2. Developed by
Name Koustav Mandal
Stream InformationTechnology
SectionA
Roll no. 18
Enrolment no. 12021002004036
Name Biswajit Adhikary
Stream InformationTechnology
SectionA
Roll no. 14
Enrolment no. 12021002004030
Name Rebanta Bera
Stream InformationTechnology
SectionA
Roll no. 22
Enrolment no. 12021002004045
Name Saurav Kumar Nayak
Stream InformationTechnology
SectionA
Roll no. 16
Enrolment no. 12021002004034
3. Project Description
As we all know that in today’s generation humans are too much busy and they want that their work
should be done in an easy way and with less amount of time taken.
Just like teachers had a lot of burden on their head and a lot of responsibilities as well. So, to make
their stress level low and increase their productivity, we have come with an idea of making a site
where they can store all the data required for giving result of a particular student.
The best part of this site is that, it can predict the final result and the site can be used both
by student and teacher where students can also fill their marks and predict what can be their
grade for current semester.
This project basically aims to create a transparent environment between teachers and students in
which a student can see their future result and teachers can easily make their result by predicting it.
4. Machine Learning Model Used :
Here in this project we will use linear regression model. Using a linear regression model for a student
grade prediction system is a straightforward and effective approach. Linear regression is a supervised
learning algorithm used to model the relationship between a dependent variable (in this case, student
grades) and one or more independent variables (features or attributes related to students). Here’s we
can go about using a linear regression model for our student grade prediction system:
1.Data Collection and Preprocessing:
1. Collect relevant data: We shall gather data about students, including features such as study
hours, attendance, previous grades, extracurricular activities, etc., and their corresponding
grades.
2. Preprocess the data: We shall clean the data, handle missing values, and ensure that the data
is in a suitable format for analysis.
2.Feature Selection:
Choose features: We shall select the features that we believe have a direct or indirect impact on a
student's grades. This selection will be based on domain knowledge and data analysis.
3.Data Splitting:
Split the data: We shall divide our dataset into two parts: a training set and a testing set. The
training set is used to train the model, and the testing set is used to evaluate its performance.
5. 4.Model Training:
1. Choose a library: We can use popular machine learning libraries like scikit-learn in Python to implement
linear
regression.
2. Train the model: We shall use the training data to fit the linear regression model. The model will learn the
relationship between the selected features and the student grades.
5.Model Evaluation:
Evaluate the model: We shall use the testing set to evaluate the model's performance. Common evaluation
metrics for regression tasks include Mean Absolute Error (MAE), Mean Squared Error (MSE), and Root
Mean Squared Error (RMSE).
6.Model Interpretation:
Interpret coefficients: In a linear regression model, the coefficients assigned to each feature indicate the
strength and direction of the relationship between the feature and the target variable (grades). Positive
coefficients indicate a positive relationship, while negative coefficients indicate a negative relationship.
7.Making Predictions:
Predict grades: Once the model is trained and evaluated, we can use it to make predictions for new, unseen
data. We should provide the relevant features of a student, and the model will predict their likely grade.
8.Deployment:
Deploy the model: Once we are satisfied with the model's performance, we can deploy it to a production
environment where it can be used to predict student grades.
We should remember that while linear regression is a simple and interpretable model, it assumes a linear
relationship between features and the target variable. If the relationship is more complex, we might need to
6. SDLC Model used in this project
For a project like student grade prediction using machine learning, we can consider using the "Agile" software
development lifecycle (SDLC) model. The Agile model is well-suited for projects that involve iterative and
incremental development, which aligns with the nature of machine learning projects where experimentation, model
tuning, and continuous improvement are common. Regular user testing and incremental updates can help ensure
that the website meets the needs of both students and professors.
• Initiate the project by clearly outlining the goals, objectives, and requirements of the student grade prediction
system. We also try to understand who will be using the system, such as teachers, students, administrators, etc.
• Divide the project into smaller iterations or sprints, each focusing on a specific aspect of the grade prediction
system.
• Decide which features or functionalities are most critical and should be implemented in the early iterations.
• Develop the selected features and functionalities using machine learning techniques. This might involve data
preprocessing, model training, and evaluation.
• At the end of each iteration, we shall showcase the developed features to stakeholders for feedback and
validation.
• Continuously integrate new features and improvements into the system and deploy them in a controlled
environment. In subsequent iterations, we shall build upon the existing features, add new ones, and refine the
models based on performance feedback.
• As the system becomes more robust and accurate, we shall release it to a wider audience, starting with a
smaller user group.
Using the Agile model allows us to adapt to changes and incorporate feedback as the project progresses. Since
machine learning projects often involve experimentation and continuous improvement, Agile's iterative approach
aligns well with these characteristics.
7. SWOT Analysis
STRENGTH WEAKNESS THREAT OPPORTUNITY
Grade accurately reflects
personal work.
Difficult to administer. Inconsistent grading process
because of different
evaluators.
Early warning system for
potential problem.
Direct observation of
individual effort.
Excessive time requirements. Student perception unfair. Increased emphasis on
teaching skills.
Simplify grading. Eliminates other course
content.
Weaker group skills. Motivate individuals to
contribute
8. Feasibility Study
Technical Feasibility:
The technical components we have in our project
User friendly Interface for both student and Teacher
where both can log in using their log in id and password.
Machine learning algorithm for predicting the grade,
here we use linear regression algorithm.
Teacher can edit the grade of any particular student.
The technologies that would be used in this project are as follows:
React(.js) — a client-side JavaScript framework.
Linear regression algorithm — a supervised machine learning
algorithm
Python
9. Feasibility Study(contd.)
Development costs
Hiring web developers, ML
developers, designers, and
accessibility experts.
Testing for usability, functionality,
and security.
Maintenance cost
• Security Updates and monitoring to
protect user data
• Bug fixes and technical support for users
• Updates to improve user experience.
10. DecisionTree
User Login
Are
credential
correct?
Yes No
View: Can view their obtained
grade for any semester.
Predict: Can predict their grade for
current semester.
View: Can view the grade of any student
Predict: Can predict the grade of any particular student for current
semester.
Edit:Can edit the grade of any particular student , after consulting with
HOD.
Whether
student
or
teacher
Student Teacher
Error