CAPSTONE PROJECT REPORT

Project Term August-December 2019

ACADEMIC PERFORMANCE PREDICTOR

Submitted by

DEEPANSHU PATHAK Registration Number: 11612065

SHUBHAM SHARMA Registration Number: 11616269
SAKET KUMAR SANKRIT Registration Number: 11601496
SAKET TRIPATHI Registration Number: 11614487

Project Group Number - CSERGC0547

Course Code CSE 439

Under the Guidance of

(Robin Prakash Mathur, Assistant Professor )

School of Computer Science and Engineering

1
 TOPIC APPROVAL PERFORMA

School of Computer Science and Engineering (SCSE)

Program : P133::B.Tech. (Information Technology)

COURSE CODE : INT439 REGULAR/BACKLOG : Regular GROUP NUMBER : CSERGC0547

Supervisor Name : Robin Prakash UID : 14597 Designation : Assistant Professor

Mathur

Qualification : Research Experience :

SR.NO. NAME OF STUDENT REGISTRATION NO BATCH SECTION CONTACT NUMBER

1 Shubham Sharma 11616269 2016 K1609 9660283690

2 Saket Tripathi 11614487 2016 K1620 9530594825
3 Deepanshu Pathak 11612065 2016 K1620 9872259457
4 Saket kumar Sankrit 11601496 2016 K1620 7017331698

SPECIALIZATION AREA : Database Systems Supervisor Signature: _ _ _ _ _

PROPOSED TOPIC : Academic Performance Predictor

Qualitative Assessment of Proposed Topic by PAC

Sr.No. Parameter Rating (out of 10)

1 Project Novelty: Potential of the project to create new knowledge 5.67

2 Project Feasibility: Project can be timely carried out in-house with low-cost and available resources in 6.44
the University by the students.
3 Project Academic Inputs: Project topic is relevant and makes extensive use of academic inputs in UG 6.22
program and serves as a culminating effort for core study area of the degree program.
4 Project Supervision: Project supervisor’s is technically competent to guide students, resolve any issues, 7.22
and impart necessary skills.
5 Social Applicability: Project work intends to solve a practical problem. 6.22

6 Future Scope: Project has potential to become basis of future research work, publication or patent. 5.89

PAC Committee Members

PAC Member (HOD/Chairperson) Name: Kewal Krishan UID: 11179 Recommended (Y/N): Yes
PAC Member (Allied) Name: Gurpreet Singh UID: 17671 Recommended (Y/N): Yes
PAC Member 3 Name: Savleen Kaur UID: 18306 Recommended (Y/N): Yes

Final Topic Approved by PAC: Academic Performance Predictor

Overall Remarks: Approved

PAC CHAIRPERSON Name: 11024::Amandeep Nagpal Approval Date: 29 Apr 2019

ii
 DECLARATION
We hereby declare that the project work entitled “ACADEMIC
PERFORMANCE PREDICTOR” is an authentic record of our own work
carried out as requirements of Capstone Project for the award of B. Tech
degree in Computer Science & Engineering from Lovely Professional
University, Phagwara, under the guidance of Mr. Robin Prakash Mathur,
during August to December 2019. All the information furnished in this
capstone project report is based on our own intensive work and is genuine.

Project Group Number: CSERGC0547

DEEPANSHU PATHAK Registration Number: 11612065

SHUBHAM SHARMA Registration Number: 11616269
SAKET KUMAR SANKRIT Registration Number: 11601496
SAKET TRIPATHI Registration Number: 11614487

(Signature of Student 1)
Date:

(Signature of Student 2)
Date:

(Signature of Student 3)
Date:

(Signature of Student 4)
Date:

iii
 CERTIFICATE
This is to certify that the declaration statement made by this group of students
correct to the best of my knowledge and belief. They have completed this
Capstone Project under my guidance and supervision. The present work is the
result of their original investigation, effort and study. No part of the work has
ever been submitted for any other degree at any University. The Capstone
Project is fit for the submission and partial fulfillment of the conditions for
the award of B Tech Degree in Computer Science & Engineering from Lovely
Professional University, Phagwara.

