Paper Group NANR 98
Cost Weighting for Neural Machine Translation Domain Adaptation. Opinion Mining in Social Networks versus Electoral Polls. Proceedings of the 1st Workshop on Natural Language Processing and Information Retrieval associated with RANLP 2017. ECNU at SemEval-2017 Task 1: Leverage Kernel-based Traditional NLP features and Neural Networks to Build a Uni …
Cost Weighting for Neural Machine Translation Domain Adaptation
| Title | Cost Weighting for Neural Machine Translation Domain Adaptation |
| Authors | Boxing Chen, Colin Cherry, George Foster, Samuel Larkin |
| Abstract | In this paper, we propose a new domain adaptation technique for neural machine translation called cost weighting, which is appropriate for adaptation scenarios in which a small in-domain data set and a large general-domain data set are available. Cost weighting incorporates a domain classifier into the neural machine translation training algorithm, using features derived from the encoder representation in order to distinguish in-domain from out-of-domain data. Classifier probabilities are used to weight sentences according to their domain similarity when updating the parameters of the neural translation model. We compare cost weighting to two traditional domain adaptation techniques developed for statistical machine translation: data selection and sub-corpus weighting. Experiments on two large-data tasks show that both the traditional techniques and our novel proposal lead to significant gains, with cost weighting outperforming the traditional methods. |
| Tasks | Domain Adaptation, Machine Translation |
| Published | 2017-08-01 |
| URL | https://www.aclweb.org/anthology/W17-3205/ |
| https://www.aclweb.org/anthology/W17-3205 | |
| PWC | https://paperswithcode.com/paper/cost-weighting-for-neural-machine-translation |
| Repo | |
| Framework | |
Opinion Mining in Social Networks versus Electoral Polls
| Title | Opinion Mining in Social Networks versus Electoral Polls |
| Authors | Javi Fern{'a}ndez, Fern Llopis, o, Yoan Guti{'e}rrez, Patricio Mart{'\i}nez-Barco, {'A}lvaro D{'\i}ez |
| Abstract | The recent failures of traditional poll models, like the predictions in United Kingdom with the Brexit, or in United States presidential elections with the victory of Donald Trump, have been noteworthy. With the decline of traditional poll models and the growth of the social networks, automatic tools are gaining popularity to make predictions in this context. In this paper we present our approximation and compare it with a real case: the 2017 French presidential election. |
| Tasks | Opinion Mining |
| Published | 2017-09-01 |
| URL | https://www.aclweb.org/anthology/R17-1032/ |
| https://doi.org/10.26615/978-954-452-049-6_032 | |
| PWC | https://paperswithcode.com/paper/opinion-mining-in-social-networks-versus |
| Repo | |
| Framework | |
Proceedings of the 1st Workshop on Natural Language Processing and Information Retrieval associated with RANLP 2017
| Title | Proceedings of the 1st Workshop on Natural Language Processing and Information Retrieval associated with RANLP 2017 |
| Authors | Mireille Makary, Michael Oakes |
| Abstract | |
| Tasks | Information Retrieval |
| Published | 2017-09-01 |
| URL | https://www.aclweb.org/anthology/papers/W/W17/W17-7700/ |
| https://doi.org/10.26615/978-954-452-038-0_ | |
| PWC | https://paperswithcode.com/paper/proceedings-of-the-1st-workshop-on-natural-1 |
| Repo | |
| Framework | |
ECNU at SemEval-2017 Task 1: Leverage Kernel-based Traditional NLP features and Neural Networks to Build a Universal Model for Multilingual and Cross-lingual Semantic Textual Similarity
| Title | ECNU at SemEval-2017 Task 1: Leverage Kernel-based Traditional NLP features and Neural Networks to Build a Universal Model for Multilingual and Cross-lingual Semantic Textual Similarity |
| Authors | Junfeng Tian, Zhiheng Zhou, Man Lan, Yuanbin Wu |
| Abstract | To address semantic similarity on multilingual and cross-lingual sentences, we firstly translate other foreign languages into English, and then feed our monolingual English system with various interactive features. Our system is further supported by combining with deep learning semantic similarity and our best run achieves the mean Pearson correlation 73.16{%} in primary track. |
