|
Title:
|
MINING EPISODE RULES WITH A DISTANT CONSEQUENT |
|
Author(s):
|
Lina Fahed, Armelle Brun, Anne Boyer |
|
ISBN:
|
978-989-8533-25-8 |
|
Editors:
|
Hans Weghorn |
|
Year:
|
2014 |
|
Edition:
|
Single |
|
Keywords:
|
Data mining, episode rules mining, minimal rules, event prediction. |
|
Type:
|
Full Paper |
|
First Page:
|
175 |
|
Last Page:
|
182 |
|
Language:
|
English |
|
Cover:
|
|
|
Full Contents:
|
click to dowload 
|
|
Paper Abstract:
|
This work focuses on event prediction in an event sequence, where we aim at predicting temporally distant events. We propose an algorithm for mining episode rules, which are minimal and have a consequent temporally distant from the antecedent. As state of the art algorithms are not able to mine rules with such characteristics, we propose an original way to mine these rules. Our proposed algorithm determines the consequent of an episode rule at an early stage in the mining process, it applies a span constraint on the antecedent and a gap constraint between the antecedent and the consequent of the rule. The algorithm is validated on an event sequence of social networks messages. Experimental results show that minimal rules with a distant consequent are actually formed and that they can be used to accurately predict distant events. |
|
|
|
|
|
|
