Semantic similarity measure for topic modeling using latent Dirichlet allocation and collapsed Gibbs sampling

Automatically extracting topics from large amounts of text is one the main uses natural language processing (NLP). The latent Dirichlet allocation (LDA) technique frequently used to extract pre-processed materials based on word frequency. One problems LDA that extracted are poor quality if document does not coherently belong a single topic. However, Gibbs sampling operates word-by-word basis, which allows it be documents with variety and modifies topic assignment word. To improve extracted, this paper developed hybrid-based semantic similarity measure for modeling combining maximize coherence score. verify effectiveness suggested model, an unstructured dataset was taken public repository. evaluation carried out shows proposed LDA-Gibbs had score 0.52650 as against 0.46504. multi-level model provides better extracted.