The purpose of this study is to identify the types of social problems in Gyeonggi-do through topic modeling from the perspective of legal data science. Through the newspaper articles related to the 'law', which is the subject of analysis, the social issues of Gyeonggi-do were typified and the meaning was placed on finding policy implications. Considering that the scope of research has been decided according to the researcher's subjective perception of the social issues of the region in a fragmentary manner, all issues appearing in the local media in Gyeonggi-do have been analyzed to explore the trends of social issues in Gyeonggi-do.
Topic modeling is an algorithm for finding a topic by finding a certain pattern in a vast unstructured literature group. It is proposed as an alternative to solving problems such as sparsity Problem, polysemy, synonyms, and semantic hierarchical structure that occur in the existing word frequency analysis. It is a methodology for inferring topics by clustering words with similar meanings using context-related clues. In this study, using topic modeling, topics were extracted through newspaper articles from four local media outlets in Gyeonggi-do, and how these topics changed over time was identified.