Modeling the Acquisition of English: An Intelligent CALL Approach Export

@article{citeulike:219676, abstract = {In this paper, we present a methodology for the development of a user model for CALL which captures various levels of language acquisition using individualized overlays supported with stereotypes. Our current focus is the empirical analysis of the order of written English grammatical structure acquisition in our learner population used to develop stereotype layers in our model.}, author = {Michaud, Lisa N. and Mccoy, Kathleen F. and Stark, Litza A.}, citeulike-article-id = {219676}, citeulike-linkout-0 = {http://citeseerx.ist.psu.edu/viewdoc/summary?doi=10.1.1.21.2148}, comment = {Goal is to bootstrap "stereotype-model" of stages in language acquisition. Use parser + human judge of correctness The stereotype model looks at: * Order of acquisition * "clumps" of acquisition (what features are learned around the same time) Stereotype model is useful as a bootstrap for forming a model of learner speech when we haven't yet gathered enough data on the learner. ---=note-separator=--- Results: Corpus 106 samples; 76\% annotator agreement. 1. There are indeed trends 2. Trends are not enough to precisely characterize learner ---=note-separator=--- ICICLE uses batche-processed interaction with learners (learner submits long composition, software reports aggregate (over multiple submissions) judgement. ---=note-separator=--- They used an English parser that has additional rules ("mal-rules") to model common L2A mistakes. Priors on this model are boosted by past learner performance (err, seems like this would create positive feedback effects). ---=note-separator=--- Feedback: uses ZPD theory to select errors that will be useful given the state of the learner.}, journal = {Lecture Notes in Computer Science}, keywords = {asl, call, error\_modeling, icicle, non\_pron\_errors, pedagogy, sla, syntax, zpd}, posted-at = {2005-06-04 19:57:24}, priority = {0}, title = {Modeling the Acquisition of English: An Intelligent CALL Approach}, url = {http://citeseerx.ist.psu.edu/viewdoc/summary?doi=10.1.1.21.2148}, volume = {2109}, year = {2001} }

TY - JOUR ID - citeulike:219676 L3 - citeulike-article-id:219676 N2 - In this paper, we present a methodology for the development of a user model for CALL which captures various levels of language acquisition using individualized overlays supported with stereotypes. Our current focus is the empirical analysis of the order of written English grammatical structure acquisition in our learner population used to develop stereotype layers in our model. JF - Lecture Notes in Computer Science TI - Modeling the Acquisition of English: An Intelligent CALL Approach VL - 2109 KW - asl KW - call KW - error_modeling KW - icicle KW - non_pron_errors KW - pedagogy KW - sla KW - syntax KW - zpd N1 - Goal is to bootstrap "stereotype-model" of stages in language acquisition. Use parser + human judge of correctness The stereotype model looks at: * Order of acquisition * "clumps" of acquisition (what features are learned around the same time) Stereotype model is useful as a bootstrap for forming a model of learner speech when we haven't yet gathered enough data on the learner. N1 - Results: Corpus 106 samples; 76% annotator agreement. 1. There are indeed trends 2. Trends are not enough to precisely characterize learner N1 - ICICLE uses batche-processed interaction with learners (learner submits long composition, software reports aggregate (over multiple submissions) judgement. N1 - They used an English parser that has additional rules ("mal-rules") to model common L2A mistakes. Priors on this model are boosted by past learner performance (err, seems like this would create positive feedback effects). N1 - Feedback: uses ZPD theory to select errors that will be useful given the state of the learner. AU - Michaud, Lisa AU - Mccoy, Kathleen AU - Stark, Litza PY - 2001/// UR - http://citeseerx.ist.psu.edu/viewdoc/summary?doi=10.1.1.21.2148 ER -