Summarizing approach for efficient search by k-medoids method

In past days, although we have focused on to collect required data, we can get required information since many data are storage and disclosed. Therefore, it has become a new task to search efficiently required information. Nowadays, the search engine such as Google, Bing and Baidu help us to search information in the internet. However, enormous number of search results is listed. In some cases, the number of search results can be more than a hundred million that are ranked by their relevancies to the search key words. It is difficult to find out the desired information because user's time and effort are required. In order to efficiently attain user's required information reach, although it is an effective way to rank data by their relevancies to the search key words, sometime it is better the way to summarize information. In this work, we propose summarizing approach for efficient search by k-medoids method. Without defining categories in advance, k-medoids method generates clusters with less susceptibility to noise. Experimental results verify our method's feasibility and effectiveness.