American Journal of Computer Science and Information Engineering 2017; 4(3): 24-30 http://www.aascit.org/journal/ajcsie ISSN: 2381-1110 (Print); ISSN: 2381-1129 (Online) Keywords Search Engine, Pseudo Relevance Feedback, Fuzzy, Recall, Precision Received: April 16, 2017 Accepted: April 27, 2017 Published: July 5, 2017 Automatic Query Expansion for Information Retrieval: A Survey and Problem Definition Olalere A. Abass 1 , Olusegun Folorunso 2 , Babafemi O. Samuel 1 1 Department of Computer Science, Tai Solarin College of Education, Ijebu Ode, Nigeria 2 Department of Computer Science, Federal University of Agriculture, Abeokuta, Nigeria Email address olaabas@gmail.com (O. A. Abass), folorusoo@funaab.edu.ng (O. Folorunso), princefm@live.com (B. O. Samuel) Citation Olalere A. Abass, Olusegun Folorunso, Babafemi O. Samuel. Automatic Query Expansion for Information Retrieval: A Survey and Problem Definition. American Journal of Computer Science and Information Engineering. Vol. 4, No. 3, 2017, pp. 24-30. Abstract An ideal information retrieval system is expected to retrieve only the relevant documents while irrelevant ones are ignored towards ensuring throughput of the retrieval system and reduce the time user spend on the search engines as well as serving a motivation for continue the search. The process of IR consists of locating relevant documents on the basis of user query, such as keywords. One of the most fundamental research questions in information retrieval is how to operationally define the notion of relevance so that we can score a document with respect to a query appropriately. The most critical language issue for retrieval effectiveness is the term mismatch problem because both the indexers and the users do often not use the same words. This scenario is called vocabulary problem. Consequently, IRS users spend much time and resources to obtain their information need after querying the system. One solution to this problem is known as query expansion via pseudo relevance feedback which is intelligent technique for boosting the overall performance in IR. This paper reviews the intelligent method of query expansion and fashion out future work on the implementation of intelligent information retrieval for the purpose of removing “noise” (irrelevant documents) from the lists of retrieved documents. 1. Introduction In the current era of advancement in Information and Communication Technology (ICT), Information Retrieval (IR), a subfield of computer science, has emerged as an important research area that is concerned with the searching and retrieving of knowledge-based information from database [1] and also deals with the representation, storage, and access of information [2]. Hence, IR focuses on the organization and retrieval of information from large database collections [3]. The field of IR has developed along with the field of databases. In the traditional IR model, it is assumed that there exist a large number of documents and data contained in such unstructured documents are without any associated schema. The process of IR consists of locating relevant documents on the basis of user query, such as keywords. The World Wide Web (WWW) provides a convenient way to interact with information sources across the Internet. IR has played a critical role in making the web a productive and useful tool, especially for researchers [4] for the purpose of efficiently retrieving relevant documents. Examples of information retrieval system (IRS) are online library catalogs and online document-management systems like storing newspaper articles. Data