| With the rapid development of the Semantic Web, there are emerging and accumulating much semantic information. However, because of the Semantic Web's distributed nature, data on it will inevitably come from many different ontologies. Information processing across ontologies is not possible without knowing the semantic mappings between their elements. Manually finding such mappings is tendious, error-prone, and clearly not possible at the Web scale. Through ontologies, agents (both human and software) can interact in a (semi) automatic way in order to find information by exploiting semantics. Hence, ontology mapping become one of the pivotal problem in Semantic technology. The development of tools to assist in the ontology mapping process is crucial to the success of the Semantic Web, it is the effective method to data collection and management.Ontology Mapping algorithm is basing on two Ontology as import, then building corresponding semantic relationship for each element(concept,attribute or relationship) of the two Ontology. This paper focuses on the Ontology Mapping problem and in particular it considers ontologies written in RDF or OWL. Even being in some features completely different, approaches to ontology mapping share some common characteristics. For the most part they exploit both linguistic and structural information to find out mappings between URIrefs (Universal Resource Identifier Reference) belonging to two different ontologies (a source ontology and a target ontology). Although structural matching is important, it is basically supported by linguistic matching.In this paper first briefly introduced Ontology and Ontology mapping, and talked the main method of Ontology mapping, and listed several the most popular Ontology mapping methods and Ontology mapping tools in the international. And we explained the mechanism of the Lucene search engine library in details, made a thorough discussion on its index principle. In the back of the chapters, we give emphasize on the search engine based on Lucene library ontology mapping method, known as the LselOM (Lucene search engine library Ontology Mapping). At last, it was evaluated by a series of experiments on its performance.It makes use of the feature and the character of the Lucene, firstly, parsing the source ontology; then constructing an index for the Source Ontology, which composes Lucene Document and many different types of information about URIrefs. We use the URIrefs of the Target Ontology as the query against the index of the Source Ontology to get the Mapping; lastly, exporting the result to the Lucene Result Set.During the searching process, LselOM combines two linguistic matching approaches: semantic-based matching method and string-based matching method. This paper uses the I-Sub Algorithm to achieve the string-based matching. And using the feature of the Lucene, we combine the WordNet synonym engine to write the synonym of the source ontology into the index. Then we make use of the similarity to evaluator the success or no of the searching. If the similarity is higher than a set score value, then we consider the mission is successful. This paper use the feature of the Lucene Mechanism score, making some improve to it, we add some factors to make the score being more abroad and more universality. In this paper, we have made a series of experiments. Through these experiments, we find the most proper score value, and make the value of each factor of the improved Lucene Mechanism score. The Experiment result indicates that, this approach is compared to the other approaches, it has some advantages in the Precision, Recall, F-Measure and the execute time. And its complexity is almost hidden to the users. |
PDF Full Text Request