USAAR-CHRONOS: Crawling the Web for Temporal Annotations
22600Authors
nan
Abstract
This paper describes the USAAR-CHRONOS participation in the Diachronic Text Evalua- tion task of SemEval-2015 to identify the time period of historical text snippets. We adapt a web crawler to retrieve the original source of the text snippets and determine the publi- cation year of the retrieved texts from their URLs. We report a precision score of >90% in identifying the text epoch. Additionally, by crawling and cleaning the website that hosts the source of the text snippets, we present Daikon, a corpus that can be used for fu- ture work on epoch identification from a di- achronic perspective. 1
Keywords
Full Content
Proceedings of the 9th International Workshop on Semantic Evaluation (SemEval 2015), pages 846–850, Denver, Colorado, June 4-5, 2015. c⃝2015 Association for Computational Linguistics USAAR-CHRONOS: Crawling the Web for Temporal Annotations Liling Tan and Noam Ordan Universit¨at des Saarlandes Campus A2.2, Saarbr¨ucken, Germany alvations@gmail.com, noam.ordan@uni-saarland.de Abstract This paper describes the USAAR-CHRONOS participation in the Diachronic Text Evalua- tion task of SemEval-2015 to identify the time period of historical text snippets. We adapt a web crawler to retrieve the original source of the text snippets and determine the publi- cation year of the retrieved texts from their URLs. We report a precision score of >90% in identifying the text epoch. Additionally, by crawling and cleaning the website that hosts the source of the text snippets, we present Daikon, a corpus that can be used for fu- ture work on epoch identification from a di- achronic perspective. 1 Introduction ”Time changes all things: there is no reason why language should escape this universal law” (De Saussure, 1959). Traditionally, there are two ways to collect linguistic data to explore how words change over time, viz. (i) the ‘armchair’ method and (ii) the ‘tape-recorder’ method (Aitchison, 2001). In the first, the linguist cross-examines numerous doc- uments from bygone years and in the latter, the lin- guist goes around recording language and studies the changes as they happen. With the ingress of historical data provided by Google (Michel et al. 2011), the ‘armchair’ method goes into warp speed as computational linguists ex- plore the different facets of lexical changes in En- glish (Mihalcea and Nastase, 2012; Popescu and Strapparava, 2013; Niculae et al., 2014). This paper presents the Saarland University (USAAR-CHRONOS) participation in the Di- achronic Text Evaluation task in SemEval-2015. We participated in Subtask 1 that requires participants to identify the year of publication for texts with clear reference to time anchors (i.e. explicit references to famous persons or events). 1.1 Task Definition In Subtask 1 of the Diachronic Text Evaluation par- ticipants are required to identify the epoch (i.e. time period) of a text snippet with clear reference to cer- tain famous persons or events. The text snippets may not necessarily contain temporal information such as year or date but it has clear reference to a historical event that can be identified from external knowledge bases. For instance, given the following text, partic- ipants are required to identify its epoch: “Dictator Saddam Hussein ordered his troops to march into Kuwait. After the invasion is condemned by the UN Security Council, the US has forged a coalition with allies. Today American troops are sent to Saudi Arabia in Operation Desert Shield, protecting Saudi Arabia from possible attack.” The text has clear temporal evidence with refer- ence to a historical figure (“Saddam Hussein”), a notable organization (“UN Security Council”) and a factual event (“Operation Desert Shield”). Histori- cally, we know that Saddam Hussein lived between 1937 to 2006, that the UN Security Council has ex- isted since 1946 and that Operation Desert Shield (i.e. the Gulf …
Content truncated to 500 words
Labeled Text
This paper describes the USAAR-CHRONOS participation in the Diachronic Text Evaluation task of SemEval-2015 to identify the time period of historical text snippets. We adapt a web crawler to retrieve the original source of the text snippets and determine the publication year of the retrieved texts from their URLs. We report a precision score of >90% in identifying the text epoch. Additionally, by crawling and cleaning the website that hosts the source of the text snippets, we present Daikon, a corpus that can be used for future work on epoch identification from a diachronic perspective. 1
Publication Details
- Paper ID:
- 22600
- Year:
- 2015
- Publication:
- TIPSTER
- Publication Type:
- conf
- DOI:
- nan