Study of Ontology or Thesaurus Based Document Clustering and Information Retrieval

Abstract

Document clustering generates clusters from the whole document collection automatically and is used in many fields, including data mining and information retrieval. Clustering text data faces a number of new challenges. Among others, the volume of text data, dimensionality, sparsity and complex semantics are the most important ones. These characteristics of text data require clustering techniques to be scalable to large and high dimensional data, and able to handle sparsity and semantics. In the traditional vector space model, the unique words occurring in the document set are used as the features. But because of the synonym problem and the polysemous problem, such a bag of original words cannot represent the content of a document precisely. Most of the existing text clustering methods use clustering techniques which depend only on term strength and document frequency where single terms are used as features for representing the documents and they are treated independently which can be easily applied to non-ontological clustering. To overcome the above issues, this paper makes a survey of recent research done on ontology or thesaurus based document clustering.

Multi-Step Iterative Algorithm for Feature Selection on Dynamic Documents

“…The algorithm was executed with important step of synonym grouping and without this important step. The outcome of basic TF-IDF algorithm is also presented as a basis for comparison and also compared with the algorithm that extends term set by adding synonyms as suggested by Bharathi (Bharathi et al, 2012;Sedding et al, 2004). The resulted entropy in Table 4 shows significant improvement with grouping measure.…”

“…One way to solve this problem is inclusion of synonyms and finding out hidden relationship between the documents. and G. Bharathi et al (2012) have given thought, in which synonyms are included and existing term set prepared by TF-IDF are enriched. This gives slightly improved results on some data sets.…”

A Probe on Document Clustering Methodologies and Its Performance Metrics

“…Ontology approach clusters [19] the input text files based on idea and relations to express learning and goal without semantic vagueness. It centers around the semantics of the language structure which gives quality groups.…”

Fast and reliable inference of semantic clusters

“…Some other studies are nevertheless closer to the scope of our work. Some researchers have for instance studied the impact of integrating knowledge base information in clustering algorithms [7] . To the best of our knowledge, Hotho et al [8][9][10] have been the first to consider this kind of approach.…”