Skip to Main Content
Search personalization is an area of considerable research interest. In this paper, we propose a framework for personalizing cross-language search using user models. Our work extends existing studies in two directions. First, the framework extends to the area of cross-language information retrieval. Second, the study aims to elicit features of cross-language search behavior from multilingual search logs. We argue that we can infer a user model, that describes individual user interests and behavior, which can be partially bootstrapped based on choice of interface language. Our experiments involved mining multilingual search logs for interesting patterns of cross-language search behavior. Different patterns were exhibited for users of different languages. The results suggest that there is scope for further investigation on the use of log analysis to improve personalization of cross-language search.