CSIRO at 2017 TREC Precision Medicine Track
نویسندگان
چکیده
We report on our participation as the CSIROmed1 team in the TREC 2017 Precision Medicine track. We submitted five runs for the scientific abstracts collection (MEDLINE and Cancer Proceedings), and five runs for the clinical trials collection. We experimented with a number of query expansion and search result re-ranking techniques. We used citation and MeSH-based re-ranking methods, as well as reranking based on a merging algorithm proposed for federated search. Our results show that boosting the gene variant in the query increases the relevance of the retrieved results. One of our five runs for clinical trials task was ranked in top 10 runs out of 133 runs submitted for this task.
منابع مشابه
UTD HLTRI at TREC 2017: Precision Medicine Track
In this paper, we describe the system designed for the TREC 2017 Precision Medicine track by the University of Texas at Dallas (UTD) Human Language Technology Research Institute (HLTRI). Our system incorporates an aspect-based retrieval paradigm wherein each of the four structured components of the topic is cast as a separate aspect, along with two “hidden” aspects encoding the need that retrie...
متن کاملRetrieving documents based on gene name variations: MedIER at TREC 2017 Precision Medicine Track
The TREC 2017 Precision Medicine Track focused on finding relevant medical documents – scientific abstracts and clinical trials – for cancer patient cases based on specific genetic variation and demographic information. We focused on the genetic variations mentioned in the query and explored ways to modify the search query and the retrieval ranking using this information. Further, we explored f...
متن کاملUD_GU_BioTM at TREC 2017: Precision Medicine Track
This paper describes the system developed for the TREC 2017 PM track. We employed a two-part system to generate the ranked list of clinical trials and scientific abstracts. The first part pertains to query expansion and document retrieval from document index. The second part pertains to generating the final ranked list by implementing a heuristic scoring method. The scoring for clinical trials ...
متن کاملPatient selection for clinical trials based on concept-based retrieval and result filtering and ranking
For our participation at the clinical trials task in the TREC 2017 Precision Medicine track 2017, we investigated retrieving and matching clinical trial documents with patient information based on text and concept annotations of the text, filtering results for demographic information such as gender and age, and re-ranking results based on patient information. Experimental results show a competi...
متن کاملThe University of Amsterdam at the TREC 2007 Enterprise Track
We describe our participation in the TREC 2007 Enterprise track and detail our language modeling-based approaches. For document search, our focus was on estimating a mixture model using a standard web collection, and on constructing query models by employing blind relevance feedback and using the example documents provided with the topics. We found that settings performing well on a web collect...
متن کامل