Disambiguation of Named Entity with Supervised Technique Over a Knowledge Base
Aarthi D1, Viswanathan V2
1Aarthi D, School of Computing Science and Engineering, VIT University, Chennai (Tamil Nadu), India.
2Viswanathan V, School of Computing Science and Engineering, VIT University, Chennai (Tamil Nadu), India.
Manuscript received on 19 February 2019 | Revised Manuscript received on 10 March 2019 | Manuscript Published on 08 June 2019 | PP: 740-745 | Volume-7 Issue-5S4, February 2019 | Retrieval Number: E11530275S419/19©BEIESP
Open Access | Editorial and Publishing Policies | Cite | Mendeley | Indexing and Abstracting
© The Authors. Blue Eyes Intelligence Engineering and Sciences Publication (BEIESP). This is an open access article under the CC-BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/)
Abstract: Named Entity Disambiguation (Entity linking) is the task to link the entity mentioned in the query search with the appropriate entity in the repository without any name disambiguation. It can facilitate many tasks such as list of people or population in the repository and query processing and information retrieval. This task is a very challenging because of ambiguity and name conventions. In this paper we address a problem of named entity matching. In order to overcome this challenge we use Query search technique that implemented here is Name Dictionary based technique. The search key is extracted and compared with all the keys from the dictionary and the appropriate value is fetched and system throws as a result. Entity linking provides the information both explicitly and implicitly. Explicit linking provides the information beyond the knowledge base whereas implicit linking provides the information only from the knowledge base. Based on the information obtained we can also add the ratings and the comments. Based on the comments and the ratings the data that provided in the repository can also be managed. Our experiments shows the promising results in extracting the Candidate entities and graph based outcome if the user performs sequence of query search on single namesake.
Keywords: Named Entity, Candidate Entity, Disambiguation, Explicit Linking, Implicit Linking.
Scope of the Article: Knowledge Discovery