Does Personalization Benefit Everyone in the Same Way? Multilingual Search Personalization for English vs. Non-English Users

The web community is witnessing an increase in the amount of available multilingual content and the number of multilingual web users. With this variety, personalized search systems are needed to connect people with relevant content, regardless of the language in which the content is provided, and taking into consideration the user’s language capabilities and preferences. Therefore, search personalization algorithms should be developed with the aspect of multilinguality in mind, part of which involves understanding the effect of personalization algorithms on the user’s search experience. This leads to an important question: given that users come from different linguistic backgrounds and have different language preferences, would personalization benefit all search users in the same way? This paper addresses this question by conducting an experiment to: (1) evaluate the effectiveness of the multilingual personalization algorithms (multilingual user modeling and multilingual result adaptation); and (2) determine whether multilingual personalization algorithms achieve the same degree of effectiveness for users who have different language preferences.