COLING 2016 论文列表

We propose a method for learning semantic CCG parsers by projecting annotations via a parallel corpus. The method opens an avenue towards cheaply creating multilingual semantic parsers mapping open-domain text to formal meaning representations. A first cross-lingually learned Dutch (from English) semantic parser obtains f-scores ranging from 42.99% to 69.22% depending on the level of label informativity taken into account, compared to 58.40% to 78.88% for the underlying source-language system. These are promising numbers compared to state-of-the-art semantic parsing in open domains.

Abstractive summarisation is not yet common amongst today’s deployed and research systems. Most existing systems either extract sentences or compress individual sentences. In this paper, we present a summariser that works by a different paradigm. It is a further development of an existing summariser that has an incremental, proposition-based content selection process but lacks a natural language (NL) generator for the final output. Using an NL generator, we can now produce the summary text to directly reflect the selected propositions. Our evaluation compares textual quality of our system to the earlier preliminary output method, and also uses ROUGE to compare to various summarisers that use the traditional method of sentence extraction, followed by compression. Our results suggest that cutting out the middle-man of sentence extraction can lead to better abstractive summaries.

A news article summary usually consists of 2-3 key sentences that reflect the gist of that news article. In this paper we explore using public posts following a new article to improve automatic summary generation for the news article. We propose different approaches to incorporate information from public posts, including using frequency information from the posts to re-estimate bigram weights in the ILP-based summarization model and to re-weight a dependency tree edge’s importance for sentence compression, directly selecting sentences from posts as the final summary, and finally a strategy to combine the summarization results generated from news articles and posts. Our experiments on data collected from Facebook show that relevant public posts provide useful information and can be effectively leveraged to improve news article summarization results.

Query relevance ranking and sentence saliency ranking are the two main tasks in extractive query-focused summarization. Previous supervised summarization systems often perform the two tasks in isolation. However, since reference summaries are the trade-off between relevance and saliency, using them as supervision, neither of the two rankers could be trained well. This paper proposes a novel summarization system called AttSum, which tackles the two tasks jointly. It automatically learns distributed representations for sentences as well as the document cluster. Meanwhile, it applies the attention mechanism to simulate the attentive reading of human behavior when a query is given. Extensive experiments are conducted on DUC query-focused summarization benchmark datasets. Without using any hand-crafted features, AttSum achieves competitive performance. We also observe that the sentences recognized to focus on the query indeed meet the query need.

Product classification is the task of automatically predicting a taxonomy path for a product in a predefined taxonomy hierarchy given a textual product description or title. For efficient product classification we require a suitable representation for a document (the textual description of a product) feature vector and efficient and fast algorithms for prediction. To address the above challenges, we propose a new distributional semantics representation for document vector formation. We also develop a new two-level ensemble approach utilising (with respect to the taxonomy tree) path-wise, node-wise and depth-wise classifiers to reduce error in the final product classification task. Our experiments show the effectiveness of the distributional representation and the ensemble approach on data sets from a leading e-commerce platform and achieve improved results on various evaluation metrics compared to earlier approaches.

In order to organize the large number of products listed in e-commerce sites, each product is usually assigned to one of the multi-level categories in the taxonomy tree. It is a time-consuming and difficult task for merchants to select proper categories within thousands of options for the products they sell. In this work, we propose an automatic classification tool to predict the matching category for a given product title and description. We used a combination of two different neural models, i.e., deep belief nets and deep autoencoders, for both titles and descriptions. We implemented a selective reconstruction approach for the input layer during the training of the deep neural networks, in order to scale-out for large-sized sparse feature vectors. GPUs are utilized in order to train neural networks in a reasonable time. We have trained our models for around 150 million products with a taxonomy tree with at most 5 levels that contains 28,338 leaf categories. Tests with millions of products show that our first predictions matches 81% of merchants’ assignments, when “others” categories are excluded.

When making clinical decisions, practitioners need to rely on the most relevant evidence available. However, accessing a vast body of medical evidence and confronting with the issue of information overload can be challenging and time consuming. This paper proposes an effective summarizer for medical evidence by utilizing both UMLS and WordNet. Given a clinical query and a set of relevant abstracts, our aim is to generate a fluent, well-organized, and compact summary that answers the query. Analysis via ROUGE metrics shows that using WordNet as a general-purpose lexicon helps to capture the concepts not covered by the UMLS Metathesaurus, and hence significantly increases the performance. The effectiveness of our proposed approach is demonstrated by conducting a set of experiments over a specialized evidence-based medicine (EBM) corpus - which has been gathered and annotated for the purpose of biomedical text summarization.

