Corpus creation date: 2018-04-22 Corpus release date: 2020-05-14 Version 1.0 Authors: Manolis Kapernaros, mkapernaros@di.uoa.gr The WikiWarsEL corpus ##################### 1. Introduction --------------- This is the README file for the distribution of the WikiWarsEL corpus, a Greek corpus containing Wikipedia [1] articles with annotations of temporal expressions. The creation of the corpus was motivated by the English, German and Vietnamese versions of the corpus, the WikiWars [2], WikiWarsDE [3] and WikiWarsVN [4] corpora. For the creation of WikiWarsEL as well as for the distribution, we followed the developers of WikiWarsVN corpus who in turn followed the developers of WikiWars corpus. Thus, a lot of the information mentioned in this file follows the information provided by them in their README file. Thanks to all of the above! 2. Licensing Terms ------------------ The licensing terms are according to the licensing terms of the WikiWars corpus: All the documents in the corpus are sources from Greek Wikipedia. As a consequence, the corpus is released under the Attribution-ShareAlike 3.0 Unported (CC BY-SA 3.0) License - To view a copy of this license, visit http://creativecommons.org/licenses/by-sa/3.0/ or send a letter to Creative Commons, PO Box 1866, Mountain View, CA 94042, USA. See also the Terms of Use at http://wikimediafoundation.org/wiki/Terms_of_Use. It is important to note that each time 'you distribute or publicly perform the work or a collection, the licensor offers to the recipient a license to the work on the same terms and conditions as the license granted to you under this license.' 3. Dictionary Structure ----------------------- The WikiWarsEL package is structured as follows: WikiWarsEL/readme.txt This is the readme file you are currently reading. WikiWarsEL/untagged/ This directory contains the 19 documents in TML format without annotations of temporal expressions. WikiWarsEL/tagged/ This directory contains the 19 documents with inline annotations of temporal expressions. 4. Corpus Creation ------------------ Here, we describe the Data Collection (Section 4.1), the Text Extraction and Preprocessing (Section 4.2) and the Annotation Process (Seciton 4.3). 4.1 Data Collection ------------------- For data collection, we used the cross-language links from the Wikipedia sites of the documents contained in the WikiWars corpus [2]. Thus the WikiWarsEL corpus contains sections of the following Wikipedia articles: NAME OF THE WAR YEAR SPAN PAGE NAME AT https://el.wikipedia.org/wiki/ World War II 1939–1945 Β΄_Παγκόσμιος_Πόλεμος World War I 1914–1918 Α΄_Παγκόσμιος_Πόλεμος American Civil War 1861–1865 Αμερικανικός_Εμφύλιος_Πόλεμος American Revolution 1775–1783 Αμερικανική_Επανάσταση Vietnam War 1955–1975 Πόλεμος_του_Βιετνάμ Korean War 1950–1953 Πόλεμος_της_Κορέας Iraq War 2003–... Πόλεμος_στο_Ιράκ French Revolution 1789–1799 Γαλλική_Επανάσταση Persian Wars 499–450 BC Περσικοί_Πόλεμοι Punic Wars 264–146 BC Καρχηδονιακοί_Πόλεμοι Chinese Civil War 1945–1949 Κινέζικος_Εμφύλιος_Πόλεμος Iran-Iraq War 1980–1988 Πόλεμος_Ιράν-Ιράκ Russian Civil War 1917–1923 Ρωσικός_Εμφύλιος_Πόλεμος French Indochina War 1946–1954 - N/A Mexican Revolution 1911–1920 - N/A Spanish Civil War 1936–1939 Ισπανικός_εμφύλιος_πόλεμος French-Algerian War 1954–1962 Πόλεμος_της_Αλγερίας Soviet-Afghanistan War 1979–1989 Σοβιετικός_πόλεμος_στο_Αφγανιστάν Russo-Japanese War 1904–1905 Ρωσοϊαπωνικός_Πόλεμος Russo-Polish War 1919–1921 Πολωνο-Σοβιετικός_Πόλεμος Biafran War 1967–1970 - N/A Abyssinian War 1935–1936 Β΄_Ιταλο-Αιθιοπικός_Πόλεμος For several of the English documents, there were no Greek versions available and as a result, the WikiWarsEL corpus contains only 19 and not the 22 documents that WikiWars and WikiWarsDE contain. 4.2 Text Extraction and Preprocessing ------------------------------------- Following Mazur and Dale (2010), we manually copied sections of the documents that describe the course of the wars. All pictures, cross-page references, and citations were removed. In contrast to WikiWars and WikiWarsDE, all text files were then converted into the TimeML format with TIMEX3 annotations, the same format used in WikiWarsVN. Due to this format, the corpus can be evaluated using the evaluation tools supplied in HeidelTime's website [5] Corpora Preparation and Evaluation Scripts. 6. Versions & Changes --------------------- Version 1.0: Initial release. 7. References ------------- [1] http://www.wikipedia.org/ [2] Pawel Mazur and Robert Dale (2010). WikiWars: A New Corpus for Research on Temporal Expressions. In the Proceedings of the Conference on Empirical Methods in Natural Language Processing (EMNLP), 9-11 October 2010, Massachusetts, USA. [3] Jannik Strötgen and Michael Gertz: WikiWarsDE: A German Corpus of Narratives Annotated with Temporal Expressions. In the Proceedings of the Conference of the German Society for Computational Linguistics and Language Technology (GSCL). Pages 129-134, 2011. based Vietnamese Text Processing Tool", http://jvntextpro.sourceforge.net/, 2010 [4] Jannik Strötgen, Ayser Armiti, Tran Van Canh, Julian Zell, and Michael Gertz. Time for More Languages: Temporal Tagging of Arabic, Italian, Spanish, and Vietnamese. In: ACM Transactions on Asian Language Information Processing (TALIP), 13(1), pages 1-21, 2014, ACM. [5] https://dbs.ifi.uni-heidelberg.de/resources/heideltime-corpora-dl/