Chung-chi Huang

Carnegie Mellon University, Language Technologies Institute, Post-Doc

Followers

Following

Co-authors

Public Views

Exprience:
0. Currently a visiting scholar at Language Technologies Institute, CMU.
1. Post doc at Institute of Information Science, Academia Sinica, Taiwan
2. Lecturer of the class “Computer Systems & Applications” and “Intro. to Computer Systems & Applications” at Department of Computer Science and Foreign Languages and Literature, National Tsing Hua University 清華大學 (NTHU 清大 for short)
3. Teacher assistant of the class “Natural Language Processing”, “Academic Writing”, and “Natural Language Processing Lab” (in English) at NTHU
4. Received travel grant to present papers at conference EACL (2012), ACL (2011), AMTA (2010), PACLIC (2009), ISUC (2008) and RANLP (2007).
5. The NLP-related prototype systems I am currently working on:

 GRASP (http://koromiko.cs.nthu.edu.tw/grasp/), a demo system summarizing the grammatical/lexical contexts for collocations/phrases (accepts English query)
 GRASP_ch (http://koromiko.cs.nthu.edu.tw/grasp/ch), a demo system summarizing the grammatical/lexical contexts (with Chinese translations) for collocations/phrases
 TransAhead_prot (http://140.114.214.80/theSite/TransAhead/), a writing assistant suggesting following text and grammar of users' current translation given a source text. Lexical translations and phraseological tendencies are offered expected to improve language learners' and student translators' productivity and help language learning
 GRASP_prot (http://140.114.214.80/theSite/GRASP_v552/), a demo system summarizing the grammatical/lexical contexts for collocations /phrases (accepts English query)
 cross-lingual_GRASP_prot (http://140.114.214.80/theSite/bGRASP_v552/), a prototype system summarizing the grammatical/lexical contexts for collocations/phrases (accepts Traditional Chinese query)
 EdIt_prot (http://140.114.214.80/theSite/EdIt_demo2/), a prototype system correcting writing errors in English sentences with grammatical suggestions
 PatternAhead_prot (http://140.114.214.80/theSite/PatternAhead/), a prototype writing assistant providing phraseological tendencies both lexically (lexical phrases/bundles) and grammatically (grammatical constructions)

Specialties
1. Natural Language Processing,
2. Machine Translation,
3. Computer-Assisted Translation,
4. Computer-Assisted Language Learning,
5. Data Mining,
6. Information Extraction/Retrieval,
7. Sentiment Analysis (Reader Interest Analysis),
8. Machine Learning
Supervisors: Jaime Carbonell, Jason S. Chang, and Lun-Wei Ku
Phone: +886922777501

less

InterestsView All (16)

Uploads

Papers by Chung-chi Huang

Using a Paraphrase Reference Tool to Improve EFL Learners’ Writing Skills

by Meihua Chen and Chung-chi Huang

Paraphrasing, or restating information in other words, is an important writing skill for the acad... more Paraphrasing, or restating information in other words, is an important writing skill for the academic genre. Mastering paraphrasing skills helps language learners write and self-edit their works. However, there has been little research on developing automatic reference tools to assist language learners’ paraphrasing for better writing quality. In the light of this pressing need, we developed an automatic paraphrase reference tool, PREFER (PREFabricate Expression Recognizer) to help English learners vary their expressions and further improve their writing skills. PREFER is designed to automatically generate and display phrasal paraphrases for EFL learners’ reference (with Chinese as their first language). The method involves using phrasal translations in a bilingual parallel corpus and machine translation techniques. We infer the semantic equivalence between English phrases if they are aligned to the same Chinese phrase (i.e., the “pivot”). For example, given a phrase “on the whole”,...

