The paper introduces Scatter/Gather, a cluster-based document browsing method, as an alternative to ranked titles for organizing and viewing retrieval results. The authors systematically evaluate Scatter/Gather and find significant improvements over similarity search ranking alone, providing evidence that supports the Cluster Hypothesis, which states that relevant documents tend to be more similar to each other than to non-relevant documents. The system employs an automatic clustering algorithm, Fractionation, to organize documents into topic-coherent groups, and presents descriptive textual summaries to users. Users can select clusters for iterative examination, and the clustering is dynamic, adapting to different query results. The evaluation includes a comparison of Scatter/Gather with ranked titles and a user study, showing that Scatter/Gather significantly improves retrieval results and that users can effectively use the system to navigate and select relevant documents. The authors conclude that Scatter/Gather provides a valuable tool for improving information retrieval and suggest future research to further understand how users interact with the clustering representation.The paper introduces Scatter/Gather, a cluster-based document browsing method, as an alternative to ranked titles for organizing and viewing retrieval results. The authors systematically evaluate Scatter/Gather and find significant improvements over similarity search ranking alone, providing evidence that supports the Cluster Hypothesis, which states that relevant documents tend to be more similar to each other than to non-relevant documents. The system employs an automatic clustering algorithm, Fractionation, to organize documents into topic-coherent groups, and presents descriptive textual summaries to users. Users can select clusters for iterative examination, and the clustering is dynamic, adapting to different query results. The evaluation includes a comparison of Scatter/Gather with ranked titles and a user study, showing that Scatter/Gather significantly improves retrieval results and that users can effectively use the system to navigate and select relevant documents. The authors conclude that Scatter/Gather provides a valuable tool for improving information retrieval and suggest future research to further understand how users interact with the clustering representation.