Signature of the Mentor:

Name of the Mentor:

Date:

UID:

Designation:

School of Computer Science & Engineering,

Lovely Professional University, Phagwara,

Punjab,

INDIA.

iv
 ACKNOWLEDGEMENT
Completing a task is never a one-man effort. It is the results of valuable
contribution of several individuals in a direct or indirect manner that helps
you shape and achieve an objective. This project would not have taken shape,
without the guidance provided by Mr. Robin Prakash Mathur, our Mentor,
who helped us in our project and resolved all the technical problems and
helped in understanding technical aspects of the project. We profusely thank
her for the support provided to us.

v
 TABLE OF CONTENTS

1. Introduction 1

1.1 Overview 1

1.2 Project Description 1

2. Profile of the Problem 3

2.1 Aim 3

2.2 The Problem of the Statement 3

3. Existing System 4

3.1 Introduction 4

4. Problem Analysis 6

4.1 Product Definition 6

4.2 Feasibility Analysis 6

4.3 Project Plan 7

4.4 Development Schedule 9

vi
5. Software Requirement Analysis 10

5.1 Introduction 10

5.2 General Description 11

5.3 Specific Requirements 12

5.4 System Analysis 13

6. Design 13

6.1 System Design 15

6.2 Data Flow Diagram (DFD) 16

7. Testing 17

7.1 Functional Testing 17

7.2 Usability Testing 18

7.3 Interface Testing 19

7.4 Test Case 20

8. Implementation 21

8.1 Project Implementation 22

8.2 Post Implementation 23

8.3 Software Maintenance 24

vii
 9. Project Legacy 25

9.1 Current Status of the Project 25

9.2 Remaining Areas of Concern 26

9.3 Technical and Managerial Lessons Learned 26

10. User Manual 27

10.1 How to Use 27

11. System Snapshots 28

11.1 Landing Page 28

11.2 Form 29

11.3 Code Snippets 31

11.4 Algorithm Comparison 35

Bibliography 36

viii
 CHAPTER 1

INTRODUCTION
1.1 Overview

Nowadays data and machine learning are the hot topics, in every field data analysis
and machine learning are playing very important and crucial role, with the help of
machine learning algorithms we can predict what is going to happen in the future
basically what will be the possible outcomes if some particular are conditions
matched. In our project we have also used some machine learning algorithms to give
the results, we have developed a machine learning model which is unique one of its
kind, it predicts cumulative grade point average of the university students, it is
unique because the prediction is going to happen on the basis of some very
interesting features like whether student live inside hostel or not, eating in mess or
not, drink/smoke or not, in relationship or not, physical activity time investment.
We have trained our machine learning algorithm on the dataset prepared by us only,
we have collected the data from the existing university students of their last
completed semesters that is those semesters of which results are already there with
them, data is genuine and reliable then data is being normalized and standardized so
that machine learning algorithms can be trained effectively and efficiently. We have
trained the algorithms with high accuracy and we have got very good results, our
model is predicting the cgpa of the student with quite good accuracy, ultimately this
project is going to help the student to analyze there semester in advance so that they
can plan their semester in accordance to that. This project will help students of any
university to get better results.

1
1.2 Project Description

Our model is made like that it predicts the CGPA of the student for the upcoming
semester by taking the input of lifestyle details he/she is having. Machine learning
algorithms have been used, to train machine learning algorithms we have collected
and prepared the genuine data of more than 200 students from different college and
universities using google form. We have trained and tested around 7 algorithms for
our model and then we have selected the best suitable algorithm for the project, the
code is written in python we have used number of different machine learning
libraries for the same. Now to make project available to every individual we have
created a website for the same, so that any individual can visit the website and can
check the probable CGPA for the next semester, for that we have designed the
webpage of the website using HTML/CSS, for backend flask has been used.
Ultimate goal of this project is to make students aware about their possible upcoming
results with high rate of accuracy so that student can prepare and make changes in
his daily lifestyle to get better results in the exams.

Academic Performance Predictor

It Includes following tools:
• Anaconda/Jupyter Lab
• Microsoft Excel
• Flask
• Scikit Learn
• Microsoft Visual Code Studio

2
 CHAPTER 2

THE PROBLEM STATEMENT

2.1 Problem Description

While the time we are studying computer science and engineering in LPU, we have
always been interested in exploring how new things. And the solution to the real-
world challenges. We have analyzed that how tough it is for the students to find out
why their grades are low, even after studying for hours they are no getting the
wanted results because many times results depends on the daily life they are
following and it becomes necessary to make changes in their daily routine, on which
aspect of their routine they have to change to get better results.