An Automatic Reference Aid for Improving EFL Learners’ Formulaic Sequences in Productive Language Use

by Mei-Hua Chen and Chung-chi Huang

Formulaic language is important to language acquisition; however, English language learners are o... more Formulaic language is important to language acquisition; however, English language learners are often reported to have problems with formulaic expressions. Several lists of formulaic sequences have been proposed, mainly for developing teaching and testing materials. However, their limited numbers and insufficient usage information seem unable to benefit formulaic language use. To address these issues we have developed GRASP, a reference aid for formulaic expressions, to promote learners’ productive competence. Users are allowed multi-word inputs to target their desired phrases or collocations. Utilizing natural language processing techniques, our system categorizes and displays the structures and sequences in a hierarchical way. The corresponding example sentences are also provided. The formulaic structures serve as a quick access index. The formulaic sequences and corpus examples illustrate the real world language use. Importantly, automatic summarization from language data lends support to the idea of data-driven learning. A single-group pre-posttest design was adopted to assess the effectiveness of GRASP on 150 Chinese-speaking college freshmen. The results indicated that our reference aid made a substantial contribution to students’ performance on formulaic expression use in a sentence completion task, compared with the existing tools. Notably, the less proficient students showed marked improvement.

Download

A Computer-Assisted Translation and Writing System.

We introduce a method for learning to predict text and grammatical construction in a computer-ass... more We introduce a method for learning to predict text and grammatical construction in a computer-assisted translation and writing framework. In our approach, predictions are offered on the fly to help the user make appropriate lexical and grammar choices during the translation of a source text, thus improving translation quality and productivity. The method involves automatically generating general-to-specific word usage summaries (i.e., writing suggestion module), and automatically learning high-confidence word- or phrase-level translation equivalents (i.e., translation suggestion module). At runtime, the source text and its translation prefix entered by the user are broken down into n-grams to generate grammar and translation predictions, which are further combined and ranked via translation and language models. These ranked prediction candidates are iteratively and interactively displayed to the user in a pop-up menu as translation or writing hints. We present a prototype writing assistant, TransAhead, that applies the method to a human-computer collaborative environment. Automatic and human evaluations show that novice translators or language learners substantially benefit from our system in terms of translation performance (i.e., translation accuracy and productivity) and language learning (i.e., collocation usage and grammar). In general, our methodology of inline grammar and text predictions or suggestions has great potential in the field of computer-assisted translation, writing, or even language learning.

Interest Analysis using Social Interaction Content with Sentiments

We introduce a method for learning to predict reader interest. In our approach, interest analysis... more We introduce a method for learning to predict reader interest. In our approach, interest analysis bases on PageRank and social interaction content (e.g., reader feedback in social media). The method involves automatically estimating topical interest preferences and automatically determining the sentiments for social content. In interest prediction, difference content sources of articles and reader feedback representing readers' viewpoints are weighted accordingly and transformed into content-word weighted word graph. Then, PageRank suggests reader interest with the help of word interestingness scores. We present the prototype system, InterestFinder, that applies the method to interest analysis. Experimental evaluation shows that content source and content word weighting, and scores of interest preferences for words inferred across articles are quite helpful. And our system benefits more from subjective social interaction content than objective one and using none, in covering general readers' interest spans.

Interest Analysis using PageRank and Social Interaction Content

Using Sublexical Translations to Handle the OOV Problem in Machine Translation

We introduce a method for learning to translate out-of-vocabulary (OOV) words. The method focuses... more We introduce a method for learning to translate out-of-vocabulary (OOV) words. The method focuses on combining sublexical/constituent translations of an OOV to generate its translation candidates. In our approach, wildcard searches are formulated based on our OOV analysis, aimed at maximizing the probability of retrieving OOVs’ sublexical translations from existing resources of Machine Translation (MT) systems. At run-time, translation candidates of the unknown words are generated from their suitable sublexical translations and ranked based on monolingual and bilingual information. We have incorporated the OOV model into a state-of-the-art machine translation system and experimental results show that our model indeed helps to ease the impact of OOVs on translation quality, especially for sentences containing more OOVs (significant improvement).

TransAhead: A Computer-Assisted Translation and Writing Tool

We introduce a method for learning to predict text completion given a source text and partial tra... more We introduce a method for learning to predict text completion given a source text and partial translation. In our approach, predictions are offered aimed at alleviating users’ burden on lexical and grammar choices, and improving productivity. The method involves learning syntax-based phraseology and translation equivalents. At run-time, the source and its translation prefix are sliced into ngrams to generate and rank completion candidates, which are then displayed to users. We present a prototype writing assistant, TransAhead, that applies the method to computer-assisted translation and language learning. The preliminary results show that the method has great potentials in CAT and CALL with significant improvement in translation quality across users.

