Main content

Home

Menu

Loading wiki pages...

View
Wiki Version:
# SENTiVENT Economic Event Annotation Guidelines Annotation Guidelines for economic Events in the SENTiVENT project for economic news text mining. The goal of this annotation scheme is to produce a gold-standard labeled dataset for enabling supervised event extraction in the company-specific news text domain. The guidelines are based on the [Rich-ERE Guidelines][1] for [Events][2] and [Argument Fillers][3] but adapted to a corpus of business and financial news articles. We exclusively annotate event structures, unlike Rich ERE which annotates Entities and Relations separately. Version 1.1. of these Guidelines were used for SENTiVENT English Event 1.0 dataset release. ## Contents This repository contains the original LaTeX files as well as the report in .pdf. This Technical Report contains the Annotation Guidelines are described in detail. Rationale is also given for the event taxonomy as well insights in the iterative annotation process. 1. Introduction: discusses the motivation behind the SENTiVENT project (sec. 1.1), as well as the annotator workflow (sec. 1.2), annotator training and annotation tools used (sec. 1.3.) 2. Terminology and Typography: explains basic concepts and Terminology. 3. Event Trigger and Type: defines event triggers and their attributes, rules for annotating event triggers and how they are realized. 4. Event Arguments: defines event arguments and specifies rules for annotation. 5. Event Coreference: specifies the annotation of event coreference. 6. Event Typology: describes the event typology/taxonomy. Describes event types and subtypes in detail with examples and potential sources of confusion. 7. Guidelines Improvement Plan: Describes issues encountered during annotation and mitigations for future refence. Guidelines were frozen for annotation at v1.0., this section was added in v1.1. 8. Bibliography ## References [1]: Linguistic Data Consortium. (2016). Rich ERE Annotation Guidelines Overview V4.2, 1–14. Retrieved from http://www.nist.gov/tac/2016/KBP/guidelines/summary_rich_ere_v4.2.pdf on September 2018 [2]: Linguistic Data Consortium. (2015). DEFT Rich ERE Annotation Guidelines: Events V3.0, 1–14. [3]: Linguistic Data Consortium. (2015). DEFT Rich ERE Annotation Guidelines: Argument Filler V2.3, 1–14. ## LaTeX Compilation XeLaTeX engine recommended, LuaLatex and pdfLatex produce errors on Overleaf.com editor. ## Author contact - Gilles Jacobs: GillesM.Jacobs@ugent.be, gilles@jacobsgill.es, gillesjacobs@pm.me - Veronique Hoste: Veronique.Hoste@ugent.be ## Mirrors This Technical Report is stored and mirrored in the following locations: - Primary source: https://github.com/GillesJ/sentivent-event-annotation-guidelines (@commit:ddff83aa8fcba0df328cfc6ebc4dc9b8148aa8f1 on date of publication.) - Mirror: osf.io/xqzw5 ## How to cite ``` @techreport{jacobs2020sentiventeventguidelines, author = {{Gilles Jacobs}}, title = {{SENTiVENT Event Annotation Guidelines v1.1}}, year = {2020}, month = {1}, publisher = {{Language and Translation Technology Team, Ghent University}}, url = {{https://osf.io/xqzw5}}, pages = {69}, doi = {10.17605/OSF.IO/XQZW5} } ```
OSF does not support the use of Internet Explorer. For optimal performance, please switch to another browser.
Accept
This website relies on cookies to help provide a better user experience. By clicking Accept or continuing to use the site, you agree. For more information, see our Privacy Policy and information on cookie use.
Accept
×

Start managing your projects on the OSF today.

Free and easy to use, the Open Science Framework supports the entire research lifecycle: planning, execution, reporting, archiving, and discovery.