서강대학교

초록 (요약문)

This study investigates the Russian-speaking migrants’ topics about Korea in Facebook. Through text mining, this study reveals what kinds of problems Russian-speaking migrants encounter while living in South Korea. The paper pursues two main research questions: 1) what kind of topics are popular among Russian-speaking migrants and 2) how these topics can be determined through textmining techniques. The few studies that investigated migrants have been conducted within frameworks provided by computational linguistic analysis. This thesis complements and extends these studies by assessing Russian-speaking migrants in Korea by analyzing Facebook groups which are open to Russian-speaking people who hold an interest in life in South Korea. The analysis for this study is composed of three steps: first, data collection and preprocessing; second, indicate co-occurrence matrix and identifying the popular topics; and third, analysis of results. The underlying topic structure was analysed using Latent Dirichlet Allocation (LDA), a topic modelling method widely adopted by text mining academics and researchers. Five latent topics were generated from the LDA model, defined as COVID-19, employment, health, visa issues and adaptation period for migrants from Commonwealth of Independent States (CIS) in Ansan city. This thesis suggests that the current needs of Russian speaking migrants can be implied from their comments, daily life stories and statements, which they share broadly with each other in Facebook.