Article Preview
TopIntroduction
Digitization of today era has suddenly increased the amount of data, and to mange this unstructured and semi-structured data is the need of the hour. Data which is constantly generating from social sites and search engines requires proper maintenance and it evoke a problem for the researches to maintain it properly. As the size of data on the web is increasing at a fast rate, in the same way the users are also increasing which require efficient and accurate data in minimum time duration. On the web, there is a great challenge to maintain the up-to-dateness of data so that the exact information can be retrieved easily. Most of the users on the web are lay persons who do not possess correct knowledge to formulate their query so that the correct information can be obtained. The information stored should be properly indexed and ranked so that the user can get the result of their query in limited time constraint.
With the increased amount of documents on the web the researchers are faced with challenge to locate their document of interest. Different algorithms are being proposed to accomplish this task. It has been found out that millions of web pages are added daily and this explosive growth of web pages place a huge pressure on the search engines which uses different algorithms for classifying and retrieving the much needed information in stipulated time duration. In spite of various algorithms present, still the problem persists for accessing exact information which should be highly relevant in the interest of the users.
The rapid growth of information on the web brings the work of search engines into limelight. They are almost used by billions of people daily and regarded as one of the most widely implementation of information retrieval technique (Lewandowski, 2005). The web pages are interacted by various numbers of users to quench their thirst for information and most of the time is spend by the user for searching the relevant data.
For accessing information from the World Wide Web, search engines normally employs content perspective where they figure the content of the page and link perspective where the interconnectivity of the pages are considered. For ranking information various factors are considered by the search engines but the most common among them is the query-independent and query-dependent approach. On one hand query-dependent take into consideration the content of the document, language, anchoring point and other parameters to retrieve information, while query-independent measure consider the link as the major factor for accessing the popularity of the site apart from length of the document, current document, filetype etc.
For solving computer related problems evolutionary computational models plays a leading role. They are used in many learning tasks and in many cases are considered as optimization technique for solving the problems. It is not only used in matching function of Information Retrieval but can also be used for clustering, indexing, classification and searching task (Cordon et al., 2003). The development of various approaches help in retrieving the information efficiently and in this regard nature-inspired algorithms like swarm intelligence, genetic algorithm play a major role. These various optimization methods help to retrieve resources fully by capturing the behavior of swarm of insects and animals (Abdullah and Hadi, 2014).
The structure of the paper is arranged as follows: Section 2 provides details about the related work of using Artificial Bee Colony in ranking pages. Information Retrieval process and Web Mining Process is discussed in Section 3 and Section 4. Ranking Algorithms are covered in Section 5 and Section 6 provides detail about Nature-Inspired Algorithms. Section 7 give an introduction of Artificial Bee Colony and Section 8 presents the working of the proposed approach the BeeRank algorithm for web page ranking. Section 9 is Experimental results, Section 10 is the Result Analysis part and finally Section 11 is the conclusion part.