Noh Jiho, Kavuluru Ramakanth
Department of Computer Science, University of Kentucky, Kentucky, USA.
Division of Biomedical Informatics, University of Kentucky, Kentucky, USA.
Proc Conf Empir Methods Nat Lang Process. 2020 Nov;2020:3389-3399. doi: 10.18653/v1/2020.findings-emnlp.304.
Information retrieval (IR) for precision medicine (PM) often involves looking for multiple pieces of evidence that characterize a patient case. This typically includes at least the name of a condition and a genetic variation that applies to the patient. Other factors such as demographic attributes, comorbidities, and social determinants may also be pertinent. As such, the retrieval problem is often formulated as search but with multiple facets (e.g., disease, mutation) that may need to be incorporated. In this paper, we present a document reranking approach that combines neural query-document matching and text summarization toward such retrieval scenarios. Our architecture builds on the basic BERT model with three specific components for reranking: (a). document-query matching (b). keyword extraction and (c). facet-conditioned abstractive summarization. The outcomes of (b) and (c) are used to essentially transform a candidate document into a concise summary that can be compared with the query at hand to compute a relevance score. Component (a) directly generates a matching score of a candidate document for a query. The full architecture benefits from the complementary potential of document-query matching and the novel document transformation approach based on summarization along PM facets. Evaluations using NIST's TREC-PM track datasets (2017-2019) show that our model achieves state-of-the-art performance. To foster reproducibility, our code is made available here: https://github.com/bionlproc/text-summ-for-doc-retrieval.
精准医学（PM）中的信息检索（IR）通常涉及寻找表征患者病例的多条证据。这通常至少包括一种病症的名称以及适用于该患者的基因变异。其他因素，如人口统计学属性、合并症和社会决定因素也可能相关。因此，检索问题通常被表述为具有多个方面（例如疾病、突变）的搜索，这些方面可能需要被纳入。在本文中，我们提出了一种文档重排方法，该方法结合了神经查询 - 文档匹配和文本摘要来处理此类检索场景。我们的架构基于基本的BERT模型构建，具有三个用于重排的特定组件：（a）文档 - 查询匹配；（b）关键词提取；（c）方面条件下的抽象摘要。（b）和（c）的结果用于将候选文档本质上转换为一个简洁的摘要，该摘要可与手头的查询进行比较以计算相关性得分。组件（a）直接生成候选文档与查询的匹配得分。完整的架构受益于文档 - 查询匹配的互补潜力以及基于沿PM方面的摘要的新颖文档转换方法。使用美国国家标准与技术研究院（NIST）的TREC - PM跟踪数据集（2017 - 2019）进行的评估表明，我们的模型实现了当前的最优性能。为了促进可重复性，我们的代码可在此处获取：https://github.com/bionlproc/text-summ-for-doc-retrieval 。
