Emerging trend prediction in biomedical literature

Fabian Moerchen; Dmitriy Fradkin; Mathaeus Dejori; Bernd Wachmann

Emerging trend prediction in biomedical literature

AMIA Annu Symp Proc. 2008 Nov 6:2008:485-9.

Authors

Fabian Moerchen¹, Dmitriy Fradkin, Mathaeus Dejori, Bernd Wachmann

Affiliation

¹ Integrated Data Systems, Siemens Corporate Research, Princeton, NJ, USA.

PMID: 18999010
PMCID: PMC2656084

Abstract

We present a study on how to predict new emerging trends in the biomedical domain based on textual data. We thereby propose a way of anticipating the transformation of arbitrary information into ground truth knowledge by predicting the inclusion of new terms into the MeSH ontology. We also discuss the preparation of a dataset for the evaluation of emerging trend prediction algorithms that is based on PubMed abstracts and related MeSH terms. The results suggest that early prediction of emerging trends is possible.

MeSH terms

Algorithms
Artificial Intelligence
Data Interpretation, Statistical
Information Storage and Retrieval / methods*
Journal Impact Factor*
Medical Subject Headings / statistics & numerical data*
Natural Language Processing*
Pattern Recognition, Automated / methods*
Periodicals as Topic / classification*
Periodicals as Topic / trends*
Terminology as Topic*
United States