PERFORMANCE EVALUATION of SEMANTIC REASONERS

Abstract-

As the performance of semantic reasoners change significantly with respect to all included characteristics, and therefore requires assessment and evaluation before selecting an appropriate reasoner for a given application. There are number of inference engines like Pellet, FaCT++, Hermit, RacerPro, KaON2, F-OWL and BaseVISor. Some of them are reviewed and tested for few prebuilt ontologies. This paper proposes performance evaluation and comparison of semantic reasoner for the ontologies of Health and Anatomy domain. Reasoners are characterized based on reasoning method, reasoning algorithm, computational complexity, classification, scalability, query and rule support.

 

Read Full paper: west2013_submission_45 (404 KB)

Classifying Problem in Inference Engine for Different Version Based Ontology

An ability of reasoning capability of inference engines a useful to derive new & useful information from existing knowledge-base. Classifying algorithms is use to classifying an ontology which improves quality of search on web. Currently Inference engines are able to classify the small ontology completely. For Large and complex version based ontology size is been increase
practically. The main aim of paper is necessary to evaluating the performance of inference engine which focus on classification parameter for large and complex version based ontology for
different domain. Result might be useful to select inference engine for practically on version based ontology for different domain.

View full paper: IJCSI_Classifying Problem in Inference Engine for Different Version Based Ontology

PERFORMANCE EVALUATION of SEMANTIC REASONERS

Abstract- As the performance of semantic reasoners change significantly with respect to all included characteristics, and therefore requires assessment and evaluation before selecting an appropriate reasoner for a given application. There are number of inference engines like Pellet, FaCT++, Hermit, RacerPro, KaON2, F-OWL and BaseVISor. Some of them are reviewed and tested for few pre-built ontologies. This paper proposes performance evaluation and comparison of semantic reasoner for the ontologies of Health and Anatomy domain. Reasoners are characterized based on reasoning method, reasoning algorithm, computational complexity, classification, scalability, query and rule support.

View full paper: west2013_submission_45

Performance Evaluation of Inference Engine in Static and Changeable Environment

The semantic web has brought many challenges for knowledge representation and inference system [2].Inference Engine play important role to extract additional information implicitly by using fact and ontologies [3]. Standard benchmark practices are used to analyze the performance of different inference engine for different version of ontology. This paper aims to execute different parameters for variety of inference engines and generate statistics based on suitability of inference engine with respect to domain under consideration. The results may be useful in choosing the inference engine for different version of ontology and domain.

 

View Full Paper: new Performance Evaluation of Inference Engine in Static and Changeable Environment (size 717 KB)

Optimize visual image search for semantic web

Abstract: In today’s web image search engines find more irrelevances in the search result. By adding semantic meaning to the document this irrelevance can be reduced. SIEVE image search algorithm combine the text based and content based method and shows the result. Also “IN-Picture” search algorithm mixing the images higher level and lower level contents. In this paper it shows some image searching framework like “SAFE” describe how image are searched using its attributes. Also describes some semantic web technology, which helps in image search and shows how detailed indexing system can use SPARQL query and ontology of an image to build semantic web based framework.

View Full Paper : ijcsit_survey optimize visual image search for semantic web (size 375 KB)

 

Optimizing Query execution over Linked Data

Abstract Linked open data also known as web of linked data is a globally distributed database. The four fold increase in the use of linked open data shows its need in the future. This linked data can be queried with SPARQL protocol and RDF query language also known as SPARQL. Various optimization techniques have been proposed but just a couple implemented as yet. In this paper two most feasible query optimization methods are proposed. Solution one is applying federation and query rewriting. Solution two is expressing the RDF structure as a context graph and pruning intermediate results based on probability or selectivity of the result. The paper gives a brief understanding for both solutions.

View Full Paper: IJETAE_Optimizing Query execution over Linked Data (size 354 KB)

SEMANTIC WEB and COMPARATIVE ANALYSIS of INFERENCE ENGINES

Abstract Semantic Web is an emerging technology for efficient reasoning support over the knowledge represented on the Web. This paper presents the semantic web standards and survey a number of Inference Engines that supports reasoning with OWL. Also analyzed the reasoner with set of ontologies and based on supported features.

View Full paper: Survey Paper on Semantic Web and Inference Engine  (Size 567KB)