Related Experiment Video
Updated: Apr 28, 2026
07:50
A Metadata Extraction Approach for Clinical Case Reports to Enable Advanced Understanding of Biomedical Concepts
Published on: September 20, 2018
15.8K
A resource-saving collective approach to biomedical semantic role labeling
Richard Tzong-Han Tsai1, Po-Ting Lai
1Department of Computer Science and Information Engineering, National Central University, Taoyuan, Taiwan, Republic of China. thtsai@csie.ncu.edu.tw.
BMC Bioinformatics
|June 3, 2014
Summary
A new collective biomedical semantic role labeling (BioSRL) system, CBIOSMILE, improves accuracy. A resource-saving version, RCBIOSMILE, achieves similar results with significantly less memory and training time.
Area of Science:
- Natural Language Processing
- Bioinformatics
- Computational Biology
Background:
- Biomedical semantic role labeling (BioSRL) identifies predicate-argument structures in biological texts.
- Current BioSRL systems independently label nodes, ignoring dependencies.
- Collective approaches using Markov Logic Networks (MLNs) are effective in general SRL but underexplored in BioSRL due to data and complexity concerns.
Purpose of the Study:
- To develop a collective BioSRL system using MLNs.
- To create a resource-saving version of the collective BioSRL system to reduce training demands.
- To evaluate the performance and efficiency of the proposed systems.
Main Methods:
- Constructed a collective BioSRL system (CBIOSMILE) based on MLN, trained on the BioProp corpus.
- Implemented a tree-pruning filter and argument candidate identifiers to optimize the training data.
- Developed a resource-saving collective BioSRL system (RCBIOSMILE) using pruned parse trees for MLN training.
Main Results:
- The proposed CBIOSMILE system outperformed the existing top BioSRL system (BIOSMILE).
- RCBIOSMILE achieved comparable accuracy to CBIOSMILE.
- RCBIOSMILE demonstrated significant efficiency gains, using 92% less memory and 57% less training time.
Conclusions:
- The efficiency of RCBIOSMILE makes it suitable for training on larger BioSRL corpora.
- Current BioProp corpus is limited for practical applications like pathway construction.
- Future work should focus on SRL training using extensive biomedical corpora.
Related Concept Videos
02:16Stereotype Content Model
13.0K
The Stereotype Content Model (SCM) was first proposed by Susan Fiske and her colleagues (Fiske, Cuddy, Glick & Xu, 2002; see also Fiske, 2012 and Fiske, 2017). The SCM specifies that when someone encounters a new group, they will stereotype them based on two metrics: warmth—or that group’s perceived intent, and how likely they are to provide help or inflict harm—and competence—or their ability to carry out that objective. Depending on the warmth-competence...
13.0K
00:49Ligand Binding and Linkage
3.0K
3.0K
00:49Ligand Binding and Linkage
4.4K
Allosteric proteins have more than one ligand binding site; the binding of a ligand to any of these sites influences the binding of ligands to the other sites. When a protein is allosteric, its binding sites are called coupled or linked. In the case of enzymes, the site that binds to the substrate is known as the active site and the other site is known as the regulatory site. When a ligand binds to the regulatory site, this leads to conformational changes in the protein that can influence...
4.4K