795 Research Article International Journal of Current Engineering and Technology ISSN 2277 - 4106 © 2013 INPRESSCO. All Rights Reserved. Available at http://inpressco.com/category/ijcet Performance and Effectiveness Examination of the IQE and AQE with Application on Arabic Content Abdel Salam Obeidat a , Sameh Ghwanmeh a* , Ali Al-Ibrahim a , Nidhal K. El-Omari a and Adel Mohammad a a WISE University, Faculty of Information Technology, Amman, Jordan Accepted 01 May 2013, Available online 01 August 2013, Vol.3, No.3 (August 2013) Abstract Literature search show that information retrieval (IR) systems on Arabic are little compared with English language. Additionally, IR systems face many problems when used with Arabic language, including, complexity and ambiguity. The performance and effectiveness of interactive query expansion (IQE) and automatic query expansion (AQE) represent a key towards improved IR systems. The performance and effectiveness of IQE compared with AQE have been examined via a series of search experiments on Arabic content. Compared with no query expansion, the experimental results showed that AQE provides enhanced performance and effectiveness, with 54% query improvement and average precision of 42.1. However, results revealed that IQE provides high performance and effectiveness compared with AQE, with 84% query improvement and average precision of 43.4. Keywords: Arabic language, Arabic content, evaluation of information retrieval, query expansion, IQE, AQE. 1. Introduction 1 Searching and Retrieving Information is defined as the process of finding the relevant documents in a specific database based on the user requests. Statistical and other common methods have been used to implement the information retrieval (IR) systems. Usually, such IR systems contain select terms (phrases and others) from the searched documents and an indexed file to enable easy access to documents. The search process can be effective if the searched results contain the maximum number of related documents and minimum number of non-related documents (Hani. S et al,2006; Chang, Y.C et al,1997). Adding additional query terms to existing query is called a query expansion and it is used to enhance the performance and correctness of the search process. This can be achieved by employing the user’s related information created from the user assessment to the relevant documents. The expansion terms can be determined and ranked from those documents (Strzalkowski T. et al,1998). The created expansion terms can be used in the query by IQE or AQE (Al-Kharashi et al,1994). It is the user decision to select the appropriate terms in the expansion query (Robertson E, 1990; Fowkes, H et al,2000). Arabic Language is a common language in the world. IR applications on Arabic are little compared with English language IR applications. Additionally, there is a shortage in the large test databases deals with Arabic content. IR *Corresponding author: Sameh Ghwanmeh systems face many problems when used with Arabic language, including, Orthographic variations, complexity, broken plurals, ambiguity, short vowels (Moukdad, H., 2001; Ruthven, I., 2006; Beaulieu, M,2004). In this paper, the performance and effectiveness of IQE and AQE have been examined. A chain of experiments were carried out using 242 Arabic abstracts from the Saudi Arabian National Computer Conference. The experiments have been conducted to provide a clear comparison between AQE and IQE techniques. Performance evaluation process has been performed to reveal the best value of n in AQE that gives the optimal value of average precision for the whole query. 2. Research Methodology The research experiments were carried out on the Arabic collection, details of which are given in Table 1. For each query, the top 10 retrieved documents are used to offer a list of probable expansion terms. The wpq method (Al- Kharashi,1994; Efthimiadis, E., 1999) of ranking terms for query expansion has been employed in this research; this has been shown to provide acceptable results for both AQE and IQE as explained in (Strzalkowski, T,1998, Noaman, A,2012; Jinxi, X.,2002). As the main objective of this research is to measure the performance and effectiveness of the IQE and AQE based on Arabic documents, the research methodology employs the algorithm presented in (Al-Kharashi,1994) for each query. The recall and precision values are calculated using a full-freezing method which is a standard method of