Knowledge-based query expansion to support scenario-specific retrieval of medical free text
Knowledge-based query expansion to support scenario-specific retrieval of medical free text
复制标题
DOI：
10.1007/s10791-006-9020-6
复制
发表时间：
2005-03
期刊：
影响因子：
--
通讯作者：
Zhenyu Liu;W. Chu
中科院分区：
文献类型：
--
作者：
Zhenyu Liu;W. Chu
在检索医疗免费文本时，用户通常对与某些场景相关的答案感兴趣，这些场景对应于医疗实践中的常见任务，例如疾病的“治疗”或“诊断”。因此，他们提出的查询通常是特定于场景的，例如，“肺癌，治疗”。处理此类查询的一个基本挑战是查询中的场景术语(例如：“治疗”)过于笼统，无法匹配相关文件中的专门术语(例如：“肺切除”)。在本文中，我们提出了一种基于知识的查询扩展方法，该方法利用UMLS知识来源在原始查询中添加与查询场景特别相关的附加术语。我们将所提出的方法与仅探索统计项相关性并扩展不一定特定于场景的术语的统计展开方法进行了比较。我们对OHSUMED测试平台的研究表明，基于知识的方法导致特定场景的扩展，平均能够比统计方法提高5%以上，对于提到某些场景的查询，例如“疾病的治疗”和“症状/疾病的鉴别诊断”，能够提高约10%。
In retrieving medical free text, users are often interested in answers relevant to certain scenarios, scenarios that correspond to common tasks in medical practice, e.g., "treatment" or "diagnosis" of a disease. Consequently, the queries they pose are often scenario-specific, e.g., "lung cancer, treatment." A fundamental challenge in handling such queries is that scenario terms in the query (e.g. "treatment") are too general to match specialized terms in relevant documents (e.g. "lung excision"). In this paper we propose a knowledge-based query expansion method that exploits the UMLS knowledge source to append the original query with additional terms that are specifically relevant to the query's scenario(s). We compare the proposed method with statistical expansion that only explores statistical term correlation and expands terms that are not necessarily scenario specific. Our study on the OHSUMED testbed shows that the knowledge-based method which results in scenario-specific expansion is able to improve more than 5% over the statistical method on average, and about 10% for queries that mention certain scenarios, such as "treatment of a disease" and "differential diagnosis of a symptom/disease."