ACADEMIC PERFORMANCE PREDICTION OF UNIVERSITY STUDENTS USING MACHINE LEARNING TECHNIQUES

Project Title

ACADEMIC PERFORMANCE PREDICTION OF UNIVERSITY STUDENTS USING MACHINE LEARNING TECHNIQUES

Project Area of Specialization Artificial IntelligenceProject Summary

Students’ academic performance plays a vital role in higher education and their careers. Many high-quality educational institutes have a record of their student’s performances and it is used to identify students who are at risk of poor performance and in investigating the instructor. Students should be advised well in advance to concentrate their efforts in a specific area in order to improve their academic achievement. This type of analysis assists an institute in lowering its failure rates.

Data science has been adopted by many researchers to solve real world problem in telecommunication, marketing, health care, medical, industrial and customer relationships. Nowadays evaluating the student performance of any organization is going to play a vital role to train the students. Education data mining refers to the collection of data mining applications in the field of education. These applications are concerned with the analysis of data from students and teachers. The analysis might be used for categorization or prediction.

Student attrition is a widespread phenomenon and it is caused due to lack of academic preparation, motivational commitment to the institution, psychosocial and financial problems. Students enrolled in a program sometimes lose hope regarding their fields and it is caused due a lot of reasons. This problem can be fixed using data analysis and machine learning.

• A questionnaire will be used to take data from multiple students.
• The data will be stored in a data-set.
• Data cleaning will be used to remove missing or incomplete data resulting in cleaned data.
• The cleaned data will then be normalized where numerical values like GPA will be transformed into categorial classes.
• In order to classify every item in a dataset, classification process is used.
• The data of male and female students will be separated.
• The final student’s academic performance dataset will be split into different parts having 4:1 ratio. 80% will be used to train the dataset and 20% will be used to test it.
• Pattern extraction, model training and evaluation process will be done on the training dataset using classification techniques.
• The testing dataset will be used to test the program and results acquired from this stage will be then evaluated and depicted as knowledge.

LIMITATION:

• Our project will work with the limited amount of dataset.
• The available dataset will be managed in csv format mostly.
• Project is built on open-source so code is public.
• At a time, project will work for single organization.

This project will help collecting data of students and give us complete information of students on a single platform.

Once an institute find outs the performance of students, they will be able to improve their specific students who become reason in increasing failures rate.

This project can be applied worldwide and a machine learning model that can be accessed by anyone with basic knowledge of computers.

Project Objectives

OBJECTIVES:

• To collect students’ data using a questionnaire and store them in a well-managed dataset.
• To apply different data model algorithms on the dataset and analyze the best one.
• To train the model using machine learning which would be capable of predicting future performance of students.
• To develop a user interface that predicts future performance of a student based on given input by the user.
• To suggest users what steps they can take to improve their future academic performance in case of a poor academic record.

Project Implementation Method

PROPOSED METHODOLOGY:

• A questionnaire will be used to take data from multiple students.
• The data will be stored in a data-set.
• Data cleaning will be used to remove missing or incomplete data resulting in cleaned data.
• The cleaned data will then be normalized where numerical values like GPA will be transformed into categorial classes.
• In order to classify every item in a dataset, classification process is used.
• The data of male and female students will be separated.
• The final student’s academic performance dataset will be split into different parts having 4:1 ratio. 80% will be used to train the dataset and 20% will be used to test it.
• Pattern extraction, model training and evaluation process will be done on the training dataset using classification techniques.
• The testing dataset will be used to test the program and results acquired from this stage will be then evaluated and depicted as knowledge.

Benefits of the Project

• This will hopefully have an impact on students’ future performance as they will know what pattern is leading them where and how they can improve their performance.
• This project will help collecting data of students and give us complete information of students on a single platform.
• Once an institute find outs the performance of students, they will be able to improve their specific students who become reason in increasing failures rate.
• This project can be applied worldwide and a machine learning model that can be accessed by anyone with basic knowledge of computers.

Technical Details of Final Deliverable

Students visit the web based application to identify their existence with all other students who have similar abilities and their performance. Due to research we are at the end where we can clearly inform which type of student perform good and which perform low.

Student will interact with user friendly UI and fill the required field and identify their performance.

This infromation also helps institute to warn students who have low performance or those students who are elgible for scholarship according to their family conditon or their performance.

This project uses technologies

• Python programming language, to write the code.
• Jupyter notebook IDE, to clean the data and exploratory data analysis, data visualization, machine learning.
• Flask, it is lightweight Python web framework.
• Firebase Realtime Database, is a cloud-hosted NoSQL database that lets you store and sync data between your users in realtime.
• Heroku, is a container-based cloud Platform as a Service (PaaS). use Heroku to deploy, manage, and scale App.

Final Deliverable of the Project Software SystemCore Industry EducationOther IndustriesCore Technology Artificial Intelligence(AI)Other TechnologiesSustainable Development Goals Quality EducationRequired Resources

Item Name	Type	No. of Units	Per Unit Cost (in Rs)	Total (in Rs)
			Total in (Rs)	17500
SSD	Equipment	1	11000	11000
Form Photocopies	Miscellaneous	500	3	1500
RAM	Equipment	2	2500	5000