| Title: | PREDICTING RARE EVENTS IN TIME SERIES |
| DOI No: | 10.1142/9789812701527_0019 |
| Source: | KNOWLEDGE MANAGEMENT: NURTURING CULTURE, INNOVATION AND TECHNOLOGY (pp 221-229)
|
| Author(s): | JUNJIE HOU
School of Software, Tsinghua University, 100084, Beijing, China
CHUNPING LI
Data Mining Group, Institute of Information System & Engineering, School of Software, Tsinghua University, 100084, Beijing, China
|
| Abstract: | Event prediction is one of the main purposes in association rule mining. In this paper, we propose an approach for predicting rare events efficiently, like engine failures, stock price situations and market analysis etc. We associate a transaction as ordinal event series that occur in equal length intervals and formulate the problem of predicting rare events. Furthermore, an algorithm for discovering valuable patterns in time series is given. |
| Full Text: | View full text in PDF format (457KB) |
| TOC: | Back to Table of Contents |
|
|
