Ontology-assisted mining of RDF documents
Resource description framework (RDF) is becoming a popular encoding language for describing and interchanging metadata of web resources. In this paper, we propose an Apriori-based algorithm for mining association rules (AR) from RDF documents. We treat relations (RDF statements) as items in traditio...
Saved in:
| Main Authors: | , |
|---|---|
| Format: | text |
| Language: | English |
| Published: |
Institutional Knowledge at Singapore Management University
2005
|
| Subjects: | |
| Online Access: | https://ink.library.smu.edu.sg/sis_research/5232 https://ink.library.smu.edu.sg/context/sis_research/article/6235/viewcontent/Advanced_Methods_for_Knowledge_Discovery_from_Complex_Data__Springer__2005_.pdf |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| Institution: | Singapore Management University |
| Language: | English |
| Summary: | Resource description framework (RDF) is becoming a popular encoding language for describing and interchanging metadata of web resources. In this paper, we propose an Apriori-based algorithm for mining association rules (AR) from RDF documents. We treat relations (RDF statements) as items in traditional AR mining to mine associations among relations. The algorithm further makes use of a domain ontology to provide generalization of relations. To obtain compact rule sets, we present a generalized pruning method for removing uninteresting rules. We illustrate a potential usage of AR mining on RDF documents for detecting patterns of terrorist activities. Experiments conducted based on a synthetic set of terrorist events have shown that the proposed methods were able to derive a reasonably small set of association rules capturing the key underlying associations. |
|---|