2.2 Solution of The Problem

Now we have understood that sometimes getting cgpa becomes matter of daily
routine and human behavior, machine learning algorithms are very muh
capable of understanding the pattern between the human behavior and the results
attain by them. We have used machine learning algorithms to get a model which
can predict the cgpa of the upcoming semester on the basis of the daily routine the
student is following, we have trained the algorithm with the genuine data collected
from number of university going students of their past semesters so that model can
for a pattern between the daily habits of the student and the CGPA they have scored
under the influence of those habits, so then model can easily predict the CGPA.

3
 CHAPTER 3

EXISTING SYSTEMS

3.1 Introduction

Verto 2.0 Application (Available on Play Store)

This application is basically for the students of Lovely Professional University

where a student (Verto) can check their Attendance, Time Table, Result.

Predictor feature, here they can predict their TGPA or CGPA by changing their
grade to the grade they want and, on that basis, their updated CGPA and TGPA
will be shown.

Attendance feature will let you find that how many classes you’ve attended out of
how many classes. It will also let you decide how many more classes you can bunk
for specific attendance percentage.

Time Table feature shows about daily schedule of classes consisting of room no,
block no, course name.
Result section will let you know about your present grades of specific semester.
This application Verto app is basically works as a cgpa calculator by taking the
Actual marks as input and it also help the students to find out how much marks
They will have to score in end term to get the desired cgpa, on the other hand
Our Academic Performance Predictor will give the prediction of cgpa on the
Basis of the lifestyle a student is living it uses machine learning techniques to
Give the future cgpa of the student.

4
Here are some snapshots of described application-
Menu Page Attendance Page CGPA Predictor

Fig 3.1 Fig 3.2 Fig 3.3

3.2 What’s New in The System to Be Developed

Predicting score/grade by changing the desire grade of that specific course is simple
math for obtaining CGPA & TGPA but with help of Machine Learning we can take
predicting CGPA & TGPA to a whole new level where we are having features,
habits etc. We will feed the data to Machine Learning model and hence model can
be train with that set of data and can help in finding the future grade of different
student with his/her data. Some of the features are such as bad habits, physical
activity, club activity, self-study time etc. The already developed which are there
in the market are not based on machine learning they are basically developed and
programmed to calculate the cgpa by taking the marks of class test, mid terms
exams or any other marks if required whereas our model is completely different
thing it dose not calculate cgpa it basically predicts cgpa, which means we don’t
need your marks in class test or mid terms we simply ask you some daily lifestyle
questions and will give you the cgpa of that semester with high degree of accuracy

5
 CHAPTER 4

PROBLEM ANALYSIS

4.1 Product Definition:

Introduction:

We have made a website Academic Performance Predictor through which

user can predict the result of student based on previous performances and
activities of student.
In this project, we used big data tools to clean and structuralized the user input data
after that we also used machine learning algorithms to predict his/her result and we
can also evaluate the student performance and we can indicate him/her where they
are lacking , what are the problem they are facing and how can they overcome from
this problem means what are the various solutions by which they can choose the
right path.

4.2 Feasibility Analysis:

There are 4 steps to describe the feasibility analysis of our system:
1: Describe all possible problems:
In our project the problem is to make a system, Academic Performance Predictor
through with user can get all information relevant to student and they can predict the
student’s result of next year/term, based on the previous performances.

2: Describe Possible Solutions:

We get the steps of our solution of our problem:

6
 Data cleaning:
We have a large amount of data that is not in proper format.so that we are cleaning
the data and structured our data with the help of big data tools.

Apply machine learning algorithms:

After cleaning part, we classified our data and apply some algorithms in order to get
maximum accuracy in our system that also helped us to predict the actual result of
user input data.

