MUMBAI, India, Aug. 31 -- Intellectual Property India has published a patent application (202621081988 A) filed by Symbiosis International Deemed University on July 02, 2026, for Machine Learning Based System And Method For Early Prediction Of Student Academic Performance.
Inventors include Shantanu Jumale; Piyush Agrawal; Om Shendre; and Dr. Priya Dasarwar.
The application for the patent was published on August 28, 2026, under issue no. 35/2026.
Abstract: ABSTRACT MACHINE LEARNING BASED SYSTEM AND METHOD FOR EARLY PREDICTION OF STUDENT ACADEMIC PERFORMANCE The present invention discloses a machine learning based system and method (100) for early prediction of student academic performance. The system comprises a data ingestion module (110) for receiving student records comprising demographic, behavioral, and academic attributes; a data preprocessing module (120) for performing missing value imputation, outlier removal using interquartile range analysis, feature encoding, and standardization; a feature engineering module (130) for generating composite performance metrics and stratified dataset splitting; a multi-model training and evaluation engine (140) configured to simultaneously train and evaluate three classifiers including Logistic Regression (141), Random Forest (142) employing ensemble decision trees with bagging, and XGBoost (143) employing gradient boosting with regularization; and an interactive web- based prediction interface (150) implemented using Streamlit for real-time prediction with comparative model analysis. The system enables early identification of at-risk students through multi-model consensus and supports both individual and bulk predictions for institutional academic risk assessment. [
Disclaimer: Curated by HT Syndication.
