Lahore Garrison University

Department of Software Engineering

Course: Software Requirement Engineering

Exam: Project # 01
Date: Wednesday, December 20, 2023 Semester: Fall-2023
Maximum Marks: 100 Due Date: 32-01-2024

1
ROBLEM STATEMENT

It is extremely critical to select a field to pursue a degree in nowadays, students do not spend time
thinking about this decision or they are not properly guided in doing so. They end up choosing fields
of their parent’s liking or fields that their friends are pursuing. A well-designed prediction system
can identify struggling students early on, allowing educators to provide timely support and
interventions to help them improve their performance. For some students, knowing that their
performance is being monitored and predicted could motivate them to put in more effort and strive
for better grades .This Web-Application software would guide and take students’ academic, personal
background, and personality traits into consideration and recommend a suitable degree or career path
to pursue, will recommend courses they should undertake in a degree of their choosing and also
predict future academic progression (GPA).

EXECUTIVE SUMMARY

Students do not think long and hard about career choices and choose fields carelessly due to a lack
of career counselling. In this project, we will develop a web application that will take into account
students’ academic background (Matric/O-level to Intermediate/A-level), personality traits, and
personal background, ask questions about different career paths and how they feel about them to
recommend a degree program they can pursue, will then give a choice to select a degree to further
explore if they want. It will recommend courses to undertake by asking a few more questions related
to the degree and also help undergraduate students of computer science and engineering to make
career choices based on the collected data. Data will be collected mainly on google forms and trained
using algorithms such as Random forest, decision tree, SVM, K-nearest neighbor, and Naïve Bayes.
75 percent will be used to train and the rest will be used for testing. Web app will ask information
using a chatbot and it will store this data in a MySQL database.

2
INTRODUCTION

Nowadays AI is going on a rapid streak to achieve the future meaning that humans are making
programs and algorithms that are more efficient and precise as compared to the programs that were
built before. So, it is beneficial for programmers to integrate AI and Machine Learning into everyday
programs to make human life easy. As our project Degree Aptitude System suggests easy ways for
students to choose the best degree for them as well as get a prediction of future Academics.
It is very crucial to choose which career to pursue. In academics choosing the right career is the first
step to success. To pursue a career a person needs guidance which can be done by various methods.
These methods mostly include career counseling and degree aptitude test. Career counseling helps
you understand yourself, your career, and your educational path to help you to decide in making a
life-changing decision.
A degree Aptitude test helps you understand where the student stands academically. A well-designed
prediction system can identify struggling students early on, allowing educators to provide timely
support and interventions to help students improve their performance. Students can allocate
resources more effectively by identifying the subjects they need extra help in, thereby optimizing
the use of time and resources. For some students, knowing that their performance is being monitored
and predicted could motivate them to put in more effort and strive for better grades. The system is
also cost-effective as the students would not have to travel to various places for guidance, they just
have to interact with the website and can know where they stand mentally and academically as well
as get guidance from the website.
The website is easy to use as it has a user-friendly interface. It consists of a login/signup page to
keep students data safe and secure so that no other entity can access the account. Then students
choose what they want to do whether get suggestions about universities which will be done with
interaction with a chatbot that will chat with them and provide basic information about universities
or get predictions of future academics. Both of these would be done by only providing academic
background and completing questionnaires about aptitude tests and personality tests.
The website uses machine-learning models to predict the future academics of students by analysing
the data provided by students. Such that students easily understand the results, so they can analyse
their academic standing and give suggestions for improvements. The analysis done by the model is
not meant for students to feel demotivated but rather provides a means that they need improvements
in desired subjects, such that they get good grades. By analysing individual students' data, the
website can provide insights into their strengths and weaknesses. Educators can then improve their
teaching methods to suit each student's learning style and needs, resulting in a more effective and
engaging learning experience. Knowing that the performance/academics of students are being
monitored and predicted the students can feel more motivated to achieve their goals. The insights
given by the website can provide information about the subjects the students are weak in and suggest
ways to improve them by providing resources. By understanding the academic trajectory, students
can make more informed decisions about their future career paths and the educational requirements
they need to meet for achieving their goals.
Students provide their academic background and attempt the questionnaires provided by the website
such as aptitude tests, and personality tests to know the standing of students. Information provided
by the website can provide better communication between parents and students, enabling parents to
support their children's interest/education more effectively, which will be beneficial for students to
make more informed decisions about their field of study and career.