In this paper, we first build a manually annotated named entity corpus of Mongolian. Then, we propose three morphological processing methods and study comprehensive features, including syllable features, lexical features, context features, morphological features and semantic features in Mongolian named entity recognition. Moreover, we also evaluate the influence of word cluster features on the system and combine all features together eventually. The experimental result shows that segmenting each suffix into an individual token achieves better results than deleting suffixes or using the suffixes as feature. The system based on segmenting suffixes with all proposed features yields benchmark result of F-measure=84.65 on this corpus.

In Sanskrit, the phonemes at the word boundaries undergo changes to form new phonemes through a process called as sandhi. A fused sentence can be segmented into multiple possible segmentations. We propose a word segmentation approach that predicts the most semantically valid segmentation for a given sentence. We treat the problem as a query expansion problem and use the path-constrained random walks framework to predict the correct segments.

In this paper, we propose a novel hybrid deep learning archtecture which is highly efficient for sentiment analysis in resource-poor languages. We learn sentiment embedded vectors from the Convolutional Neural Network (CNN). These are augmented to a set of optimized features selected through a multi-objective optimization (MOO) framework. The sentiment augmented optimized vector obtained at the end is used for the training of SVM for sentiment classification. We evaluate our proposed approach for coarse-grained (i.e. sentence level) as well as fine-grained (i.e. aspect level) sentiment analysis on four Hindi datasets covering varying domains. In order to show that our proposed method is generic in nature we also evaluate it on two benchmark English datasets. Evaluation shows that the results of the proposed method are consistent across all the datasets and often outperforms the state-of-art systems. To the best of our knowledge, this is the very first attempt where such a deep learning model is used for less-resourced languages such as Hindi.

Much previous research on multiword expressions (MWEs) has focused on the token- and type-level tasks of MWE identification and extraction, respectively. Such studies typically target known prevalent MWE types in a given language. This paper describes the first attempt to learn the MWE inventory of a “surprise” language for which we have no explicit prior knowledge of MWE patterns, certainly no annotated MWE data, and not even a parallel corpus. Our proposed model is trained on a treebank with MWE relations of a source language, and can be applied to the monolingual corpus of the surprise language to identify its MWE construction types.

Annotation projection is a practical method to deal with the low resource problem in incident languages (IL) processing. Previous methods on annotation projection mainly relied on word alignment results without any training process, which led to noise propagation caused by word alignment errors. In this paper, we focus on the named entity recognition (NER) task and propose a weakly-supervised framework to project entity annotations from English to IL through bitexts. Instead of directly relying on word alignment results, this framework combines advantages of rule-based methods and deep learning methods by implementing two steps: First, generates a high-confidence entity annotation set on IL side with strict searching methods; Second, uses this high-confidence set to weakly supervise the model training. The model is finally used to accomplish the projecting process. Experimental results on two low-resource ILs show that the proposed method can generate better annotations projected from English-IL parallel corpora. The performance of IL name tagger can also be improved significantly by training on the newly projected IL annotation set.

This work focuses on the development of linguistic analysis tools for resource-poor languages. We use a parallel corpus to produce a multilingual word representation based only on sentence level alignment. This representation is combined with the annotated source side (resource-rich language) of the parallel corpus to train text analysis tools for resource-poor languages. Our approach is based on Recurrent Neural Networks (RNN) and has the following advantages: (a) it does not use word alignment information, (b) it does not assume any knowledge about foreign languages, which makes it applicable to a wide range of resource-poor languages, (c) it provides truly multilingual taggers. In a previous study, we proposed a method based on Simple RNN to automatically induce a Part-Of-Speech (POS) tagger. In this paper, we propose an improvement of our neural model. We investigate the Bidirectional RNN and the inclusion of external information (for instance low level information from Part-Of-Speech tags) in the RNN to train a more complex tagger (for instance, a multilingual super sense tagger). We demonstrate the validity and genericity of our method by using parallel corpora (obtained by manual or automatic translation). Our experiments are conducted to induce cross-lingual POS and super sense taggers.

