Kroha+06

テキスト分析

http://csdl2.computer.org/persagen/DLAbsToc.jsp?resourcePath=/dl/proceedings/&toc=comp/proceedings/dexa/2006/2641/00/2641toc.xml&DOI=10.1109/DEXA.2006.135

Text Mining of Business News for Forecasting Kroha, P.; Baeza-Yates, R.; Krellner, B. Database and Expert Systems Applications, 2006. DEXA apos;06. 17th International Conference on Volume , Issue , 04-08 Sept. 2006 Page(s): 171 - 175 Digital Object Identifier 10.1109/DEXA.2006.135

Summary: In this paper, we analyze the relation between the content of business news and long-term market trends. We describe cleansing and classification of business news, we investigate how much similarity good news and bad news have, and how their ratio behaves in context of long-terms market trends. We have processed more than 400 thousand business news coming from the years 1999 to 2005. We present results of our experiments and their possible impact on forecasting of long-term market trends.