Bing’s way of improving search accuracy
In networks class, we have learned how PageRank uses the power of the web as a network to display most accurate search results. Nowadays, besides popular search algorithms like PageRank, search engines are also using other methods to improve their search results. Bing uses clickstream/surf-stream to see what people are browsing, specially people who are using the Bing Toolbar. So if a user does a search on Google, on Yahoo, on Amazon or any other site, and then clicks on a search result, Bing can see the new site you are headed to. Bing would associate the next page you went to after doing those searches as being a possible “answer” to those searches. Bing uses this information to test against its search results to improve the accuracy of its search results. This method also takes advantage of the fact that the web is organized as a network with links acting as edges connecting nodes (webpages).
Google recently accused Bing of copying their search results, as the results for Google and Bing for many queries were extremely similar. However, in reality, this is not because Bing was copying Google’s results or using a similar PageRank algorithm, but because Bing started using clickstream data to monitor the URLs a user has visited and since Google searches are used very frequently, their results factored in quite heavily in Bings data.
Here’s an article with more details on Bings response to Google’s plagiarism accusations: http://searchengineland.com/bing-why-googles-wrong-in-its-accusations-63279