Study on classification of domain-oriented user exploration process for exploratory search

In the study of exploratory search, different supporting methods have already been proposed to meet users´ different information needs. However, these methods are mostly targeted at different domains of user exploration process. In this paper, we study on a classification method of domain-oriented user exploration process for exploratory search to support determine user´s search domains for search engines who will well-directed help users to complete exploratory tasks. First, we seek to define and built exploration process model base on tree structure. Second, we study on domain-oriented classification method based on user exploration process model and design a parameter training method based on ListNet. In the experimental stage, we compare the abilities between our method and two `baseline´ methods. Results show that the precision, recall and F1-measure of our method is found to increase by 7 percent to 15 percent over these previous methods.