A system for finding biological entities that satisfy certain conditions from texts

Abstract

Finding biological entities (such as genes or proteins) that satisfy certain conditions from texts is an important and challenging task in biomedical information retrieval and text mining. It is essential for many biomedical applications, such as drug discovery which normally requires collecting existing scientific facts from documents. This paper presents an effective IR system for this task, in which 1) domain knowledge is incorporated to improve retrieval effectiveness; 2) query expansion with related concepts on multiple semantic levels is employed; 3) a gene symbol disambiguation technique is implemented. We evaluated these techniques and examined two different concept-based IR models. Experiments based upon the proposed framework yield significant improvement (22% for automatic and 16.7% for nonautomatic) over the best reported results of passage retrieval in the Genomics track of TREC 2007.

Building optimal information systems automatically

“…Recently, researchers in biomedical information retrieval and question answering continue to leverage the TREC Genomics data set for evaluation [37,14,32,20,23,17,5]. Given the absence of an easy-to-use framework for building baseline systems for new tasks and exploring large parts of the configuration space, iterative research typically focuses on perturbations to a single module while keeping modules and parameters elsewhere in the system frozen.…”

Disambiguating biomedical acronyms using EMIM

“…However, they reported that their approach was insufficient to deal with the ambiguity of acronyms. Similarly, Zhou et al [6] used the ADAM acronym database to expand long-forms in the query with the corresponding acronyms as query concepts for their conceptbased retrieval framework. Still, they encountered problems with the ambiguity of acronyms and gene names.…”