• M.MUTHULAKSHMI
• K.PENNITA JASMINE
• V.PERIYA SATHIYA
• K.SANTHIYA
• R.S.SATHYARAJ

This project introduces a comprehensive predictive analytics solution aimed at identifying students at risk of dropping out of school education. Leveraging the power of machine learning, the solution integrates a Random Forest classifier with a Feature Tokenizer Transformer (FTT) to effectively analyze and interpret both structured and unstructured data. The primary objective is to accurately predict the likelihood of student dropouts based on a wide range of influential factors, including academic performance, attendance records, behavioral patterns, socioeconomic background, and qualitative teacher feedback. The Random Forest algorithm is utilized for its robustness and ability to handle high-dimensional data, building a reliable predictive model that can classify students into risk categories. To enhance the model’s capability, the Feature Tokenizer Transformer is employed to convert complex textual data—particularly teacher comments and feedback—into quantitative features that can be incorporated into the prediction pipeline. This enables the system to capture subtle behavioral cues and emotional indicators that may not be reflected in numerical data alone. By combining traditional data analytics with advanced natural language processing (NLP) techniques, this project aims to offer an early-warning system for educators and administrators. The end goal is to support timely interventions, reduce dropout rates, and ultimately contribute to better educational outcomes and equity in student support systems.

"Implement Software Solutions to Reduce Student Dropout Rates at Various Education Stages", International Journal of Emerging Technologies and Innovative Research (www.jetir.org), ISSN:2349-5162, Vol.12, Issue 5, page no.g170-g174, May-2025, Available :http://www.jetir.org/papers/JETIR2505720.pdf

"Implement Software Solutions to Reduce Student Dropout Rates at Various Education Stages", International Journal of Emerging Technologies and Innovative Research (www.jetir.org | UGC and issn Approved), ISSN:2349-5162, Vol.12, Issue 5, page no. ppg170-g174, May-2025, Available at : http://www.jetir.org/papers/JETIR2505720.pdf