|
Title:
|
INTRODUCTION OF TIME CONSTRAINTS TO A SEQUENTIAL MINING METHOD |
|
Author(s):
|
Shigeaki Sakurai , Ken Ueno , Ryohei Orihara |
|
ISBN:
|
972-8924-02-X |
|
Editors:
|
Pedro Isaías and Miguel Baptista Nunes |
|
Year:
|
2005 |
|
Edition:
|
2 |
|
Keywords:
|
Text Mining, Sequential Mining, Time Series Event Patterns, Time Constraints, Daily Business Reports. |
|
Type:
|
Short Paper |
|
First Page:
|
328 |
|
Last Page:
|
332 |
|
Language:
|
English |
|
Cover:
|
|
|
Full Contents:
|
click to dowload 
|
|
Paper Abstract:
|
A sequential mining method is proposed. The method extracts events from textual data that includes time information and discovers interesting sequences of events as time series event patterns. However, the method does not evaluate time intervals between events. The method may extract patterns with invalid sub-sequences of events. Because it does not care the fact that one event cannot influence another event if their time interval is too long. In this paper, we introduce 7 types of time constraints and propose a new sequential mining method including the time constraints. The method is able to define flexible time constraints and to discover valid interesting patterns. In addition, the paper applies the method to daily business reports collected by a sales force automation system and verifies its effectiveness through numerical experiments. |
|
|
|
|
|
|