Connect with Online platform:

After the work of backhand now we just added our system code to front view with
the help of python tools and some other online platforms

3. List Evaluation Criteria:

Now it’s time to evaluate our all solutions to get the best solution for our problem.
There are following parameters in our system to check whether the solution is
accurate and passing all the parameters.
1: accuracy of algorithms
2: efficiency of our system
3: time frame and cost

4. Propose the Most Feasible Solution:

Once the criteria of list evaluation have been done after that we will write the
solution and its attributes that is better than others. In our system we have gone with
solution which have more accurate answer for the CGPA predicting process and the
accuracy and efficiency will matter most in our solution.

7
5. Conclusion:

Now the whole description of our project is that we made a system that will predict
the student academic performance and we can get the point where they are lacking.
And based on result we can share a common feedback and report to students, and
they can improve their self in prospective area.in this website system we are
choosing solution that will produce more accurate and efficient result for input data.
And that has less space complexity and time complexity among all solutions.

4.3: Project plan:

According to problem description and its feasibility study report this project is

dividing into 4 major parts:

1: Data cleaning and structuralized the data: In this section we just cleaned and
managed the Input data and making it in structured format (normalized) in order to
get best accuracy.
Tools: R-studio, excel file format for store the data, google form to collect the
information.
2: Machine learning tools and algorithms: after cleaning the data we have applied
some machine learning algorithms on our dataset to predict the actual output of our
problem statement, machine learning basically understands the human behavior
and can predict the result on some provided inputs, this simple technique has been
used because life style of a student and marks scored by the student are very much
correlated there is a pattern between both of it has been proved by many different
studies, machine learning algorithms can detect that pattern easily.

8
Tools: We have done our coding part in python language, anaconda platform that
support all the libraries for python and machine learning, Jupyter notebook for
compilation and run the code.
3: Online website (platform): in other hand we are also design interface for user to
give input data and with the help of our backhand work (work no .1 & 2) we have
produced efficient output after various evaluation criteria as per for the particular
input data.
.

9
 CHAPTER 5
SOFTWARE REQUIREMENT ANALYSIS

5.1 Introduction

Requirement analysis is critical to the success of the system or any

application or any website. The requirements should be measurable,
documented, tested and should be related to needs of the all kinds of users
and should be satisfying design.
Any application developed will be having a vision and mission to serve the
user efficiently, effectively, properly, responsively and user friendly. So, for
the development of Packers and Transport Management System, first we
identified what users are expecting from us.

5.2 General Description

‘Academic Performance Predictor’ is website which helps to college student

to predict their upcoming academic score. That predictions depends upon
various factors. To predict upcoming academic score, we have used machine
learning algorithms. That algorithms works on every factor and gives
beneficial results to students that will helps them to grow in their future. We
have used regression techniques for predicting the cgpa of the student because
cgpa is something which is continuous it can have any value between 0 to 10
so regression is used for continuous values prediction, we have trained more
than seven regression algorithms and then we have chose the best performing
algorithm on our dataset that was SVM Regressor, so we have used SVM
regressor for our model. The dataset which we have used to train our
algorithms is also prepared by our team, the data is hundred percent genuine
it is of the student whose semester results was already out.

10
 5.3 Specific Requirements

5.3.1 TOOL DEVELOPMENT

5.3.1.1 Flask (Web framework)

Flask (web framework) Flask is a micro web framework written in Python. It is

classified as a microframework because it does not require particular tools or
libraries. It has no database abstraction layer, form validation, or any other
components where pre-existing third-party libraries provide common functions.

5.3.1.2 Anaconda (Python distribution)

Anaconda is a free and open-source distribution of the Python and R programming

languages for scientific computing (data science, machine learning applications,
large-scale data processing, predictive analytics, etc.), that aims to simplify package
management and deployment.

5.2.1 Website Development

5.2.1.1 HTML
Hypertext Markup Language is the standard markup language for documents
designed to be displayed in a web browser. It can be assisted by technologies such
as Cascading Style Sheets and scripting languages such as JavaScript.

5.2.1.2 CSS
Cascading Style Sheets is a style sheet language used for describing the
presentation of a document written in a markup language like HTML. CSS is a
cornerstone technology of the World Wide Web, alongside HTML and JavaScript.

