Title page for etd-0807116-200608


[Back to Results | New Search]

URN etd-0807116-200608
Author Hsin-Ching Huang
Author's Email Address No Public.
Statistics This thesis had been viewed 5636 times. Download 120 times.
Department Information Management
Year 2015
Semester 2
Degree Master
Type of Document
Language English
Title Time Series forecast of Company Revenue Trend Using Financial News
Date of Defense 2016-07-07
Page Count 41
Keyword
  • sentiment analyze
  • financial news
  • revenue prediction
  • time series
  • ARIMA model
  • Abstract The application nowadays of text mining are very extensive, and our study focuses on the field of finance. Most recent text mining studies of finance research the prediction of the stock market trend or the forecast of the stock prices, and other important economic indicators of the companies, as appeared in their financial statements, are seldom addressed. Yet these indicators could be quite important and reflect the financial status of the companies’ cash flow and market share. Most of these studies use the large amount of text on the Internet and obtain useful information through sentiment analysis.
    In this study, we adopt an automatically expanded finance sentiment dictionary and aggregate the sentiment values of the financial news. Furthermore, we combine text mining with the ARIMA model for time series analysis of the company's revenue. Experimental result shows that the forecasting quarterly revenue trend from analyzing news articles appeared in the last quality is quite effective with the precision up to 84%.
    Advisory Committee
  • Wan-Shiou Yang - chair
  • Wei-Che Tsai - co-chair
  • San-Yih Hwang - advisor
  • Files
  • etd-0807116-200608.pdf
  • Indicate in-campus at 3 year and off-campus access at 3 year.
    Date of Submission 2016-09-07

    [Back to Results | New Search]


    Browse | Search All Available ETDs

    If you have more questions or technical problems, please contact eThesys