variables, followed by identifying and eliminating out-
liers.

8.3.1 Procedural Steps

8.3.1.1 Model Training:

The model is trained using the prepared training data.

8.3.1.2 Model Refinement:

The model is fine-tuned, and the training data is further categorized based on speci-
fied parameters.

8.3.1.3 Processing the Testing Dataset:

After model training, the previously stored testing dataset is processed through the
model for prediction.

8.3.1.4 Confirmation of Predictions:

If the predicted values from the testing dataset align with the expected outcomes,
these values are confirmed as the final predictions.

NBNSTIC, Department of Computer Engineering 2023-24 55

8.3.1.5 Handling Unsatisfactory Predictions:

In cases where the predicted values are deemed unsatisfactory or false, indicating a
lack of alignment with expectations, the values are returned for additional fine-tuning
of the model.

8.3.2 Utilized Algorithms

• Logistic Regression (Binary Classification): Logistic regression, a widely-

used statistical technique, is employed for binary outcome prediction, such as
determining the likelihood of an individual securing a job post-training. This
method involves collecting information on individuals who have completed
the training program, including educational background, professional experi-
ence, and training performance. Predictions are generated by inputting pre-
dictor values for a new individual, with the model estimating the probability
of job placement. If the probability exceeds a predefined threshold successful
placement is anticipated; otherwise, non-placement is predicted.

• Random Forest: Random Forest, a versatile algorithm suitable for multiclass

classification, is utilized to predict students’ placement ranges.

1. Data Preparation: Comprehensive data on academic performance, skills,

and relevant parameters is collected.

2. Feature Selection: Essential features correlated with the target variable

(placement range) are identified using techniques like correlation analy-
sis.

3. Model Training: An ensemble of decision trees is iteratively constructed

to minimize the loss function.

4. Hyperparameter Tuning: Fine-tuning of hyperparameters, including

the learning rate, number of trees, and maximum depth, is undertaken.

5. Model Evaluation: A comprehensive assessment of model performance

is conducted, incorporating metrics such as accuracy, precision, recall,
and F1-score.

NBNSTIC, Department of Computer Engineering 2023-24 56

• Decision Tree: Decision tree algorithm is strategically applied in multiclass
classification to forecast student placement.

1. Data Processing: Collection and preprocessing of data, including ad-

dressing missing values, outliers, and encoding categorical variables, is
meticulously executed.

2. Feature Importance: Identifying pivotal features is achieved using tech-

niques such as Gini impurity or information gain.

3. Train-Test Split: Strategic division of data into training and testing sets
(e.g., 80-20 split) is implemented.

4. Model Training: Training of decision tree classifiers on the provided

dataset is systematically carried out.

5. Hyperparameter Tuning: Optimization of hyperparameters specific to

the decision tree algorithm is diligently performed. Hyperparameters
may include maximum depth, minimum samples split, and minimum
samples leaf.

6. Evaluation: Rigorous assessment of performance metrics, including pre-

cision, recall, F1-score, and accuracy, is executed.

Our future endeavors involve leveraging the advanced capabilities of XG-

Boost, also known as eXtreme Gradient Boosting, to significantly enhance predic-
tive accuracy. By capitalizing on the strengths of this sophisticated machine learning
algorithm, we expect to achieve more precise and reliable predictions in our forth-
coming analytical endeavors.

NBNSTIC, Department of Computer Engineering 2023-24 57

CHAPTER 9

SOFTWARE TESTING
9.1 TYPE OF TESTING USED

The ”C2C- Predictive Analysis of Student Campus Placement” project underwent

rigorous testing to ensure its functionality, reliability, and performance. Various
testing methodologies were employed, including:

1. Unit Testing: Individual components and modules of the system were tested
in isolation to verify their correctness and functionality.

2. Integration Testing: The interaction between different modules and compo-

nents was tested to ensure seamless integration and communication.

3. System Testing: The system as a whole underwent testing to validate its com-
pliance with specified requirements and to identify any defects or inconsisten-
cies.

4. User Acceptance Testing (UAT): End-users, including students, recruiters,

and academic administrators, participated in UAT to evaluate the system’s us-
ability, functionality, and overall satisfaction.