3
Not only students but educators and institutes can also use this website to improve teaching
techniques as they will know the teaching trends.

COMPETITORS/COMPETITIVE ANALYSIS
• [Link] :

Offers the user 15 sets of pictures of four different career activities. You choose the one you like the
best by check-marking it and the least appealing one by crossing it. No academic background of
students is taken into consideration. No personal background or aptitude test of students is taken
into consideration or inquired about. Whereas our web app will inquire about both these things.
Based on Holland code.

• [Link] :

Users are given 7 skills(Artistic, interpersonal, mathematics, mechanical, communication, science,

managerial) and by a slider, they are to select their expertise in said skill from low to high. They can
filter salary expectations and also expected growth in the field they might pursue. Does not take into
consideration personality traits. Academic history of students is not inquired. Aptitude test of
students is taken into consideration either.

• [Link] :

This test consists of 48 questions that you will have to rate based on how much you would
enjoy performing a task each on a scale of (1) strongly dislike to (5) strongly enjoy. This
test will help you choose a career path based on your personality type and your interests in
a particular field or subject. Suggested fields are too many and not narrowed down enough.
Does not ask previous marks, aptitude or personal background. These fields are covered b y
our web app which includes a chatbot.

• [Link] :

Hosts 4 types of tests for career prediction Holland code-based, Myles and Briggs theory,
Scientifically validated, and Pictures based. In Holland and Myles’s theory, users have phrases, from
each pair, they choose the one that describes them the best. In a scientifically validated test, an
activity is described, and users have to choose how much they like or dislike the activity. The
pictures-based test provides a pair of photos from which users select the one most appealing to them.
Does not evaluate aptitude or personal background.

4
OBJECTIVES

• Provide Personalized Recommendations: The project aims to generate personalized degree

recommendations for students based on their interests, skills, and career aspirations. The
objective is to help students make informed decisions about their academic paths and select
degree programs that align with their strengths and passions also.
• Enhance Academic Success: The project's objective is to enhance students' academic success
by providing personalized degree recommendations. By guiding students toward degree
programs that align with their abilities and interests, the project aims to increase their
motivation and engagement in their chosen fields of study.
• Reducing Information Gap: The project seeks to reduce the information gap by collecting
and analyzing relevant data about academic programs, career outcomes, and industry trends.
The objective is to provide students with accurate, up-to-date information that helps them
understand the potential career paths and opportunities associated with different
degree programs.
• Improve User Experience: The objective is to enhance user satisfaction by developing a user-
friendly and intuitive interface for students to interact with the degree aptitude software. The
project aims to create a positive user experience by simplifying the process of inputting
preferences, accessing degree recommendations, and navigating through the system,
ensuring ease of use and smooth interactions.

5
MOTIVATION

Career counselling is the best choice in selecting the major of your career. Many university services
are significant help in choosing the right path for a student’s career. Revealing the crucial factors
that influence students’ career planning poses a significant challenge. Psychologically speaking,
adopting a cognitive approach that involves gathering, analyzing, and assessing pertinent personal
information is essential for effective career counseling services. However, given the intricacy of
individual aspirations and visions, students often struggle to determine their post-graduation paths
with absolute clarity. The ultimate goal is to guide students toward degree programs that align with
their strengths and passions, opening doors to diverse career opportunities and ensuring long-term
success and fulfillment.

