Supervised Machine Learning

This repository is a display of my capabilities in data analytics, featuring contributions to Kaggle competitions and work on various machine learning algorithms.

Recognitions

Chaifetz School of Business Data Science Machine Learning Badge

This badge is awarded to individuals who have shown exceptional ability in creating supervised machine learning models. The recipient of this badge is proficient in applying techniques for feature construction, feature extraction, feature selection, and hyperparameter tuning to enhance model performance. The badge also acknowledges the successful participation in one classification and one regression competition, illustrating a balanced skill set in both domains.

Kaggle Competition Insights

Titanic Survival Prediction

Summary: Predict survival on the Titanic and get familiar with ML basics.

• Titanic Survival Exploration

House Prices: Advanced Regression Techniques

Summary: Predict sales prices and practice feature engineering.

• House Price Prediction Analysis

MNIST Digit Recognizer

Summary: Classify handwritten digits using Support Vector Machines, Principal Component Analysis, Convolutional Neural Networks, and more.

• MNIST Digit Recognizer

Techniques explored

Linear Regression

Concept: Model the relationship between a scalar response and explanatory variables.

Decision Tree & Random Forest

Concept: Use tree-like models for classification or regression.

Support Vector Machine (SVM) & Naive Bayes

Concept: Explore SVM's margin maximization and Naive Bayes for probabilistic classification.

Ensemble & Boosting Methods

Concept: Improve predictions by combining several models.

Feature Engineering

Concept: Enhance model performance with better feature representations.