Evaluation and Refinement of Clustered Search Results with the Crowd

Explore a 22-minute conference talk from the 24th International Conference on Intelligent User Interfaces that delves into using crowd-based human evaluation to improve clustered search results. Learn about a novel workflow that employs multiple clustering algorithms, followed by crowd workers' assessments to identify and refine the best clustered results. Discover how this approach can scale the process of creating high-quality clustered search results for large numbers of queries, potentially revolutionizing web and app search experiences. Examine the effectiveness of this method through an evaluation conducted on 120 top Google Play Store search queries, comparing the results with expert judgments.