Text Similarity Approach based on Semantic Networks and Words Description
Abstract
Finding similarity in texts is important in many areas such as information retrieval, document compilation, word clarification, automatic recording of articles, and classification of short answers, machine translation, and text summarization. Calculating sentence similarity is not an easy task due to the differences and variations in natural language. Methods for calculating the similarity between texts depend on the semantic or grammatical aspects. This paper discusses an approach for calculating the similarity between two sentences using semantic networks and word description. The text represented by a semantic network, which is composed of nodes and relationships. Similarity found by taking into consideration the description of the word in terms of its meaning and definition, and the grammatical aspects by measuring the similarity of parts of speech between the sentences.