MOTH FLAME BASED SCHOLAR AND TEACHER ACADEMIC DATA ANALYSIS FOR INSTITUTES GRADE PREDICTION

Abstract

Educational data analytics is used to study the data available in the educational field and bring out the hidden knowledge from it. Many of researchers work to predict scholar grades. It was found that school, college, university performance also need to be analyze for improvement. This paper has developed a STFOIAGP (Scholar and Teacher Feature Optimization for Institute Academic Grade Prediction) model that identifies the scholar and teacher features. In this work teacher adaptability for institute system makes a different learning for grade prediction of institutes. Feature optimization was done by the Moth Flame algorithm and learning. Optimized features were transformed by Principal Component Analysis to get better learning parameter. Experiment was done on real dataset of Maharashtra Schools from 16 districts. Result shows that proposed STFOIAGP (Scholar and Teacher Feature Optimization for Institute Academic Grade Prediction) model optimized feature set has improved the prediction accuracy as compared to existing models