Mining Stock and News Data for Prediction

The Hong Kong University of Science and Technology
Department of Computer Science and Engineering

FYT Presentation and Demonstration

Title: "Mining Stock and News Data for Prediction"

by

Mr. LEE Mang Lung

Predicting stock market is not an easy task due to the gigantic volume of 
information available in the market. It requires advanced knowledge of 
markets, politics, history, banking and human behavior. It is further 
complicated by the rapidly growing volume of daily news - both financial 
and geopolitical - which is not easily handled. Even financial 
professionals are also required long hours spent daily in following market 
data and reading news. This method is time-consuming and error-prone.

In this study, we aim to find ways to simplify the process of reading 
financial news using data mining techniques. We developed a tool to 
simplify the process of reading financial news. It involves the 
preliminary filtering of the financial news related to the stock market. 
The tool can discover hidden information and summarize the results in a 
comprehensive and user-friendly format. And we investigated the accuracy 
of the algorithms used in this paper.


Date		:  14 May 2010 (Friday)

Time		:  9am to 9:40am

Venue		:  Room 3416

Advisor		:  Dr. Chen Lei

2nd Reader	:  Dr. Yi Ke