Skip to content

Latest commit

 

History

History
executable file
·
49 lines (28 loc) · 2.87 KB

README_factuality.md

File metadata and controls

executable file
·
49 lines (28 loc) · 2.87 KB
Raw

Bio-SCoRes-Factuality

Assessing factuality of semantic relations

Bio-SCoRes now incorporates a component for factuality prediction of semantic relations. It has so far used semantic predications extracted by SemRep (Rindflesch and Fiszman, 2003) as the basis, but it is capable of using relations extracted by other systems, as well, provided that the relations and their arguments are associated with textual mentions. See Kilicoglu et al. (2017) for more details.

Prerequisites

  • Java (jar files have been generated with 1.8, though it is possible to recompile with 1.7)
  • Ant (needed for recompilation, version 1.8 was used for compilation)

Directory Structure

dist directory contains libraries relevant to Bio-SCoRes-Factuality. These are the following:

  • ling.jar: Contains the core linguistic components used by Bio-SCoRes-Factuality.
  • factualitytasks.jar: Contains the factuality-related tasks.

To use Bio-SCoRes-Factuality from your application, ensure that ling.jar is included in your classpath. factualitytasks.jar is required if you plan to use/adapt the example factuality assessment pipeline described in Kilicoglu et al. (2017).

Data used for experiments in Kilicoglu et al. (2017) can be downloaded from https://skr3.nlm.nih.gov/Factuality. A UMLS license is required.

lib directory contains third-party libraries required by the system (see Note regarding Stanford Core NLP below.)

resources directory contains WordNet dictionary files that are required by the system as well as the factuality trigger file (factuality_dist.xml).

The top level directory contains ant build file as well as properties files used by the pipelines.

  • build.xml: Ant build file for all components.
  • factuality_semrep.properties: Properties file for SemRep factuality pipeline

Usage

If you're interested in incorporating Bio-SCoRes-Factuality into your NLP pipeline, a good starting point is the source code for the SemRep factuality pipeline (tasks.factuality.semrep.SemRepFactualityPipeline).

Note on Named Entity Recognition

Bio-SCoRes does not provide a named entity recognition module, but it can extract relations using a compositional interpretation method (Kilicoglu et al., 2015; 2017), if provided with named entities and relation triggers. However, for the experiments in Kilicoglu et al. (2017), it uses named entities and relations provided by SemRep and only attempts to address factuality of these semantic relations.

Note on Stanford CoreNLP package

Stanford CoreNLP model jar file that is needed for processing raw text for lexical and syntactic information (stanford-corenlp-3.3.1-models.jar) is not included with the distribution due to its size. It can be downloaded from http://stanfordnlp.github.io/CoreNLP/ and copied to lib directory.

Contact