Book chapter
Chapter 3 - Natural Language Processing and Ontology-enhanced Biomedical Literature Mining for Systems Biology
Computational Systems Biology
2006
Abstract
This chapter discusses some of the latest natural language processing and data mining techniques in systems biology research, and demonstrates their usefulness in chromatin protein interaction experiments and microarray data analysis. A variety of approaches to provide a biological explanation of gene clusters have been developed. TextQuest is geared toward summarizing documents retrieved in response to keywords-based search on PubMed. MedMiner can provide summarized literature information on genes but is limited to finding relations between two genes only. Estimating the reliability of the pattern to ignore patterns that tend to generate bogus relationships is one of the problems one needs to address. The standard clustering algorithms can be categorized into partitioning algorithms such as k-means or k-medoid and hierarchical algorithms such as single-link or average-link. The rationale behind this approach is that a small overlap of the clusters will result in a small classification error when the clustering is later used for classifying new documents.
Metrics
19 Record Views
3 citations in Scopus
Details
- Title
- Chapter 3 - Natural Language Processing and Ontology-enhanced Biomedical Literature Mining for Systems Biology
- Creators
- Xiaohua Hu - Drexel University
- Publication Details
- Computational Systems Biology
- Publisher
- Elsevier
- Resource Type
- Book chapter
- Language
- English
- Academic Unit
- Information Science
- Scopus ID
- 2-s2.0-84882317727
- Other Identifier
- 991019170440804721