Download

EdIt: A Broad-Coverage Grammar Checker Using Pattern Grammar

We introduce a new method for learning to detect grammatical errors in learner’s writing and prov... more We introduce a new method for learning to detect grammatical errors in learner’s writing and provide suggestions. The method involves parsing a reference corpus and inferring grammar patterns in the form of a sequence of content words, function words, and parts-of-speech (e.g., “play ~ role in Ving” and “look forward to Ving”). At run-time, the given passage submitted by the learner is matched using an extended Levenshtein algorithm against the set of pattern rules in order to detect errors and provide suggestions. We present a prototype implementation of the proposed method, EdIt, that can handle a broad range of errors. Promising results are illustrated with three common types of errors in non-native writing.

Download

GRASP: Grammar- and Syntax-based Pattern-Finder in CALL

We introduce a method for learning to describe the attendant contexts of a given query for langua... more We introduce a method for learning to describe the attendant contexts of a given query for language learning. In our approach, its phraseology is represented as a summary of general pattern grammar (anchored with query words) to specific lexical phrases. The method involves syntactical analyses and inverted file construction. At run-time, grammatical constructions and their lexical instantiations characterizing the query’s usages are generated and organized expected to enhance users’ deep vocabulary knowledge. We present a prototype system, GRASP, that applies the method to collocation and phrase learning. Constrained experiments preliminarily show that language learners benefit more from GRASP context summarization than conventional lookup approaches, and that learner error correction system is within reach with our lexicalized pattern grammar.

Download

Learning to Mine Pattern Grammar for Computer-Assisted Language Learning

We introduce a method for learning to mine the attendant contexts of a given query for language l... more We introduce a method for learning to mine the attendant contexts of a given query for language learning. In our approach, it phraseology is represented as a summary of general pattern grammar to specific lexical phrases. The method involves syntactic analyses and inverted file construction. At run-time, grammatical constructions and their lexical instantiations characterizing the query’s usages are generated and organized expected to enhance users’ deep vocabulary knowledge. We present three prototype systems that apply the method to computer-assisted language learning. Experimental results show that language learners benefit more from our context summarization than conventional lookup approaches, and that learner error correction system and writing assistant are within reach with our grammatically- and lexically-motivated usage summaries.

Automatically Generating Grammar Patterns and Lexical Bundles for Assisting Language Learners’ Collocation Learning

We introduce a system, GRASP (Huang et al, 2011), that provides a general-to-specific usage summa... more We introduce a system, GRASP (Huang et al, 2011), that provides a general-to-specific usage summary of the query collocation or phrase to accelerate learners’ grasp of word usages. The summary consists of generalized contexts of collocations or phrases using parts-of-speech and their corresponding specific lexical instantiation (i.e., lexical bundles). Consider the collocation "reach agreement". GRASP characterizes its surrounding contexts using patterns “reach ARTICLE ADJ agreement” (as in “reach a preliminary agreement”) and “reach ~ agreement PREP DETERMINER (as in “reach ~ agreement on the”). Such grammar patterns as well as lexical bundles depict multi-word query’s phraseology lexically and grammatically.

An evaluation of GRASP applied to CALL was conducted with classroom students. In-between pre- and post-test of an English-to-Chinese sentence composition/translation task, half of the participants were instructed to use GRASP while the other half used traditional tools for word usage learning. Test items contain blanks to test participants’ knowledge on article, adjective, and preposition collocates. Encouragingly, the results show that GRASP boosted participants’ achievements three-times as much as the gain using the traditional tools and participants benefited from GRASP whether the level they were at. Overall, GRASP is promising and effective in assisting language learners in collocation and phrase learning.

Using A Paraphrase Reference Tool to Improve EFL Learners Writing Skills

Paraphrasing is an important writing skill for the academic genre. Mastering paraphrasing skills ... more Paraphrasing is an important writing skill for the academic genre. Mastering paraphrasing skills helps language learners write and self-edit their works. However, there has been little research on developing automatic reference tools to assist language learners’ paraphrasing for better writing quality. In the light of this pressing need, we developed PREFER, an automatic paraphrase reference tool, to help English learners vary their expressions and further improve their writing skills. PREFER automatically generate
phrasal paraphrases with a bilingual corpus using machine translation techniques. For example, given a phrase “on the whole”, PREFER generates its paraphrases “in a nutshell” which shares the same syntactic pattern “Prep_phrase_Noun” as well as
“generally speaking” which appears in different syntactic structures, also along with their corresponding example sentences for learners’ reference.

