Machine Learning: Multiple Regression, Regularization, & NLP
Project Overview
Project 1
This project focuses on building a multiple regression model to predict outcomes using various features in the dataset. The project also explores regularization techniques such as Ridge and Lasso to avoid overfitting and improve the model's generalizability. Different error metrics, including mean squared error, mean absolute error, and max error, are used to evaluate model performance.
Project 2
This project delves into Natural Language Processing (NLP), employing techniques such as tokenization, text preprocessing, and feature extraction using principal component analysis (PCA). The goal was to analyze textual data and extract meaningful patterns to inform machine learning models. The project also emphasizes visualization techniques and performance analysis using NLP-specific metrics.
Languages, Platform, and Tools
Python, Pandas, NumPy, Matplotlib, Seaborn, Scikit-learn, NLTK
Data Viz