11
5.3 System Analysis

System analysis is an important activity that takes place when new

information systems are being built and existing ones are changed.
Therefore, application development is central to the whole of the system
development. It includes gathering the necessary data and developing plans
for new system by-system analyst.
It includes two main facts:

i) Problem Identification

ii) User Requirement

5.3.1 Problem Identification

In problem identification, without any clear understanding of the problem in

the application, any further work done will lead to wastage of time and effort
at later stages. Hence various questions must be posed identifying the correct
problem like:
i) What is the actual problem?

ii) Causes of problem

iii) Is it important to solve the problem or not?

iv) Types of benefits expected after the problem is solved.

5.3.1 User Requirements

Shared, complete and accurate information requirements are essential.

Unfortunately, determining the information each user need is particularly a
different task in the system development.

12
 CHAPTER 6

DESIGN

Design is defined as a multistage process that primarily focuses on four

distinct features of a program:
1) Data structure

2) Software architecture

3) Interface representation,

4) Procedural detail.

The design process translates the requirements into a representation of the

software that can be accessed for quality product before code generation
begins. It is the most creative and challenging stage of the SDLC (Software
Development Life Cycle). It refers to the technical specifications that will be
applied to the system.
The selected architectural design defines all the components that needs to be
developed, communications with third party services, user flows and
database communications as well as front-end representations and behavior
of each components. The design is usually kept in the Design Specification
Document. In design phase basically what happened we had defined the blue
print of how we will going to do the coding part of the model which coding
languages will be used, which machine learning algorithms will be used
what will be the features for training the model and how it will done, the
flow charts were prepared pseudo code were written. Design made the
coding part easier as it become easy for the coders to decide what they have
to do in the coding part they just have to code in the required language .

13
6.1 System Design

Design goals basically describe the qualities of the system that developers
should keep in mind. Such goals are normally derived from the non-
functional requirements of the system. Design goals are grouped into five
categories. These are-

● User Interface

● Performance

● Dependability

● Maintenance

● End user criteria

● Usability

● Availability

● Safety

User Interface

The user interface will be a application program. This module will contain the
following:Multiple factors (set of questions for users) Result generator.

Performance

The system should have a fast response time (real time) with maximum
throughput. The admin has chosen fast response time over throughput and
hence the system should try to be more interactive. .
14
Dependability

The users need the system to be highly dependable as it is expected to be used

by end users who wants a better knowledge about information gathering.

Maintenance

The system should be easy to manage and easily modifiable to make

changes to the features and functionalities of the system. As it is an open
source code return in python
End User Criteria

End User criteria means to define a set of methods, techniques and tools that
help the nonprofessional software developers to create, modify or extend a
software artifact.

Usability

Usability is the extent to which a product can be used by specified users to

achieve goals with effectiveness, reliability, efficiency and satisfaction in a
specified context of use. From the end user’s perspective, the system is
designed in such a way that it is easy to learn and use, efficient and having
no to very few errors and bugs if any.

Availability

The services of the interface will be available as website it will be up and

running at every point. The rest full interfaces will also be available for
other developers.

15
Safety

The transmission of data is secured, as the security is provided by data servers.

The data will be stored in data storage and all the safety measures like data
encryption and data retrieval will be taken care.

Data Flow Diagram

Fig6.1

16
 CHAPTER 7

TESTING

Web Application testing is a critical section of software quality assurance and

represents the ultimate review of specification, design and coding.
Testing presents an intriguing perspective on a framework utilizing different
test information or cases. Planning of the test information assumes a
significant job in the framework testing. Subsequent to setting up the test
information, the framework under study is tried that test information.
Blunders and bugs are found and revised by utilizing the accompanying
testing steps and redresses are recorded for future references. In this way,
arrangement of testing is performed on the framework before it is prepared
for usage.

7.1. Functionality Testing

This is utilized to check if your item is according to the particulars you

proposed for it just as the utilitarian prerequisites you outlined for it in
your formative documentation. Testing Activities Included:
Test all parts in your website and model are working correctly and make
sure there are no broken links. Links to be checked will include –
• Home Page
• Model Page
• About Us Page
• Form in Model Page
• Google Form Page

17
7.1.1 Forms are working as expected. This will include-

• Scripting checks on the form are working as expected. For example- if a

user does not fill a mandatory field in a form no output is shown.
• Check default values are checked or not?

• When form is submitted with help of necessary radio buttons, system

will direct data to the backend for prediction.
• Forms are ideally organized for better meaningfulness.