9.2 TEST CASES AND TEST RESULTS:

1. Unit Testing:

• Test Case 1: Verify that the predictive models accurately predict student
placements based on historical data. Result: Passed

• Test Case 2: Ensure that the user authentication module securely man-
ages user credentials. Result: Passed

2. Integration Testing:

• Test Case 1: Verify that the predictive models seamlessly integrate with
the user interface for displaying placement predictions. Result: Passed

• Test Case 2: Ensure that the database management system effectively

stores and retrieves student and placement data. Result: Passed

NBNSTIC, Department of Computer Engineering 2023-24 59

3. System Testing:

• Test Case 1: Validate that the system allows administrators to input new
data and update existing records. Result: Passed

• Test Case 2: Confirm that the reporting module generates accurate and
insightful analytics on placement trends. Result: Passed

4. User Acceptance Testing (UAT):

• Test Case 1: Evaluate user experience in navigating the system’s inter-

face and accessing placement predictions. Result: Passed with minor
feedback for UI enhancements.

• Test Case 2: Assess the satisfaction level of recruiters with the accuracy
and relevance of placement predictions. Result: Passed with positive
feedback for predictive accuracy and usefulness.

Overall, the testing phase confirmed the functionality, reliability, and user sat-
isfaction of the ”C2C- Predictive Analysis of Student Campus Placement” system,
ensuring its successful implementation and deployment.

NBNSTIC, Department of Computer Engineering 2023-24 60

CHAPTER 10

RESULTS
10.1 SCREEN SHOTS

NBNSTIC, Department of Computer Engineering 2023-24 62

NBNSTIC, Department of Computer Engineering 2023-24 63
OUTPUTS / SNAPSHOTS OF THE RESULTS:

Campus Analyzer

NBNSTIC, Department of Computer Engineering 2023-24 64

Student Placement Analyzer

NBNSTIC, Department of Computer Engineering 2023-24 65

CHAPTER 11

DEPLOYMENT AND MAINTENANCE

11.1 INSTALLATION AND UNINSTALLATION

11.1.1 Installation

1. Download the Installation Package: Access the official ”C2C- Predictive

Analysis of Student Campus Placement” project repository or website. Down-
load the installation package, which may be available as a zip file, executable,
or source code.

2. Extract the Package (if applicable): If the package is compressed, extract its
contents to a suitable location on your system.

3. Check System Requirements: Ensure that your system meets the minimum
requirements specified for running the project.

4. Install Dependencies (if any): Review the project documentation for any re-
quired dependencies. Install necessary software or libraries if they are not
already installed on your system.

5. Run the Installation Script (if applicable): Execute any provided installa-
tion script to set up the project. Follow the instructions provided during the
installation process.

6. Configure the Project: If configuration is required, such as specifying database

connections or API keys, provide the necessary information as per the project
documentation.

7. Start the Application: Launch the ”C2C- Predictive Analysis of Student

Campus Placement” project according to the provided instructions. This may
involve running a command, executing an executable file, or accessing a web
URL.

8. Verify Installation: Test the functionality of the project to ensure it is running

correctly. Verify that all features can be accessed without encountering any
errors.

NBNSTIC, Department of Computer Engineering 2023-24 67

11.1.2 Uninstallation

1. Stop the Application: Ensure that the ”C2C- Predictive Analysis of Student
Campus Placemens” project is not running to avoid conflicts during uninstal-
lation. Terminate any associated processes or services.

2. Backup Data (if needed): If important data is stored within the project, back
it up before proceeding with uninstallation.

3. Locate Uninstallation Instructions: Refer to the project documentation for

specific uninstallation instructions.

4. Execute Uninstallation Script (if provided): If an uninstallation script is

available, run it to remove project files and configurations.

5. Manual Removal (if necessary): If no uninstallation script is provided, man-

ually remove project files and directories, ensuring not to delete essential sys-
tem files.

6. Remove Dependencies (if applicable): Uninstall any additional dependen-

cies installed during the installation process using the appropriate method for
your operating system.

7. Clean Up Configuration Files (if needed): Remove any configuration files

or settings created during installation.

8. Verify Uninstallation: Ensure that all project-related files and components

have been successfully removed from your system. Confirm that no leftover
directories or files remain.

9. Restore Data (if needed): If data was backed up before uninstallation, restore
it to its original location.

NBNSTIC, Department of Computer Engineering 2023-24 68

CHAPTER 12

CONCLUSION AND FUTURE SCOPE

Key aspects:

• User Interface (UI): Provides an intuitive interface for users to input data, view
predictions, and manage accounts.

• Application Layer: Contains the core business logic, handling user input, ini-
tiating predictive analyses, and managing system components.

• Database: Stores student data, prediction results, user information, and other
relevant data, ensuring data integrity and accessibility.

• Predictive Analysis Module: Encapsulates the predictive analysis algorithm,

processing input data to generate placement predictions.

• User Management Module: Handles user authentication, authorization, and

account management to ensure secure user access.

