Supervised Learning Based Hypothesis Generation from Biomedical Literature.

Shengtian Sang,Hongfei Lin,Zhihao Yang,Zongyao Li

doi:10.1155/2015/698527

Shengtian Sang, Hongfei Lin + Show 2 more

Open Access

https://doi.org/10.1155/2015/698527

Copy DOI

Journal: BioMed research international	Publication Date: Jan 1, 2015
Citations: 33	License type: CC BY 3.0

Affiliation: Dalian University of Technology

Abstract

Nowadays, the amount of biomedical literatures is growing at an explosive speed, and there is much useful knowledge undiscovered in this literature. Researchers can form biomedical hypotheses through mining these works. In this paper, we propose a supervised learning based approach to generate hypotheses from biomedical literature. This approach splits the traditional processing of hypothesis generation with classic ABC model into AB model and BC model which are constructed with supervised learning method. Compared with the concept cooccurrence and grammar engineering-based approaches like SemRep, machine learning based models usually can achieve better performance in information extraction (IE) from texts. Then through combining the two models, the approach reconstructs the ABC model and generates biomedical hypotheses from literature. The experimental results on the three classic Swanson hypotheses show that our approach outperforms SemRep system.

Highlights

Literature-based discovery (LBD) was pioneered by Swanson in the 1980s and it focuses on finding new relationships in existing knowledge from unrelated literatures and provides logical explanations [1,2,3]
Swanson’s method is to find a bridge that links two conceptually related topics that ought to have been studied together but never have been. In his initial work [1], Swanson found there are some types of blood disorders in patients with Raynaud’s phenomenon (A), such as high blood viscosity (B), and at the same time he found that fish oil (C) can reduce blood viscosity (B) in other literatures
TP is the number of correctly predicted pieces of data, FP is the number of false positives, and FN is the number of false negatives in the test set

Summary

Introduction

Literature-based discovery (LBD) was pioneered by Swanson in the 1980s and it focuses on finding new relationships in existing knowledge from unrelated literatures and provides logical explanations [1,2,3]. Swanson’s method is to find a bridge that links two conceptually related topics that ought to have been studied together but never have been In his initial work [1], Swanson found there are some types of blood disorders in patients with Raynaud’s phenomenon (A), such as high blood viscosity (B), and at the same time he found that fish oil (C) can reduce blood viscosity (B) in other literatures. Since no literature had proved the relationship between fish oil (C) and Raynaud’s phenomenon (A) at that time, Swanson proposed that fish oil (C) might treat Raynaud’s phenomenon (A), and this hypothesis was verified in medical experiments two years later In this hypothesis, the topic of blood viscosity (B) served as a bridge between the topics of Raynaud’s phenomenon (A) and dietary fish oil (C). Swanson summarized this method as ABC model and he has published several other medical discoveries using this methodology [2, 3]

Objectives

Results

Conclusion