| Tasks | Cross-Lingual Semantic Textual Similarity, Machine Translation, Semantic Similarity, Semantic Textual Similarity |
| Published | 2017-08-01 |
| URL | https://www.aclweb.org/anthology/S17-2028/ |
| https://www.aclweb.org/anthology/S17-2028 | |
| PWC | https://paperswithcode.com/paper/ecnu-at-semeval-2017-task-1-leverage-kernel |
| Repo | |
| Framework | |
Evaluating hypotheses in geolocation on a very large sample of Twitter
| Title | Evaluating hypotheses in geolocation on a very large sample of Twitter |
| Authors | Bahar Salehi, Anders S{\o}gaard |
| Abstract | Recent work in geolocation has made several hypotheses about what linguistic markers are relevant to detect where people write from. In this paper, we examine six hypotheses against a corpus consisting of all geo-tagged tweets from the US, or whose geo-tags could be inferred, in a 19{%} sample of Twitter history. Our experiments lend support to all six hypotheses, including that spelling variants and hashtags are strong predictors of location. We also study what kinds of common nouns are predictive of location after controlling for named entities such as dolphins or sharks |
| Tasks | Fraud Detection |
| Published | 2017-09-01 |
| URL | https://www.aclweb.org/anthology/W17-4409/ |
| https://www.aclweb.org/anthology/W17-4409 | |
| PWC | https://paperswithcode.com/paper/evaluating-hypotheses-in-geolocation-on-a |
| Repo | |
| Framework | |
Connotation Frames of Power and Agency in Modern Films
| Title | Connotation Frames of Power and Agency in Modern Films |
| Authors | Maarten Sap, Marcella Cindy Prasettio, Ari Holtzman, Hannah Rashkin, Yejin Choi |
| Abstract | The framing of an action influences how we perceive its actor. We introduce connotation frames of power and agency, a pragmatic formalism organized using frame semantic representations, to model how different levels of power and agency are implicitly projected on actors through their actions. We use the new power and agency frames to measure the subtle, but prevalent, gender bias in the portrayal of modern film characters and provide insights that deviate from the well-known Bechdel test. Our contributions include an extended lexicon of connotation frames along with a web interface that provides a comprehensive analysis through the lens of connotation frames. |
| Tasks | |
| Published | 2017-09-01 |
| URL | https://www.aclweb.org/anthology/D17-1247/ |
| https://www.aclweb.org/anthology/D17-1247 | |
| PWC | https://paperswithcode.com/paper/connotation-frames-of-power-and-agency-in |
| Repo | |
| Framework | |
SSN_MLRG1 at SemEval-2017 Task 4: Sentiment Analysis in Twitter Using Multi-Kernel Gaussian Process Classifier
| Title | SSN_MLRG1 at SemEval-2017 Task 4: Sentiment Analysis in Twitter Using Multi-Kernel Gaussian Process Classifier |
| Authors | Angel Deborah S, S Milton Rajendram, T T Mirnalinee |
| Abstract | The SSN MLRG1 team for Semeval-2017 task 4 has applied Gaussian Process, with bag of words feature vectors and fixed rule multi-kernel learning, for sentiment analysis of tweets. Since tweets on the same topic, made at different times, may exhibit different emotions, their properties such as smoothness and periodicity also vary with time. Our experiments show that, compared to single kernel, multiple kernels are effective in learning the simultaneous presence of multiple properties. |
| Tasks | Sentiment Analysis |
| Published | 2017-08-01 |
| URL | https://www.aclweb.org/anthology/S17-2118/ |
| https://www.aclweb.org/anthology/S17-2118 | |
| PWC | https://paperswithcode.com/paper/ssn_mlrg1-at-semeval-2017-task-4-sentiment |
| Repo | |
| Framework | |
Efficiency-aware Answering of Compositional Questions using Answer Type Prediction
| Title | Efficiency-aware Answering of Compositional Questions using Answer Type Prediction |
| Authors | David Ziegler, Abdalghani Abujabal, Rishiraj Saha Roy, Gerhard Weikum |
