- unsupervised Topic Modelling algorithm
- Generative statistical model
- Assumptions
- each document is a mixture of topics
- each word is attributable to at least one of the topics
- number of topics k to be pre-defined
- Output
- list of words (ranked by probability) associated with a topic of interest
- probability that a word of interest is associated with a topic
- proportion of a document that pertains to a topic
- does not output the topic itself