As a Data Science guy with one year of hands-on experience in the field, I am passionate about using data to solve complex problems and drive strategic decisions. My expertise spans statistical analysis, machine learning, data visualization, and programming, with a proven ability to transform raw data into actionable insights.
Compared three PyTorch CNNs—a VGG-style dual-conv, a lightweight single-conv, and a pretrained ResNet-50—on a 10-class animal dataset (limited dataset). The simple CNN beat the deeper custom net on limited data, while ResNet-50 converged in five epochs to over 95% validation accuracy, showcasing transfer learning’s power.
Explore my Customer Review Analysis and Sentiment Prediction project, where I leverage Python and machine learning to uncover insights and predict customer sentiments from reviews.
Performed on the E-Commerce data to explore transactional patterns through RFM Analysis and based on K-Means to cluster customer segments.
Using SARIMA regression model to forecast the time series total sales data for the next year.
This is a Decision Support System Project about recommend songs. Using K-mean to cluster all of the key attributes of songs and give a suggestion setlist relevantly and similar to that cluster. The data used is a piece of data made public by Spotify on Kaggle.