بناء نماذج machine learning

The project "Predictive Analytics for Student Success" aims to address the issues of student dropout and underperformance by leveraging predictive analytics. Using the CRISP-DM methodology (Business Understanding, Data Understanding, Data Preparation, Modeling, Evaluation, and Deployment), it identifies at-risk students and enables timely interventions to improve retention and success rates. The business problem involves mitigating wasted potential and negative consequences, ultimately leading to better academic outcomes. The goal is to identify students at risk of dropping out or underperforming.

The data analysis includes exploratory data analysis (EDA) with visualizations like box plots, scatter plots, and pie charts to understand the distribution of grades (e.g., Assignment01, Assignment02, Final Exam) across classes (G and W). The dataset, loaded from "Student_Performance_Prediction-B.csv," contains 486 entries with 10 columns, including student IDs and various assessment scores.

For deployment, you have utilized Gradio to create an interface and hosted it on Hugging Face, making the predictive model accessible for real-time use. This allows stakeholders to input student data and receive predictions on their performance risk.