A scoping review of topic modelling on online data

With <span>the increasing prevalence of unstructured online data generated (e.g., social media, forums), mining them is important since they provide a genuine viewpoint the public. Due to this significant advantage, topic modelling has become more than ever. Topic natural language processing (NLP) technique that mainly reveals relevant topics hidden in text corpora. This paper aims review recent research trends and state-of-the-art techniques used when dealing with data. Preferred reporting items for systematic reviews meta-analysis (PRISMA) methodology was scoping review. study conducted on works published from 2020 2022. We constructed 5 questions interest many researchers. 36 papers revealed work non-English languages needed, common pre-processing were applied all datasets regardless e.g., stop word removal; latent dirichlet allocation (LDA) most also one best performing; produced result evaluated using coherence. In conclusion, largely benefited LDA, thus, it interesting see if trend continues future across languages.</span>