Colin Cherry

Dependency Treelet Translation: Syntactically Informed Phrasal SMT

We describe a novel approach to statistical machine translation that combines syntactic information in the source language with recent advances in phrasal translation. This method requires a source-language dependency parser, target language word segmentation and an unsupervised word alignment component. We align a parallel corpus, project the source dependency parse onto the target sentence, extract dependency treelet translation pairs, and train a tree-based ordering model. We describe an efficient decoder and show that using these tree-based models in combination with conventional SMT models provides a promising approach that incorporates the power of phrasal SMT with the linguistic generality available in a parser.

Machine-learned solutions for three stages of clinical information extraction: the state of the art at i2b2 2010.

Objective As clinical text mining continues to mature, its potential as an enabling technology for innovations in patient care and clinical research is becoming a reality. A critical part of that process is rigid benchmark testing of natural language processing methods on realistic clinical narrative. In this paper, the authors describe the design and performance of three state-of-the-art text-mining applications from the National Research Council of Canada on evaluations within the 2010 i2b2 challenge. Design The three systems perform three key steps in clinical information extraction: (1) extraction of medical problems, tests, and treatments, from discharge summaries and progress notes; (2) classification of assertions made on the medical problems; (3) classification of relations between medical concepts. Machine learning systems performed these tasks using large-dimensional bags of features, as derived from both the text itself and from external sources: UMLS, cTAKES, and Medline. Measurements Performance was measured per subtask, using micro-averaged F-scores, as calculated by comparing system annotations with ground-truth annotations on a test set. Results The systems ranked high among all submitted systems in the competition, with the following F-scores: concept extraction 0.8523 (ranked first); assertion detection 0.9362 (ranked first); relationship detection 0.7313 (ranked second). Conclusion For all tasks, we found that the introduction of a wide range of features was crucial to success. Importantly, our choice of machine learning algorithms allowed us to be versatile in our feature design, and to introduce a large number of features without overfitting and without encountering computing-resource bottlenecks.

SemEval-2016 Task 6: Detecting Stance in Tweets

Here for the first time we present a shared task on detecting stance from tweets: given a tweet and a target entity (person, organization, etc.), automatic natural language systems must determine whether the tweeter is in favor of the given target, against the given target, or whether neither inference is likely. The target of interest may or may not be referred to in the tweet, and it may or may not be the target of opinion. Two tasks are proposed. Task A is a traditional supervised classification task where 70% of the annotated data for a target is used as training and the rest for testing. For Task B, we use as test data all of the instances for a new target (not used in task A) and no training data is provided. Our shared task received submissions from 19 teams for Task A and from 9 teams for Task B. The highest classification F-score obtained was 67.82 for Task A and 56.28 for Task B. However, systems found it markedly more difficult to infer stance towards the target of interest from tweets that express opinion towards another entity.

Unsupervised Morphological Segmentation with Log-Linear Models

Morphological segmentation breaks words into morphemes (the basic semantic units). It is a key component for natural language processing systems. Unsupervised morphological segmentation is attractive, because in every language there are virtually unlimited supplies of text, but very few labeled resources. However, most existing model-based systems for unsupervised morphological segmentation use directed generative models, making it difficult to leverage arbitrary overlapping features that are potentially helpful to learning. In this paper, we present the first log-linear model for unsupervised morphological segmentation. Our model uses overlapping features such as morphemes and their contexts, and incorporates exponential priors inspired by the minimum description length (MDL) principle. We present efficient algorithms for learning and inference by combining contrastive estimation with sampling. Our system, based on monolingual features only, outperforms a state-of-the-art system by a large margin, even when the latter uses bilingual information such as phrasal alignment and phonetic correspondence. On the Arabic Penn Treebank, our system reduces F1 error by 11% compared to Morfessor.

A Probability Model to Improve Word Alignment

Word alignment plays a crucial role in statistical machine translation. Word-aligned corpora have been found to be an excellent source of translation-related knowledge. We present a statistical model for computing the probability of an alignment given a sentence pair. This model allows easy integration of context-specific features. Our experiments show that this model can be an effective tool for improving an existing word alignment.

NRC-Canada-2014: Detecting Aspects and Sentiment in Customer Reviews

Reviews depict sentiments of customers towards various aspects of a product or service. Some of these aspects can be grouped into coarser aspect categories. SemEval-2014 had a shared task (Task 4) on aspect-level sentiment analysis, with over 30 teams participated. In this paper, we describe our submissions, which stood first in detecting aspect categories, first in detecting sentiment towards aspect categories, third in detecting aspect terms, and first and second in detecting sentiment towards aspect terms in the laptop and restaurant domains, respectively.

Inversion transduction grammar for joint phrasal translation modeling

We present a phrasal inversion transduction grammar as an alternative to joint phrasal translation models. This syntactic model is similar to its flat-string phrasal predecessors, but admits polynomial-time algorithms for Viterbi alignment and EM training. We demonstrate that the consistency constraints that allow flat phrasal models to scale also help ITG algorithms, producing an 80-times faster inside-outside algorithm. We also show that the phrasal translation tables produced by the ITG are superior to those of the flat joint phrasal model, producing up to a 2.5 point improvement in BLEU score. Finally, we explore, for the first time, the utility of a joint phrasal translation model as a word alignment method.

Soft Syntactic Constraints for Word Alignment through Discriminative Training

Word alignment methods can gain valuable guidance by ensuring that their alignments maintain cohesion with respect to the phrases specified by a monolingual dependency tree. However, this hard constraint can also rule out correct alignments, and its utility decreases as alignment models become more complex. We use a publicly available structured output SVM to create a max-margin syntactic aligner with a soft cohesion constraint. The resulting aligner is the first, to our knowledge, to use a discriminative learning method to train an ITG bitext parser.

On the Syllabification of Phonemes

Syllables play an important role in speech synthesis and recognition. We present several different approaches to the syllabification of phonemes. We investigate approaches based on linguistic theories of syllabification, as well as a discriminative learning technique that combines Support Vector Machine and Hidden Markov Model technologies. Our experiments on English, Dutch and German demonstrate that our transparent implementation of the sonority sequencing principle is more accurate than previous implementations, and that our language-independent SVM-based approach advances the current state-of-the-art, achieving word accuracy of over 98% in English and 99% in German and Dutch.

An Expectation Maximization Approach to Pronoun Resolution

We propose an unsupervised Expectation Maximization approach to pronoun resolution. The system learns from a fixed list of potential antecedents for each pronoun. We show that unsupervised learning is possible in this context, as the performance of our system is comparable to supervised methods. Our results indicate that a probabilistic gender/number model, determined automatically from unlabeled text, is a powerful feature for this task.