Ontology-based enriched concept graphs for medical document classification

Shanavas, Niloofer; Wang, Hui; Lin, Zhiwei; Hawe, Glenn I.

doi:10.1016/j.ins.2020.03.006

Cited by 21 publications

(7 citation statements)

References 24 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…It is a web-based tool that uses a list of genes from the Gene Ontology and enriches them using statistical methods. Recently, Shanavas et al (2020) presented a method to enrich the UMLS concepts with related documents from a pool of professional healthcare documents. Their aim was to provide retrieval systems with more information about medical concepts.…”

Section: Related Workmentioning

confidence: 99%

An automated method to enrich consumer health vocabularies using GloVe word embeddings and an auxiliary lexical resource

Ibrahim¹,

Gauch²,

Salman³

et al. 2021

PeerJ Computer Science

View full text Add to dashboard Cite

Background Clear language makes communication easier between any two parties. A layman may have difficulty communicating with a professional due to not understanding the specialized terms common to the domain. In healthcare, it is rare to find a layman knowledgeable in medical terminology which can lead to poor understanding of their condition and/or treatment. To bridge this gap, several professional vocabularies and ontologies have been created to map laymen medical terms to professional medical terms and vice versa. Objective Many of the presented vocabularies are built manually or semi-automatically requiring large investments of time and human effort and consequently the slow growth of these vocabularies. In this paper, we present an automatic method to enrich laymen’s vocabularies that has the benefit of being able to be applied to vocabularies in any domain. Methods Our entirely automatic approach uses machine learning, specifically Global Vectors for Word Embeddings (GloVe), on a corpus collected from a social media healthcare platform to extend and enhance consumer health vocabularies. Our approach further improves the consumer health vocabularies by incorporating synonyms and hyponyms from the WordNet ontology. The basic GloVe and our novel algorithms incorporating WordNet were evaluated using two laymen datasets from the National Library of Medicine (NLM), Open-Access Consumer Health Vocabulary (OAC CHV) and MedlinePlus Healthcare Vocabulary. Results The results show that GloVe was able to find new laymen terms with an F-score of 48.44%. Furthermore, our enhanced GloVe approach outperformed basic GloVe with an average F-score of 61%, a relative improvement of 25%. Furthermore, the enhanced GloVe showed a statistical significance over the two ground truth datasets with P < 0.001. Conclusions This paper presents an automatic approach to enrich consumer health vocabularies using the GloVe word embeddings and an auxiliary lexical source, WordNet. Our approach was evaluated used healthcare text downloaded from MedHelp.org, a healthcare social media platform using two standard laymen vocabularies, OAC CHV, and MedlinePlus. We used the WordNet ontology to expand the healthcare corpus by including synonyms, hyponyms, and hypernyms for each layman term occurrence in the corpus. Given a seed term selected from a concept in the ontology, we measured our algorithms’ ability to automatically extract synonyms for those terms that appeared in the ground truth concept. We found that enhanced GloVe outperformed GloVe with a relative improvement of 25% in the F-score.

show abstract

Section: Related Workmentioning

confidence: 99%

An automated method to enrich consumer health vocabularies using GloVe word embeddings and an auxiliary lexical resource

Ibrahim¹,

Gauch²,

Salman³

et al. 2021

PeerJ Computer Science

View full text Add to dashboard Cite

show abstract

“…Considering this assumption, a new document embedding method is proposed in Reference [23] using the hyperlinks and relations between documents for classification. Open domain knowledge bases are less useful in complex domains, like medical document classification [24]. Incorporating domain knowledge with relevant closed-domain ontologies are used in these situations to enrich document representations.…”

