Topic Modelling with LDA

Simple, no-frills mixed-topic modelling. The implementation primarily follows the steps outlined in this technical report. We randomly draws 60 quotes from a pool and feed them into LDA. The output has 4 topics. Each topic's top 5 word tokens are shown, and below we show the topic distribution across each quotes, the higher scored topic is more opaque while the low scored topic is more transparent.






Quotes sourced from https://www.huffingtonpost.com/lolly-daskal-/100-motivational-quotes-t_b_4505356.html. || Porter-Stemmer came from https://github.com/kristopolous/Porter-Stemmer
Fork me on GitHub