MACHINE LEARNING // DECISION TREE // 09

Decision Tree Classification.

Built a Streamlit classification application that trains a Decision Tree on income data and exposes prediction probabilities, decision paths, evaluation output, and tree structure.

CLIENT

Academic ML Project

ROLE

Machine Learning Developer

CHALLENGE

A classification result alone does not explain how a tree model reaches a decision or whether the evaluation setup contains data leakage.

SOLUTION

Implemented data loading, categorical and numerical preprocessing, model training, interactive prediction, probability output, decision-path extraction, confusion matrix, classification report, tree visualization, and a target-leakage warning.

KEY FEATURES

01Dataset loading and preprocessing
02Decision Tree training
03Interactive prediction input
04Class probability output
05Decision-path inspection
06Confusion matrix
07Classification report
08Tree structure visualization
09Target-leakage warning

TECHNOLOGY

Pythonscikit-learnPandasNumPyMatplotlib

RESULTS

Delivered an explainable classification demo that connects model input, tree decisions, prediction output, and evaluation evidence in one interface.

PROJECT LINKS

LIVE DEMOVISIT

GITHUBVIEW

NOTE

The application is an educational model demonstration. Its predictions and evaluation metrics depend on the supplied dataset and preprocessing configuration and must not be used for high-impact decisions.

NEXT PROJECT

Income Prediction ML App.

MACHINE LEARNING // MODEL INFERENCE