Access Type

Open Access Dissertation

Date of Award

January 2019

Degree Type


Degree Name



Computer Science

First Advisor

Saeid Balaneshinkordan


Accurately answering queries that describe a clinical case and aim at finding articles in a collection of medical literature requires utilizing knowledge bases in capturing many explicit and latent aspects of such queries. Proper representation of these aspects needs knowledge-based query understanding methods that identify the most important query concepts as well as knowledge-based query reformulation methods that add new concepts to a query. In the tasks of Clinical Decision Support (CDS) and Precision Medicine (PM), the query and collection documents may have a complex structure with different components, such as disease and genetic variants that should be transformed to enable an effective information retrieval. In this work, we propose methods for representing domain-specific queries based on weighted concepts of different types whether exist in the query itself or extracted from the knowledge bases and top retrieved documents. Besides, we propose an optimization framework, which allows unifying query analysis and expansion by jointly determining the importance weights for the query and expansion concepts depending on their type and source. We also propose a probabilistic model to reformulate the query given genetic information in the query and collection documents. We observe significant improvement of retrieval accuracy will be obtained for our proposed methods over state-of-the-art baselines for the tasks of clinical decision support and precision medicine.