A grammar of spoken Chinese = 中國話的文法

Learning Vocabulary in Another Language: Appendixes

A phrase-based statistical machine translation approach — the alignment template approach — is described. This translation approach allows for general many-to-many relations between words. Thereby, the context of words is taken into account in the translation model, and local changes in word order from source to target language can be learned explicitly. The model is described using a log-linear modeling approach, which is a generalization of the often used source–channel approach. Thereby, the model is easier to extend than classical statistical machine translation systems. We describe in detail the process for learning phrasal translations, the feature functions used, and the search algorithm. The evaluation of this approach is performed on three different tasks. For the German–English speech VERBMOBIL task, we analyze the effect of various system components. On the French–English Canadian HANSARDS task, the alignment template system obtains significantly better results than a single-word-based translation model. In the Chinese–English 2002 National Institute of Standards and Technology (NIST) machine translation evaluation it yields statistically significantly better NIST scores than all competing research and commercial translation systems.

https://files.ifi.uzh.ch/dbtg/sdbs13/T07.0.pdf

The Alignment Template Approach to Statistical Machine Translation

Community-run, formal evaluations and manually annotated text corpora are critically important for advancing biomedical text-mining research. Recently in BioCreative V, a new challenge was organized for the tasks of disease named entity recognition (DNER) and chemical-induced disease (CID) relation extraction. Given the nature of both tasks, a test collection is required to contain both disease/chemical annotations and relation annotations in the same set of articles. Despite previous efforts in biomedical corpus construction, none was found to be sufficient for the task. Thus, we developed our own corpus called BC5CDR during the challenge by inviting a team of Medical Subject Headings (MeSH) indexers for disease/chemical entity annotation and Comparative Toxicogenomics Database (CTD) curators for CID relation annotation. To ensure high annotation quality and productivity, detailed annotation guidelines and automatic annotation tools were provided. The resulting BC5CDR corpus consists of 1500 PubMed articles with 4409 annotated chemicals, 5818 diseases and 3116 chemical-disease interactions. Each entity annotation includes both the mention text spans and normalized concept identifiers, using MeSH as the controlled vocabulary. To ensure accuracy, the entities were first captured independently by two annotators followed by a consensus annotation: The average inter-annotator agreement (IAA) scores were 87.49% and 96.05% for the disease and chemicals, respectively, in the test set according to the Jaccard similarity coefficient. Our corpus was successfully used for the BioCreative V challenge tasks and should serve as a valuable resource for the text-mining research community.Database URL: http://www.biocreative.org/tasks/biocreative-v/track-3-cdr/.

BioCreative V CDR task corpus: a resource for chemical disease relation extraction

Inspired by the success of the General Language Understanding Evaluation benchmark, we introduce the Biomedical Language Understanding Evaluation (BLUE) benchmark to facilitate research in the development of pre-training language representations in the biomedicine domain. The benchmark consists of five tasks with ten datasets that cover both biomedical and clinical texts with different dataset sizes and difficulties. We also evaluate several baselines based on BERT and ELMo and find that the BERT model pre-trained on PubMed abstracts and MIMIC-III clinical notes achieves the best results. We make the datasets, pre-trained models, and codes publicly available at https://github.com/ ncbi-nlp/BLUE_Benchmark.

/pdf/transfer-learning-in-biomedical-natural-language-processing-1af16rz1tc.pdf

Transfer Learning in Biomedical Natural Language Processing: An Evaluation of BERT and ELMo on Ten Benchmarking Datasets.

One effective way to improve the state of the art is through competitions. Following the success of the Critical Assessment of protein Structure Prediction (CASP) in bioinformatics research, a number of challenge evaluations have been organized by the text-mining research community to assess and advance natural language processing (NLP) research for biomedicine. In this article, we review the different community challenge evaluations held from 2002 to 2014 and their respective tasks. Furthermore, we examine these challenge tasks through their targeted problems in NLP research and biomedical applications, respectively. Next, we describe the general workflow of organizing a Biomedical NLP (BioNLP) challenge and involved stakeholders (task organizers, task data producers, task participants and end users). Finally, we summarize the impact and contributions by taking into account different BioNLP challenges as a whole, followed by a discussion of their limitations and difficulties. We conclude with future trends in BioNLP challenge evaluations.

/pdf/community-challenges-in-biomedical-text-mining-over-10-years-3wp65pimir.pdf

Community challenges in biomedical text mining over 10 years: success, failure and the future

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).

Using Sublexical Translations to Handle the OOV Problem in Machine Translation

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.

/pdf/a-thesaurus-based-semantic-classification-of-english-2t9rxii1m7.pdf

A Thesaurus-Based Semantic Classification of English Collocations

Social media has long been a popular resource for sentiment analysis and data mining. In this paper, we learn to predict reader interest after article reading using social interaction content in social media. The abundant interaction content (e.g., reader feedback) aims to replace typically private reader profile and browse history. Our method involves estimating interest preferences with respect to article topics and identifying quality social content concerning informativity. During interest analysis, we combine and transform articles and their reader responses into PageRank word graph to balance author- and reader-end influence. Semantic features of words, such as their content sources (authors vs. readers), syntactic parts-of-speech, and degrees of references (i.e., significances) among authors and readers, are used to weight PageRank word graph. We present the prototype system, Interest Finder, that applies the method to reader interest prediction by calculating word interestingness scores. Two sets of evaluation show that traditional, local Page Rank can more accurately cover more span of reader interest with the help of topical interest preferences learned globally, word nodes' semantic information, and, most important of all, quality social interaction content such as reader feedback.

/pdf/interest-analysis-using-semantic-pagerank-and-social-16mmadoxzv.pdf

Interest Analysis Using Semantic PageRank and Social Interaction Content

We introduce a method for learning to describe the attendant contexts of a given query for language learning. In our approach, we display phraseological information in the form of a summary of general patterns as well as lexical bundles anchored at the query. The method involves syntactical analyses and inverted file construction. At run-time, grammatical constructions and their lexical instantiations characterizing the usage of the given query are generated and displayed, aimed at improving learners' deep vocabulary knowledge. We present a prototype system, GRASP, that applies the proposed method for enhanced collocation learning. Preliminary experiments show that language learners benefit more from GRASP than conventional dictionary lookup. In addition, the information produced by GRASP is potentially useful information for automatic or manual editing process.

Chung-Chi Huang

Papers

Community challenges in biomedical text mining over 10 years: success, failure and the future

Using Sublexical Translations to Handle the OOV Problem in Machine Translation

A Thesaurus-Based Semantic Classification of English Collocations

Interest Analysis Using Semantic PageRank and Social Interaction Content

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