• External Integrations Module: Manages interfaces with external systems, fa-

cilitating data exchange with career guidance services, student databases, and
reporting tools.

• Security Module: Enforces security measures, including encryption, access

controls, and audit logging.

• Reporting and Analytics Module: Generates comprehensive reports based on

prediction results, providing insights for decision-making.

Conclusion:

The designed system aims to provide a reliable, scalable, and user-friendly platform
for predictive analysis of student placement. By incorporating various architectural
styles, deployment considerations, and security measures, the system is poised to
meet the needs of students, administrators, and career counselors while adhering to
ethical standards and regulatory requirements. Ongoing monitoring, updates, and
user feedback will contribute to the system’s continuous improvement and effective-
ness. The proposed system, designed for comprehensive student evaluation, inte-
grates meticulous data preprocessing and iterative model refinement. Looking ahead,

NBNSTIC, Department of Computer Engineering 2023-24 70

the integration of XGBoost represents a promising avenue for further enhancing pre-
dictive accuracy. Our study significantly contributes to the progression of placement
predictions, providing tangible benefits to both students and institutions. Future re-
search directions involve harnessing XGBoost’s advanced capabilities for precise
predictions and exploring its broader applications in analytics and decision-making
processes within educational contexts.

Future Scope:

The future scope for the Campus Placement Analysis project is dynamic and opens
doors for continuous innovation, expansion, and integration of emerging technolo-
gies. As the project evolves, it has the potential to reshape not only campus place-
ments but also the entire landscape of career development and education-to-employment
transition.

1. Integration of Real-time Data:

Incorporating real-time job market data, economic indicators, and industry

trends will enhance the accuracy of predictions and recommendations, keeping
the system up-to-date with dynamic market shifts.

2. NLP and Sentiment Analysis:

Implementing Natural Language Processing (NLP) and sentiment analysis can

help the system extract insights from job descriptions, company reviews, and
industry news, enriching the quality of recommendations and predictions.

3. Multi-Dimensional Analysis:

Expanding beyond branch-wise analysis, the project can incorporate multi-

dimensional insights, such as gender, region, and diversity, to create more
inclusive placement strategies.

4. Personalized Learning Pathways:

Integrating learning platforms with personalized pathways based on placement

trends can enable students to proactively develop skills for upcoming job roles.

NBNSTIC, Department of Computer Engineering 2023-24 71

REFERENCES
1 M. Siva Surya, Dr. M. Sathish Kumar, Dr. D. Gandhimathi”Student Placement
Prediction Using Supervised MachineLearning”

DOI: 10.1109/ICACITE53722.2022.9823648.

2 Gurivireddy Mary Spandana, L Pallavi ”PlacementPrediction System using

Machine Learning” DOI: 10.1109/ICECAA58104.2023.10212409.

3 Navuluri Divya, Sravya Namburu, Rajalakshmi Raja”Student Placement Anal-

ysis using Machine Learning”DOI:10.1109/ICCES57224.2023.10192633.

4 Jyoti Khandelwal Gaurav Pareek, Ratul Dey, ShilpaPareek ”The Study of

Machine Learning ClassificationAlgorithm for Student Placement Prediction”
DOI: 10.1109/IEMECON56962.2023.10092294.

5 K. Rai, Predicting Covid-19 based on symptomsusingmachine learning tech-

niques implemented inPython, AIPConference Proceedings, vol. 2555, pp.
050002-1–050002-10,October 2022.

6 Manoj Manike, Priyanshu Singh, Purna Sai Madala, SteveAbraham Varghese,

Saleti Sumalatha ”Student PlacementChance Prediction Model using Machine
LearningTechniques” DOI: 10.1109/CICT53865.2020.9672372.

7 B. Singh, S. K. Henge, A. Sharma, C. Menaka, P. Kumar,S. K. Mandal, and

B. Debtera, ML-Based InterconnectedAffecting Factors with Supporting Ma-
trices for AssessmentofRisk in Stock Market, Wireless Communications and-
MobileComputing, vol. 2022, 15 pages, Dec. 2022.

8 Nitesh Kumar Sharma, Aniket Kumar Singh, ShubhamSalvi ”College Kart

and k-NN AlgorithmbasedPlacementPrediction”

DOI: 10.1109/ICESC51422.2021.9532987

NBNSTIC, Department of Computer Engineering 2023-24 73

ANNEXURE A

LABORATORY ASSIGNMENTS ON
PROJECT ANALYSIS OF ALGORITHMIC
DESIGN
A.1 PROBLEM STATEMENT DEVELOPMENT

The project aims to create a predictive system for student placement in academic
institutions. The problem involves developing a machine learning-based model to
predict the likelihood of students being placed in specific job roles or companies
after completing their education. This predictive system targets enhancing career
guidance and optimizing the placement process for students and educational institu-
tions.

