Text Mining and Topic Modelling

Li, Yulei; Shan, Shan; Lin, Zhibin

Abstract

Social media platforms have become a prevalent place where customers can share their real opinions about products, services, or brands. This encourages businesses to invest abounding resources to analyse and understand what their customers are discussing on social media. This chapter will attempt to introduce one application of natural language processing (NLP) or text mining in business research. This chapter focuses on understanding (i) what is the Topic Modelling in Text Mining?, (ii) how to Collect Textual Data on Social Media?, (iii) what are latent Dirichlet Allocation (LDA) and hierarchical latent Dirichlet Allocation (hLDA)?, (iv) how to visualise the hierarchical topics generated by hLDA?, (v) how to interpret the hLDA results?, (vi) how to write the results or findings section for hLDA results?, and (vii) what are the limitations of topic modelling?

Citation

Li, Y., Shan, S., & Lin, Z. (2023). Text Mining and Topic Modelling. In P. Foroudi, & C. Dennis (Eds.), Researching and Analysing Business: Research Methods in Practice. London: Routledge. https://doi.org/10.4324/9781003107774-13

• Online Publication Date: Dec 14, 2023
• Publication Date: Dec 14, 2023
• Deposit Date: Dec 8, 2023
• Publicly Available Date: Jun 15, 2025
• Publisher: Routledge
• Edition: 1st Edition
• Book Title: Researching and Analysing Business: Research Methods in Practice
• Chapter Number: 11
• DOI: https://doi.org/10.4324/9781003107774-13
• Public URL: https://durham-repository.worktribe.com/output/1985013