● Boston housing using Machine Learning (Personal Project) – (Jan 2024)

• Role: Python developer.
• Features: Predicting predict housing prices in the Boston area using various machine learning models.
• Python Programming: Utilized Jupyter Notebook for project workflow and data visualization. Data manipulation with pandas. Data visualization with seaborn and matplotlib (box plots, distribution plots, heatmaps, regression plots). Data preprocessing with normalization and standardization. Model training and evaluation. Enhanced model accuracy.
• Skills demonstrated: • Machine learning model development in Python. Data cleaning and processing with Excel and Jupyter Notebook. • Data preprocessing techniques. • Machine learning model implementation (Linear regression, Decision trees, Ensemble methods). • Model evaluation and comparison. • Feature importance analysis.
• Mathematical and Statistical theories used: • Boxplots and Distribution plots: Outliers, Central tendency, Spread of data and relationship between features. • Correlation matrix: Aiding in feature selection and identifying potential multicollinearity, Measures the linear relationship between pairs of features. • Min-Max normalization: Scales features to a common range (0-1) for improved convergence during model training. • Linear Regression: Uses a linear equation to model the relationship between features and the target variable, providing interpretable coefficients. • Decision Trees and Ensemble Methods: Divide the feature space into regions for prediction, offering non-linear modeling capabilities and robust against outliers. • Cross-Validation: Splits the data into training and testing sets to assess model generalizability and prevent overfitting. • Mean Squared Error (MSE): Measures the average squared difference between predicted and actual values, quantifying model performance.
• Project Outcome: Successfully training and evaluation of multiple regressing models for housing price prediction. Insights into feature importance and their impact on housing prices.