Improving Biomedical Analogical Retrieval with Embedding of Structural Dependencies

@inproceedings{Paullada2020ImprovingBA,
  title={Improving Biomedical Analogical Retrieval with Embedding of Structural Dependencies},
  author={Amandalynne Paullada and Bethany L Percha and Trevor A. Cohen},
  booktitle={BIONLP},
  year={2020}
}
Inferring the nature of the relationships between biomedical entities from text is an important problem due to the difficulty of maintaining human-curated knowledge bases in rapidly evolving fields. Neural word embeddings have earned attention for an apparent ability to encode relational information. However, word embedding models that disregard syntax during training are limited in their ability to encode the structural relationships fundamental to cognitive theories of analogy. In this paper… 

Figures and Tables from this paper

Supervised learning with word embeddings derived from PubMed captures latent knowledge about protein kinases and cancer
TLDR
An approach based on natural language processing and machine learning to investigate the relations between PKs and cancers, predicting PKs whose inhibition would be efficacious to treat a certain cancer with good accuracy is presented.
Supervised learning with word embeddings derived from PubMed captures latent knowledge about protein kinases and cancer
TLDR
An approach based on natural language processing and machine learning to the relations between PKs and cancers, predicting PKs whose inhibition would be efficacious to treat a certain cancer.
M ay 2 02 1 Should Semantic Vector Composition be Explicit ? Can it be Linear ? Semantic Spaces at the Intersection of NLP
TLDR
A survey of whether semantic vector composition should be explicitly described mathematically, or whether it can be a model-internal side-effect of training a neural network, and whether a compositional model can be implemented on a quantum computer.
Should Semantic Vector Composition be Explicit? Can it be Linear?
TLDR
A survey of whether semantic vector composition should be explicitly described mathematically, or whether it can be a model-internal side-effect of training a neural network, and whether a compositional model can be implemented on a quantum computer.

References

SHOWING 1-10 OF 38 REFERENCES
Insights into Analogy Completion from the Biomedical Domain
TLDR
BMASS is presented, a novel dataset for evaluating linguistic regularities in biomedical embeddings, and it is demonstrated that the relationships described in the dataset pose significant semantic challenges to current word embedding methods.
Medical Semantic Similarity with a Neural Language Model
TLDR
The demonstrated superiority of this model for providing an effective semantic similarity measure is promising in that this may translate into effectiveness gains for techniques in medical information retrieval and medical informatics (e.g., query expansion and literature-based discovery).
A global network of biomedical relationships derived from text
TLDR
This work combined a published ensemble biclustering algorithm (EBC) with hierarchical clustering to group the dependency paths into semantically‐related categories, which were annotated with labels, or ‘themes’ (‘inhibition’ and ‘activation’, for example).
Predicting Adverse Drug-Drug Interactions with Neural Embedding of Semantic Predications
TLDR
This work applies Embedding of Semantic Predications to the problem of predicting polypharmacy side-effects for particular drug combinations, building on a recent reconceptualization of this problem as a network of drug nodes connected by side effect edges.
Integrating shortest dependency path and sentence sequence into a deep learning framework for relation extraction in clinical text
TLDR
A novel neural approach to model shortest dependency path (SDP) between target entities together with the sentence sequence for clinical relation extraction is proposed by modeling SDP along with sentence sequence to extract multi-relations from clinical text.
Reasoning with vectors: A continuous model for fast robust inference
TLDR
The paper explains the way logical connectives can be used in semantic vector models, and summarizes the development of Predication-based Semantic Indexing, which involves the use of Vector Symbolic Architectures to represent the concepts and relationships from a knowledge base of subject-predicate-object triples.
Empirical distributional semantics: Methods and biomedical applications
A hybrid model based on neural networks for biomedical relation extraction
Dissecting Contextual Word Embeddings: Architecture and Representation
TLDR
There is a tradeoff between speed and accuracy, but all architectures learn high quality contextual representations that outperform word embeddings for four challenging NLP tasks, suggesting that unsupervised biLMs, independent of architecture, are learning much more about the structure of language than previously appreciated.
...
1
2
3
4
...