[FONT=Arial, Helvetica, sans-serif]
Free registration and access to corpus resources at
http://candle.cs.nthu.edu.tw/candle/
[/FONT]
Corpora And NLP for Digital Learning of English
T[FONT=Arial, Helvetica, sans-serif]he project uses various corpora-including a Chinese-English corpus, Sinorama, and various natural language processing (NLP) tools to construct advanced English learning tasks for intermediate learners in Taiwan such as college freshman students or those with a similar English proficiency level. It is thus named, Corpora And NLP for Digital Learning of English (CANDLE). The NLP tools used in CANDLE include:
[FONT=Arial, Helvetica, sans-serif] 1. Monolingual, Bilingual, Speech and Learner Corpora
2. Part of Speech Tagging, Chunking and Parsing
3. Bilingual Concordance, Semantic concordance and WordNet
4. Text-to-Speech Server
5. Phrase/Collocation Translator
6. Automatic Generation of CALL activities
7. Automatic Evaluation/Grader of Translation Quality Based on BLEU.[/FONT]
[FONT=Arial, Helvetica, sans-serif]Chunking helps English reading/listening, so do automatic word profiling, test generator, grader of free-text response, concordance-based language references, and text-to-speech generator on various language learning skills. The project is unique because it adopts a unique approach in Taiwan and internationally by using the Sinorama bilingual corpus and building on learners' first language background knowledge to empower learners with culture-based materials. Both first language and its culture provide a scaffold for learners to use while learning the new language. The project aims at various types of English learning activities by emphasizing structural knowledge, and complex problem solving learning: reading, writing, listening, speaking, and translation. Its major features include e-practice that adapts to learners' levels, automatically assessing/monitoring learner's progress, and profiling of learners' preferences.[/FONT]
[/FONT]
Free registration and access to corpus resources at
http://candle.cs.nthu.edu.tw/candle/
[/FONT]
Corpora And NLP for Digital Learning of English
T[FONT=Arial, Helvetica, sans-serif]he project uses various corpora-including a Chinese-English corpus, Sinorama, and various natural language processing (NLP) tools to construct advanced English learning tasks for intermediate learners in Taiwan such as college freshman students or those with a similar English proficiency level. It is thus named, Corpora And NLP for Digital Learning of English (CANDLE). The NLP tools used in CANDLE include:
[FONT=Arial, Helvetica, sans-serif] 1. Monolingual, Bilingual, Speech and Learner Corpora
2. Part of Speech Tagging, Chunking and Parsing
3. Bilingual Concordance, Semantic concordance and WordNet
4. Text-to-Speech Server
5. Phrase/Collocation Translator
6. Automatic Generation of CALL activities
7. Automatic Evaluation/Grader of Translation Quality Based on BLEU.[/FONT]
[FONT=Arial, Helvetica, sans-serif]Chunking helps English reading/listening, so do automatic word profiling, test generator, grader of free-text response, concordance-based language references, and text-to-speech generator on various language learning skills. The project is unique because it adopts a unique approach in Taiwan and internationally by using the Sinorama bilingual corpus and building on learners' first language background knowledge to empower learners with culture-based materials. Both first language and its culture provide a scaffold for learners to use while learning the new language. The project aims at various types of English learning activities by emphasizing structural knowledge, and complex problem solving learning: reading, writing, listening, speaking, and translation. Its major features include e-practice that adapts to learners' levels, automatically assessing/monitoring learner's progress, and profiling of learners' preferences.[/FONT]
[/FONT]