International Research Journal of Engineering and Technology (IRJET) e-ISSN: 2395 -0056 Volume: 04 Issue: 01 | Jan -2017 www.irjet.net p-ISSN: 2395-0072 © 2017, IRJET | Impact Factor value: 5.181 | ISO 9001:2008 Certified Journal | Page 156 SEMANTIC SIMILARITY BETWEEN SENTENCES PANTULKAR SRAVANTHI 1 , DR. B. SRINIVASU 2 1 M.tech Scholar Dept. of Computer Science and Engineering, Stanley College of Engineering and Technology for Women, Telangana- Hyderabad, India 2 Associate Professor - Dept. of Computer Science and Engineering, Stanley College of Engineering and Technology for Women, Telangana- Hyderabad, India ---------------------------------------------------------------------***--------------------------------------------------------------------- Abstract - The task of measuring sentence similarity is defined as determining how similar the meanings of two sentences are. Computing sentence similarity is not a trivial task, due to the variability of natural language - expressions. Measuring semantic similarity of sentences is closely related to semantic similarity between words. It makes a relationship between a word and the sentence through their meanings. The intention is to enhance the concepts of semantics over the syntactic measures that are able to categorize the pair of sentences effectively. Semantic similarity plays a vital role in Natural language processing, Informational Retrieval, Text Mining, Q & A systems, text-related research and application area. Traditional similarity measures are based on the syntactic features and other path based measures. In this project, we evaluated and tested three different semantic similarity approaches like cosine similarity, path based approach (wu – palmer and shortest path based), and feature based approach. Our proposed approaches exploits preprocessing of pair of sentences which identifies the bag of words and then applying the similarity measures like cosine similarity, path based similarity measures. In our approach the main contributions are comparison of existing similarity measures and feature based measure based on Wordnet. In feature based approach we perform the tagging and lemmatization and generates the similarity score based on the nouns and verbs. We evaluate our project output by comparing the existing measures based on different thresholds and comparison between three approaches. Finally we conclude that feature based measure generates better semantic score. Key Words: WordNet, Path based similarity, Features based Similarity, Word Overlap, Cosine similarity, Word order similarity, Semantic similarity. 1. INTRODUCTION Sentence similarity measures are becoming increasingly more important in text-related research and other application areas. Some dictionary-based measures to capture the semantic similarity between two sentences, which is heavily based on the WordNet semantic dictionary [1].Sentence similarity is one of the core elements of Natural Language Processing (NLP) tasks such as Recognizing Textual Entailment (RTE)[2] and Paraphrase Recognition[3]. Given two sentences, the task of measuring sentence similarity is defined as determining how similar the meaning of two sentences is. The higher the score, the more similar the meaning of the two sentences. WordNet and similarity measures play an important role in sentence level similarity than document level[4]. 1.1 Problem Description Determining the similarity between sentences is one of the crucial tasks in natural language processing (NLP). To estimate the accurate score generated from syntactic similarity to semantic similarity. Computing sentence similarity is not a trivial task, due to the variability of natural language expressions. Measuring semantic similarity of sentences is closely related to semantic similarity between words. In information retrieval, similarity measure is used to assign a ranking score between a query and texts in a corpus [5]. 1.2 Basics and background knowledge In the background we have defined the basic definitions and different strategies that can be used. 1.2.1 WordNet WordNet is the product of a research project at Princeton University. It is a large lexical database of English. In WordNet nouns, verbs, adverbs and adjectives are organized by a variety of semantic relations into synonym sets (synsets), which represent one concept. Examples of relations are synonymy, autonomy, hyponymy, member, similar, domain and cause and so on. In this paper, we are only concerned about the similarity measure based on nouns and synonym relation of WordNet. 1.2.2 Semantic Similarity The semantic similarity sometimes called as topological similarity. Semantic similarity is calculated at document level, term level and sentence level. The document and sentence level is calculated based on the terms which