a framework for evaluating snippet generation for dataset search

• DOI: 10.48550/arxiv.1907.01183
• Publication Date: 2019-01-01
• Authors: Gong Cheng; Yuzhong Qu; Shuxin Li; Evgeny Kharlamov; Xiaxia Wang; Jeff Z. Pan; Jinchi Chen
• Keywords: FOS: Computer and information sciences; Computer Science - Databases; 0202 electrical engineering, electronic engineering, information engineering; Databases (cs.DB); 02 engineering and technology; Information Retrieval (cs.IR); Computer Science - Information Retrieval

Reusing existing datasets is of considerable significance to researchers and developers. Dataset search engines help a user find relevant datasets for reuse. They can present a snippet for each retrieved dataset to explain its relevance to the user's data needs. This emerging problem of snippet generation for dataset search has not received much research attention. To provide a basis for future research, we introduce a framework for quantitatively evaluating the quality of a dataset snippet. The proposed metrics assess the extent to which a snippet matches the query intent and covers the main content of the dataset. To establish a baseline, we adapt four state-of-the-art methods from related fields to our problem, and perform an empirical evaluation based on real-world datasets and queries. We also conduct a user study to verify our findings. The results demonstrate the effectiveness of our evaluation framework, and suggest directions for future research.<br/>17 pages, to appear at the research track of the 18th International Semantic Web Conference (ISWC 2019)<br/>