Author:
Yushi Xu, and Stephanie Seneff
Abstract:
This paper concerns a framework for building interactive speech-based language learning games. The core of the framework, the �筂ialogue manager,�� controls the game procedure via a control script. The control script allows the developers to have easy access to the natural language process capabilities provided by six core building blocks. Using the framework, three games for Mandarin learning were implemented: a reading game, a translation game, and a question-answering game. We verified the effectiveness and usefulness of the framework by evaluating the three games. In the in-lab and public evaluation phases, we collected a total of 4025 utterances from 31 subjects. The evaluation showed that the game systems responded to the users�� utterances appropriately about 89% of the time, and assessment of the users�� performances correlated well with their human-judged proficiency.
Keywords:
Computer Aided Language Learning, Machine Translation, Automatic Question Generation, Automatic Answer Judging
Author:
Hao-Jan Howard Chen
Abstract:
Many ESL students need to improve writing skills to pass various language tests; thus, writing teachers need to read many compositions and provide feedback. To help ESL teachers reduce their teaching load and to give students faster feedback, various English grammar checkers have been developed. Few of these PC-based grammar checkers, however, are widely available to ESL learners. As the Internet has become an important tool for language education, web-based grammar checkers have begun to emerge. In this paper, we first introduce two new web-based grammar checkers (Microsoft ESL Assistant and NTNU statistical grammar checker) and then compare their performance. Ten common EFL errors selected from a large Chinese EFL learner corpus were used to test these two grammar checkers. The test results showed that the NTNU statistical checker was far more sensitive to various learner errors, and it could detect eight types of selected errors. Microsoft ESL Assistant could only deal with five types of errors. Moreover, these two checkers both could not deal with fragments and run-on sentences errors. It seems clear that both checkers have room for improvement before they can provide satisfactory service to ESL learners. The Microsoft ESL Assistant should expand its coverage to detect more learner errors. NTNU checker should reduce false alarms and indicate the locations of errors more accurately. Learner errors are indeed complicated for developers of grammar checkers, but the strong need for a functional grammar checker deserves CALL researchers�� special attention.
Keywords:
ESL Writing, Errors, Grammar Checker, Ngrams, Rules.
Author:
Sophia Skoufaki
Abstract:
This paper presents an initial attempt to examine whether Rhetorical Structure Theory (RST) (Mann & Thompson, 1988) can be fruitfully applied to the detection of the coherence errors made by Taiwanese low-intermediate learners of English. This investigation is considered warranted for three reasons. First, other methods for bottom-up coherence analysis have proved ineffective (e.g., Watson Todd et al., 2007). Second, this research provides a preliminary categorization of the coherence errors made by first language (L1) Chinese learners of English. Third, second language discourse errors in general have received little attention in applied linguistic research. The data are 45 written samples from the LTTC English Learner Corpus, a Taiwanese learner corpus of English currently under construction. The rationale of this study is that diagrams which violate some of the rules of RST diagram formation will point to coherence errors. No reliability test has been conducted since this work is at an initial stage. Therefore, this study is exploratory and results are preliminary. Results are discussed in terms of the practicality of using this method to detect coherence errors, their possible consequences about claims for a typical inductive content order in the writing of L1 Chinese learners of English, and their potential implications for Automated Writing Evaluation (AWE) software, since discourse organization is one of the essay characteristics assessed by this software. In particular, the extent to which the kinds of errors detected through the RST analysis match those located by Criterion (Burstein, Chodorow, & Leachock, 2004), a well-known AWE software by Educational Testing Service (ETS), is discussed.
Keywords:
Automated Writing Evaluation, Discourse Organization, Coherence Errors, Rhetorical Structure Theory
Author:
Ching-Ying Lee, and Jyi-Shane Liu
Abstract:
One of the most common lexical misuse problems in the second language context concerns near synonyms. Dictionaries and thesauri often overlook the nuances of near synonyms and make reference to near synonyms in providing definitions. The semantic differences and implications of near synonyms are not easily recognized and often fail to be acquired by L2 learners. This study addressed the distinctions of synonymous semantics in the context of second language learning and use. The purpose is to examine the effects of lexical collocation behaviors on identifying salient semantic features and revealing subtle difference between near synonyms. We conducted both analytical evaluation and empirical evaluation to verify that proper use of collocation information leads to learners�� successful comprehension of lexical semantics. Both results suggest that the process of organizing and identifying salient semantic features is favorable for and is accessible to a good portion of L2 learners, and thereby, improving near-synonym distinction.
Keywords:
Lexical Semantics, Near-synonym Distinction, Lexical Collocation Behavior
Author:
Siaw-Fong Chung
Abstract:
Using a corpus-based approach, this paper analyzes figurative language through observing the Chinese five elements (鈭磰�) of �� �𤉖etal,�� �� �姫ood,�� 瘞� �姫ater,�� �� �侰ire�� and �� �𨧡arth.�� This work found that there are at least two types of figurative language in Mandarin Chinese �� one of which occurs at the morphosyntactic level and the other occurs during the mappings between two domains (between the body part terms and these five elements). When the figurative uses of the co-occurring five elements with body part terms were tested in a psycholinguistic experiment composed of two groups of subjects (non-native and native speakers of Mandarin), a majority of the non-native speakers were unable to comprehend these figurative uses. This study attempts to prove that a linguistically-driven understanding of the five elements will be of great help to teaching or learning figurative language in a Mandarin L2 context.
Keywords:
Corpus, Five Elements, Figurative Language, Body Part, Learners of Chinese, Psycholinguistic Experiment