Enhancing the Coherence of a Meta Search Engine using Genetic Algorithm

A meta search engine is tool that retrieves information by integrating results from several search engines. The results are put through a ranking process defined by the meta search engine and they are merged into a single consolidated rank list. Result aggregation and ranking process are crucial phases which are important factors for comparison between any two meta search engines. It is a crucial procedure that influences a meta search engine's performance. The earlier developed meta search engine models considered majorly the rank assigned by the underlying search engines and not many other criteria. The naive models were not capable enough to handle the dynamic nature of web. This was the main motive to propose a novel approach that is adaptable to the dynamic nature of the web. The proposed approach uses Best Worst Method (BWM) to assign weights to criteria and Genetic Algorithm with Cluster Dependent Ordered Weighted Average (DOWA) as fitness function to obtain documents list sorted in decreasing order of relevance. This approach has higher precision when compared to the existing models.