I am a MS CS student, interested in machine learning, data science, and its real-world applications. I have industry experience in building informational retrieval and search software, leveraging NLP for finance and legal industries (see projects).
Topic Modeling (Comparison of BERT, BOW and Top2vec)
Project links
Skills
About this project
the project took articles from US news media outlets and modeled topics using models like BERTopic, Top2vec, and BOW LDA. Best performance is achieved by LDA using 14 topics. The unique issues identified and similarly grouped cases can be found at:
https://github.com/purbid/topic_modeling