Textual Mining- Evaluation of Mann Ki Baat Repository

摘要

Background and Objective: As computers and the Internet are broadly utilized in nearly every region, numerous computerized text data is produced each day. It becomes a fundamental task to explore and effectively search such massive data. The main aim of the present study is to emphasize the recurrence of topics and identifying main ideas from a popular monthly addressing radio program Mann Ki Baat by using topic modeling technique. Data and Method: The present study utilizes the unstructured data of Mann ki Baat from January 2020 to March 2020, collected from the PMINDIA website. This program was initiated by the Honorable Prime Minister of India, Mr. Narendra Modi. This examination uses a popular technique Topic modeling based on LDA (Latent Dirichlet Allocation). Findings: The results show that the method automatically extracts the main ideas and issues discussed. Besides it provides information about the most likely topics and themes discussed in each month that left an impact on people and helped in raising awareness. Novelty: This is a first study of the application of popular technique topic modelling on Mann ki Baat. Further, this is the first attempt to extract the ideas discussed in a social campaign using a statistical model.