Neural Question Generation 2017


Notes:

  • Neural network
  • Neural networks

References:

See also:

100 Best Convolutional Neural Network Videos | 100 Best Java Neural Network Videos | 100 Best MATLAB Neural Network Videos | 100 Best Neural Network Training Videos | 100 Best Neural Network Tutorial Videos | 100 Best Recurrent Neural Network VideosCNN (Convolutional Neural Network) & Dialog Systems 2016 | DNN (Deep Neural Network) & Human Language Technology 2014 | DNN (Deep Neural Network) & Human Language Technology 2015 | Neural Conversation Models 2016 | Neural Network & Dialog Systems 2016 | Neural Turing Machines 2016 | PNN (Probabilistic Neural Network) & Dialog Systems | RNN (Recurrent Neural Network) & Dialog Systems 2016 | RNN (Recurrent Neural Network) & Question Answering Systems 2016


CLEVR: A diagnostic dataset for compositional language and elementary visual reasoning
J Johnson, B Hariharan… – … Vision and Pattern …, 2017 – openaccess.thecvf.com
… nearly one million questions in CLEVR, more than 853k are unique. CLEVR can easily be extended by adding new question families. Question generation. Generating a question for an im- age is conceptually simple: we choose …

Video question answering via hierarchical spatio-temporal attention networks
Z Zhao, Q Yang, D Cai, X He, Y Zhuang – International Joint Conference on …, 2017 – ijcai.org
… where w (o) ky and w (0) kh are parameter vectors. The h (0) j is the j-th latent state of the decoder neural network … Following the state-of-the-art question generation method [Heilman and Smith, 2010], we generate the question-answer pairs from the movie descriptions …

Image-grounded conversations: Multimodal context for natural question and response generation
N Mostafazadeh, C Brockett, B Dolan, M Galley… – arXiv preprint arXiv …, 2017 – arxiv.org
… words (V&T.BOW-Gen) repre- sentation, or an RNN (V&T.RNN-Gen), as depicted in Figure 7. The textual feature vector is then con- catenated to the vision vector and fed into a fully connected (FC) feed forward neural network. As Page 7. Figure 6: Question generation using the …

Video question answering via attribute-augmented attention network learning
Y Ye, Z Zhao, Y Li, L Chen, J Xiao… – Proceedings of the 40th …, 2017 – dl.acm.org
… where 1[·] is the indicator function. We denote all the model co- efficients including neural network parameters and the result em- beddings by ? … Following the state-of-the-art question generation method, we generate the question- answer pairs from the video descriptions …

Lc-quad: A corpus for complex question answering over knowledge graphs
P Trivedi, G Maheshwari, M Dubey… – International Semantic Web …, 2017 – Springer
… We frame our question generation problem as a transduction problem, similar to [10], in which KB subgraphs generated by the seed … Recent approaches in the second category (information retrieval based) are based on neural networks and have achieved promising results [2, 8 …

Visual Question Generation as Dual Task of Visual Question Answering
Y Li, N Duan, B Zhou, X Chu, W Ouyang… – arXiv preprint arXiv …, 2017 – arxiv.org
… and the image feature v to get the question embedding q, and then generates a ques- tion sentence with recurrent neural network (RNN … Intuitively, learn- ing to answer questions may boost the question generation and vice versa, as both of them require similar abilities: im- age …

Question asking as program generation
A Rothe, BM Lake, T Gureckis – Advances in Neural Information …, 2017 – papers.nips.cc
… The Visual Question Generation (VQG) data set [16] contains images paired with interesting, human-generated questions. For instance, an image of a car wreck might be paired with the question, “What caused the accident?” Deep neural networks, similar to those used for …

Leveraging Video Descriptions to Learn Video Question Answering.
KH Zeng, TH Chen, CY Chuang, YH Liao, JC Niebles… – AAAI, 2017 – aaai.org
… The state-of- the-art question generation method (Heilman and Smith 2010) can only generate Yes/No questions with the answer Yes … 2015) to directly learn the sentence semantic and syntac- tic properties in a Recurrent Neural Network framework …

Learning to Ask: Neural Question Generation for Reading Comprehension
X Du, J Shao, C Cardie – arXiv preprint arXiv:1705.00106, 2017 – arxiv.org
… 3 Task Definition In this section, we define the question generation task … (2) with ht being the recurrent neural networks state variable at time step t, and ct being the attention- based encoding of x at decoding time step t (Sec- tion 4.2). Ws and Wt are parameters to be learned …

Visual reference resolution using attention memory for visual dialog
PH Seo, A Lehrmann, B Han, L Sigal – Advances in neural …, 2017 – papers.nips.cc
… based on recurrent (RNN with long-short term memory units), hierarchical recurrent (HRNN)2 and convolu- tional (CNN) neural networks, followed by … During question generation, the target digits for a question is selected based on a subset of the previous targets referred to by …

