Optimizing Mobile App Recommendations Using Crowdsourced Educational Data

With the rapid growth of mobile learning applications, students often face difficulty selecting the most relevant educational apps that suit their academic needs. This research proposes an intelligent recommendation system that optimizes mobile app suggestions using crowdsourced educational data. The system analyses application usage patterns and textual descriptions to recommend suitable apps for Undergraduate (UG), Postgraduate (PG), and Graduate students. Natural Language Processing (NLP) techniques such as text cleaning, tokenization, and feature extraction are applied to preprocess app metadata and user interaction data. A Random Forest Classifier is employed as the primary machine learning model due to its robustness, scalability, and strong performance in classification tasks. The model analyses behavioural patterns and content features to generate personalized recommendations. Experimental evaluation using metrics such as accuracy, precision, recall, and F1-score demonstrates that the proposed model produces reliable and interpretable results compared with traditional recommendation approaches. The system enhances personalization, reduces irrelevant suggestions, and improves user engagement in educational environments. This research contributes to the development of intelligent, AI-driven educational recommendation systems that support effective digital learning.