Generating association rules from semi-structured documents using an extended concept hierarchy
CIKM '97 Proceedings of the sixth international conference on Information and knowledge management
A new and versatile method for association generation
Information Systems
Design of Database Structures
Fast Algorithms for Mining Association Rules in Large Databases
VLDB '94 Proceedings of the 20th International Conference on Very Large Data Bases
A new logic correlation rule for HIV-1 protease mutation
Expert Systems with Applications: An International Journal
Mining interesting XML-enabled association rules with templates
KDID'04 Proceedings of the Third international conference on Knowledge Discovery in Inductive Databases
| Hi-index | 0.00 |
The need for sophisticated analysis of textual documents is becoming more apparent as data is being placed on the Web and digital libraries are surfacing. This paper presents an algorithm for generating constrained association rules from textual documents. The user specifies a set of constraints, concepts and/or structured values. Our algorithm creates matrices and lists based on these prespecified constraints and uses them to generate large itemsets. Because these matrices are small and sparse, we are able to quickly generate higher order large itemsets. Further, since we maintain concept relationship information in a concept library, we can also generate rulesets involving concepts related to the initial set of constraints.