Ontology alignment in the biomedical domain using entity definitions and context

@inproceedings{Wang2018OntologyAI,
  title={Ontology alignment in the biomedical domain using entity definitions and context},
  author={Lucy Lu Wang and Chandra Bhagavatula and Mark Neumann and Kyle Lo and Christopher Wilhelm and Waleed Ammar},
  booktitle={BioNLP},
  year={2018}
}
Ontology alignment is the task of identifying semantically equivalent entities from two given ontologies. Different ontologies have different representations of the same entity, resulting in a need to de-duplicate entities when merging ontologies. We propose a method for enriching entities in an ontology with external definition and context information, and use this additional information for ontology alignment. We develop a neural architecture capable of encoding the additional information… 

Figures and Tables from this paper

Large-Scale Ontology Alignment- An Extraction Based Method to Support Information System Interoperability

TLDR
This article proposes ONTEM as an effective alignment method for large-scale ontology based on the ontology entities extraction and conducts a comprehensive evaluation using the datasets of the OAEI 2018 campaign to reveal that onTEM is one of the most effective systems.

Multi-view Embedding for Biomedical Ontology Matching

TLDR
An alternative ontology matching framework called MultiOM is proposed, which models the matching process by embedding techniques from multiple views and designs different loss functions based on cross-entropy to learn the vector representations of concepts, and proposes a novel negative sampling skill tailored for the structural relations asserted in ontologies.

ONTOCONNECT: Domain-Agnostic Ontology Alignment using Graph Embedding with Negative Sampling

TLDR
This paper presents an ontology alignment approach using graph embedding with negative sampling that is independent of the domain and does not require any human intervention.

Ontology Matching by Jointly Encoding Terminological Description and Network Structure

TLDR
This work proposes an alternative ontology matching framework, which models the matching process by embedding techniques with jointly encoding ontology terminological description and network structure and improves the iterative final alignment method by introducing an automatic adjustment of threshold method.

Biomedical ontology alignment with BERT

TLDR
A novel ontology alignment system based on a contextual embedding model named BERT, aiming to sufficiently utilize the text semantics implied by ontologies, is proposed, demonstrating that, despite using the to-be-aligned classes alone as the input, the system outperforms the leading systems.

DAEOM: A Deep Attentional Embedding Approach for Biomedical Ontology Matching

TLDR
An alternative ontology matching framework called Deep Attentional Embedded Ontology Matching (DAEOM), which models the matching process by embedding techniques with jointly encoding ontology terminological description and network structure, and is competitive with several OAEI top-ranked systems in terms of F-measure.

Augmenting Ontology Alignment by Semantic Embedding and Distant Supervision

TLDR
A machine learning based extension to traditional ontology alignment systems, using distant supervision for training, ontology embedding and Siamese Neural Networks for incorporating richer semantics, finds that the extension recalls many additional valid mappings and also avoids some false positive mappings.

BERTMap: A BERT-based Ontology Alignment System

TLDR
This paper proposes a novel OM system named BERTMap which can support both unsupervised and semi-supervised settings and first predicts mappings using a classifier based on fine-tuning the contextual embedding model BERT on text semantics corpora extracted from ontologies, and then refines the mappings through extension and repair by utilizing the ontology structure and logic.

Background knowledge in ontology matching: A survey

TLDR
This survey introduces the reader to the general ontology matching problem and presents a classification system for external background knowledge, concept linking strategies, as well as for background knowledge exploitation approaches.

Word Embedding based Heterogeneous Entity Matching on Web of Things

TLDR
A word embedding based matching technique that models the word’s semantic in the vector space, and uses two vectors’ cosine angle to measure the corresponding words’ similarity and results show that this approach outperforms other advanced matching methods.

References

SHOWING 1-10 OF 25 REFERENCES

Evolution of biomedical ontologies and mappings: Overview of recent approaches

Use artificial neural network to align biological ontologies

TLDR
An artificial neural network approach is taken to learn and adjust weights for semantic aspects learned from a subset of concepts, and thereby support a new ontology alignment algorithm, customized for biological ontologies, with the purpose of avoiding some disadvantages in both rule-based and learning-based aligning algorithms.

Knowledge entity learning and representation for ontology matching based on deep neural networks

TLDR
A representation learning method based on deep neural networks which aim at learning the high level abstract representations of the input entity from its descriptive information to better measure the similarity between entities.

A Neural-Networks-Based Approach for Ontology Alignment

TLDR
This paper reports a new method for compound metric creation based on a supervised learning approach in data mining where a training set is used to create a neural network model, performs sensitivity analysis on it to select appropriate metrics among a set of existing ones, and finally constructs a Neural network model to combine the result metrics into a compound one.

The AgreementMakerLight Ontology Matching System

TLDR
A new core framework, AgreementMakerLight, focused on computational efficiency and designed to handle very large ontologies, while preserving most of the flexibility and extensibility of the original AgreementMaker framework is developed.

Ontology alignment using artificial neural network for large-scale ontologies

TLDR
An intelligent combination using Artificial Neural Network ANN as a machine learning-based method to ascertain how to combine multiple similarity measures into a single aggregated metric with the final aim of improving the ontology alignment quality is proposed.

Review of ontology matching with background knowledge

TLDR
The motivation that described the urgency of ontology matching, the various background knowledge and their strengths, also indirect matching process are reviewed and the comparison of indirect matching system is provided.

LogMap: Logic-Based and Scalable Ontology Matching

TLDR
This paper presents LogMap--a highly scalable ontology matching system with 'built-in' reasoning and diagnosis capabilities, and is the only matching system that can deal with semantically rich ontologies containing tens (and even hundreds of thousands of classes).

SemMedDB: a PubMed-scale repository of biomedical semantic predications

TLDR
SemMedDB is described, a repository of semantic predications (subject-predicate-object triples) extracted from the entire set of PubMed citations that can assist in hypothesis generation and literature-based discovery in biomedicine as well as in clinical decision-making support.

OAEI 2016 results of AML

TLDR
The development of AML was focused primarily on tackling instance matching, an aspect of ontology matching that was missing from its portfolio, however, it also made several developments with regard to class matching, namely with the use of OBO logical definitions.