| Abstract | This paper investigates the problem of answering compositional factoid questions over knowledge bases (KB) under efficiency constraints. The method, called TIPI, (i) decomposes compositional questions, (ii) predicts answer types for individual sub-questions, (iii) reasons over the compatibility of joint types, and finally, (iv) formulates compositional SPARQL queries respecting type constraints. TIPI{'}s answer type predictor is trained using distant supervision, and exploits lexical, syntactic and embedding-based features to compute context- and hierarchy-aware candidate answer types for an input question. Experiments on a recent benchmark show that TIPI results in state-of-the-art performance under the real-world assumption that only a single SPARQL query can be executed over the KB, and substantial reduction in the number of queries in the more general case. |
| Tasks | Question Answering |
| Published | 2017-11-01 |
| URL | https://www.aclweb.org/anthology/I17-2038/ |
| https://www.aclweb.org/anthology/I17-2038 | |
| PWC | https://paperswithcode.com/paper/efficiency-aware-answering-of-compositional |
| Repo | |
| Framework | |
Probabilistic Models for Integration Error in the Assessment of Functional Cardiac Models
| Title | Probabilistic Models for Integration Error in the Assessment of Functional Cardiac Models |
| Authors | Chris Oates, Steven Niederer, Angela Lee, François-Xavier Briol, Mark Girolami |
| Abstract | This paper studies the numerical computation of integrals, representing estimates or predictions, over the output $f(x)$ of a computational model with respect to a distribution $p(\mathrm{d}x)$ over uncertain inputs $x$ to the model. For the functional cardiac models that motivate this work, neither $f$ nor $p$ possess a closed-form expression and evaluation of either requires $\approx$ 100 CPU hours, precluding standard numerical integration methods. Our proposal is to treat integration as an estimation problem, with a joint model for both the a priori unknown function $f$ and the a priori unknown distribution $p$. The result is a posterior distribution over the integral that explicitly accounts for dual sources of numerical approximation error due to a severely limited computational budget. This construction is applied to account, in a statistically principled manner, for the impact of numerical errors that (at present) are confounding factors in functional cardiac model assessment. |
| Tasks | |
| Published | 2017-12-01 |
| URL | http://papers.nips.cc/paper/6616-probabilistic-models-for-integration-error-in-the-assessment-of-functional-cardiac-models |
| http://papers.nips.cc/paper/6616-probabilistic-models-for-integration-error-in-the-assessment-of-functional-cardiac-models.pdf | |
| PWC | https://paperswithcode.com/paper/probabilistic-models-for-integration-error-in |
| Repo | |
| Framework | |
Chinese Spelling Check based on N-gram and String Matching Algorithm
| Title | Chinese Spelling Check based on N-gram and String Matching Algorithm |
| Authors | Jui-Feng Yeh, Li-Ting Chang, Chan-Yi Liu, Tsung-Wei Hsu |
| Abstract | This paper presents a Chinese spelling check approach based on language models combined with string match algorithm to treat the problems resulted from the influence caused by Cantonese mother tone. N-grams first used to detecting the probability of sentence constructed by the writers, a string matching algorithm called Knuth-Morris-Pratt (KMP) Algorithm is used to detect and correct the error. According to the experimental results, the proposed approach can detect the error and provide the corresponding correction. |
| Tasks | Language Modelling, Spelling Correction |
| Published | 2017-12-01 |
| URL | https://www.aclweb.org/anthology/W17-5906/ |
| https://www.aclweb.org/anthology/W17-5906 | |
| PWC | https://paperswithcode.com/paper/chinese-spelling-check-based-on-n-gram-and |
| Repo | |
| Framework | |
LIG-CRIStAL Submission for the WMT 2017 Automatic Post-Editing Task
| Title | LIG-CRIStAL Submission for the WMT 2017 Automatic Post-Editing Task |
| Authors | Alex B{'e}rard, re, Laurent Besacier, Olivier Pietquin |
| Abstract | |
| Tasks | Automatic Post-Editing, Machine Translation, Spelling Correction |
| Published | 2017-09-01 |
| URL | https://www.aclweb.org/anthology/W17-4772/ |
