post cover image

[Completed] SDS CP #15 - Student Performance Prediction

Project Objectives

  • Build a model which accurately predicts the performance of students

  • Deploy the model using web app interface


Project Phases & Timeline

Data Cleaning & Analysis (Week 1)

  • Handling null values, fixing data types, data inconsistencies

  • EDA, understanding distributions, outliers, relationships of features with target variable

  • Feature Selection using correlation analysis, ANOVA tests, F-test, etc.

Feature Engineering & Model Selection (Week 2 & 3)

  • Building new features, one hot encoding, feature scaling

  • Handling outliers through statistical and heuristic methods

  • Normalisation or Standardisation of input features is the ML algorithm in the pipeline requires it

  • Model training, comparison and selection

  • Model evaluation and Optimization using K-fold cross validation, hyperparameter tuning

Deployment (Week 4)

  • Building a streamlit app

  • Deploying model to app

  • Deploying app to streamlit cloud


Link to GitHub: https://github.com/SuperDataScience-Community-Projects/SDS-CP015-student-performance-pred

Post a comment