應用知識運算技術建構財金知識入口網站之雛形系統
Date Issued
2003
Date
2003
Author(s)
DOI
912416H002021
Abstract
Many organizations have knowledge repository or data warehouses; however, information or
knowledge is scattered everywhere without proper use. Worse, the rapid growth of Internet
accelerates the creation of unstructured and unclassified information. It has resulted in
information overload. Though we can reach a flood of information through general-purpose
search engines, the effort of browsing through them is quite tedious and painstaking. Since most
of us are unable to utilize information effectively, there is a need for technology to solve this
issue. The purpose of this research is to explore text mining and data mining techniques to
address the problem. We discuss the major components of these knowledge computing techniques
required for building up a professional financial knowledge portal.
In the practical implementation, the laws and regulations related to securities and futures
markets were utilized to demonstrate the usage of SOM for topic categorization (topic map). We
also applied support vector machines in credit rating problems and stock market indices
prediction, with much better results than the traditionally statistics approaches and neural
network.
Subjects
knowledge management
financial knowledge portal
document categorization
topic
map
map
self-organizing map
support vector machines
Publisher
臺北市：國立臺灣大學工商管理學系
Type
report
File(s)![Thumbnail Image]()
Loading...
Name
912416H002021.pdf
Size
267.55 KB
Format
Adobe PDF
Checksum
(MD5):05f35dc54b5b459f567976287b740066