COLING 2016 论文列表

Prepositions are very common and very ambiguous, and understanding their sense is critical for understanding the meaning of the sentence. Supervised corpora for the preposition-sense disambiguation task are small, suggesting a semi-supervised approach to the task. We show that signals from unannotated multilingual data can be used to improve supervised preposition-sense disambiguation. Our approach pre-trains an LSTM encoder for predicting the translation of a preposition, and then incorporates the pre-trained encoder as a component in a supervised classification system, and fine-tunes it for the task. The multilingual signals consistently improve results on two preposition-sense datasets.

Semantic classification of words using distributional features is usually based on the semantic similarity of words. We show on two different datasets that a trained classifier using the distributional features directly gives better results. We use Support Vector Machines (SVM) and Multi-relational Matrix Factorization (MRMF) to train classifiers. Both give similar results. However, MRMF, that was not used for semantic classification with distributional features before, can easily be extended with more matrices containing more information from different sources on the same problem. We demonstrate the effectiveness of the novel approach by including information from WordNet. Thus we show, that MRMF provides an interesting approach for building semantic classifiers that (1) gives better results than unsupervised approaches based on vector similarity, (2) gives similar results as other supervised methods and (3) can naturally be extended with other sources of information in order to improve the results.

Word embeddings have been demonstrated to benefit NLP tasks impressively. Yet, there is room for improvements in the vector representations, because current word embeddings typically contain unnecessary information, i.e., noise. We propose two novel models to improve word embeddings by unsupervised learning, in order to yield word denoising embeddings. The word denoising embeddings are obtained by strengthening salient information and weakening noise in the original word embeddings, based on a deep feed-forward neural network filter. Results from benchmark tasks show that the filtered word denoising embeddings outperform the original word embeddings.

Topic modeling and word embedding are two important techniques for deriving latent semantics from data. General-purpose topic models typically work in coarse granularity by capturing word co-occurrence at the document/sentence level. In contrast, word embedding models usually work in much finer granularity by modeling word co-occurrence within small sliding windows. With the aim of deriving latent semantics by considering word co-occurrence at different levels of granularity, we propose a novel model named Latent Topic Embedding (LTE), which seamlessly integrates topic generation and embedding learning in one unified framework. We further propose an efficient Monte Carlo EM algorithm to estimate the parameters of interest. By retaining the individual advantages of topic modeling and word embedding, LTE results in better latent topics and word embedding. Extensive experiments verify the superiority of LTE over the state-of-the-arts.

Existing work learning distributed representations of knowledge base entities has largely failed to incorporate rich categorical structure, and is unable to induce category representations. We propose a new framework that embeds entities and categories jointly into a semantic space, by integrating structured knowledge and taxonomy hierarchy from large knowledge bases. Our framework enables to compute meaningful semantic relatedness between entities and categories in a principled way, and can handle both single-word and multiple-word concepts. Our method shows significant improvement on the tasks of concept categorization and dataless hierarchical classification.

An important difference between traditional AI systems and human intelligence is the human ability to harness commonsense knowledge gleaned from a lifetime of learning and experience to make informed decisions. This allows humans to adapt easily to novel situations where AI fails catastrophically due to a lack of situation-specific rules and generalization capabilities. Commonsense knowledge also provides background information that enables humans to successfully operate in social situations where such knowledge is typically assumed. Since commonsense consists of information that humans take for granted, gathering it is an extremely difficult task. Previous versions of SenticNet were focused on collecting this kind of knowledge for sentiment analysis but they were heavily limited by their inability to generalize. SenticNet 4 overcomes such limitations by leveraging on conceptual primitives automatically generated by means of hierarchical clustering and dimensionality reduction.

Classifying the stance expressed in online microblogging social media is an emerging problem in opinion mining. We propose a probabilistic approach to stance classification in tweets, which models stance, target of stance, and sentiment of tweet, jointly. Instead of simply conjoining the sentiment or target variables as extra variables to the feature space, we use a novel formulation to incorporate three-way interactions among sentiment-stance-input variables and three-way interactions among target-stance-input variables. The proposed specification intuitively aims to discriminate sentiment features from target features for stance classification. In addition, regularizing a single stance classifier, which handles all targets, acts as a soft weight-sharing among them. We demonstrate that discriminative training of this model achieves the state-of-the-art results in supervised stance classification, and its generative training obtains competitive results in the weakly supervised setting.

