Fast Information Retrieval from Web Pages HAZEM M. EL-BAKRY Faculty of Computer Science & Information Systems, Mansoura University, EGYPT E-mail: helbakry20@yahoo.com NIKOS MASTORAKIS Department of Computer Science, Military Institutions of University Education (MIUE) -Hellenic Naval Academy, Greece Abstract: In this paper, a new fast algorithm for information retrieval is presented. Such algorithm relies on performing cross correlation in the frequency domain between input data and the input weights of fast neural networks (FNNs). It is proved mathematically and practically that the number of computation steps required for the presented FNNs is less than that needed by conventional neural networks (CNNs). The main objective of Internet users is to find the required information with high efficiency and effectiveness. Finding information on an object’s visual features is useful when specific keywords for the object are not known. Since intelligent mobile agent technology is expected to be a promising technology for information retrieval, there is a number of intelligent mobile agent based-information retrieval approaches have been proposed in recent years. Here, the work presented in [25] for image-based information retrieval using mobile agents is greatly enhanced. Multiple information agents continuously traverse the Internet and collect images that are subsequently indexed based on image information such as the URL location, size, type and the date of indexation. In the search phase, the intelligent mobile agent receives the image of object as a query and searches the set of web pages that contain information about the object. This is done by matching the query to images on web pages faster than the work presented in [25]. Keywords: Fast information retrieval, Content-based image retrieval, Image clustering, and Intelligent mobile agent I. Introduction With the development of Internet technology, the fast growing World Wide Web has become one of the most important sources of information and knowledge. When searching the Web, a user can be overwhelmed by thousands of results retrieved by a search engine, of which few are valuable. The problem for search engines is not only to find topic relevant results, but results consistent with the user’s information need. How to retrieve desired information from the Internet with high efficiency and good effectiveness is become the main concern of internet user-based [1]. Users generally find information using search engines. The input to these search engines often consists of keywords or other written text like a question. But the Web contains not only text, but also information in other modalities such as images. They are however not yet being used as input for general Web searches. When the user wants to query about an object that has unique visual features, the image of the object that represents the visual properties can be used as a query. By enabling searches on object appearance provide users with a new, more convenient and direct means for finding information about objects encountered in everyday life. Many existing world wide search engines, for example Google and Yahoo, rely on meta-data (annotations) or the context in which the image is found and the query is performed using text-based information retrieval method, the matching is performed based on image captions or file names, thus the performance of image retrieval is based on the similarity between the user’s text query and image text annotation not on the image content itself [2]. Content Based Image Retrieval (CBIR) searches for images by using asset of visual features such as color, shape and texture, are extracted from the images that characterize the image content [3]. These techniques have been used in many areas, such as geographic information system, biomedical image processing and digital libraries [4]. One of the main advantages of the CBIR approach is the possibility of an automatic retrieval process, instead of the traditional keyword- based approach, which usually requires very difficult and time-consuming previous annotation of database images. Comparatively speaking, CBIR more objectively reflects the content of images [6]. Most of CBIR systems are only operate on a local demo database of a few thousand images stored at the host web site [7]. Unlike image retrieval from a fixed database, where each image is treated as an independent object, for Searching for image on the Web basically comes down to locating an appropriate Web page and to retrieve relevant information about the image from that site[8,9]. Intelligent mobile agent is a program that traverses the Internet, moves to different sites with different characteristics, searches for the desired information, and returns the search report to the user query [10,11]. When large quantities of data are stored at distributed sites, moving the computations to the data is a more realistic and feasible approach, compared with migrating data to the computations. In other words; instead of gathering information from Proc. of the 7th WSEAS Int. Conf. on COMPUTATIONAL INTELLIGENCE, MAN-MACHINE SYSTEMS and CYBERNETICS (CIMMACS '08) ISSN: 1790-5117 229 ISBN: 978-960-474-049-9