A Comparison of Statistical and Data Mining Techniques for Enrichment Ontology with Instances
Aurawan Imsombut, Jesada Kajornrit
Available Online January 2017.
- https://doi.org/10.2991/icefs-17.2017.52How to use a DOI?
- Ontology Enrichment, Statistical Technique, Classification, Conditional Random Fields (CRFs), Feature-weighted k-Nearest Neighbor
- To enrich instances into an ontology is an important task. That is because the process does extend the knowledge in ontology to cover more the domain of interest, as a result, more benefits can be gained. There are many techniques to classify instances of concepts, however, two popular ones are the statistic and the data mining methods. This paper compares the use of these two methods to classify instances to enrich ontology having more domain knowledge. This paper selects conditional random field for the statistic method and feature-weight k-nearest neighbor classification for data mining method. The experiments were conducted on the tourism ontology. The results pointed out that conditional random fields methods provided more precision and recall value than the other, specifically, F1-measure is 74.09% for conditional random fields and 60.04% for feature-weight k-nearest neighbor classification.
- Open Access
- This is an open access article distributed under the CC BY-NC license.
Cite this article
TY - CONF AU - Aurawan Imsombut AU - Jesada Kajornrit PY - 2017/01 DA - 2017/01 TI - A Comparison of Statistical and Data Mining Techniques for Enrichment Ontology with Instances BT - 2017 International Conference on Economics, Finance and Statistics (ICEFS 2017) PB - Atlantis Press SP - 408 EP - 413 SN - 2352-5428 UR - https://doi.org/10.2991/icefs-17.2017.52 DO - https://doi.org/10.2991/icefs-17.2017.52 ID - Imsombut2017/01 ER -