Limits...
Event-based text mining for biology and functional genomics.

Ananiadou S, Thompson P, Nawaz R, McNaught J, Kell DB - Brief Funct Genomics (2014)

Bottom Line: This article provides an overview of recent research into event extraction.We cover annotated corpora on which systems are trained, systems that achieve state-of-the-art performance and details of the community shared tasks that have been instrumental in increasing the quality, coverage and scalability of recent systems.Finally, several concrete applications of event extraction are covered, together with emerging directions of research.

View Article: PubMed Central - PubMed

No MeSH data available.


Interface to EVEX database, showing results after searching for the gene ATR.
© Copyright Policy - creative-commons
Related In: Results  -  Collection

License
getmorefigures.php?uid=PMC4499874&req=5

elu015-F8: Interface to EVEX database, showing results after searching for the gene ATR.

Mentions: Another event-based system offers a user interface over the EVEX database [94], allowing search based on the 40 million bio-molecular events extracted from 21.9 million PubMed abstracts and 460 000 PubMed Central open access full-text articles. Selecting a particular gene causes the event types in which it participates to be identified. In Figure 8, the events displayed involve the gene ATR. The statement ‘ATR regulates 82 genes or proteins’ denotes that ATR has been identified as the Cause of regulation events, in which 82 unique genes or proteins have been identified as the Theme. An example of an event involving each of these genes/proteins is displayed. For each gene/protein, links allow the user to further ‘drill down’ to information of interest, e.g. to find further examples of the given event type with a specific Cause and Theme, or to discover further event types involving a specific pair of genes/proteins. The events displayed in Figure 8 provide further evidence of how discourse contexts are important in distinguishing between different event interpretations (as explained in the section ‘Interpretation of Bio-Events’ above), and thus that such search systems could benefit from taking this information into account. For example, in the first row, which describes an interaction between ATR and Nor1, the word ‘find’ denotes that the event is stated based on experimental observations, while the word ‘weakly’ denotes that intensity of the regulation is very low.Figure 8:


Event-based text mining for biology and functional genomics.

Ananiadou S, Thompson P, Nawaz R, McNaught J, Kell DB - Brief Funct Genomics (2014)

Interface to EVEX database, showing results after searching for the gene ATR.
© Copyright Policy - creative-commons
Related In: Results  -  Collection

License
Show All Figures
getmorefigures.php?uid=PMC4499874&req=5

elu015-F8: Interface to EVEX database, showing results after searching for the gene ATR.
Mentions: Another event-based system offers a user interface over the EVEX database [94], allowing search based on the 40 million bio-molecular events extracted from 21.9 million PubMed abstracts and 460 000 PubMed Central open access full-text articles. Selecting a particular gene causes the event types in which it participates to be identified. In Figure 8, the events displayed involve the gene ATR. The statement ‘ATR regulates 82 genes or proteins’ denotes that ATR has been identified as the Cause of regulation events, in which 82 unique genes or proteins have been identified as the Theme. An example of an event involving each of these genes/proteins is displayed. For each gene/protein, links allow the user to further ‘drill down’ to information of interest, e.g. to find further examples of the given event type with a specific Cause and Theme, or to discover further event types involving a specific pair of genes/proteins. The events displayed in Figure 8 provide further evidence of how discourse contexts are important in distinguishing between different event interpretations (as explained in the section ‘Interpretation of Bio-Events’ above), and thus that such search systems could benefit from taking this information into account. For example, in the first row, which describes an interaction between ATR and Nor1, the word ‘find’ denotes that the event is stated based on experimental observations, while the word ‘weakly’ denotes that intensity of the regulation is very low.Figure 8:

Bottom Line: This article provides an overview of recent research into event extraction.We cover annotated corpora on which systems are trained, systems that achieve state-of-the-art performance and details of the community shared tasks that have been instrumental in increasing the quality, coverage and scalability of recent systems.Finally, several concrete applications of event extraction are covered, together with emerging directions of research.

View Article: PubMed Central - PubMed

No MeSH data available.