Automatic generation of english reference question by utilising nonrestrictive relative clause
AY Satria, T Tokunaga – Proceedings of the 9th …, 2017 – pdfs.semanticscholar.org
… Chen, D. and Manning, C. (2014). A fast and accurate de- pendency parser using neural networks. In Proceed- ings of the 2014 Conference on Empirical Methods in … Heilman, M. (2011). Automatic factual question generation from text. PhD thesis, Carnegie Mellon University …

Towards an open-domain social dialog system
M Schmidt, J Niehues, A Waibel – Dialogues with Social Robots, 2017 – Springer
… This module is similar to the function of the OOV component in [6]. 3.3.3 Question Generation … Comput. Speech Lang. 24(2), 150–174 (2010). 8. Chen, D., Manning, CD: A fast and accurate dependency parser using neural networks …

VQABQ: Visual Question Answering by Basic Questions
JH Huang, M Alfadly, B Ghanem – arXiv preprint arXiv:1703.06492, 2017 – arxiv.org
… In [24], they exploit RNN and Convolutional Neural Network (CNN) to build a question generation algorithm, but the generated question sometimes has invalid grammar. The input in [18] is the concatenation of each word embedding with the same feature vector of image …

Question Generation for Language Learning: From ensuring texts are read to supporting learning
M Chinkina, D Meurers – Proceedings of the 12th Workshop on …, 2017 – aclweb.org
… A typical text-based Question Generation (QG) system consists of three components: target se- lection (sentences and words), generation of ques- tions (and answers), and the generation of distrac- tors … Finally, QG is not an exception to the wave of neural networks, and Du et al …

iVQA: Inverse visual question answering
F Liu, T Xiang, TM Hospedales, W Yang… – arXiv preprint arXiv …, 2017 – arxiv.org
… We pose question generation as a multi-modal dynamic inference pro- cess and propose an iVQA model that can gradually … Potemkin villages (Goodfellow, Shlens, and Szegedy 2015), with neural networks finding unexpected correlates that pro- vide shortcuts to give away the …

Proceedings of the 12th Workshop on Innovative Use of NLP for Building Educational Applications
J Tetreault, J Burstein, C Leacock… – Proceedings of the 12th …, 2017 – aclweb.org
… Transparent text quality assessment with convolutional neural networks Robert Östling and Gintare Grigonyte … Multiple Choice Question Generation Utilizing An Ontology Katherine Stasaski and Marti A. Hearst xvii Page 18. Friday, September 8, 2017 (continued) …

Deep active learning for dialogue generation
N Asghar, P Poupart, X Jiang, H Li – … of the 6th Joint Conference on …, 2017 – aclweb.org
… In this work we demonstrate how online Deep Active Learning can be integrated with standard neural network based dialogue … BS and other di- verse decoding techniques on several NLP tasks, including image captioning, machine translation and visual question generation …

Multiple Choice Question Generation Utilizing An Ontology
K Stasaski, MA Hearst – Proceedings of the 12th Workshop on Innovative …, 2017 – aclweb.org
… cO2017 Association for Computational Linguistics Multiple Choice Question Generation Utilizing An Ontology … A subsequent study with three science teachers on the results of a modified question generation algo- rithm finds significant improvements …

Video Question Answering via Gradually Refined Attention over Appearance and Motion
D Xu, Z Zhao, J Xiao, F Wu, H Zhang… – ACM …, 2017 – pdfs.semanticscholar.org
… KEYWORDS Video Question Answering; Attention Mechanism; Neural Network … [26] samples several frames from the video and extracts the fc7 layer’s activations of the convolutional neural network (CNN) as the feature for each frame …

Creativity: Generating Diverse Questions using Variational Autoencoders
U Jain, Z Zhang, A Schwing – arXiv preprint arXiv:1704.03493, 2017 – arxiv.org
… In this paper we propose a creative algorithm for visual question generation which combines the advantages of variational autoencoders with long short-term memory networks … We think that visual question generation is an important task for two reasons …

Robustness Analysis of Visual QA Models by Basic Questions
JH Huang, M Alfadly, B Ghanem – arXiv preprint arXiv:1709.04625, 2017 – arxiv.org
… Recurrent Neural Networks (RNN) can handle the se- quences of flexible length … In (Ren, Kiros, and Zemel 2015), they exploit RNN and Convolutional Neu- ral Network (CNN) to build a question generation algorithm, but the generated question sometimes has invalid gram- mar …

