Search

This study introduced a novel algorithm to compute similarities between natural languages. Using syntactical relationships derived from natural languages, the algorithm proposed a semantic pattern model and quantified natural languages using the word similarity technology based on WordNet and the Stanford Parser. Traditional IR technologies may not always determine the perfect matching without obvious relation or concept overlap between two natural language sentences. Some approaches deal with this problem via determining the order of words and the evaluation of semantic vectors; however, they were hard to be applied to compare the sentences with complex syntax as well as long sentences and sentences with arbitrary patterns and grammars. The proposed approach takes advantage of corpus-based ontology and grammatical rules to overcome this problem. The experimental results indicated that the algorithm could yield optimal results in semantic recognition when applied to sentences or short texts that are grammatically complex.

Traditionally information retrieval consists mainly of determining which documents of a collection contain the keywords in the user query. However, a growing number of tasks, especially those related to Semantic Web technologies and applications rely on accurately measuring the similarity between documents and online texts. Instead of giving the absolute similarity degree of two documents, this paper presents a semantic corpus and Formal Concept Analysis-based procedure to build the concept map for a given set of documents and quantify the semantic relations between the concepts. The proposed approach includes three algorithms - a) the concept lattice constructing algorithm, b) the concept similarity measure, and c) the sub-lattice similarity measure.