Domain classification for celebrities using spreading activation and reasoning on semantic network

Author

Dongsheng Wang ; Kyunglag Kwon ; In-Jeong Chung

Author_Institution

Dept. of Comput. & Inf. Sci., Korea Univ., Sejong, South Korea

fYear

2013

fDate

2-5 July 2013

Firstpage

744

Lastpage

749

Abstract

Celebrities are influential nodes in Social Network Services (SNSs) whose domain classification can assist researchers to better analyze normal users´ interests. In this paper, we propose an automatic classification method that extracts the celebrity´s name and description as the input and the associated domain can be mapped to a domain ontology by performing Spreading activation on the knowledge base and reasoning on the domain ontology. The knowledge base is merged from YAGO ( Wikipedia and WordNet) and WordNet Domain ontology. We fetch 15,879 celebrity profiles, map them to WordNet Domain ontology, and demonstrate a satisfactory result of the mapping ratio.

Keywords

knowledge based systems; ontologies (artificial intelligence); semantic networks; social networking (online); SNS; Wikipedia; WordNet domain ontology; YAGO; automatic classification method; celebrity profiles; domain classification; knowledge base; mapping ratio; reasoning; semantic network; social network services; spreading activation; user interests; Cognition; Electronic publishing; Encyclopedias; Internet; Ontologies; Semantics; SNS celebrity; Semantic Network; Spreading Activation; WordNet; doman classification and mapping;

fLanguage

English

Publisher

ieee

Conference_Titel

Ubiquitous and Future Networks (ICUFN), 2013 Fifth International Conference on

Conference_Location

Da Nang

ISSN

2165-8528

Type

conf

DOI

10.1109/ICUFN.2013.6614919

Filename

6614919