ShapeWorld-A new test methodology for multimodal language understanding
A Kuhnle, A Copestake – arXiv preprint arXiv:1704.04517, 2017 – arxiv.org
… This kind of task is particularly common for neural network models (see, for instance, Ben- gio et al … 1 However, there have been experiments in which parts of the data are artificial and/or generated automatically, for instance, automatic question generation from annotation (Ren …

End-to-end optimization of goal-driven and visually grounded dialogue systems
F Strub, H de Vries, J Mary, B Piot, A Courville… – arXiv preprint arXiv …, 2017 – arxiv.org
… The question generation task requires to produce a new question qj+1, given an im- age I and a history of j questions and answers (q,a)1:j. We … We outline here the neural network ar- chitecture that achieved the best performance and refer to [de Vries et al., 2016] for a thorough …

Learning to Disambiguate by Asking Discriminative Questions
Y Li, C Huang, X Tang, CC Loy – arXiv preprint arXiv …, 2017 – openaccess.thecvf.com
… av- erage. In addition, we present an effective method for vi- sual discriminative question generation. The … the reference. We define the aforementioned problem as Visual Dis- criminative Question Generation (VDQG). Specifically …

Reinforced mnemonic reader for machine comprehension
M Hu, Y Peng, X Qiu – CoRR, abs/1705.02798, 2017 – pdfs.semanticscholar.org
… 2016; Joshi et al. 2017), end-to-end neural networks have achieved promising results on MC tasks (Wang et al. 2016; Xiong, Zhong, and Socher 2017; Seo et al … For each query, we lookup its query-category embedding and use a feedforward neural network for projec- tion …

Attentive Convolution
W Yin, H Schütze – arXiv preprint arXiv:1710.00519, 2017 – arxiv.org
… de Abstract In NLP, convolution neural networks (CNNs) have benefited less than recurrent neural net- works (RNNs) from attention mechanisms. We hypothesize … tasks. Convolution neural networks (CNNs, LeCun et al. (1998)) and …

The 2016 RWTH keyword search system for low-resource languages
P Golik, Z Tüske, K Irie, E Beck, R Schlüter… – … Conference on Speech …, 2017 – Springer
… question generation for decision tree based state tying. In: ICASSP, pp. 805–808 (1998)Google Scholar. 4. Doetsch, P., Zeyer, A., Voigtlaender, P., Kulikov, I., Schlüter, R., Ney, H.: RETURNN: the RWTH extensible training framework for universal recurrent neural networks …

Crowdsourcing multiple choice science questions
J Welbl, NF Liu, M Gardner – arXiv preprint arXiv:1707.06209, 2017 – arxiv.org
… There has not been much work applying neural networks to this task, likely because of the paucity of training data; this paper is an attempt to address this issue by constructing a much larger dataset than was previously available, and we present … Automatic Question Generation …

MLP-GA based algorithm to detect application layer DDoS attack
KJ Singh, T De – Journal of Information Security and Applications, 2017 – Elsevier
… test module is to challenge the sender to response for a correct answer to the text based question generated by the question generation module … One of the main benefits of neural networks (NN) is their ability to represent a relationship which may be both linear and nonlinear …

An Analysis of Visual Question Answering Algorithms
K Kafle, C Kanan – arXiv preprint arXiv:1703.09684, 2017 – arxiv.org
… categories. [3, 23, 5, 25, 8, 14, 10, 18, 22, 24, 27, 29, 30, 32, 9, 20]. Almost all systems use CNN features to represent the image and either a recur- rent neural network (RNN) or a bag-of-words model for the question. We briefly …

Generating questions asked by facilitator agents using preceding context in web-based discussion
Y Ikeda, S Shiramatsu – Agents (ICA), 2017 IEEE International …, 2017 – ieeexplore.ieee.org
… We focused on superficial case structures of extracted phrases suitable for question generation … For future work, we will consider deep neural network- based methods for extracting important phrases from preceding context …

Identifying where to focus in reading comprehension for neural question generation
X Du, C Cardie – Proceedings of the 2017 Conference on Empirical …, 2017 – aclweb.org
… distribution of yt (? {0,1}), P1 (yt|D;?) = softmax ( MLP ( tanh ( [ ?? ht; ?? ht] ))) where MLP is multi-layer neural network and tanh is the activation function. Question Generation Similar to Du et al. (2017), we implement the sentence …

Machine Comprehension by Text-to-Text Neural Question Generation
X Yuan, T Wang, C Gulcehre, A Sordoni… – arXiv preprint arXiv …, 2017 – arxiv.org
… These approaches comprise pipelines of independent components that are difficult to tune for final performance measures. More recently, neural networks have enabled end-to-end training of question generation systems. Ser- ban et al …

