Article Preview
Top1. Introduction
The intelligent real-time system design needs to incorporate an autonomic feature selection in their runs to achieve the unexpected criticalities of systems and its environment (Nair & Persya, 2015).
With the growing number of information sources available via the internet, the problem of how to combine distributed and heterogeneous information sources is becoming increasingly difficult (YIGAL et al., 1993). Enabling seamless integration of multiple data sources while retrieving web data, an efficient web search mechanism that fulfils the customer requirement is always needed (Bagrudeen, 2016). Large area, high bandwidth network such as World Wide Web creates a number of opportunities and challenges for effective retrieval of information. Such a network makes users to gain access to huge amounts of data in a wide variety of types. In realistic setting, without effective information management tools most of this data is worthless since users are unable to find data of interest (Knoblock & Ambite, 1997; Knoblock, 1996). Today, Google's search engine is merely the front end to its advertising business and the more it knows about us, the more easily it can target specific advertisements when we search (Shroff, 2013). The available source of information in World Wide Web is classified domains into particular topics, including: health, travel, shopping, etc. The classical contents of web, such as URL, title, headers, and body; as well as, the classical ranking, such as global rank, local rank, insight rank, knowledge bases rank, and so forth are important for making information in its distributed resources are comparable. In recent years, traditional approaches to building distributed or aggregated systems do not scale well (Lenat & Guha, 1990). Current systems e.g. search services on the World Wide Web provide limited capabilities for locating, combining, processing, and organizing information. The advent of large area networks connecting many diverse repositories of data creates obstacles in finding a particular data easily. More contrast, providing access to the large number of information sources and organizing them into a network of information agents is a big challenge. Moreover, due to the amount and heterogeneity of the data, it is challenging to perform data analysis directly; especially when the data is captured from a large number of distributed sources (Zhou et al., 2016). Some agents provide expertise on a specific topic and sometimes drawing on relevant information from other information agents. To build such topographical network, we need an infrastructure of a single agent system that can be instantiated to provide accessing to multiple agents. Information mediator that provides access to heterogeneous data and knowledge base is crucial (Al-akashi, 2014). We need to consider a unique aspect that is critical for any agent-based system: how to draw and simulate knowledge in the network and how to categorize or classify data in the network based on their topic similarity, and finally, how to combining them. Formulating the results retrieved from the distributed agents and computing the federated rank based on that aspect is the last goal of our approach.