Predictive model for improving academic performance using random Forest algorithm

Academic performance is a key indicator of students' academic success and future professional development. Factors such as lack of resources, inadequate educational strategies and socio-economic differences often have a negative impact on academic outcomes, affecting students' personal and professional development for the future. To address this problem, a predictive model based on the Random Forest algorithm is proposed, which seeks to identify patterns and critical factors that impact students' learning. The model consists of four phases: Phase 1 is academic data collection, where relevant academic and demographic data are collected. Phase 2 is data cleansing, where duplicate data are removed, inconsistencies corrected and the data prepared for analysis. Phase 3 is the training of the predictive technique, where the Random Forest algorithm is trained using the collected data. And, in step 4, is the model evaluation, the results will be validated using metrics such as precision, recall and F1-score. After the implementation of the project, the predictive model achieved a precision of 91%, which means an improvement in education towards students and improvements in academic performance; also providing teachers and administrators with an effective tool to implement personalized educational interventions and thus improve academic performance.