An Ontology Based Model for Document Clustering

An Ontology Based Model for Document Clustering

Sridevi U. K. (Sridevi U. K.Sri Krishna College of Engineering and Technology, India) and Nagaveni N. (Nagaveni N.Coimbatore Institute of Technology, India)
DOI: 10.4018/978-1-4666-2047-6.ch013
OnDemand PDF Download:
List Price: $37.50


Clustering is an important topic to find relevant content from a document collection and it also reduces the search space. The current clustering research emphasizes the development of a more efficient clustering method without considering the domain knowledge and user’s need. In recent years the semantics of documents have been utilized in document clustering. The discussed work focuses on the clustering model where ontology approach is applied. The major challenge is to use the background knowledge in the similarity measure. This paper presents an ontology based annotation of documents and clustering system. The semi-automatic document annotation and concept weighting scheme is used to create an ontology based knowledge base. The Particle Swarm Optimization (PSO) clustering algorithm can be applied to obtain the clustering solution. The accuracy of clustering has been computed before and after combining ontology with Vector Space Model (VSM). The proposed ontology based framework gives improved performance and better clustering compared to the traditional vector space model. The result using ontology was significant and promising.
Chapter Preview


Searching the Web has become more challenging due to the rapid growth in information. The query response time and the scalability issue in the information retrieval can be reduced by cluster retrieval approach. Clustering of retrieved result is used to present more organized result to the user. Clustering simplifies web search engine work by grouping large amount of documents, retrieved according to a given query (Madyloval & Öğüdücü, 2009). Generally, text document clustering methods attempt to group the document based on some topic that is different than those topics represented by the other groups (Frakes & Yates, 1992). For the document clustering, the current keyword-based methodologies tend to be inconsistent and ineffective when the terms are used for cluster analysis. Most of document clustering algorithms use Vector Space Model (VSM) for document representation (Salton, 1989). The vector space model is a widely used data representation for document classification and clustering (Salton & McGill, 1983; Salton, Wong, & Yang, 1975). VSM represents documents as vectors in the space in terms, and measures using inner-products. VSM ignores semantic relations between terms. The research problem of improving relevance in search and ranking of document can be done by considering the semantics of relations. Using ontology as background knowledge can improve document clustering quality with its concept hierarchy knowledge (Zhang et al., 2008).

Ontology based information retrieval matches the relevance of a user generated query against an ontology-based knowledge-base. The ontological information retrieval utilizes the relations between the keywords (Castells, Fernandez, & Vallet, 2007). The vector space model can be combined with the ontology based information retrieval to retrieve the relevant documents (Vallet, Fernande, & Castells, 2005). The efficiency of the information retrieval process can be increased by annotating documents with semantic information.

The motivation of our research is that the terms in the document have multiple meanings. Thus, providing ontology based similarity certainly helps to formulate more effective clustering according to the user’s needs. The objective of the research is to define a model for the ontology based annotation and clustering using particle swarm optimization. In this paper we have defined how to apply ontology based annotation method to improve the clustering of the documents. The quality of the solution obtained can be improved by using annotated weights and optimized clustering algorithm.

The idea of the approach is based on concepts and relations extracted from the documents. Our approach consists of three major steps. First, the concepts and their relations are extracted from the document based on ontology based annotation. Second, ontology based indexing is done on the corpus. Third, the PSO-based clustering algorithm finds the optimized clusters.

The main contributions of this paper are:

  • 1.

    Most of the document clustering approaches do not consider the semantic similarity of terms.

  • 2.

    We propose a PSO based clustering method on the document represented based on semantic similarity.

  • 3.

    We conduct experiments to evaluate the ontology based method and the traditional vector space model. The results show that the ontology based model yields better precision and recall.

In the next section, related literature from the field of ontology based methodologies and clustering using particle swarm algorithm is surveyed. The section following describes the proposed methodology that includes the document annotation, similarity measure and clustering framework. The chapter then presents the system architecture that implements the ontology based model for document clustering. A description of the data sets and experimental results follows. The last section of the chapter presents the conclusion and future work.

Complete Chapter List

Search this Book: