International Journal of Computer Applications (0975 – 8887) Volume 184– No.34, October 2022 1 Development of Hybrid Intelligent based Information Retreival Technique Gregory Gabriel James Department of Computer Science, Rhema University, Aba, Abia State, Nigeria Abugor Ejaita Okpako Department of CyberSecurity, Faculty of Computing, University of Delta, Agbor, Delta State, Nigeria C. Ituma Department of Computer Science, Ebonyi State University, Abakili, Ebonyi State,Nigeria J.E. Asuquo Department of Chemistry, University of Uyo, Nigeria ABSTRACT To find information over the internet to a certain level, depends on our capacity to track all related subjects and classify them into bunches of comparative themes. As the domain of information is enlarging over the internet , the time consumption and the difficulties experienced by researchers to find a relevant material that meets the user’s specified request increases, thereby putting the researchers into a state of dilemma at the cause of searching for relevant information that meets their need. The pursuit to trim down the challenges of impasse faced by researchers as well as time exhausted to filter relevant materials in the pools of irrelevant materials have motivated this research. The work aims at developing a Neuro- fuzzy intelligent search framework for tracking and recovery of web archives. The method used was Object-Oriented analysis and Design (OOAD). A hybrid intelligent framework – based tracking system was utilized as the finest choice for tracking archives, since the shortcomings of Neural Network and Fuzzy Logic based tracking system were complemented while their individual qualities are upgraded. This paper expands prior Fuzzy-based information retrieval approaches through increasing the Fuzzy variables and their linguistic values by utilizing distinctive rules and functions that characterized the record. The mapping of input to output parameters was achieved by applying the triangular membership’s functions. Adaptive neural fuzzy inference system model also utilized the Takagi Sugeno inference mechanism. It was observed that using ANFIS improved the hybrid intelligent framework – based tracking system performance slightly with 0.22641 representing 22.64% over the Fuzzy Inference System (FIS) results, thereby guarantee retrieval of most relevant documents that met the user’s request. Keywords Intelligence, ANFIS model, Neuro-fuzzy, Geno Method 1. INTRODUCTION Finding materials in the internet is constantly a complicated job among researchers. Generally, most researchers travel a very long way to gather required materials to meet their academic needs and this makes research quit expensive. Information technology changed the practice and brought about intelligent search methods. This intelligent search methods use the internet in facilitating the information search. The paper introduced an intelligent based paradigm for document tracking and retrieval. The classification of documents into similar topics and specific knowledge groups was done using neuro- fuzzy clustering method. The reason for utilizing this strategy is to construct a versatile intelligent data recovery framework. The framework will cluster web records into specific knowledge groups and similar themes by applying unsupervised machine learning procedure which decreases the rate of unimportant records recovered and displayed. 2. ANALYSIS OF THE PROPOSED SYSTEM This new method is based on Neuro-Fuzzy approach which enhances the search for document. The merging of two advance technologies to form a single hybrid brilliant model makes the systems the leading technology for archive tracking and recovery. The reason for embracing the Neuro-Fuzzy strategy is to plan a versatile intelligent data recovery framework that utilized unsupervised learning procedures in diminishing the rate in which insignificant reports are recovered and displayed. 2.1 Architecture of the System The Figure 1 shows the model of the proposed NFDTRS: 2.2 Conceptual Designs for the Software The framework of this intelligent model in figure 1 is an improved work of [6]. The above figure is made up of different elements of the entire system with their functions clearly stated. The framework comprises the following components: 1. Search Interface 2. Database Model Fig. 1: The Proposed Neuro-Fuzzy Architecture User’s Query SEARCH INTERFACE Tracking Query Analyzer/Feature Extraction Crawler Database Web Indexer Document Classification & Retrieval ANFIS Fuzzy Inference Engine Neural Network Training Search Output