The requirements for effective search and management of the WWW are stronger than ever. Currently Web documents are classified based on their content not taking into account the fact that these documents are connected to each other by links. We claim that a page’s classification is enriched by the detection of its incoming links’ semantics. This would enable effective browsing and enhance the validity of search results in the WWW context. Another aspect that is underaddressed and strictly related to the tasks of browsing and searching is the similarity of documents at the semantic level. The above observations lead us to the adoption of a hierarchy of concepts (ontology) and a thesaurus to exploit links and provide a better characterization of Web documents. The enhancement of document characterization makes operations such as clustering and labeling very interesting. To this end, we devised a system called THESUS. The system deals with an initial sets of Web documents, extracts keywords from all pages’ incoming links, and converts them to semantics by mapping them to a domain’s ontology. Then a clustering algorithm is applied to discover groups of Web documents. The effectiveness of the clustering process is based on the use of a novel similarity measure between documents characterized by sets of terms. Web documents are organized into thematic subsets based on their semantics. The subsets are then labeled, thereby enabling easier management (browsing, searching, querying) of the Web. In this article, we detail the process of this system and give an experimental analysis of its results.

http://www.db-net.aueb.gr/files/HNVV03_VLDBJ.pdf

THESUS: Organizing Web document collections based on link semantics

In order to overcome this problem, we propose, in this paper, a new similarity measure giving realistic results and closer relations to reality for concepts not located in the same path. The paper presents a similarity measure that is suited for comparing concepts in ontology. Although finding similar concepts is a core task in the area of ontology alignment/merging [5][6]. The proposed measure can be adopted effectively in this field.

/pdf/a-new-similarity-measure-based-on-edge-counting-1pen8pu6w0.pdf

A New Similarity Measure Based On Edge Counting

This paper introduces a new similarity measure based on edge counting in a taxonomy like WorldNet or Ontology. Measurement of similarity between text segments or concepts is very useful for many applications like information retrieval, ontology matching, text mining, and question answering and so on. Several measures have been developed for measuring similarity between two concepts: out of these we see that the measure given by Wu and Palmer [1] is simple, and gives good performance. Our measure is based on their measure but strengthens it. Wu and Palmer [1] measure has a disadvantage that it does not consider how far the concepts are semantically. In our measure we include the shortest path between the concepts and the depth of whole taxonomy together with the distances used in Wu and Palmer [1]. Also the measure has following disadvantage i.e. in some situations, the similarity of two elements of an IS-A ontology contained in the neighborhood exceeds the similarity value of two elements contained in the same hierarchy. Our measure introduces a penalization factor for this case based upon shortest length between the concepts and depth of whole taxonomy.

A New Similarity Measure for Taxonomy Based on Edge Counting

Resume: Le domaine de l’identification de la similarite a ete considere comme un sujet de recherche fortement recommande dans les domaines du Web semantique, de l’intelligence artificielle et de la litterature linguistique. Dans le domaine du Web semantique ou les ontologies interviennent pour la modelisation des connaissances, la mesure de [Wup 94] a l’avantage d’etre simple a implementer et d’avoir aussi de bonnes performances comparativement a d’autres mesures de similarite [Lin 98]. Neanmoins, cette mesure presente l’inconvenient suivant : dans certaines situations, la valeur de similarite de deux elements d’une ontologie contenus dans le voisinage depasse la valeur de similarite de deux elements contenus dans la meme hierarchie. Cette situation est inadequate dans le cadre de la recherche de l’information. Afin de pallier a ce probleme, nous proposons une nouvelle mesure de similarite en se basant sur la mesure de Wu et Palmer. Nous avons applique notre mesure sur une ontologie de domaine pedagogique. Les resultats obtenus montrent que notre mesure offre un temps d’execution presentant un gain de temps par rapport a l’approche de Wu et Palmer. En plus, nous avons compare la pertinence des valeurs de similarite obtenues, il en ressort que notre mesure est avantageuse et offre un apport pour la resolution du probleme cite ci-dessus.

/pdf/une-extension-de-mesure-de-similarite-entre-les-concepts-d-ibtlkr01v8.pdf

Une extension de mesure de similarité entre les concepts d'une ontologie

This paper introduces a new similarity measure based on edge counting in a taxonomy like WorldNet or Ontology. Measurement of similarity between text segments or concepts is very useful for many applications like information retrieval, ontology matching, text mining, and question answering and so on. Several measures have been developed for measuring similarity between two concepts: out of these we see that the measure given by Wu and Palmer [1] is simple, and gives good performance. Our measure is
based on their measure but strengthens it. Wu and Palmer [1] measure has a disadvantage that it does not
consider how far the concepts are semantically. In our measure we include the shortest path between the
concepts and the depth of whole taxonomy together with the distances used in Wu and Palmer [1]. Also the
measure has following disadvantage i.e. in some situations, the similarity of two elements of an IS-A ontology contained in the neighbourhood exceeds the similarity value of two elements contained in the same hierarchy. Our measure introduces a penalization factor for this case based upon shortest length between the concepts and depth of whole taxonomy.

/pdf/a-new-similarity-measure-for-taxonomy-based-on-edge-counting-vhb2lzn5kf.pdf

A New Similarity Measure for Taxonomy based on Edge Counting

The purpose of this paper is to define and investigate document retrieval strategies based on rough sets. A subject classification index is modelled by means of a family of set-theoretic information systems. The family induces a power set hierarchy built from the set of keywords. In the paper, the task of information retrieval in response to a query that refers to several levels of that hierarchy is defined. Retrieval strategies are proposed and discussed. The strategies are, roughly speaking, based on various types of similarity between sets of keywords included in a query and in a document. It is not necessarily required that the document matches the query; we allow for approximate matching. Measures of relevance of documents to a query are proposed for the given strategies.

Paul Siemens

Papers

A rough set model of information retrieval