7.1.2 Testing GUI

Testing whether-

• Fonts color should be readable.

• Readable font size.

• Navigation bar is clearly functionable.

7.2. Usability Testing:

Usability Testing has now become a vital part of any Machine Learning
related project. It can be carried out by testers like you or a small focus group
like the target audience to the user.

7.2.1 Test Navigation:

• Menus, fastens or Links to various Options on your device ought to be

effectively obvious and steady all through the apparatus

7.2.3 Test the Content:

• Content ought to be neat with no spelling or linguistic mistakes.

18
7.3 Interface Testing:

Three areas to be tested here are – ML Model, Web and Dataset

• Application: Test requests are sent correctly to the background code of

the tool and output at the client side is displayed correctly. Errors if any
must be caught by the application and must be only shown to the
administrator and not the end user.
7.5. Compatibility Testing

Similarity tests guarantees that your site shows effectively crosswise over
various stages. This would include-
Browser Compatibility Test: Same website is accessible from different
browsers.
You need to test if your web pages are being displayed correctly across
browsers.

Make sure your website fine for various combination of Operating systems
such as Windows, Linux, Mac and website is compatible with Browsers such
as Firefox, Internet Explorer, Safari etc.

7.6. Performance Testing:

This will ensure your site works under all loads. Testing activities will
include but not limited to –

• Form response times at different connection speeds.

• Load test your website to determine its behavior under normal and peak
loads on the targeted data.
• Stress test your site to decide its break moment that pushed to past
19
 typical burdens at top time.
• Test if an accident happens because of pinnacle load, how does the
instrument recoup from such an occasion

Project Name: Academic Performance Predictor

Test Cases

Test Priority (Low/Medium/High): Med

Module Name: Form Module
Test Title: Providing input via form in module page
Description: Prediction of CGPA as output

Pre-conditions: User should have valid personal details

Dependencies:

S. Test Steps Test Data Expected Result Actual Result Status(Pass/Fail)

N
o
Selecting details in Form I/P as CGPA(Integer) 7.33 Pass
1 Integer
Selecting no values(default) I/P as Error Message NULL Pass
2 NULL
Selecting last value of I/P as CGPA(Integer) 8.11 Pass
3 features Integer

20
 CHAPTER 8

IMPLEMENTATION

8.1 Project Implementation

Procedure whereby "Project inputs are changed over to extend outputs". May
be looked at as:
1. Placing the action, the activities of the project.

2. Putting into practice what was proposed in the project

document (i.e. changing the project proposal into the actual
project).
3. Managing of the project or executing the project outcomes.
Project Implementation phase involves:
• Project Activation
• Project Operations

This implies choosing to have the venture begun. Along these lines, every
one of the necessities required for making the product valuable and address
every one of the issues of the customer was accumulated by having sittings with
the customer. The achievability study was made about the undertaking that
whether the venture requested by the customer was conceivable to actualize or
not. At that point, likewise the language or the stage that was required for the
advancement of the task was chosen with the goal that the most basic, secure,
hearty stage ought to be picked so all that what was required by the customer
can be actualized by the designers effectively.

21
 8.1.1 Project Activation

Project enactment basically implies intending to have the undertaking begun. It is

the fundamental or establishment phase of task execution and it includes the
coordination and designation of assets (reserves, materials) to make the project
operational.
The following activities are undertaken during project activation:

1) Selection of technology and language to use to make project functional.

2) Division of work among group members according to their area of interest.
3) Preparation of possibility plans.
4) Executing of the plan -

This is down to earth the board of a task. Here, project inputs are changed into
yields to accomplish quick destinations. In this the fundamental necessities
being taken from the customer, at that point those were executed into the
practical structure by giving the contribution to the framework and getting the
ideal yield from the framework. Fundamentally, the Black Box strategy was
received for the customer, as the customer being a layman thinking nothing
about the coding and the strategies embraced for getting the yield from the
framework, so the yield that was wanted by the customer was going.

22
8.1.2 Project Operation

Project implementation plan includes:

i) What is the arrangement of these activities?

