Download presentation
Presentation is loading. Please wait.
Published byDaniel Warren Modified over 9 years ago
1
Mining fuzzy domain ontology based on concept Vector from wikipedia category network
2
Outline Abstract Introduction Fuzzy on tology generation Empirical experiments and result s Conclusion References
3
Abstract Ontology is essential in the formalization of domain knowledge for effective human-computer interactions (i.e.,expert-finding). Many researchers have proposed approaches to measure the similarity between concepts by accessing fuzzy domain ontology. However, engineering of the construction of domain ontologies turns out to be labor intensive and tedious. In this paper, we propose an approach to mine domain concepts from Wikipedia Category Network, and to generate the fuzzy relation based on a concept vector extraction method to measure the relatedness between a single term and a concept.
4
Abstract Our methodology can conceptualize domain knowledge by mining Wikipedia Category Network. An empirical experiment is conducted to evaluate the robustness by using TREC dataset. Experiment results show the constructed fuzzy domain ontology derived by proposed approach can discover robust fuzzy domain ontologywith satisfactory accuracy in information retrieval tasks.
5
Introduction The contribution of this paper is to propose an approach to mine fuzzy domain ontology which contains two parts. First, an approach is proposed to conceptualize domain knowledge by using Wikipedia Category Network. Second, fuzzy relation is generated to calculate the semantic relatedness among terms, concepts, and domains. Especially ontology-based systems can be implemented by our fuzzy domain ontology, because domain knowledge is categorized, and a term is mapped to the domain knowledge by using term-domain fuzzy relation. The underlying principles of the proposed approach will be elaborated in the following section
6
Fuzzy on tology generation
7
The purpose of the proposed system is to mine fuzzy domain ontology from Wikipedia Category Network. The fuzzy domain ontology is a representation of domain knowledge which indicates how much a term is related to a domain. Actually, Wikipedia is not only neither a tree-based structure nor a DAG structure (Directed Acyclic Graph), but also the directed graph with cycles. In fact, Wikipedia permits such paradoxes as a category being its own “grandparent” [2]. Ontology Building Stage can handle this kind of conflict of Wikipedia Category Network. The notations in this paper are defined as follow.
8
Fuzzy on tology generation A. Pre-Processing Stage and Wiki Mapping Stage: Pre-Processing Stage retrieves a set of key terms from a set of articles, where each document belongs to one or more domains. Wiki Mapping Stage uses the search engine to map each term to its Wikipedia pages, and the Wikipedia pages are mapped to its Wikipedia categories.
9
Fuzzy on tology generation B. Ontology Building Stage: In this stage, the fuzzy relation is generated to connect Wikipedia categories and predefined domains. First, concept representation finder summarizes several concepts to represent a specific domain, each concept exists a unique concept representation which is a Wikipedia category. Second, fuzzy relation generator has two fuzzy relations that building relationship between Wikipedia categories and domains. Fuzzy relation RW C is represented the semantic relatedness of Wikipedia categories and concepts. Fuzzy relation RC D is represented the semantic relatedness between concepts and domain
10
Empirical experiments and result
13
Conclusion In this paper we propose a fuzzy domain ontology generation methodology which uses concept vector to traverse Wikipedia Category Network for calculating semantic relatedness in the expert-finding system of National Science Council of Taiwan. The proposed fuzzy domain ontology is composed of domain conceptualization and term-domain fuzzy relation generation. The proposed approach can transfer a domain to a set concepts from Wikipedia Category Network, and overcome Wikipedia conflict (Cyclic Graph). The methodology can be used for ontology-based classification problems.
14
References
Similar presentations
© 2024 SlidePlayer.com. Inc.
All rights reserved.