Cal Poly CSC 365 Intro to Database Systems Spring 2023

This project predicted health insurance charges using two methods: Averaging and Linear Regression. Linear Regression delivered more accurate predictions, highlighting its superior performance.

This project repository comprises an exploratory data analysis (EDA) of the "Breast Cancer Gene Expression Profiles (METABRIC)" dataset, employing Python libraries for data manipulation and visualization. It includes data loading, preparation, and EDA tasks, aiming to gain insights into breast cancer data.

👨‍👩‍👧‍👦 Estimate the number of people using WiFi emanations

This Jupyter Notebook focuses on credit risk prediction using a Random Forest Classifier. It covers data preprocessing, exploratory data analysis (EDA), model training, and handling class imbalance. Additionally, essential metrics such as precision, recall, F1-score, and confusion matrices are computed to evaluate the model's performance.

This GitHub repository hosts a bike rental demand prediction project. It utilizes Python libraries for data analysis and builds regression models to forecast demand based on various attributes. The project covers data preprocessing, exploratory analysis, and model evaluation.

This repository includes an NLP (Natural Language Processing) project that focuses on text classification for news articles sourced from an Iranian newspaper. It encompasses various stages of the NLP pipeline, including data exploration, text preprocessing, vectorization, train-test data splitting, and machine learning model training.

This GitHub repository houses a water treatment plant classification project. Leveraging Python and machine techniques, it categorizes the plant's operation state using sensor data. The project covers data preprocessing, EDA, model training, and hyperparameter tuning, highlighting expertise in data science, ML, and data visualization.

This repository predicts body fat percentage using Lasso and Ridge regression models. It compares their accuracy and performance metrics, offering detailed code and model evaluations in the Jupyter Notebook.

This project analyzes 1976 Los Angeles ozone and meteorological data, using linear regression in three stages: all features, the most correlated feature (temperature), and temperature-based data splitting. It underscores feature selection and data stratification's significance.

This repository hosts a case study on Transfer Learning aimed at improving the accuracy of metal nut inspection systems. The project explores the practical application of Transfer Learning methods to optimize quality control in the manufacturing industry, providing valuable insights into their effectiveness and potential benefits.