A.2 JUSTIFYING FEASIBILITY USING IDEA MATRIX

• Increase:

– Increase Prediction Accuracy: Enhance the accuracy of placement pre-

dictions by leveraging advanced machine learning algorithms and com-
prehensive data analysis of historical placement records.

• Deliver:

– Deliver Customization: Ensure the system provides customized recom-

mendations based on individual student profiles, academic records, skills,
and industry trends.

• Evaluate:

– Evaluate Algorithmic Robustness: Assess the robustness and performance

of various machine learning algorithms like decision trees, random forests,
and neural networks to determine the most suitable model for accurate
predictions.

• Accelerate:

– Accelerate System Scalability: Optimize the system’s scalability to han-

dle a large volume of student data efficiently while maintaining predic-
tion accuracy.

NBNSTIC, Department of Computer Engineering 2023-24 75

A.3 FEASIBILITY ASSESSMENT USING MATHEMATICAL MODELS

A.3.1 NP-Hard, NP-Complete, or Satisfiability Issues

Utilize mathematical models to analyze the computational complexity of the predic-

tion algorithms. Assess whether any aspects of the placement prediction problem
fall into NP-Hard, NP-Complete, or satisfiability problem domains.

A.3.2 Input-Output Relationship (y=f(x))

Establish a clear mapping between input data (student profiles, academic records)
and output predictions (placement outcomes) using mathematical modeling, ensur-
ing a well-defined input-output relationship.

A.4 INTEGRATION OF KNOWLEDGE CANVAS

A.4.1 Identification of Opportunity for Product

• Identify Key Placement Factors: Recognize crucial factors influencing student

placements, such as academic performance, skills, internships, and industry
demand, emphasizing the importance of accurate prediction models.

A.4.2 Feasibility from a Business Perspective

• Assess Industry Demand: Conduct market analysis and surveys to gauge the
demand for predictive placement systems in educational institutions and the
job market.

A.4.3 Integration of Concepts

• Alignment of User Needs and Feasibility: Integrate insights from market anal-
ysis and user needs assessment with feasibility evaluations, ensuring the pre-
dictive system aligns with both computational efficiency and market demand.

By combining these elements, the project aims to create an efficient and accu-
rate student placement prediction system that caters to the needs of both educational
institutions and students seeking career opportunities.

NBNSTIC, Department of Computer Engineering 2023-24 76

ANNEXURE B

LABORATORY ASSIGNMENTS ON
PROJECT QUALITY AND RELIABILITY
TESTING OF PROJECT DESIGN
B.1 QUALITY ASSURANCE PLAN

Objective:

• Develop a comprehensive Quality Assurance (QA) plan for the Student Place-
ment Prediction System.

Tasks:

• Define QA objectives, standards, and methodologies.

• Create test cases for each module (speech recognition, NLP, email composi-
tion).

• Establish criteria for evaluating system performance and accuracy.

B.2 RELIABILITY TESTING

Objective:

• Assess the system’s reliability and robustness under various conditions.

Tasks:

• Conduct stress testing on the system (e.g., heavy user load, simultaneous com-
mands).

• Perform reliability testing by subjecting the system to prolonged usage.

• Simulate adverse conditions (e.g., network interruptions, background noise) to

gauge system response.

B.3 USABILITY TESTING FOR ACCESSIBILITY

Objective:

• Evaluate the system’s usability and accessibility for visually impaired users.

Tasks:

• Recruit participants with visual impairments for user testing.

NBNSTIC, Department of Computer Engineering 2023-24 78

• Collect feedback on the system’s intuitiveness, effectiveness, and accessibility
features.

• Identify areas for improvement based on user feedback and observations.

B.4 INTEGRATION TESTING AND INTEROPERABILITY

Objective:

• Ensure seamless integration of system components and compatibility with ex-

ternal systems.

Tasks:

• Perform integration testing to verify the interaction between speech recogni-

tion, NLP, and email modules.

• Assess interoperability with existing assistive technologies (e.g., screen read-

ers).

• Document any compatibility issues and propose solutions or adjustments.

B.5 SECURITY AND PRIVACY TESTING

Objective:

• Assess the system’s security measures and privacy protocols.

Tasks:

• Conduct vulnerability assessments and penetration testing.

• Evaluate user authentication mechanisms for security loopholes.

• Review privacy policies and ensure compliance with data protection regula-
tions.

