Interacting with Explanations through Critiquing
This addresses the need for interactive recommendation systems that allow users to refine preferences through explanation critiquing, representing an incremental advance in personalization.
The paper tackles the problem of improving recommendations by enabling users to critique textual explanations, allowing them to remove or add aspects they dislike or like, which updates the user model and recommendations. It shows that human users significantly prefer these explanations over state-of-the-art techniques, and the system achieves good performance in adapting to multi-step critiquing on real-world datasets.
Using personalized explanations to support recommendations has been shown to increase trust and perceived quality. However, to actually obtain better recommendations, there needs to be a means for users to modify the recommendation criteria by interacting with the explanation. We present a novel technique using aspect markers that learns to generate personalized explanations of recommendations from review texts, and we show that human users significantly prefer these explanations over those produced by state-of-the-art techniques. Our work's most important innovation is that it allows users to react to a recommendation by critiquing the textual explanation: removing (symmetrically adding) certain aspects they dislike or that are no longer relevant (symmetrically that are of interest). The system updates its user model and the resulting recommendations according to the critique. This is based on a novel unsupervised critiquing method for single- and multi-step critiquing with textual explanations. Experiments on two real-world datasets show that our system is the first to achieve good performance in adapting to the preferences expressed in multi-step critiquing.