Search and Graph Database Technologies for Biomedical Semantic Indexing: Experimental Analysis

Abstract

BackgroundBiomedical semantic indexing is a very useful support tool for human curators in their efforts for indexing and cataloging the biomedical literature.ObjectiveThe aim of this study was to describe a system to automatically assign Medical Subject Headings (MeSH) to biomedical articles from MEDLINE.MethodsOur approach relies on the assumption that similar documents should be classified by similar MeSH terms. Although previous work has already exploited the document similarity by using a k-nearest neighbors algorithm, we represent documents as document vectors by search engine indexing and then compute the similarity between documents using cosine similarity. Once the most similar documents for a given input document are retrieved, we rank their MeSH terms to choose the most suitable set for the input document. To do this, we define a scoring function that takes into account the frequency of the term into the set of retrieved documents and the similarity between the input document and each retrieved document. In addition, we implement guidelines proposed by human curators to annotate MEDLINE articles; in particular, the heuristic that says if 3 MeSH terms are proposed to classify an article and they share the same ancestor, they should be replaced by this ancestor. The representation of the MeSH thesaurus as a graph database allows us to employ graph search algorithms to quickly and easily capture hierarchical relationships such as the lowest common ancestor between terms.ResultsOur experiments show promising results with an F1 of 69% on the test dataset.ConclusionsTo the best of our knowledge, this is the first work that combines search and graph database technologies for the task of biomedical semantic indexing. Due to its horizontal scalability, ElasticSearch becomes a real solution to index large collections of documents (such as the bibliographic database MEDLINE). Moreover, the use of graph search algorithms for accessing MeSH information could provide a support tool for cataloging MEDLINE abstracts in real time.

Semantic Classification and Indexing of Open Educational Resources with Word Embeddings and Ontologies

“…Also, most research that deals with the classification problem of a large number of classes [2,[24][25][26][27][28][29] appears to rely on machine learning classifiers or deep neural networks. In our work, each thesaurus term has been moved to a vector space after training a doc2vec model with some corpus and using MeSH headings as tags.…”

“…Taking into account the assumption that similar documents are classified with similar MeSH terms, authors in [29] have proceeded to an implementation with an F-score of 0.69. This work starts with the conversion of documents into vectors by search engine indexing (Elastic Search) and the identification of the most similar documents based on cosine similarity.…”

Improving Large-Scale k-Nearest Neighbor Text Categorization with Label Autoencoders

“…Approaches based on k-NN have been widely used in large-scale multi-label categorization in many domains, including MEDLINE documents [37,39,40]. This preference for this lazy learning method is mainly due to its scalability, minimum parameter tuning and, despite its simplicity, its ability to deliver acceptable results when a large number of training samples are available.…”

“…Several state-of-the-art methods for MeSH indexing were introduced by teams participating in this challenge, most of them modeling the task as a multi-label learning problem [30]. Some relevant recent developments in MeSH indexing are MeSHLabeler [31], DeepMeSH [32], MeSH Now [33], AttentionMeSH [34], MeSHProbeNet [35], FullMeSH [26], BERTMeSH [36] and k-NN methods using ElasticSearch and MTI such as [37].…”

Thesaurus-based word embeddings for automated biomedical literature classification

“…Another approach [ 25 ] for solving this issue uses the cosine similarity metric and a representation of the thesaurus as a graph database. It starts with the use of Elastic Search to convert texts into vectors, then with the cosine similarity metric it identifies the most similar texts.…”