A Students Project MCA Department

Academic Performance Analytics Using ML

Project Team

• Mr. Ratikant Ashok Patil

• Mr. Ruturaj Shahu Deshmukh

• Mr. Ganesh Baburao Shirtode

Project Guide: Ms. N. B. Gajul
Department: Master of Computer Applications (MCA)
Institution: SVERI’s College of Engineering, Pandharpur

Introduction

Educational institutions generate academic data through examinations, attendance, assignments and learning management systems. Analysing this information can help educators understand academic performance and identify important trends.

The project “Academic Performance Analytics Using ML” explores how Big Data Analytics and machine-learning concepts can be applied to academic data. It focuses on the process of collecting, cleaning, storing, processing and analysing data to generate meaningful insights.

Project Objectives

• To understand the concepts and characteristics of Big Data.

• To explore academic data from attendance, assessments and other educational sources.

• To understand distributed storage and processing using Hadoop and Apache Spark.

• To explore data cleaning, transformation and analytical techniques.

• To present analytical information through charts, dashboards and reports.

Technologies Used

The project report discusses Hadoop, HDFS, MapReduce and Apache Spark for Big Data storage and processing concepts. It also explores machine-learning techniques and visualization methods for understanding academic performance.

Flowchart

Data Preprocessing

Need of Project

Educational institutions generate large amounts of data through student registration, examinations, attendance, assignments, learning management systems, online activities and feedback. Managing and analyzing this information using traditional methods can become difficult when the volume and variety of data increase.

Different academic systems may store information in different formats. Some information may be structured, while other information may be semi-structured or unstructured. Combining these sources is therefore an important requirement for meaningful academic analysis.

Manual analysis of large datasets can also be time-consuming. Big Data technologies provide distributed storage and parallel processing capabilities that can help process large datasets more efficiently.

There is therefore a need for a systematic Big Data Analytics approach that can collect information from multiple sources, store it in a scalable environment, clean and transform the data, perform analytics and present the results through understandable visualizations.

The proposed project fulfills this academic need by demonstrating a complete Big Data Analytics lifecycle from data collection to analytical insights and visualization.