Bees Algorithm And Its Application To Vertical Web Search

Recently several bionic computing methods and intelligent optimization algorithms yielded, which is inspired by the feature of swarm intelligence of bees in nature. These methods are efficient and less dependent on the problems, so it is thoroughly concerned and applied in many field. This work collected some of these methods and algorithms and gives a specific description of three intelligent optimization algoritms in these methods. The three algorithm are Bees Algorithm (BA),Artificial Bee Colony Algorithm (ABC) and The Marriage in Honey-Bees Optimization Algorithm (MBO).Comparing with orther evolutionary algorithm, Bees Algorithm is simple,easy to implement and don't need specific information of the problem, so it is fit for solving the complex optimization problems. However, in the late stage of evolution the convergence speed of BA become slow and it is hard to enhance the accuracy of solution. In this paper, to overcome the problems of bees algorithm (BA), two method to improve the performance of BA is proposed. A dynamic radius bees algorithm (BA-DR) is first presented by dynamically adjusting neighborhood search radius. Then another enhanced algorithm is proposed by combining the chaotic-local-search strategy. The experiment results demonstrate that the proposed algorithm is better in the convergence velocity and precision.Currently general search engine can not meet the demand of professional and accurate information. So vertical search that based on the topic of some field become focus and it is impotant for the design of net crawler. Net crawler is a kind of intelligent software and it can crawl pages in the Web automatically. Recently, several heuristic search strategy have been applied in the research of net crawler. In this paper, as a try of solving the combinatorial optimization problem, A net crawler search agorithm based on the search strategy of BA is proposed and technical details to implement a net crawler system is discussed.