In the literature, various supervised learning approaches have been adopted to address the task of reader emotion classification. However, the classification performance greatly suffers when the size of the labeled data is limited. In this paper, we propose a two-view label propagation approach to semi-supervised reader emotion classification by exploiting two views, namely source text and response text in a label propagation algorithm. Specifically, our approach depends on two word-document bipartite graphs to model the relationship among the samples in the two views respectively. Besides, the two bipartite graphs are integrated by linking each source text sample with its corresponding response text sample via a length-sensitive transition probability. In this way, our two-view label propagation approach to semi-supervised reader emotion classification largely alleviates the reliance on the strong sufficiency and independence assumptions of the two views, as required in co-training. Empirical evaluation demonstrates the effectiveness of our two-view label propagation approach to semi-supervised reader emotion classification.

Aspect-level analysis of sentiments contained in a review text is important to reveal a detailed picture of consumer opinions. While a plethora of methods have been traditionally employed for this task, majority focus has been on analyzing only aspect-centered local information. However, incorporating context information from non-local aspect neighbors may capture richer structure in review text and enhance prediction. This may especially be helpful to resolve ambiguous predictions. The context around an aspect can be incorporated using semantic relations within text and inter-label dependencies in the output. On the output side, this becomes a structured prediction task. However, non-local label correlations are computationally heavy and intractable to infer for structured prediction models like Conditional Random Fields (CRF). Moreover, some prior intuition is required to incorporate non-local context. Thus, inspired by previous research on multi-stage prediction, we propose a two-level model for aspect-based analysis. The proposed model uses predicted probability estimates from first level to incorporate neighbor information in the second level. The model is evaluated on data taken from SemEval Workshops and Bing Liu’s review collection. It shows comparatively better performance against few existing methods. Overall, we get prediction accuracy in a range of 83-88% and almost 3-4 point increment against baseline (first level only) scores.

We examine the potential of recurrent neural networks for handling pragmatic inferences involving complex contextual cues for the task of article usage prediction. We train and compare several variants of Long Short-Term Memory (LSTM) networks with an attention mechanism. Our model outperforms a previous state-of-the-art system, achieving up to 96.63% accuracy on the WSJ/PTB corpus. In addition, we perform a series of analyses to understand the impact of various model choices. We find that the gain in performance can be attributed to the ability of LSTMs to pick up on contextual cues, both local and further away in distance, and that the model is able to solve cases involving reasoning about coreference and synonymy. We also show how the attention mechanism contributes to the interpretability of the model’s effectiveness.

Penn Discourse Treebank (PDTB)-style annotation focuses on labeling local discourse relations between text spans and typically ignores larger discourse contexts. In this paper we propose two approaches to infer discourse relations in a paragraph-level context from annotated PDTB labels. We investigate the utility of inferring such discourse information using the task of revision classification. Experimental results demonstrate that the inferred information can significantly improve classification performance compared to baselines, not only when PDTB annotation comes from humans but also from automatic parsers.

Discourse parsing is a popular technique widely used in text understanding, sentiment analysis and other NLP tasks. However, for most discourse parsers, the performance varies significantly across different discourse relations. In this paper, we first validate the underfitting hypothesis, i.e., the less frequent a relation is in the training data, the poorer the performance on that relation. We then explore how to increase the number of positive training instances, without resorting to manually creating additional labeled data. We propose a training data enrichment framework that relies on co-training of two different discourse parsers on unlabeled documents. Importantly, we show that co-training alone is not sufficient. The framework requires a filtering step to ensure that only “good quality” unlabeled documents can be used for enrichment and re-training. We propose and evaluate two ways to perform the filtering. The first is to use an agreement score between the two parsers. The second is to use only the confidence score of the faster parser. Our empirical results show that agreement score can help to boost the performance on infrequent relations, and that the confidence score is a viable approximation of the agreement score for infrequent relations.

We introduce a novel task Anecdote Recognition and Recommendation. An anecdote is a story with a point revealing account of an individual person. Recommending proper anecdotes can be used as evidence to support argumentative writing or as a clue for further reading. We represent an anecdote as a structured tuple — < person, story, implication >. Anecdote recognition runs on archived argumentative essays. We extract narratives containing events of a person as the anecdote story. More importantly, we uncover the anecdote implication, which reveals the meaning and topic of an anecdote. Our approach depends on discourse role identification. Discourse roles such as thesis, main ideas and support help us locate stories and their implications in essays. The experiments show that informative and interpretable anecdotes can be recognized. These anecdotes are used for anecdote recommendation. The anecdote recommender can recommend proper anecdotes in response to given topics. The anecdote implication contributes most for bridging user interested topics and relevant anecdotes.

