Best Match: New relevance search for PubMed

Fiorini, Nicolas; Canese, Kathi; Starchenko, Grisha; Kireev, Evgeny; Kim, Won; Miller, Vadim; Osipov, Maxim; Kholodov, Michael; Ismagilov, Rafis; Mohan, Sanjay; Ostell, James; Lu, Zhiyong

doi:10.1371/journal.pbio.2005343

Cited by 102 publications

(91 citation statements)

References 33 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The first step consists in a textual search process by using the Vector Space Model; nonetheless, this step may be replaced with any other well-known model like that used by the PubMed Best Match System (BM25), giving an example used in Medicine [34] . This model is the same as that implemented in the study [18] used to compare our approach.…”

Section: Discussionmentioning

confidence: 99%

A relevance and quality-based ranking algorithm applied to evidence-based medicine

Serrano-Guerrero¹,

Romero²,

Olivas³

2020

Computer Methods and Programs in Biomedicine

View full text Add to dashboard Cite

a b s t r a c tBackground: The amount of information available about millions of different subjects is growing every day. This has led to the birth of new search tools specialized in different domains, because classical information retrieval models have trouble dealing with the special characteristics of some of these domains. Evidence-based Medicine is a case of a complex domain where classical information retrieval models can help search engines retrieve documents by considering the presence or absence of terms, but these must be complemented with other specific strategies which allow retrieving and ranking documents including the best current evidence and methodological quality.Objective: The goal is to present a ranking algorithm able to select the best documents for clinicians considering aspects related to the relevance and the quality of said documents.Methods: In order to assess the effectiveness of this proposal, an experimental methodology has been followed by using Medline as a data set and the Cochrane Library as a gold standard.Results: Applying the evaluation methodology proposed, and after submitting 40 queries on the platform developed, the MAP (Mean Average Precision) obtained was 20.26%.Conclusions: Successful results have been achieved with the experiments, improving on other studies, but under different and even more complex circumstances. with real users [ 5 , 6 ]; nevertheless, there are studies that question whether technology applied to clinical information retrieval is useful and cast doubt on the ability of scientists to search appropriately [7] . Hence, it is necessary to develop algorithms to improve the quality of search results, especially as finding the best studies may suppose a better treatment for patients [8] .An example may be the algorithm proposed in [9] which models quality criteria through fuzzy prototypes in order to filter Personal Health Records (PHRs). This is only one example of documents in Medicine, but there are many other types of documents (review articles, original studies, case reports, etc.) that are important, especially in Evidence-based Medicine (EBM). According to [10] , EBM is "the conscientious, explicit, and judicious use of current best evidence in making decisions about the care of individual patients." Therefore, EBM consists of integrating the best research evidence with clinical expertise and patient values. It is based on two main principles, the application of the best available evidence from the scientific method to medical decision making [11] and the assessment of the quality of evidence of the advantages and disadvantages of treatments [12] .

show abstract

Section: Discussionmentioning

confidence: 99%

A relevance and quality-based ranking algorithm applied to evidence-based medicine

Serrano-Guerrero¹,

Romero²,

Olivas³

2020

Computer Methods and Programs in Biomedicine

View full text Add to dashboard Cite

show abstract

“…The following weighting schemes are used in our experiments to produce document rankings for an atomic clause: IDF, TF-IDF, BM25, InL2 of Divergence from Randomness, PubMed, term position, text score, publication date, and document length. The PubMed weighting scheme uses the state-of-the-art learning to rank system of Pubmed [20]. The best match ranking system of PubMed uses a three-stage ranking system: first, documents are retrieved using the Boolean query; then, documents are ranked using BM25; finally, top-ranked documents are re-ranked using LambdaMART trained on click data, using document features such as document length, publication date, and past usage.…”

Section: Methodsmentioning

confidence: 99%

You Can Teach an Old Dog New Tricks: Rank Fusion applied to Coordination Level Matching for Ranking in Systematic Reviews

Scells

Zuccon

Koopman

2020

Lecture Notes in Computer Science

View full text Add to dashboard Cite

Coordination level matching is a ranking method originally proposed to rank documents given Boolean queries that is now several decades old. Rank fusion is a relatively recent method for combining runs from multiple systems into a single ranking, and has been shown to significantly improve the ranking. This paper presents a novel extension to coordination level matching, by applying rank fusion to each sub-clause of a Boolean query. We show that, for the tasks of systematic review screening prioritisation and stopping estimation, our method significantly outperforms the state-of-the-art learning to rank and bag-ofwords-based systems for this domain. Our fully automatic, unsupervised method has (i) the potential for significant real-world cost savings (ii) does not rely on any intervention from the user, and (iii) is significantly better at ranking documents given only a Boolean query in the context of systematic reviews when compared to other approaches.

show abstract

“…Soldaini et al [31] applied query reformulation techniques to address the need of literature search based on case reports. Best Match [6] is the first relevance search algorithm for PubMed that leverages the intelligence of users and machine-learning technology as an alternative to the traditional sorting techniques. Delta [25] is a deep learning based model that applies convolution operation upon an updated document matrix in which each word is replaced with the most similar word in the query.…”

Section: Related Work 21 Biomedical Literature Retrievalmentioning

confidence: 99%

Graphene

Zhao

Sboner

et al. 2019

Proceedings of the 28th ACM International Conference on Information and Knowledge Management

View full text Add to dashboard Cite

Effective biomedical literature retrieval (BLR) plays a central role in precision medicine informatics. In this paper, we propose GRAPHENE, which is a deep learning based framework for precise BLR. GRAPHENE consists of three main different modules 1) graph-augmented document representation learning; 2) query expansion and representation learning and 3) learning to rank biomedical articles. The graph-augmented document representation learning module constructs a document-concept graph containing biomedical concept nodes and document nodes so that global biomedical related concept from external knowledge source can be captured, which is further connected to a BiLSTM so both local and global topics can be explored. Query expansion and representation learning module expands the query with abbreviations and different names, and then builds a CNN-based model to convolve the expanded query and obtain a vector representation for each query. Learning to rank minimizes a ranking loss between biomedical articles with the query to learn the retrieval function. Experimental results on applying our system to TREC Precision Medicine track data are provided to demonstrate its effectiveness.

show abstract

Best Match: New relevance search for PubMed

Cited by 102 publications

References 33 publications

A relevance and quality-based ranking algorithm applied to evidence-based medicine

A relevance and quality-based ranking algorithm applied to evidence-based medicine

You Can Teach an Old Dog New Tricks: Rank Fusion applied to Coordination Level Matching for Ranking in Systematic Reviews

Graphene

Contact Info

Product

Resources

About