Student Performance Predictor and Dashboard

Project Domain / Category

Data Analytics / Machine Learning / Educational Data Mining

Abstract / Introduction

Academic success plays a vital role in shaping a student’s career and future opportunities. However, many factors including attendance, participation, assignments, and exam results affect performance. Identifying students at risk of poor performance at an early stage allows teachers and institutions to take preventive actions. This project, “Student Performance Predictor & Dashboard”, aims to build a system that predicts student outcomes using machine learning and provides an interactive dashboard for monitoring progress. The system will utilize academic records, attendance data, and exam results to forecast performance and highlight at-risk students. By offering insights through data visualization, the dashboard will assist educators in making informed decisions and support students in improving their academic results.

Functional Requirements

            Dataset:

            Collect a dataset containing student attendance, assignment scores, test results, and other academic records.

            Possible sources: open educational datasets (e.g., UCI, Mendeley Data) or institution-specific student records.

            The dataset may require merging attendance logs with exam performance data.

            Data Pre-processing:

            Clean the dataset to handle missing values and inconsistencies.

            Normalize numeric features such as grades and attendance percentage.

            Encode categorical variables (e.g., gender, study habits).

            Split data into training and testing sets for model evaluation.

            Machine Learning Models:

            Apply classification algorithms (Decision Trees, Random Forest, Logistic Regression, Support Vector Machines).

            Compare models based on accuracy, precision, recall, and F1-score.

            Select the best-performing model for deployment in the dashboard.

            Dashboard Development:

Build an interactive dashboard to visualize performance insights.

Teachers’ view: overview of class performance, risk alerts, and comparative graphs.

Students’ view: personal academic progress, attendance trends, and predicted outcomes.

Role-based login for teachers and students.

            Recommendations:

Provide early alerts for students at risk of failing or underperforming.

Highlight key factors (low attendance, assignment performance, exam results) that influence predictions.

Page 152 of 167