| https://www.aclweb.org/anthology/W17-4772 | |
| PWC | https://paperswithcode.com/paper/lig-cristal-submission-for-the-wmt-2017 |
| Repo | |
| Framework | |
Proceedings of the International Conference Recent Advances in Natural Language Processing, {RANLP} 2017
| Title | Proceedings of the International Conference Recent Advances in Natural Language Processing, {RANLP} 2017 |
| Authors | Ruslan Mitkov, Galia Angelova |
| Abstract | |
| Tasks | |
| Published | 2017-09-01 |
| URL | https://www.aclweb.org/anthology/papers/R/R17/R17-1000/ |
| https://doi.org/10.26615/978-954-452-049-6_ | |
| PWC | https://paperswithcode.com/paper/proceedings-of-the-international-conference-3 |
| Repo | |
| Framework | |
Joint Learning for Event Coreference Resolution
| Title | Joint Learning for Event Coreference Resolution |
| Authors | Jing Lu, Vincent Ng |
| Abstract | While joint models have been developed for many NLP tasks, the vast majority of event coreference resolvers, including the top-performing resolvers competing in the recent TAC KBP 2016 Event Nugget Detection and Coreference task, are pipeline-based, where the propagation of errors from the trigger detection component to the event coreference component is a major performance limiting factor. To address this problem, we propose a model for jointly learning event coreference, trigger detection, and event anaphoricity. Our joint model is novel in its choice of tasks and its features for capturing cross-task interactions. To our knowledge, this is the first attempt to train a mention-ranking model and employ event anaphoricity for event coreference. Our model achieves the best results to date on the KBP 2016 English and Chinese datasets. |
| Tasks | Coreference Resolution |
| Published | 2017-07-01 |
| URL | https://www.aclweb.org/anthology/P17-1009/ |
| https://www.aclweb.org/anthology/P17-1009 | |
| PWC | https://paperswithcode.com/paper/joint-learning-for-event-coreference |
| Repo | |
| Framework | |
Document Embedding Generation for Cyber-Aggressive Comment Detection using Supervised Machine Learning Approach
| Title | Document Embedding Generation for Cyber-Aggressive Comment Detection using Supervised Machine Learning Approach |
| Authors | Shylaja S S, Abhishek Narayanan, Abhijith Venugopal, Abhishek Prasad |
| Abstract | |
| Tasks | Document Embedding |
| Published | 2017-12-01 |
| URL | https://www.aclweb.org/anthology/W17-7543/ |
| https://www.aclweb.org/anthology/W17-7543 | |
| PWC | https://paperswithcode.com/paper/document-embedding-generation-for-cyber |
| Repo | |
| Framework | |
SCFM: Social and crowdsourcing factorization machines for recommendation
| Title | SCFM: Social and crowdsourcing factorization machines for recommendation |
| Authors | Yue Ding a, Dong Wang b, Xin Xin b, Guoqiang Li b, ∗, Daniel Sun c, b, Xuezhi Zeng d, Rajiv Ranjan e |
| Abstract | With the rapid development of social networks, the exponential growth of social information has attracted much attention. Social information has great value in recommender systems to alleviate the sparsity and cold start problem. On the other hand, the crowd computing empowers recommender systems by utilizing human wisdom. Internal user reviews can be exploited as the wisdom of the crowd to contribute information. In this paper, we propose social and crowdsourcing factorization machines, called SCFM. Our approach fuses social and crowd computing into the factorization machine model. For social computing, we calculate the influence value between users by taking users’ social information and user similarity into account. For crowd computing, we apply LDA (Latent Dirichlet Allocation) on people review to obtain sets of underlying topic probabilities. Furthermore, we impose two important constraints called social regularization and domain inner regularization. The experimental results show that our approach outperforms other state-of-the-art methods |
| Tasks | Recommendation Systems |
| Published | 2017-10-14 |
| URL | https://doi.org/10.1016/j.asoc.2017.08.028 |
| https://doi.org/10.1016/j.asoc.2017.08.028 | |
| PWC | https://paperswithcode.com/paper/scfm-social-and-crowdsourcing-factorization |
| Repo | |
| Framework | |