Section: Related Workmentioning

confidence: 99%

Linked Data Triples Enhance Document Relevance Classification

et al. 2021

View full text Add to dashboard Cite

Standardized approaches to relevance classification in information retrieval use generative statistical models to identify the presence or absence of certain topics that might make a document relevant to the searcher. These approaches have been used to better predict relevance on the basis of what the document is “about”, rather than a simple-minded analysis of the bag of words contained within the document. In more recent times, this idea has been extended by using pre-trained deep learning models and text representations, such as GloVe or BERT. These use an external corpus as a knowledge-base that conditions the model to help predict what a document is about. This paper adopts a hybrid approach that leverages the structure of knowledge embedded in a corpus. In particular, the paper reports on experiments where linked data triples (subject-predicate-object), constructed from natural language elements are derived from deep learning. These are evaluated as additional latent semantic features for a relevant document classifier in a customized news-feed website. The research is a synthesis of current thinking in deep learning models in NLP and information retrieval and the predicate structure used in semantic web research. Our experiments indicate that linked data triples increased the F-score of the baseline GloVe representations by 6% and show significant improvement over state-of-the art models, like BERT. The findings are tested and empirically validated on an experimental dataset and on two standardized pre-classified news sources, namely the Reuters and 20 News groups datasets.

show abstract

“…The size of data used by the model was less, and also suffered from a data imbalance problem. Synonymy, Polysemy, and Multi-word concepts are the curse of medical literature [14]. Exact information can only be extracted using semantic matching between domain ontologies and concept maps from documents.…”

Section: Literature Reviewmentioning

confidence: 99%

Ontology Based Text Classifier for Information Extraction from Coronavirus Literature

Sivakami

Thangaraj

2021

Trends Sci

View full text Add to dashboard Cite

The world is fighting an unprecedented coronavirus pandemic, and no country was prepared for it. Understanding the nature of this disease, when there is no available cure, is vital to encourage accurate clinical diagnosis and drug discovery prospects. When the amount of literature available is vast, it is important to represent the disease domain as completely as possible. The system should capture the morphology, semantics, syntax, and pragmatics of the given literature, in order to extract useful information. Also, building a classifier for a particular domain suffers from a zero frequency issue. To solve this effectively, latent topics are extracted and semantically represented in ontology to build a text classifier for coronavirus literature. The classifier is equipped with 2 components- ‘ontology’ and ‘machine learning data model’. Ontology helps to model the morphology and the semantic and pragmatic aspects of the text data through Latent Drichlet Allocation (LDA). It also preserves the contextual information in the document space, providing holistic feature representation facilities. To solve zero frequency and to extract actionable insights, a machine learning algorithm, Multi class Support Vector Machine (M-SVM), is incorporated with the ontology. It encodes features and achieves a classifier with highly discriminated classes. Further, to preserve contextual information space, and to enable data model formulation, the ontology is generated as a knowledge graph with their respective predefined classes. The resulting dataset can be used for clinical diagnosis and further research on the disease. Experimental results have shown that the proposed classifier outperforms the existing systems, with better domain representation. HIGHLIGHTS When the amount of literature available is vast, it is important to represent the disease domain as completely as possible. The system should capture the morphology, semantics, syntax, and pragmatics of the given literature, in order to extract useful information The classifier is equipped with 2 components- ‘ontology’ and ‘machine learning data model’. Ontology helps to model the morphology and the semantic and pragmatic aspects of the text data through Latent Drichlet Allocation (LDA). It also preserves the contextual information in the document space, providing holistic feature representation facilities To preserve contextual information space, and to enable data model formulation, the ontology is generated as a knowledge graph with their respective predefined classes. The resulting dataset can be used for clinical diagnosis and further research on the disease GRAPHICAL ABSTRACT

show abstract

Ontology-based enriched concept graphs for medical document classification

Cited by 21 publications

References 24 publications

An automated method to enrich consumer health vocabularies using GloVe word embeddings and an auxiliary lexical resource

An automated method to enrich consumer health vocabularies using GloVe word embeddings and an auxiliary lexical resource

Linked Data Triples Enhance Document Relevance Classification

Ontology Based Text Classifier for Information Extraction from Coronavirus Literature

Contact Info

Product

Resources

About