Heterogenous Analysis of KeyBERT, BERTopic, PyCaret and LDAs methods: P53 in Ovarian Cancer use case

Abstract

In recent times, researchers with Computational background have found it easier to relate to Artificial Intelligence with the advancement of the transformer model, and unstructured medical data. This paper explores the heterogeneity of keyBERT, BERTopic, PyCaret and LDAs as key phrase generators and topic model extractors with P53 in ovarian cancer as a use case. PubMed abstract on mutant p53 was first extracted with the Entrez-global database and then preprocessed with Natural Toolkit (NLTK). keyBERT was then used for extracting keyphrases, and BERTopic modelling was used for extracting the related themes. PyCaret was further used for unigram topics and LDAs for examining the interaction among the topics in the word corpus. Lastly, Jaccard similarity index was used to check the similarity among the four methods. The results showed no relationship exists with KeyBERT, having a score of 0.0 while relationship exists among the three other topic models with score of 0.095, 0.235, 0.4 and 0.111. Based on the result, it was observed that keywords, keyphrases, similar topics, and entities embedded in the data use a closely related framework, which can give insights into medical data before modelling.