Retrieval feedback in MEDLINE.

OBJECTIVE: To investigate a new approach for query expansion based on retrieval feedback. The first objective in this study was to examine alternative query-expansion methods within the same retrieval-feedback framework. The three alternatives proposed are: expansion on the MeSH query field alone, expansion on the free-text field alone, and expansion on both the MeSH and the free-text fields. The second objective was to gain further understanding of retrieval feedback by examining possible dependencies on relevant documents during the feedback cycle. DESIGN: Comparative study of retrieval effectiveness using the original unexpanded and the alternative expanded user queries on a MEDLINE test collection of 75 queries and 2,334 MEDLINE citations. MEASUREMENTS: Retrieval effectivenesses of the original unexpanded and the alternative expanded queries were compared using 11-point-average precision scores (11-AvgP). These are averages of precision scores obtained at 11 standard recall points. RESULTS: All three expansion strategies significantly improved the original queries in terms of retrieval effectiveness. Expansion on MeSH alone was equivalent to expansion on both MeSH and the free-text fields. Expansion on the free-text field alone improved the queries significantly less than did the other two strategies. The second part of the study indicated that retrieval-feedback-based expansion yields significant performance improvements independent of the availability of relevant documents for feedback information. CONCLUSIONS: Retrieval feedback offers a robust procedure for query expansion that is most effective for MEDLINE when applied to the MeSH field.

研究目的：本研究旨在探索一种基于检索反馈的查询扩展新方法。本研究的首要目标是在同一检索反馈框架下，探究多种备选查询扩展方案。本次提出的三种备选方案分别为：仅针对医学主题词表（MeSH）字段进行扩展、仅针对自由文本字段进行扩展，以及同时针对MeSH字段与自由文本字段进行扩展。本研究的第二项目标是通过检验反馈周期中检索效果与相关文档的潜在依赖关系，进一步深化对检索反馈机制的认知。
研究设计：本研究采用对照研究设计，在包含75条查询与2334条MEDLINE文献题录的MEDLINE测试集上，对比原始未扩展查询与各备选扩展查询的检索效果。
评价指标：采用11点平均准确率（11-point-average precision scores，11-AvgP）对比原始未扩展查询与各扩展查询的检索效果。该指标为11个标准召回点处准确率得分的平均值。
研究结果：三种扩展策略均显著提升了原始查询的检索效果。仅针对MeSH字段进行扩展的效果，与同时针对MeSH字段与自由文本字段进行扩展的效果相当；仅针对自由文本字段进行扩展的效果提升幅度，显著低于另外两种策略。本研究的第二部分结果表明，基于检索反馈的查询扩展可带来显著的性能提升，且该效果不受反馈所需相关文档可用性的影响。
研究结论：检索反馈是一种鲁棒的查询扩展方法，在MEDLINE场景下，将其应用于MeSH字段时可取得最佳效果。