Many techniques are used to obtain topic models. This paper aims to demonstrate the implementation of LDA: a widely used topic modeling technique and how to visualize the obtained topics. We use the topic model Latent Dirichlet Allocation (LDA) from genism package for extracting the topics from the abstracts and titles of different research papers present in the almetric data. We show the correlation among different topics extracted from both the abstracts and titles. We have done different visualizations like frequency distribution of word counts in documents, word counts of Top N keywords in each topic, word counts of topic keywords, sentence chart colored by topic, dominant topics in the documents, pyLDAVis interactive chart visualization.
Topic Models, Visualization, Correlation
https://github.com/PSdiv/Topic_Modeling-NLP/blob/master/Topic_Modeling_of_Research_Papers.pdf