Dutch scientists improve prediction of searches




Scientists from the University of Amsterdam have developed a method to better predict where a user of a search engine is looking for. They do this by looking at the time of the year and the profile of the user.

A summary of the study by the University of Amsterdam on its website published . Researcher Fei Cai developed the method, an improvement of 7 percent in the prediction of searches should worry. Scientists compare themselves with other state-of-the-art method for autocomplete, but it is not clear what that is exactly.

The Cai method works by taking the time of year in predicting a search. Example, it is likely that a user who “hello” type in “Halloween” will look like the end of October, compared with the rest of the year, according to the University of Amsterdam. Also Cai algorithm takes personal characteristics along with autocompletion search results. The UvA as an example of the user’s location.

It is not yet clear when the technology developed in practice will be used. Search engines may have an interest in the predictive abilities of the algorithm of Cai, but this is nothing said. The results were announced at a scientific conference, but have not yet been published in a recognized journal.


In: Technology & Gadgets Asked By: [15484 Red Star Level]

Answer this Question

You must be Logged In to post an Answer.

Not a member yet? Sign Up Now »