insectatorious/hn_kaggle

Simple clustering of HN posts from this Kaggle dataset: https://www.kaggle.com/hacker-news/hacker-news-posts

Clustering Hacker News post titles

A simple method of clustering and viewing Hacker News posts.

Data obtained from https://www.kaggle.com/hacker-news/hacker-news-posts.

Example plot

This screenshot shows the first 1000 titles clustered.

For an interactive plot, see it directly on Plotly.

Requirements

• Numpy
• Pandas
• Scikit-learn
• Gensim
• A pre-trained Word2Vec model (eg this model trained on the Google News corpus)
• Plotly (this is just for visualising, feel free to use any other library).

Pip install

pip install cython numpy pandas scikit-learn gensim plotly

Alternately, see the requirements.txt file.