Machine Learning · Guided project
Student Performance Predictor (Regression) — Guided Project
Predict final exam scores from study habits and attendance with linear regression and a random forest, compare them honestly and explain which factors matter.
- Beginner
- ~5 hours
- 5 steps
The brief
Use a public student performance dataset. Frame the problem, build a baseline, train a linear regression and a random forest inside scikit-learn pipelines, compare them with cross-validation and report the final result on a held-out test set. Then interpret the model — which features matter, and what would you tell a mentor?
Skills you will use: Train/test split · Pipelines · Linear regression · Random forest · MAE / RMSE
Build it
Step by step
Tick each step as you finish it. 0 of 5 done.
-
Step 1Frame and explore
Define the target and the decision it supports; explore distributions and missing values.
-
Step 2Baseline
Measure the MAE of always predicting the training average.
-
Step 3Pipelines
Build preprocessing (impute, scale, one-hot) and train linear regression and a random forest.
-
Step 4Compare
Use 5-fold cross-validation, then evaluate the better model once on the test set.
-
Step 5Interpret and report
Show coefficients or feature importances, explain them carefully and list two improvements.
Self-review before you call it done
Tick each point once your project really meets it. These are the things an interviewer will look for.
Check all 4 points, then mark the project as finished.
Next