Intentradar: search user interface that anticipates user's search intents.

We introduce IntentRadar, an interactive search user interface that anticipates user's search intents by estimating them from user interaction. The estimated intents are represented as keywords and visualized on a radial layout that organizes the keywords as directions in the information space. IntentRadar assists users to direct their search by allowing to target relevance feedback on keywords by manipulating the position of the keywords on the radar. The system then learns and visualizes improved estimates of intents and retrieves documents corresponding to the present search intent estimate. IntentRadar has been shown to significantly improve users' task performance and the quality of retrieved information without compromising task execution time.