To assess whether PREFER benefits EFL learners, we carried out an experiment with 64 first-year EFL college students. They were asked to paraphrase seven short paragraphs in the pre-test with no e-help, and paraphrase another seven short paragraphs in the posttest using respectively three different reference tools. The results indicate that the students’ writing performance improved most with the assistance of PREFER, and approximately 75 percent of students agreed that PREFER benefits their writing task.

Bilingual Segmentation for Alignment and Translation

We propose a method that bilingually segments sentences in languages with no clear delimiter for ... more We propose a method that bilingually segments sentences in languages with no clear delimiter for word boundaries. In our model, we first convert the search for the segmentation into a sequential tagging problem, allowing for a polynomial-time dynamic-programming solution, and incorporate a control to balance monolingual and bilingual information at hand. Our bilingual segmentation algorithm, the integration of a monolingual language model and a statistical translation model, is devised to tokenize sentences more suitably for bilingual applications such as word alignment and machine translation. Empirical results show that bilingually-motivated segmenters outperform pure monolingual one in both the word-aligning (12% reduction in error rate) and the translating (5% improvement in BLEU) tasks, suggesting monolingual segmentation is useful in some aspects but, in a sense, not built for bilingual researches.

Computational Method for Collocation and Phrase Learning

We introduce a method for learning to find the representative syntax-based context of a given col... more We introduce a method for learning to find the representative syntax-based context of a given collocation/phrase. In our approach, grammatical patterns are extracted for query terms aimed at accelerating lexicographers’ and language learners’ navigation through the word usage and learning process. The method involves automatically lemmatizing, part-of-speech tagging and shallowly parsing the sentences of a large-sized general corpus, and automatically constructing inverted files for quick search. At run-time, contextual grammar patterns are retrieved and presented to users with their corresponding statistical analyses. We present a prototype system, GRASP (grammar- and syntax-based pattern-finder), that applies the method to computer-assisted language learning. Preliminary results show that the extracted patterns not only resemble phrases in grammar books (e.g., make up one’s mind) but are promising in assisting the process of language learning and lexicography.

A Thesaurus-Based Semantic Classification of English Collocations

Researchers have developed many computational tools aimed at extracting collocations for both sec... more Researchers have developed many computational tools aimed at extracting collocations for both second language learners and lexicographers. Unfortunately, the tremendously large number of collocates returned by these tools usually overwhelms language learners. In this paper, we introduce a thesaurus-based semantic classification model that automatically learns semantic relations for classifying adjective-noun (A-N) and verb-noun (V-N) collocations into different thesaurus categories. Our model is based on iterative random walking over a weighted graph derived from an integrated knowledge source of word senses in WordNet and semantic categories of a thesaurus for collocation classification. We conduct an experiment on a set of collocations whose collocates involve varying levels of abstractness in the collocation usage box of Macmillan English Dictionary. Experimental evaluation with a collection of 150 multiple-choice questions commonly used as a similarity benchmark in the TOEFL synonym test shows that a thesaurus structure is successfully imposed to help enhance collocation production for L2 learners. As a result, our methodology may improve the effectiveness of state-of-the-art collocation reference tools concerning the aspects of language understanding and learning, as well as lexicography.

Download

Fertility-based Source-Language-biased Inversion Transduction Grammar for Word Alignment

We propose a version of Inversion Transduction Grammar (ITG) model with IBM-style notation of fer... more We propose a version of Inversion Transduction Grammar (ITG) model with IBM-style notation of fertility to improve word-alignment performance. In our approach, binary context-free grammar rules of the source language, accompanied by orientation preferences of the target language and fertilities of words, are leveraged to construct a syntax-based statistical translation model. Our model, inherently possessing the characteristics of ITG restrictions and allowing for many consecutive words aligned to one and vice-versa, outperforms the Bracketing Transduction Grammar (BTG) model and GIZA++, a state-of-the-art word aligner, not only in alignment error rate (23% and 14% error reduction) but also in consistent phrase error rate (13% and 9% error reduction). Better performance in these two evaluation metrics suggests that, based on our word alignment result, more accurate phrase pairs may be acquired, leading to better machine translation quality.

Download

Chung-chi Huang

Uploads

Papers by Chung-chi Huang

Log In