Stock market prediction has always been an emerging topic for researchers.
Stock price trend forecasting based only on numerical time series data has been paid a
great attention by previous scholars. However, stock prices can be influenced by
many factors, ranging from news releases of companies and local politics to news of
superpower economy. An enormous amount of this valuable information is widely
available from news articles via the public communication media. With the rapid
development of web technology, currently a growing number of people all over the
world share information on the web. The dramatic increase of the amount of web data
has led to difficulties for investors when following and considering all available
information. Therefore, Mining textual documents and time series concurrently
become more relevant