Temporal Positive-unlabeled Learning for Biomedical Hypothesis Generation via Risk Estimation

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Akujuobi, Uchenna, Chen, Jun, Elhoseiny, Mohamed, Spranger, Michael, Zhang, Xiangliang
Format: Preprint
Published: 2020
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866917611999068160
author Akujuobi, Uchenna
Chen, Jun
Elhoseiny, Mohamed
Spranger, Michael
Zhang, Xiangliang
author_facet Akujuobi, Uchenna
Chen, Jun
Elhoseiny, Mohamed
Spranger, Michael
Zhang, Xiangliang
contents Understanding the relationships between biomedical terms like viruses, drugs, and symptoms is essential in the fight against diseases. Many attempts have been made to introduce the use of machine learning to the scientific process of hypothesis generation(HG), which refers to the discovery of meaningful implicit connections between biomedical terms. However, most existing methods fail to truly capture the temporal dynamics of scientific term relations and also assume unobserved connections to be irrelevant (i.e., in a positive-negative (PN) learning setting). To break these limits, we formulate this HG problem as future connectivity prediction task on a dynamic attributed graph via positive-unlabeled (PU) learning. Then, the key is to capture the temporal evolution of node pair (term pair) relations from just the positive and unlabeled data. We propose a variational inference model to estimate the positive prior, and incorporate it in the learning of node pair embeddings, which are then used for link prediction. Experiment results on real-world biomedical term relationship datasets and case study analyses on a COVID-19 dataset validate the effectiveness of the proposed model.
format Preprint
id arxiv_https___arxiv_org_abs_2010_01916
institution arXiv
publishDate 2020
record_format arxiv
spellingShingle Temporal Positive-unlabeled Learning for Biomedical Hypothesis Generation via Risk Estimation
Akujuobi, Uchenna
Chen, Jun
Elhoseiny, Mohamed
Spranger, Michael
Zhang, Xiangliang
Machine Learning
Artificial Intelligence
Understanding the relationships between biomedical terms like viruses, drugs, and symptoms is essential in the fight against diseases. Many attempts have been made to introduce the use of machine learning to the scientific process of hypothesis generation(HG), which refers to the discovery of meaningful implicit connections between biomedical terms. However, most existing methods fail to truly capture the temporal dynamics of scientific term relations and also assume unobserved connections to be irrelevant (i.e., in a positive-negative (PN) learning setting). To break these limits, we formulate this HG problem as future connectivity prediction task on a dynamic attributed graph via positive-unlabeled (PU) learning. Then, the key is to capture the temporal evolution of node pair (term pair) relations from just the positive and unlabeled data. We propose a variational inference model to estimate the positive prior, and incorporate it in the learning of node pair embeddings, which are then used for link prediction. Experiment results on real-world biomedical term relationship datasets and case study analyses on a COVID-19 dataset validate the effectiveness of the proposed model.
title Temporal Positive-unlabeled Learning for Biomedical Hypothesis Generation via Risk Estimation
topic Machine Learning
Artificial Intelligence
url https://arxiv.org/abs/2010.01916