NBNSTIC, Department of Computer Engineering 2023-24 79

B.6 PERFORMANCE EVALUATION AND OPTIMIZATION

Objective:

• Analyze the system’s performance and identify areas for optimization.

Tasks:

• Identify bottlenecks and areas of inefficiency within the system.

• Propose optimization strategies to enhance system performance.

• Implement recommended optimizations and measure the impact.

NBNSTIC, Department of Computer Engineering 2023-24 80

ANNEXURE C

PROJECT PLANNER
Sr. No. Name/Title Start Date End Date

1 Preliminary Survey 25/08/2023 1/09/2023

2 Introduction and Problem Statement 1/09/2023 8/09/2023

3 Literature Survey 8/09/2023 15/09/2023

4 Project Statement 15/09/2023 22/09/2023

5 Software Requirement and Specification 22/09/2023 29/09/2023

6 System Design 29/09/2023 6/10/2023

7 Partial Report Submission 6/10/2023 6/10/2023

8 Architecture Design 6/10/2023 13/10/2023

9 Implementation Plan 13/10/2023 13/10/2023

10 Deployment Plan 13/10/2023 20/10/2023

11 Testing Plan 20/10/2023 27/10/2023

12 Paper Publish 9/10/2023 19/10/2023

13 Report Submission 31/10/2023 3/11/2023

14 Presented 1st and 2nd module of the project 12/01/2024 26/01/2024

15 Completed UI for the project and integrated modules 26/01/2024 29/02/2024

16 Presented 3rd and 4th module of the project 29/02/2024 29/03/2024

17 Integrated module 3rd and 4th with the UI 29/03/2024 12/04/2024

18 Debugging and Testing 12/04/2024 19/04/2024

19 Report Submission 02/05/2024 02/05/2024

Table C.1: Project Timeline

NBNSTIC, Department of Computer Engineering 2023-24 82

ANNEXURE D

REVIEWERS COMMENTS OF PAPER

SUBMITTED
(At-least one technical paper must be submitted in Term-I on the project design
in the conferences/workshops in IITs, Central Universities or UoP Conferences or
equivalent International Conferences Sponsored by IEEE/ACM)

1. Paper Title: ”C2C - Predictive Analysis of Student Campus Placement”

• Name of the Conference/Journal where paper submitted : International

Journal of Scientific Research in Engineering and Management (IJS-
REM)

• Paper accepted/rejected : Accepted

2. Paper Title: ”Campus To Corporate Predictive Analysis.”

• Name of the Conference/Journal where paper submitted : NCCC-2024

• Paper accepted/rejected : Accepted

NBNSTIC, Department of Computer Engineering 2023-24 84

ANNEXURE E

PLAGIARISM REPORT
Figure E.1: Plagiarism Report

NBNSTIC, Department of Computer Engineering 2023-24 86

ANNEXURE F

INFORMATION OF PROJECT GROUP

MEMBERS
1. Name : Vilas Dasalkar

2. Date of Birth :04/02/2000

3. Gender : Male

4. Permanent Address : Vishnu Nagar Selu, Parbhani

5. E-Mail : [email protected]

6. Mobile/Contact No. : 7758024749

7. Paper Published:Yes

NBNSTIC, Department of Computer Engineering 2023-24 88

1. Name : Nihar Balasaheb Jachak

2. Date of Birth : 22/10/2002

3. Gender : Male

4. Permanent Address : Flt-355,Bldg-6,LIG Colony,GokhaleNagar,Pune-411016

5. E-Mail : [email protected]

6. Mobile/Contact No. : 7620280669

7. Placement Details : Placed at Hexaware Technologies

8. Paper Published:Yes

NBNSTIC, Department of Computer Engineering 2023-24 89

1. Name : Kajol Padmakar Pawar

2. Date of Birth : 18/04/2002

3. Gender : Female

4. Permanent Address : 23, prabhat colony, Jalgaon -425001

5. E-Mail : [email protected]

6. Mobile/Contact No. : 8600857462

7. Placement Details : Placed at Accenture

8. Paper Published:Yes

NBNSTIC, Department of Computer Engineering 2023-24 90

1. Name : Prasad Tanaji Kurale

2. Date of Birth : 25/08/2003

3. Gender : Male

4. Permanent Address :Sr.no.-30/6 Jijamata chowk,Ambegoan BK, Pune -411046

5. E-Mail : [email protected]

6. Mobile/Contact No. : 8421492004

7. Placement Details : Placed at Hexaware Technologies

8. Paper Published:Yes

NBNSTIC, Department of Computer Engineering 2023-24 91