In what manner or rather what sequence should be followed in
obtaining the desired requirements of the project.
ii) What is the timeline for these activities?
Each activity for the completion of the whole system need to be
completed within the required time frame so that in the required
time frame the system can be Functionally developed.
iii) What is the time limitation and remedial action?
Time taken to execute project exercises is one proportion of
effective ness of supervision or observing of task usage. In this
way, the correct consideration was taken so the task was created
on schedule and the report for the capstone was created and
arranged inside the given time period, with the goal that we could
demonstrate ourselves to be productive and savvy as designer.
iv) What is the Project activity and Time listing?
It records the planning of the considerable number of exercises
of a venture. It shows the how much time is required by every
movement/module in application. Along these lines, a movement
ought not take a lot of time which can influence the exhibition of
the entire venture.

23
 8.2 Post Implementation

• After the framework is executed and, a survey ought to be led to decide if

the framework is meeting the necessary desires and where enhancements
are required.
• A post execution survey quantifies the frameworks execution against pre-
characterized prerequisites of the framework. It decides how well the
framework will keeps on attempting to meet execution particulars.
• A post – Implementation survey is likewise characterized as an assessment
of a framework regarding the degree to which the framework achieves
expressed destinations and jobs and genuine venture costs surpass starting
gauges. It is typically a survey of serious issues that necessities
transformation and those that surfaced during the usage stage.
The post usage study starts with the survey group, which gathers and audits
demands for assessment. Startling change in the framework that influences
the client or framework execution is a worry factor that prompts framework
survey. When solicitation is documented, the client is approached to check
how well the framework is working to particulars or how well the deliberate
advantages have been figured it out. Recommendations with respect to
enhancements and changes are additionally invited.

24
 CHAPTER 9

PROJECT LEGACY

9.1 Current Status of the Project

1) We developed interfaces for ‘Academic Performance Predictor’ using python

Flask framework.
2) We train our machine with various machine learning algorithms.
3)User can use our project with the help of many compatible browsers.

9.2 Remaining Areas of Concern

1)We are working on a secure application for the user.

9.3 Technical and Managerial Lessons Learned

9.3.1 Technical
1)Machine learning algorithms
2)Flask
3) CSS
4) Excel data cleaning

9.3.2. Managerial
i) Learned Team Management.

ii) Learned how to deal with different situations when stuck.

iii) Learned how to work in group of developers.

25
iv) Learned to distribute tasks to members.

v)Learned about the leadership quality.

vi)Learned about the software skills that must be present in an Engineer.

vii)Learned about managing Time.

26
 CHAPTER 10

USER MANUAL
10.1 How to use

10.1.1. For User

1. User open the Page of the website.
2. User will find the opening page as introduction of Academic
Performance Predictor (Project Name).
3. User will access the website by the Main Menu.
4. User will predict his/her CGPA by filling up the form necessary for
the model in the “Model” page.
5. User will find information about the developers of this project by
navigating to the “About Us” page.

27
 CHAPTER 11

SYSTEM SNAPSHOTS

11.1 Landing Page (Home Page)

Fig 11.1

28
11.2 Form

Fig 11.2

29
11.3 Form

Fig 11.3

30
11.4 Code Snippet

Fig 11.4

31
11.5 Code Snippet

Fig 11.5

Fig11.6 9
32
11.6 Code Snippet

Fig 11.7
33
11.7 Code Snippet

Fig 11.8

34
 COMPARISON OF ALGORITHMS

Name of Mean R2 Score Mean Algorithm

... S.No Algorithm Absolute Squared Accuracy
Error Error Rank

1 Linear Regression 0.764 0.132 1.214 3

2 SVR (Without 0.706 0.195 1.126 2

GridSearchCV)

3 SVR (With 0.663 0.317 0.954 1

GridSearchCV)

4 RANSAC 0.767 -0.08 1.513 6

Regressor

5 AdaBoost 0.882 0.020 1.371 4

Regressor

6 ElasticNet 0.826 -0.001 1.401 5

RESULT: SVR with GridSerachCV gives the best results on our dataset.

35
 BIBLIOGRAPHY

1. https://scikit-learn.org/stable/
2. https://www.python.org/
3. https://www.palletsprojects.com/p/flask/
4. https://towardsdatascience.com/

5. Python Machine Learning By Sebastian Raschka

36
37
38
39
40
6.

41
42
41