International Journal of Computer Applications (0975 – 8887) Volume 56– No.13, October 2012 37 Web Information Retrieval using WordNet Jyotsna Gharat Asst. Professor, Xavier Institute of Engineering, Mumbai, India Jayant Gadge Asst. Professor, Thadomal Shahani Engineering College Mumbai, India ABSTRACT Information retrieval (IR) is the area of study concerned with searching documents or information within documents. The user describes information needs with a query which consists of a number of words. Finding weight of a query term is useful to determine the importance of a query. Calculating term importance is fundamental aspect of most information retrieval approaches and it is traditionally determined through Term Frequency -Inverse Document Frequency (IDF). This paper proposes a new term weighting technique called concept-based term weighting (CBW) to give a weight for each query term to determine its significance by using WordNet Ontology. General Terms Term frequency (TF), Inverse Document Frequency (IDF), Vector Space Model, Extraction Algorithm. Keywords Information Retrieval (IR), Part of Speech (POS), WordNet, Ontology, Concept-based Term Weighting (CBW). 1. INTRODUCTION The purpose of information retrieval is to provide information that changes the knowledge state of a user so that this user is better able to perform a present task. An information retrieval process begins when a user enters a query into the system. The information retrieval system compares the query with documents in the collection and returns the documents that are likely to satisfy the user’s information requirements. In information retrieval a query does not uniquely identify a single object in the collection. Instead, several objects may match the query, perhaps with different degrees of relevancy. Most IR systems compute a numeric score on how well each object in the database match the query, and rank the objects according to this value. The top ranking objects are then shown to the user. The process may then be iterated if the user wishes to refine the query. Goal of IR is to find documents relevant to an information need from a large document set. Web search engines are the most familiar example of IR systems. Knowledge representation and procedures for processing such knowledge/information [10] are major issues while dealing with information retrieval system. A fundamental weakness of current information retrieval method is that the vocabulary that searchers use is often not the same as the one by which the information has been indexed. Most of the existing textual information retrieval approaches depend on a lexical match between words in user’s requests and words in target objects. WordNet [1, 5, 7 and 8] is a lexical database which is available online and provides a large repository of English lexical items. WordNet is a machine-readable dictionary developed by George A. Miller et al. at Princeton University. In WordNet nouns, verbs, adjectives and adverbs are grouped into sets of cognitive synonyms (synsets), each expressing a distinct concept. Synsets are interlinked by means of conceptual- semantic and lexical relations. The resulting network of meaningfully related words and concepts can be navigated with the browser. WordNet can also be used for Query Expansion [3]. In proposed method WordNet is utilized to get conceptual information of each word in the given query context. Based on the extracted concepts proposed method can find the weight of a query. Then this is compared with commonly used Vector Space Model using Term-Frequency, Inverse Document Frequency (TF-IDF). The remainder of this paper is organized as follows: Section 2 introduces common approach to find weight of a query. Section 3 discusses proposed method with the help of system architecture. Experiment result is reported in section 4. Finally a conclusion regarding the idea is made in section 5. 2. COMMON APPROACH Three classic framework models have been used in the process of retrieving information: Boolean, Vector Space and Probabilistic. Boolean model matches query with precise semantics in the document collection by Boolean operations with operators AND, OR, NOT. It predicts either relevancy or non-relevancy of each document, leading to the disadvantage of retrieving very few or very large documents. The Boolean model is the lightest model having inability of partial matching which leads to poor performance in retrieval of information. Because of its Boolean nature, results may be tides, missing partial matching, while on the contrary, vector space model, considering term-frequency, inverse document frequency measures, achieves utmost relevancy in retrieving documents in information retrieval. The drawback of binary weight assignments in Boolean model is remediated in the Vector Space Model which projects a framework in which partial matching is possible. Vector space model is introduced by G. Salton in late 1960s in which partial matching is possible. TF- IDF [6] is a traditional approach which is used to find the term importance by finding weight of a term. Steps to find weight of a query using vector space model are as shown in fig 1. 1. Remove punctuation & numbers from web pages. 2. Remove stopwords. 3. Apply Porter stemming algorithm [9]. 4. Calculate term frequency (TF) of each term (q) within a query (Q) from document. 5. Calculate Inverse Document Frequency (IDF) of each term in the query (Q). 6. Compute TF-IDF of each term of query using equations (1) and (2).