기계학습을 활용한 부동산 감성지수와 서울시 아파트 매매가격의 인과관계 연구

• 영문명: A Study of Casual Relationship between Sentiment for Housing Market and Seoul Apartment Price Using Machine Learning Technique
• 발행기관: 한국부동산정책학회
• 저자명: 박재수(Park, Jae-Soo) 이재수(Lee, Jae-Su)
• 간행물 정보: 『부동산정책연구』제21집 제1호, 75~93쪽, 전체 19쪽
• 주제분류: 사회과학 > 정책학
• 파일형태: PDF
• 발행일자: 2020.06.20

영문 초록

Recently, with the development of computer hardware and software, it is possible to analyze unstructured data. Also, emotional analysis using big data is being widely utilized as an alternative to conventional approaches. The emotional analysis is a technique to find out the information such as people s attitudes, opinions, or inclinations from unstructured text. It also can be usefully applied to grasp the psychology of people participating in the real estate market. This study aims to verify the causal relationship between Seoul apartment sales price by calculating each emotional index calculated using real-time news from online newspaper articles for Seoul apartment market. It is found that the sentiment index based on on-line news articles has significant correlation and causality relationships with the apartment sales price in the Seoul metropolitan area.

목차

Ⅰ. 서론

Ⅱ. 이론 및 선행연구 검토

Ⅲ. 분석 방법

Ⅳ. 분석 결과

Ⅴ. 결론

참고문헌

키워드

Topic modeling TextRank Naive Bayes Cross corre ation Granger causality Big data Seoul