A NEW LATENT SEMANTIC ANALYSIS BASED METHODOLOGY FOR KNOWLEDGE EXTRACTION FROM BIOMEDICAL LITERATURE AND BIOLOGICAL PATHWAYS DATABASES

F. Abate, A. Acquaviva, E. Ficarra, E. Macii

2011

Abstract

Nowadays, a considerable amount of genetic and biomedical studies are mostly diffused on theWeb and freely available. This exciting capability, if from one side opens the way to new scenarios of cooperating research, on the other side makes the knowledge retrieval and extraction an extremely time consuming operation. In this context, the development of new tools and algorithms to automatically support the scientist activity to achieve a reliable interpretation of the complex interactions among biological entities is mandatory. In this paper we present a new methodology aimed at quantifying the biological degree of correlation among biomedical terms present in literature. The proposed method overcomes the limitation of current tools based on public literature information only, by exploiting the trustworthy information provided by biological pathways databases. We demonstrate how to integrate trusted pathway information in a semantic correlation extraction chain based on UMLS Metathesaurus and relying on PubMed as literature database. The effectiveness of the obtained results remarks the importance of automatically quantifying the degree of correlation among biomedical terms in order to helpfully support the scientist research activity.

Paper Citation

in Harvard Style

Abate F., Acquaviva A., Ficarra E. and Macii E. (2011). A NEW LATENT SEMANTIC ANALYSIS BASED METHODOLOGY FOR KNOWLEDGE EXTRACTION FROM BIOMEDICAL LITERATURE AND BIOLOGICAL PATHWAYS DATABASES . In Proceedings of the International Conference on Bioinformatics Models, Methods and Algorithms - Volume 1: BIOINFORMATICS, (BIOSTEC 2011) ISBN 978-989-8425-36-2, pages 66-74. DOI: 10.5220/0003171400660074

in Bibtex Style

@conference{bioinformatics11,
author={F. Abate and A. Acquaviva and E. Ficarra and E. Macii},
title={A NEW LATENT SEMANTIC ANALYSIS BASED METHODOLOGY FOR KNOWLEDGE EXTRACTION FROM BIOMEDICAL LITERATURE AND BIOLOGICAL PATHWAYS DATABASES },
booktitle={Proceedings of the International Conference on Bioinformatics Models, Methods and Algorithms - Volume 1: BIOINFORMATICS, (BIOSTEC 2011)},
year={2011},
pages={66-74},
publisher={SciTePress},
organization={INSTICC},
doi={10.5220/0003171400660074},
isbn={978-989-8425-36-2},
}

in EndNote Style

TY - CONF
JO - Proceedings of the International Conference on Bioinformatics Models, Methods and Algorithms - Volume 1: BIOINFORMATICS, (BIOSTEC 2011)
TI - A NEW LATENT SEMANTIC ANALYSIS BASED METHODOLOGY FOR KNOWLEDGE EXTRACTION FROM BIOMEDICAL LITERATURE AND BIOLOGICAL PATHWAYS DATABASES
SN - 978-989-8425-36-2
AU - Abate F.
AU - Acquaviva A.
AU - Ficarra E.
AU - Macii E.
PY - 2011
SP - 66
EP - 74
DO - 10.5220/0003171400660074