This repository contains ML models I am building to practice machine learning concepts as part of my AIML learning journey.
- Dataset: Built-in Iris dataset from sklearn
- Model Used: Decision Tree Classifier
- Accuracy: 100%
- What it does: Predicts which type of iris flower (Setosa, Versicolor, Virginica) based on sepal and petal measurements
-
Dataset: Built-in Titanic dataset from seaborn
-
Model Used: Decision Tree Classifier
-
Accuracy: ~70%
-
What it does: Predicts whether a passenger survived or not based on class, age, sex, fare etc.
-
Key learning: Preprocessing messy real world data, handling missing values, converting text to numbers
-
Dataset: Heart Disease Dataset from Kaggle
-
Model Used: Decision Tree Classifier
-
Accuracy: ~80%
-
What it does: Predicts whether a person has heart disease or not based on age, sex, chest pain type, cholesterol, blood pressure etc.
-
Key learning: Working with Kaggle CSV datasets, label encoding multiple text columns
- Importing ML libraries (pandas, sklearn, pickle)
- Loading built-in and Kaggle datasets
- Exploring data (head, shape, isnull)
- Preprocessing data (dropna, fillna, map)
- Handling missing values
- Converting text columns to numbers (Label Encoding)
- Splitting data into training and testing sets
- Training a Decision Tree Classifier
- Testing model accuracy
- Classification report and confusion matrix
- Saving and loading models using pickle
- Predicting on new unseen data
- Python
- Google Colab
- Scikit-learn
- Pandas
- Seaborn
- Pickle
- Iris Flower Classification
- Titanic Survival Prediction
- Heart Disease Prediction
- Earthquake Risk Prediction
- Flood Risk Prediction
- Cyclone Risk Prediction
To build strong ML fundamentals and apply them to a real world Disaster Management System that predicts risk levels and sends alerts to affected users.
Update your README and push to GitHub! Then come back — we start your real disaster project! 🚀😊