The phrase table is considered to be the main bilingual resource for the phrase-based statistical machine translation (PBSMT) model. During translation, a source sentence is decomposed into several phrases. The best match of each source phrase is selected among several target-side counterparts within the phrase table, and processed by the decoder to generate a sentence-level translation. The best match is chosen according to several factors, including a set of bilingual features. PBSMT engines by default provide four probability scores in phrase tables which are considered as the main set of bilingual features. Our goal is to enrich that set of features, as a better feature set should yield better translations. We propose new scores generated by a Convolutional Neural Network (CNN) which indicate the semantic relatedness of phrase pairs. We evaluate our model in different experimental settings with different language pairs. We observe significant improvements when the proposed features are incorporated into the PBSMT pipeline.

Research in statistical machine translation (SMT) is largely driven by formal translation tasks, while translating informal text is much more challenging. In this paper we focus on SMT for the informal genre of dialogues, which has rarely been addressed to date. Concretely, we investigate the effect of dialogue acts, speakers, gender, and text register on SMT quality when translating fictional dialogues. We first create and release a corpus of multilingual movie dialogues annotated with these four dialogue-specific aspects. When measuring translation performance for each of these variables, we find that BLEU fluctuations between their categories are often significantly larger than randomly expected. Following this finding, we hypothesize and show that SMT of fictional dialogues benefits from adaptation towards dialogue acts and registers. Finally, we find that male speakers are harder to translate and use more vulgar language than female speakers, and that vulgarity is often not preserved during translation.

In this paper we combine two strands of machine translation (MT) research: automatic post-editing (APE) and multi-engine (system combination) MT. APE systems learn a target-language-side second stage MT system from the data produced by human corrected output of a first stage MT system, to improve the output of the first stage MT in what is essentially a sequential MT system combination architecture. At the same time, there is a rich research literature on parallel MT system combination where the same input is fed to multiple engines and the best output is selected or smaller sections of the outputs are combined to obtain improved translation output. In the paper we show that parallel system combination in the APE stage of a sequential MT-APE combination yields substantial translation improvements both measured in terms of automatic evaluation metrics as well as in terms of productivity improvements measured in a post-editing experiment. We also show that system combination on the level of APE alignments yields further improvements. Overall our APE system yields statistically significant improvement of 5.9% relative BLEU over a strong baseline (English–Italian Google MT) and 21.76% productivity increase in a human post-editing experiment with professional translators.

Parallel sentence representations are important for bilingual and cross-lingual tasks in natural language processing. In this paper, we explore a bilingual autoencoder approach to model parallel sentences. We extract sentence-level global descriptors (e.g. min, max) from word embeddings, and construct two monolingual autoencoders over these descriptors on the source and target language. In order to tightly connect the two autoencoders with bilingual correspondences, we force them to share the same decoding parameters and minimize a corpus-level semantic distance between the two languages. Being optimized towards a joint objective function of reconstruction and semantic errors, our bilingual antoencoder is able to learn continuous-valued latent representations for parallel sentences. Experiments on both intrinsic and extrinsic evaluations on statistical machine translation tasks show that our autoencoder achieves substantial improvements over the baselines.

This paper proposes a novel context-aware joint entity and word-level relation extraction approach through semantic composition of words, introducing a Table Filling Multi-Task Recurrent Neural Network (TF-MTRNN) model that reduces the entity recognition and relation classification tasks to a table-filling problem and models their interdependencies. The proposed neural network architecture is capable of modeling multiple relation instances without knowing the corresponding relation arguments in a sentence. The experimental results show that a simple approach of piggybacking candidate entities to model the label dependencies from relations to entities improves performance. We present state-of-the-art results with improvements of 2.0% and 2.7% for entity recognition and relation classification, respectively on CoNLL04 dataset.

Nowadays, neural networks play an important role in the task of relation classification. In this paper, we propose a novel attention-based convolutional neural network architecture for this task. Our model makes full use of word embedding, part-of-speech tag embedding and position embedding information. Word level attention mechanism is able to better determine which parts of the sentence are most influential with respect to the two entities of interest. This architecture enables learning some important features from task-specific labeled data, forgoing the need for external knowledge such as explicit dependency structures. Experiments on the SemEval-2010 Task 8 benchmark dataset show that our model achieves better performances than several state-of-the-art neural network models and can achieve a competitive performance just with minimal feature engineering.