Suresh Kumar

Predictive Performance Analytics System (PPAS) Team Members Role: Suresh Kumar: Project Lead and Frontend Developer Bitrip: Machine Learning Specialist Anush: Backend Developer Table of contents- Executive Summary Introduction Scopes & Limitations FR & NFR Modelling Diagram Prototype Table of contents- Planning Testing Software Specification Group Member Contribution Feedback Executive Summary Development Journey Spanned two semesters with a phased approach: requirements analysis, prototype design, backend and frontend integration. Purpose and Goals To enable early identification of struggling students using predictive analytics, empowering data-driven interventions. Core Outcome A fully tested, web-based system offering real-time academic risk detection, dashboard insights, and grade tracking. Predictive Performance Analytics System (PPAS) • Institutional Innovation: PPAS empowers educators by forecasting academic risks through real-time analytics and data visualization. • Smart Intervention: The system enables proactive measures by identifying at-risk students before academic failure. • Technology Stack: Built using React.js, Django, MySQL, and ML algorithms to ensure scalability and predictive accuracy. Introduction & Background Modern Academic Challenges PPAS Vision Institutions face increasing difficulty in proactively supporting student success a midst rising data complexity. To provide a scalable, predictive platform for monitoring academic performance and enhancing student retention. Tech-Driven Context Built with modern web technologies, PPAS enables real-time insights through machine learning and analytics. Scope & Limitations Scope • Data Collection & Analysis: Gathers grades, attendance, Limitations • accuracy. engagement, socio-economic background. • Machine Learning Predictions: Predicts student • Real-Time Visualizations: Dashboards and charts for monitoring progress. • Early Risk Detection: Calculates risk scores to enable timely interventions. • • User Role Management: Secure access for admins, • Narrow Risk Factor Scope: May miss non-academic influences (e.g., mental health). • Tech & Resource Barriers: Requires robust infrastructure and connectivity. • Adoption Challenges: Needs proper training and user buy-in. • Privacy & Security Risks: Must comply with data regulations educators, and students. (e.g., GDPR, FERPA). System Integration: Compatible with existing student data • Prediction Accuracy: Models may not always be precise. systems. • Limited Model Generalizability: Requires retraining for different contexts. performance and identifies at-risk students. • Data Quality Dependence: Incomplete/missing data may reduce Scalability: Suitable for institutions of all sizes. Functional Requirements User & Role Management Performance Tracking & Prediction Data Visualization & Reporting Secure login, role-based access for Admins, Educators, and Students. Grade/attendance management and ML-based risk scoring for students. Dashboards and exportable reports with real-time insights. Non-Functional Requirements Performance & Scalability Security & Privacy Usability & Accessibility Fast response (<2s), handles large datasets, cloud-ready. MFA, role-based control, data protection (e.g., GDPR, FERPA). Intuitive UI, mobile support, WCAG 2.1 compliant. Modeling Diagram System Architecture Frontend Layer Built using React.js and Tailwind CSS for a responsive, user-friendly interface. Backend Layer Powered by Django for business logic, data handling, and integration with ML APIs. Machine Learning & Database Uses scikit-learn models for predictions and MySQL for structured data storage. Authentication Flow Student Risk Analysis Flow Course Management Flow Data Flow Diagram User Role Management (USE CASE DIAGRAM) Prototype Prototype Explanation Grade Prediction Dashboard This is the main analytics view for educators and administrators. It provides: • A summary of predicted student grades using machine learning models. • Visual insights into student performance trends over time. • Key indicators such as attendance percentages and grade progression. • Early alerts for students identified as at-risk based on performance data. Use Case: Helps teachers quickly identify which students may need support or intervention. Prototype Explanation Settings Page A configuration panel mostly used by administrators. It includes: • User Role Assignment – Assign or update roles like Admin, Educator, or Student. • Risk Threshold Configuration – Set limits that define which students are marked "at-risk." • Grade Prediction Cutoffs – Adjust prediction sensitivity for pass/fail analysis. • Data Export Options – Export student data and performance reports to formats like CSV or PDF. Use Case: Provides flexibility and control over how the system works and who can access what. Prototype Explanation Search Functionality An intelligent search tool for quick access to specific student profiles. It offers: • Autocomplete suggestions as users type names or IDs. • Detailed performance summaries for selected students. • Visualization of subject-wise predicted scores and risk levels. Use Case: Enables quick review of any student’s academic status without navigating through multiple pages. Prototype Explanation Student Overview Page A personalized dashboard for each student, showing: • Cumulative GPA and performance breakdown by semester or course. • Attendance rates and class participation metrics. • Highlighted areas needing improvement, such as failing subjects or low attendance. Use Case: Allows both students and educators to track progress and reflect on academic performance. Prototype Explanation Attendance & Grade Pages Detailed records presented in an easy-to-read format. Features include: •Filters to view attendance/grades by year, semester, or course. •A tabular layout displaying attendance status (e.g., Present, Absent, Late) per date. •Lists of individual grades for assessments, with feedback or remarks from educators. Use Case: Promotes transparency and makes it easy to identify attendance or academic patterns. Planning Gantt Chart of PPAS Project Work Breakdown Structure (WBS) Testing & Evaluation Unit Testing Integration Testing Validated individual modules like login, dashboard, file upload, and grade prediction. Tested interactions between frontend, backend, and ML prediction components. User Acceptance Testing Educators evaluated the system for responsiveness, usability, and visual clarity. Implementation showcase Software Specifications Frontend Backend Technology: React.js with Vite Styling: Tailwind CSS Routing: React Router Features: Responsive UI, rolebased dashboards, interactive charts Framework: Django (Python) APIs: Django REST Framework (DRF) Authentication: JWT (JSON Web Tokens) Machine Learning: Scikit-learn, SHAP for Explainable AI Database System: MySQL (Relational DB) Data Types: Grades, attendance, demographics, risk scores Design: Normalized schema with ER Diagram Integration: Supports CSV uploads and manual entry Group Contribution S.N. Project Phase 1 Requirements Analysis & Planning 2 UI/UX Design (Prototype) 3 Frontend Development 4 Responsibility Team Member(s) Functional/Non-functional specs, use-case planning Figma design, layout of dashboards & user flows React.js components, pages, styling with Tailwind CSS Suresh Kumar Backend Development Django models, APIs, ML integration Bitrip 5 Database Design & Queries Schema design, SQL queries for CRUD operations Anush 6 Machine Learning Model Data preprocessing, training, risk prediction model Bitrip 7 Integration (Frontend + Backend) API connectivity, routing, authentication Suresh Kumar 8 Testing & Debugging Unit, integration, and user acceptance testing Anush, Suresh Kumar 9 Documentation & Report Writing Final report, formatting, screenshots, references Bitrip, Anush 10 Project Management & Meetings Task allocation, milestone tracking, Gantt chart Suresh Kumar 11 Final Presentation & Walkthrough Slide preparation, demo video, delivery Suresh Kumar, Bitrip Bitrip, Oliver Suresh Kumar Conclusion System Accomplishments Delivered a scalable, user-centric academic analytics platform with predictive capabilities. Trans-formative Potential Empowers institutions to shift from reactive to proactive student support. Scalability & Next Steps Ready for institutional roll out with road-map for real-time data and mobile enhancements. Thank You