Title :
Interactive data analysis on numeric-data
Author :
Chu, Hong Ki ; Wong, Man Hong
Author_Institution :
Dept. of Comput. Sci. & Eng., Chinese Univ. of Hong Kong, Shatin, Hong Kong
Abstract :
Data mining has been a hot topic in computer science. Many researchers have been putting lots of effort into how to extract explicit knowledge from large databases. Among the problems in data mining, finding useful patterns in large databases has attracted lots of interest in recent years. However, like other data mining algorithms, most of the proposed clustering algorithms suffer from the same demerit: lack of user interaction and exploration. In this paper, a new algorithm called IDAN (Interactive Data Analysis on Numeric-data) is being introduced. IDAN is good in discovering clustering patterns from numeric data. This algorithm is incremental and provides more user interaction in the mining process. At the same time, it allows the user to explore the rules or clusters found when integrated with a visualizer
Keywords :
data analysis; data mining; data visualisation; interactive systems; pattern clustering; very large databases; IDAN algorithm; clustering algorithms; data mining; interactive data analysis; knowledge extraction; large databases; numeric data; rules; useful pattern discovery; user exploration; user interaction; visualizer; Clustering algorithms; Computer science; Data analysis; Data engineering; Data mining; Ores; Partitioning algorithms; Pattern analysis; Read only memory; Sampling methods;
Conference_Titel :
Database Engineering and Applications, 1999. IDEAS '99. International Symposium Proceedings
Conference_Location :
Montreal, Que.
Print_ISBN :
0-7695-0265-2
DOI :
10.1109/IDEAS.1999.787271