A4 Vertaisarvioitu artikkeli konferenssijulkaisussa
EVEX in ST'13: Application of a large-scale text mining resource to event extraction and network construction
Tekijät: Kai Hakala, Sofie Van Landeghem, Tapio Salakoski, Yves Van de Peer, Filip Ginter
Toimittaja: N/A
Julkaisuvuosi: 2013
Kokoomateoksen nimi: Proceedings of the BioNLP Shared Task 2013 Workshop (BioNLP-ST'13)
Aloitussivu: 26
Lopetussivu: 34
ISBN: 978-1-937284-55-8
Tiivistelmä
During the past few years, several novel text mining algorithms have been developed in the context of the BioNLP Shared Tasks on Event Extraction. These algorithms typically aim at extracting biomolecular interactions from text by inspecting only the context of one sentence. However, when humans interpret biomolecular research articles, they usually build upon extensive background knowledge of their favorite genes and pathways. To make such world knowledge available to a text mining algorithm, it could first be applied to all available literature to subsequently make a more informed decision on which predictions are consistent with the current known data. In this paper, we introduce our participation in the latest Shared Task using the large-scale text mining resource EVEX which we previously implemented using state-of-the-art algorithms, and which was applied to the whole of PubMed and PubMed Central. We participated in the Genia Event Extraction (GE) and Gene Regulation Network (GRN) tasks, ranking first in the former and fifth in the latter.
During the past few years, several novel text mining algorithms have been developed in the context of the BioNLP Shared Tasks on Event Extraction. These algorithms typically aim at extracting biomolecular interactions from text by inspecting only the context of one sentence. However, when humans interpret biomolecular research articles, they usually build upon extensive background knowledge of their favorite genes and pathways. To make such world knowledge available to a text mining algorithm, it could first be applied to all available literature to subsequently make a more informed decision on which predictions are consistent with the current known data. In this paper, we introduce our participation in the latest Shared Task using the large-scale text mining resource EVEX which we previously implemented using state-of-the-art algorithms, and which was applied to the whole of PubMed and PubMed Central. We participated in the Genia Event Extraction (GE) and Gene Regulation Network (GRN) tasks, ranking first in the former and fifth in the latter.
Ladattava julkaisu This is an electronic reprint of the original article. |  |
Turun yliopiston Tutkimustietojärjestelmän portaali