Twitter is the most generally involved online entertainment stage for sharing data. Furthermore, Twitter is quite possibly of the biggest social medium in Indonesia. The vast majority of the data or news that is distributed comes from issues or occasions that are creating in the public eye. The issue of police officers committing crimes is one that quickly becomes viral. This examination is a continuation of past exploration to foresee virality in view of Twitter information. Predicting and classifying text in the form of tweets and posts (socmed Twitter) using BERT. IndoBERT as a safeguarded model in the characterization of kinds of wrongdoing. ViralBERT to foresee tweet virality utilizing text includes and adding numeric highlights to enhance virality forecasts. Using the baseline from this study, the data set was used to train ViralBert and validate model results. The dataset was gathered utilizing the Twitter Programming interface with the watchwords 'oknum polisi' and oknum polri'. The consequences of this virality expectation will be utilized as a kind of perspective by the police in following up a case that is viral locally
Copyrights © 2023