빅데이터 분석기법을 이용한 기업의 재무성과 관리방안 연구

Amid the expansion of the areas using Big Data to make decisions in companies, this study derived the RM(Real-Estate Media Index) I standardized through text-opinion mining with NLP(Natural language processing)-based unstructured data and conducted regression and correlation analysis with social or economic indexes. In other words, this research is aimed to present the ways of application by collecting real estate market trends promptly from online media articles, making them into indexes, and conducting verification through correlation with financial data related to the sales of companies, regression modeling, and machine learning prediction techniques. According to the results of analyzing correlation between the RMI derived from the monthly positive and negative ratios of the real estate media, company A’s internal financial indexes, and social-economic indexes, the receipt amount was included as one of the internal indexes; therefore, regression modeling for the RMI and also for the receipt amount was done respectively. According to the findings, the mutual deposit and land market consumer sentiment index have positive (+) effects on the RMI. Also, the RMI and loan amount are positive (+) effects on the Company A's receipt amount, and the producer price index have negative (-) effects on the Company A's receipt amount. Based on the results of empirical research, this author suggests whether (structured / unstructured) Big Data analysis related to the real estate market can be applied to the prediction and management of financial performances in a company.