MétaCan
Menu
Back to cohort
Record W2795042751

Collocation and collocation error processing in the context of second language learning

2018· dissertation· en· W2795042751 on OpenAlex

Why this work is in the frame

A frame that forgets how it found something cannot be audited. These are the routes that admitted this work.

aboutThe title or abstract carries a Canadian signal from the geographic lexicon.
no affNo Canadian affiliation: this work is invisible to an affiliation-only frame.
No Canadian affiliation. An affiliation-only frame, the usual design, would never have seen this work. It is one of the works that make the case for inverting the frame.

Bibliographic record

VenueTDX (Tesis Doctorals en Xarxa) · 2018
Typedissertation
Languageen
FieldComputer Science
TopicText Readability and Simplification
Canadian institutionsnot available
Fundersnot available
KeywordsHumanitiesPhilosophyArtificial intelligenceComputer science
DOInot available

Abstract

fetched live from OpenAlex

La presente tesis tiene como foco el procesamiento de colocaciones y errores colocacionales. El aprendizaje de colocaciones es uno de los mayores retos a la hora de aprender una lengua, por lo que el desarrollo de herramientas computacionales avanzadas de ayuda al aprendizaje de las colocaciones es altamente deseable. En concreto, en la tesis se proponen (i) tecnicas de clasificacion semantica de colocaciones (basadas en la tipologia de las Funciones Lexicas), para la generacion automatica de recursos en los que, dada una palabra clave y un significado particular, sea posible encontrar el/los colocativo(s) que co-ocurre(n) con dicha palabra para expresar el significado dado y (ii) tecnicas de reconocimiento y clasificacion automatico de errores colocacionales en textos de aprendices de espanol, siguiendo una tipologia detallada que da cuenta de varias clases de errores lexicos y gramaticales, asi como de la localizacion del error. Tras la introduccion (Chapter 1), se presenta el marco teorico (Chapter 2), en el que se describe en detalle el concepto de colocacion, junto con las tipologias (semantica y de errores) que se usan en los experimentos. Posteriormente (Chapter 3), se detalla el estado del arte (i) sobre tecnicas para la extraccion de colocaciones, (ii) sobre tecnicas de clasificacion semantica de colocaciones, (iii) de metodos para la deteccion y correccion de errores colocacionales, (iv) de metodos para la deteccion y correccion de errores gramaticales y (v) sobre la generacion y uso de corpus artificiales para las tareas de deteccion y correccion de errores. Los capitulos 4-6 forman el cuerpo central de la tesis. Primero, se presentan dos metodos de extraccion y clasificacion semantica de colocaciones (Chapter 4). Luego, se proponen metodos basados en reglas y en tecnicas de aprendizaje automatico para la clasificacion de errores colocacionales (Chapter 5). Finalmente, se propone el uso de long short-term memory networks (LSTMs) para la deteccion y clasificacion simultanea de errores colocacionales (Chapter 6). Debido al reducido tamano del corpus de estudiantes disponible, se propone un algoritmo para la generacion automatica de un corpus de errores colocacionales artificial, y se demuestra que, entrenando el sistema con el corpus sintetico, es posible detectar y clasificar errores colocacionales de naturaleza gramatical y lexica en el corpus 'real' de aprendices. Referencias: - Margarita Alonso Ramos, Leo Wanner, Orsolya Vincze, Gerard Casamayor del Bosque, Nancy Vazquez Veiga, Estela Mosqueira Suarez, and Sabela Prieto Gonzalez. Towards a Motivated Annotation Schema of Collocation Errors in Learner Corpora. In Proceedings of the 7th International Conference on Language Resources and Evaluation (LREC), pages 3209-3214, La Valetta, Malta, 2010. - Cristobal Lozano. CEDEL2: Corpus Escrito del Espanol L2. In C.M. Bretones Callejas, editor, Applied Linguistics Now: Understanding Language and Mind, pages 197-212. Universidad de Almeria, Almeria, 2009. - Igor Mel'cuk. Lexical Functions: A Tool for the Description of Lexical Relations in a Lexicon. In L. Wanner, editor, Lexical functions in Lexicography and Natural Language Processing, 31:37-102, John Benjamins Publishing Company, 1996. - Tomas Mikolov, Ilya Sutskever, Kai Chen, Greg S. Corrado, and Jeff Dean. Distributed Representations of Words and Phrases and their Compositionality. In Advances in Neural Information Processing Systems (NIPS), pages 3111-3119, Montreeal, Canada, 2013. Articulos en relacion a la tesis: - Sara Rodriguez-Fernandez, Roberto Carlini, and Leo Wanner. Classification of Lexical Collocation Errors in the Writings of Learners of Spanish. In Proceedings of the International Conference on Recent Advances in Natural Language Processing (RANLP), pages 529-536, Hissar, Bulgaria, 2015. - Sara Rodriguez-Fernandez, Roberto Carlini, and Leo Wanner. Classification of Grammatical Collocation Errors in the Writings of Learners of Spanish. Procesamiento del Lenguaje Natural, 55:49-56, 2015. - Sara Rodriguez-Fernandez, Luis Espinosa-Anke, Roberto Carlini, and Leo Wanner. Semantics-Driven Recognition of Collocations Using Word Embeddings. In Proceedings of the 54th Annual Meeting of the Association for Computational Linguistics (ACL), pages 499-505, Berlin, Germany, 2016. - Sara Rodriguez-Fernandez, Luis Espinosa-Anke, Roberto Carlini, and Leo Wanner. Example-based Acquisition of Fine-grained Collocation Resources. In Proceedings of the 10th International Conference on Language Resources and Evaluation (LREC), pages 2317-2322, Portoroz, Slovenia, 2016. - Sara Rodriguez-Fernandez, Luis Espinosa-Anke, Roberto Carlini, and Leo Wanner. Semantics-Driven Collocation Discovery. Procesamiento del Lenguaje Natural, 57:57-64, 2016. - Sara Rodriguez-Fernandez, Roberto Carlini, and Leo Wanner. Generation of a Spanish Artificial Collocation Error Corpus. In Proceedings of the 11th International Conference on Language Resources and Evaluation (LREC), Miyazaki, Japan, 2018.

Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.

Full frame distilled prediction

Teacher imitation

Not calibrated prevalence, not ground truth. Human validation pending. Learned from the 10,348 direct Codex labels and 10,348 direct Gemma labels. Candidate is the union of thresholded teacher heads; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels or direct frontier model labels.

metaresearch head score (Codex)0.001
metaresearch head score (Gemma)0.000
Version: codex-gemma-dda1882f352aValidation status: machine_predicted_unvalidated
Candidate categoriesnone
Consensus categoriesnone
DomainCandidate signal: none · Consensus signal: none
Study designCandidate signal: Observational · Consensus signal: none
GenreCandidate signal: Empirical · Consensus signal: Empirical
Teacher disagreement score0.791
Threshold uncertainty score0.803

Codex and Gemma teacher scores by category

CategoryCodexGemma
Metaresearch0.0010.000
Meta-epidemiology (narrow)0.0000.000
Meta-epidemiology (broad)0.0000.000
Bibliometrics0.0000.001
Science and technology studies0.0000.000
Scholarly communication0.0000.001
Open science0.0010.000
Research integrity0.0000.000
Insufficient payload (model declined to judge)0.0000.000

Machine scores (provisional)

The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.

Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.

Opus teacher head0.020
GPT teacher head0.310
Teacher spread0.290 · how far apart the two teachers sit on this one work
Validation statusscore_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from it