LDA’s approach to topic modeling is, it considers each document as a collection of topics and each topic as collection of keywords. The most common ones are Latent Semantic Analysis or Indexing(LSA/LSI), Hierarchical Dirichlet process (HDP), Latent Dirichlet Allocation(LDA) the one we will be discussing in this post. There are several existing algorithms you can use to perform the topic modeling. Finding good topics depends on the quality of text processing, the choice of the topic modeling algorithm, the number of topics specified in the algorithm. Topic models are useful for purpose of document clustering, organizing large blocks of textual data, information retrieval from unstructured text and feature selection. Topic model is a probabilistic model which contain information about the text.Įx: If it is a news paper corpus it may have topics like economics, sports, politics, weather. Topic modeling is technique to extract the hidden topics from large volumes of text. This blog post is part-2 of NLP using spaCy and it mainly focus on topic modeling.ĭo check part-1 of the blog, which includes various preprocessing and feature extraction techniques using spaCy.

Python gensim text compare online