An annotated Taiwanese Learners' Corpus of Spanish, CATE

Abstract

Español, CATE: (http://140.116.245.228/ ) -is unique for being the first and largest annotated Taiwanese learners' corpus of Spanish with multi-query functions in the on-line interface. This corpus has collaborated with other L3 learners' corpora under the integrated project entitled "Multilingual Corpora" executed by the FLLD (Department of Foreign Languages and Literatures) and technically supported by the WMMKS ( Web Mining and Multilingual Knowledge S ystem) Laboratory of CSIE (Computer Science and Information Engineering) of the National Cheng Kung University in Taiwan.There are four steps in the procedure of corpus creation: data collecting, entry, annotation and programming. The written texts for our corpus have been collected since 2005, contributed by Taiwanese learners of different levels (categorized according to learning hours of Spanish) 2 from 15 universities 3 in Taiwan who take Spanish as their second foreign language and English as their first. During the first four years, 1,913 compositions (337,122 words) have been collected. Corresponding to these written texts, two types of general information have been compiled and processed in the database: (1) the authors' personal information (sex, department, university, learning hours of Spanish, special background related to Spanish learning including being immigrants, exchange students, transfers and previous learning experience) and (2) its attribution (length, task type, textual type, and theme). For the purpose of public use, a web-oriented system has been designed, and the authors of all compositions have signed a consent form agreeing that their written production can be used for academic research.In order to facilitate more advanced research of error and contrastive analysis, learners' errors have been correction-annotated. So far, the mistakes in the compiled texts in the sub-corpus CATE [2005][2006] (1,058 compositions, 181,322 words) made by the participants have been manually corrected by native speakers of Spanish and saved in an XML format by applying the Corpus Tool of UAM (Universitad Autónoma de Madrid) developed by Mick Brought to you by | Simon Fraser University Authenticated Download Date | 6/2/15 12:11 AM

Proposing a Customised Method for Extratextual Documentative Annotation on Written Text Corpus

“…In most cases, these methods are employed differently keeping in mind the different types of end application of the annotated texts. So far we have come across several types of text annotation, such as, grammatical annotation (Garside, Leech, & McEnery, 1997;Santorini, 1990;Brants et al, 2004;Schmid, 2008); syntactic annotation (Aldebazal et al, 2009); orthographic annotation (Leech, McEnery, & Wynne, 1997), prosodic annotation (Cox, 2011), semantic annotation (Rayson & Stevenson, 2008), discoursal annotation (Kipp, Neff, & Albrecht, 2007), anaphoric annotation (Lu, 2010), etymological annotation (Dash & Hussain, 2013), figurative annotation (Oostdijk & Boves, 2008), pragmatic annotation (Archer, Culpeper, & Davies, 2008); and sociopragmatic annotation (Archer & Culpeper, 2003), etc. Also we have come across a few text encoding techniques and standards, such as, Constituent Likelihood Automatic Word-tagging System (CLAWS) (Garside, 1987;McEnery & Hardie, 2011), Text Encoding Initiative (TEI) (Sperberg-McQueen & Burnard, 1994), Hidden Markov Model (HMM) (Kupiec, 1992), COCOA annotation method (McEnery & Wilson, 1996), Dynamic Programming Method (DPM) (DeRose, 1988), and EAGLE annotation guideline (Sinclair, 1996), etc.…”

Linguistic Annotation in/for Corpus Linguistics

“…One learner corpus project for which English is not the target language is the Corpus of Taiwanese Learners' Corpus of Spanish, which contains data from Taiwanese speakers (L2: English, L3: Spanish) of different levels from 15 universities. The corpus is richly annotated for parts-of-speech, lemmas, and errors made by the learners, and made available in XML format [52].…”

El corpus de aprendices japoneses CELEN y su aplicación a la docencia y la investigación en ELE

“…En este ámbito, los corpus nos sirven para conocer la frecuencia de los elementos lingüísticos (por ejemplo, con qué frecuencia los aprendices usan ciertas palabras o estructuras) o responder dudas que no podemos resolver con nuestra propia intuición (por ejemplo, qué preposiciones suelen usar con ciertos verbos). Además de este uso pedagógico, para el que el corpus fue diseñado en un principio, los datos contenidos en él pueden ser de ayuda también en la investigación sobre el aprendizaje y la adquisición de lenguas extranjeras (en el análisis de errores, la lingüística corpus de reducido tamaño compilados para llevar a cabo investigaciones particulares, existen en este momento tres que pueden ser consultados libremente (CEDEL2 (Lozano, 2022); CAES (Palacios et al, 2019); y COWS-L2H (Yamada et al, 2020)), otros tres mediante registro (Aprescrilov (Buyse y González, 2013), CATE (Lu, 2010) y CORESPI (Bailini y Frigerio, 2019)) y uno mediante compra (CORANE (Cestero Mancera et al, 2001)).…”