FEATURES OF PROJECT

• Chat bot-based interaction:

An AI chatbot will be used to ask questions about personality traits, and career preferences
to determine interests and also take info from students and give data to algorithms for making
predictions.
• Degree Prediction:
Will predict degree programs for students based on their academic backgrounds and the
careers they wish to pursue in the future.
• Courses recommendation:
After it has predicted degrees to the user, they will be given the choice to select a degree for
further exploration. It will ask for additional information related to the selected degree and
recommend courses they should choose if they decide to pursue the said degree.
• Career Prediction:
Suggesting careers and also jobs that are suitable for engineering and computer science
students.
• Assessing personality traits:
Tests that will help determine a student’s personality to choose a suitable career accordingly.
• Projecting grades:
Data collected will be trained using machine learning algorithms to project students’ GPAs.
• Institutes providing recommended degrees:
Nationwide educational institutes will be listed which provide degrees that the students
are recommended.

6
Research Papers
The Grey Prediction Algorithm can work with small or incomplete datasets, making it useful in
situations where historical data are scarce. The algorithm is relatively easy to understand and
implement, especially compared to more complex forecasting methods like neural networks or time
series models. Assumptions about system behavior: The Grey Prediction Algorithm assumes that the
development of a system can be described by a first-order ordinary differential equation, which may
not hold for all types of systems. It may not be suitable for complex systems with nonlinear behavior.
The algorithm is primarily designed for short-term or mid-term forecasting. It may not perform as
well when trying to predict far into the future due to the inherent limitations of the Grey Differential
Equation model.
Accuracy: Supervised regression models have the potential to achieve high prediction accuracy,
especially when trained on a large and representative dataset. They can capture complex
relationships between input features and GPA, allowing for accurate predictions. Flexibility: These
methods can handle a wide range of input features, including both numerical and categorical
variables. This flexibility enables the inclusion of diverse information such as previous academic
performance, standardized test scores, extracurricular activities, and personal attributes. The
accuracy and reliability of the predictions heavily rely on the quality and representativeness of the
training data. Biases, outliers, or missing data in the training set can negatively impact the model's
performance and generalize poorly to new, unseen data. More complex regression models, such as
ensemble methods or neural networks, may provide improved prediction accuracy but can be more
challenging to interpret. Balancing model complexity with interpretability is a trade-off that needs
to be considered.
The paper presents a clear objective of improving the accuracy of predicting academic performance
by focusing on oversampling small-sized educational datasets. This provides a specific and practical
application for the proposed methodology. The authors combine multiple techniques, including
Tabular Conditional GAN, NCC, and the SMOTE-IPF data balancing algorithm, to address the
challenge of small dataset size. This comprehensive approach enhances the potential for accurate
predictions. The current approach is configured specifically for binary problem-based datasets,
which may limit its applicability to other types of datasets or multi-label classification problems.
This restricts the generalizability of the findings. The paragraph mentions a 2% improvement in
accuracy compared to previous approaches. However, it does not provide additional comparison
metrics or statistical significance tests to support the significance of this improvement.
The study focuses on the predictive validity of the Differential Aptitude Test (DAT) in the context
of psychology study programs. This has practical implications for using aptitude tests to predict
academic success in specific fields of study. The study uses both test techniques and documentation
techniques for data collection, which adds to the rigor of the research. Additionally, the data analysis
employs multiple linear regression analysis, a commonly used statistical technique for examining
relationships between variables. The study focuses specifically on psychology study programs and
may not be directly applicable to other fields of study. The findings may not generalize to different
academic disciplines or contexts. The study uses GPA as the dependent variable for academic

7
success. While GPA is a commonly used measure, it may have limitations in capturing the full range
of academic performance or success in the psychology study program.
This study investigated the effects of past computer experience on the computerized aptitude test
performance of college students. The subjects were 92 undergraduates. All subjects completed an
arithmetic reasoning test by computer and a Computer Experience Questionnaire. Past computer
experience did account for a significant amount of variance on the arithmetic reasoning test.
Suggestions regarding the future use of computerized aptitude tests are given. The study focuses on
the issue of high attrition rates in the first study year of computing studies. This highlights the
importance of identifying methods to determine in advance which applicants are at risk of not
completing the first year, to reduce dropout rates. The study proposes an action plan based on the
exploration of students and the evaluation of existing activities conducted at the Faculty of
Computing. This demonstrates a proactive approach toward addressing the attrition issue by
leveraging initiatives such as the School of Young Programmers, Compensative Courses in High
School Mathematics, and Mentoring programs. The study focuses specifically on the attrition trend
in the Faculty of Computing at the University of Latvia. The findings and proposed solutions may
have limited generalizability to other institutions or fields of study, as the challenges and factors
influencing attrition can vary.

