Time Series Analysis on Stock Market for Text Mining Correlation of Economy News

• DOI: 10.48550/arxiv.1403.2002
• Publication Date: 2014-01-01
• Authors: Şadi Evren Şeker; Cihan Mert; Khaled Al-Naami; Nuri Özalp; Uğur Ayan
• Keywords: Stock (firearms)

This paper proposes an information retrieval method for the economy news. The effect of news, are researched in word level and stock market values considered as ground proof. correlation between prices news is already addressed problem most countries. well-known approach applying text mining approaches to some time series analysis techniques over closing order apply classification or clustering algorithms features extracted. study goes further tries ask question what available which one suitable? In this study, their dates collected into a database applied part has been kept simple with only term frequency-inverse document frequency method. For part, we have studied 10 different methods such random walk, moving average, acceleration, Bollinger band, price rate change, periodic difference, momentum relative strength index variation. also explained these comparative way Turkish Stock Market more than 2 year period. On other hand, on high-circulating newspapers Turkey.