September 27, 2017

Lyes Limam: Extracting user search interests using a pruning algorithm


The object of the presentation is to discuss a method to extract user search interests from query logs by taking into account the semantic properties of query terms. First, a short recall is made about how to extract a global semantic representation of a search query log which will be used as a basis for processing. Then, the presentation focuses on the pruning clustering algorithm that is applied to the log representation to identify the user interests. After that, two applications are presented in order to show how to exploit the results of the algorithm. The evaluation has been done on large real-life logs of a popular search engine