ARCHITECTURAL DESIGN (Redesign)

8
• Admin : Administrator of the website and data-base, who oversees the website and removes
bugs.
• Student: User’s of the website.
• Login / Sign-up: A login and sing-up page for student to keep student’s record and keep
students data secure and authenticated.
• Website: A collection of interconnected web pages hosted on a web server and accessible via
the internet. A digital space that provides information, services, or resources to users.
• Interface: Visual and Interactive elements that users interact with when they visit the website.
Consist of the design, layout, and user experience (UX) aspects of the website.
• Degree Predictor: A module to search for the best degree for students, by using the
information given by students in the questionnaire.
• CGPA Projection: A module to project the future academics of students, by using information
given by students in the academic background.
• Questionaire: A set of questions that will be provided on the website filled by students.
• Personal Background: A set of questions related to the personal background of students such
as interests, family background, Language and Linguistic Background, etc.
• Academic Background: A set of questions related to the academics of students such as marks
in various subjects, Academics achievement, Subjects the student likes, etc.
• Aptitude Test: A set of test questions for students such as Numerical reasoning, Logical
reasoning, Spatial reasoning, Abstract reasoning, Memory skills, etc.
• Server: It plays a crucial role in functioning and delivering the website such as Web Hosting,
Data Storage, and Database Management.
• Framework: "server-side framework" that provides a structured environment and set of tools
for developing web applications, and services.
• Authentication: The process of verifying the identity of a user who wants to access a
resource.
• Machine Learning: A branch of artificial intelligence (AI) that focuses on the use of data and
algorithms, which are used to predict future academics.
• API: APIs of universities to get desired information about the university like courses they
offer and the specialty of the university.
• Recommendation Engine: Techniques that provide personalized suggestions to users based
on their preferences, behaviors, and interactions with the platform.
• Database: Used to store track records of students their academics, and personalized
backgrounds.

9
IMPLEMENTATION TOOLS AND TECHNIQUES

Please mention these tools and techniques with logical and realistic reasoning considering your company's
strength.

These are some important questions you should consider during the software requirements
gathering and user agreement process. Answer these questions in your document.
▪ How to initiate the software requirement gathering process.
▪ How do you write a software development agreement?
▪ How would you schedule meetings and ensure that all relevant stakeholders are included despite
time zone differences?
▪ You are working on a project with a tight deadline, and a stakeholder requests additional features
that were not initially planned. How do you handle this situation?
▪ During the requirements gathering phase, you discover that some stakeholders are not actively
participating. How do you encourage their engagement?
▪ During a requirements gathering session, you encounter conflicting requirements from two important
stakeholders. How would you resolve this conflict?
▪ What techniques or tools would you use to document and communicate the inter-module data flow?
▪ How might changes in data sharing requirements impact other aspects of the project, such as security
or performance?
▪ What tools and technologies would you use to facilitate communication with a geographically
distributed team?
▪ How do you manage potential delays or misunderstandings caused by asynchronous
communication?

10
▪ How would you assess the impact of this late requirement on the project timeline and budget?
▪ What steps would you take to communicate this change to the development team and other
stakeholders?
▪ How might this impact existing features and functionalities?
▪ How would you prioritize resolving conflicts in data formats and communication protocols?
▪ How might changes in one external system's requirements impact the integration with others?
▪ How would you identify and document the specific compliance requirements relevant to the project?
▪ What measures would you implement to ensure continuous compliance throughout the development
lifecycle?
▪ How might changes in compliance standards affect the existing requirements and development
process?
▪ What tools or methodologies would you use to facilitate communication and collaboration between
development and operations teams?
▪ How do you ensure that the operations team is involved early enough in the requirements phase to
plan for infrastructure changes?
▪ What strategies would you employ to minimize downtime during the transition and ensure system
stability?

11