We present experiments in incrementally learning a dependency parser. The parser will be used in the WordsEye Linguistics Tools (WELT) (Ulinski et al., 2014) which supports field linguists documenting a language’s syntax and semantics. Our goal is to make syntactic annotation faster for field linguists. We have created a new parallel corpus of descriptions of spatial relations and motion events, based on pictures and video clips used by field linguists for elicitation of language from native speaker informants. We collected descriptions for each picture and video from native speakers in English, Spanish, German, and Egyptian Arabic. We compare the performance of MSTParser (McDonald et al., 2006) and MaltParser (Nivre et al., 2006) when trained on small amounts of this data. We find that MaltParser achieves the best performance. We also present the results of experiments using the parser to assist with annotation. We find that even when the parser is trained on a single sentence from the corpus, annotation time significantly decreases.

Multiword expressions (MWEs) are pervasive in natural languages and often have both idiomatic and compositional readings, which leads to high syntactic ambiguity. We show that for some MWE types idiomatic readings are usually the correct ones. We propose a heuristic for an A* parser for Tree Adjoining Grammars which benefits from this knowledge by promoting MWE-oriented analyses. This strategy leads to a substantial reduction in the parsing search space in case of true positive MWE occurrences, while avoiding parsing failures in case of false positives.

We present a dependency to constituent tree conversion technique that aims to improve constituent parsing accuracies by leveraging dependency treebanks available in a wide variety in many languages. The technique works in two steps. First, a partial constituent tree is derived from a dependency tree with a very simple deterministic algorithm that is both language and dependency type independent. Second, a complete high accuracy constituent tree is derived with a constraint-based parser, which uses the partial constituent tree as external constraints. Evaluated on Section 22 of the WSJ Treebank, the technique achieves the state-of-the-art conversion F-score 95.6. When applied to English Universal Dependency treebank and German CoNLL2006 treebank, the converted treebanks added to the human-annotated constituent parser training corpus improve parsing F-scores significantly for both languages.

Syntax plays an important role in the task of predicting the semantic structure of a sentence. But syntactic phenomena such as alternations, control and raising tend to obfuscate the relation between syntax and semantics. In this paper we predict the semantic structure of a sentence using a deeper syntax than what is usually done. This deep syntactic representation abstracts away from purely syntactic phenomena and proposes a structural organization of the sentence that is closer to the semantic representation. Experiments conducted on a French corpus annotated with semantic frames showed that a semantic parser reaches better performances with such a deep syntactic input.

In Computational Linguistics, Hindi and Urdu are not viewed as a monolithic entity and have received separate attention with respect to their text processing. From part-of-speech tagging to machine translation, models are separately trained for both Hindi and Urdu despite the fact that they represent the same language. The reasons mainly are their divergent literary vocabularies and separate orthographies, and probably also their political status and the social perception that they are two separate languages. In this article, we propose a simple but efficient approach to bridge the lexical and orthographic differences between Hindi and Urdu texts. With respect to text processing, addressing the differences between the Hindi and Urdu texts would be beneficial in the following ways: (a) instead of training separate models, their individual resources can be augmented to train single, unified models for better generalization, and (b) their individual text processing applications can be used interchangeably under varied resource conditions. To remove the script barrier, we learn accurate statistical transliteration models which use sentence-level decoding to resolve word ambiguity. Similarly, we learn cross-register word embeddings from the harmonized Hindi and Urdu corpora to nullify their lexical divergences. As a proof of the concept, we evaluate our approach on the Hindi and Urdu dependency parsing under two scenarios: (a) resource sharing, and (b) resource augmentation. We demonstrate that a neural network-based dependency parser trained on augmented, harmonized Hindi and Urdu resources performs significantly better than the parsing models trained separately on the individual resources. We also show that we can achieve near state-of-the-art results when the parsers are used interchangeably.

Popular techniques for domain adaptation such as the feature augmentation method of Daumé III (2009) have mostly been considered for sparse binary-valued features, but not for dense real-valued features such as those used in neural networks. In this paper, we describe simple neural extensions of these techniques. First, we propose a natural generalization of the feature augmentation method that uses K + 1 LSTMs where one model captures global patterns across all K domains and the remaining K models capture domain-specific information. Second, we propose a novel application of the framework for learning shared structures by Ando and Zhang (2005) to domain adaptation, and also provide a neural extension of their approach. In experiments on slot tagging over 17 domains, our methods give clear performance improvement over Daumé III (2009) applied on feature-rich CRFs.