Using Data Analyzing and Machine Learning tools like Orange to understand
key information and patterns in huge amount of data sources.
This project analyzes song lyrics to explore the relationship between word choices and music genres. By compiling a corpus of lyrics categorized by genre, it reveals how themes, emotions and linguistic patterns vary across genres like metal, R&B and rap.
This project uses Orange to extract features from images and group them using k-means clustering and random forest classification. It shows how machine learning can visually organize and predict image categories.
This project analyzes Harvard Art Museums’ artifact descriptions to reveal cultural biases in collections. It shows overrepresentation of Chinese, Japanese, and European cultures, and underrepresentation of South Asian and African ones. Data visualizations highlight themes and historical influences on collecting practices.