Which is the Effective Way for Gaokao: Information Retrieval or Neural Networks?
S Guo, X Zeng, S He, K Liu, J Zhao – … of the 15th Conference of the …, 2017 – aclweb.org
… Which is the Effective Way for Gaokao: Information Retrieval or Neural Networks … One is based on entity search technique (IR approach) and the other is based on a text entail- ment approach where we specifically employ deep neural networks (NN approach) …

Proceedings of ACL 2017, Student Research Workshop
A Ettinger, S Gella, M Labeau, CO Alm… – Proceedings of ACL …, 2017 – aclweb.org
… Text-based Speaker Identification on Multiparty Dialogues Using Multi-document Convolutional Neural Networks Kaixin Ma, Catherine Xiao and Jinho D. Choi … Domain Specific Automatic Question Generation from Text Katira Soleymanzadeh …

Generating Natural Language Question-Answer Pairs from a Knowledge Graph Using a RNN Based Question Generation Model
S Indurthi, D Raghu, MM Khapra, S Joshi – pdfs.semanticscholar.org
… use a Recurrent Neural Network (RNN) (Werbos, 1990; Rumelhart et al., 1988) based model with Long Short Term Mem- ory (LSTM) (Hochreiter and Schmidhuber, 1997) units to generate questions from the given set of keywords. The input to our question generation model is a …

Neural Question Generation from Text: A Preliminary Study
Q Zhou, N Yang, F Wei, C Tan, H Bao… – arXiv preprint arXiv …, 2017 – arxiv.org
… Large-scale manually annotated passage and question pairs play a crucial role in developing question generation systems … We conduct the experiments on SQuAD, and the experiment results show the neural network models can produce fluent and di- verse questions from text …

Question generation for question answering
N Duan, D Tang, P Chen, M Zhou – … of the 2017 Conference on Empirical …, 2017 – aclweb.org
… The contribution of the pa- per is 2-fold: First, two types of question generation approaches are proposed, one is a retrieval-based method using convo- lution neural network (CNN), the other is a generation-based method using recurrent neural network (RNN); Second, we …

A Joint Model for Question Answering and Question Generation
T Wang, X Yuan, A Trischler – arXiv preprint arXiv:1706.01450, 2017 – arxiv.org
… Active comprehension: Problem-solving schema with question generation for comprehension of complex short stories. Reading Re- search Quarterly, pp. 166–186, 1982. Sutskever, Ilya, Vinyals, Oriol, and Le, Quoc V. Sequence to sequence learning with neural networks …

Asking the Difficult Questions: Goal-Oriented Visual Question Generation via Intermediate Rewards
J Zhang, Q Wu, C Shen, J Zhang, J Lu… – arXiv preprint arXiv …, 2017 – arxiv.org
… learning paradigm and propose three different intermedi- ate rewards, namely the goal-achieved reward, progressive reward, and informativeness reward, to explicitly optimize the question generation … We build the neural network architecture for Oracle by referring to [10] …

Data augmentation for visual question answering
K Kafle, M Yousefhussien, C Kanan – Proceedings of the 10th …, 2017 – aclweb.org
… This is especially a problem with deep convolu- tional neural networks (CNNs) that have millions of parameters … Visual question generation was also studied in (Mostafazadeh et al., 2016), with an emphasis on generating questions about images that are beyond the literal visual …

Proceedings of the 2nd Workshop on Representation Learning for NLP
P Blunsom, A Bordes, K Cho, S Cohen, C Dyer… – Proceedings of the 2nd …, 2017 – aclweb.org
… ACL’15, which focused on vector space models of meaning, compositionality, and the application of deep neural networks and spectral … Machine Comprehension by Text-to-Text Neural Question Generation Xingdi Yuan, Tong Wang, Caglar Gulcehre, Alessandro Sordoni, Philip …

Question Answering and Question Generation as Dual Tasks
D Tang, N Duan, T Qin, M Zhou – arXiv preprint arXiv:1706.02027, 2017 – arxiv.org
… We describe the question generation (QG) model in this section … The encoder could be implemented with different neural network ar- chitectures such as convolutional neural network (Kalchbrenner and Blunsom, 2013; Meng et al., 2015) and recurrent neural network (RNN …

Eckhard Bick and Marcos Zampieri
J Koco?, M Marci?czuk, A Aghaebrahimian, F Jur?í?ek… – pdfs.semanticscholar.org
… 163 Miroslav Smatana, Ján Parali?, and Peter Butka Neural Networks for Featureless Named Entity Recognition in Czech … Automatic Question Generation Based on Analysis of Sentence Structure . . . 223 Miroslav Blšták and Viera Rozinajová XVIII Contents Page 3 …

Video Question Answering Using a Forget Memory Network
Y Ge, Y Xu, Y Han – CCF Chinese Conference on Computer Vision, 2017 – Springer
… [19] introduced a novel recurrent neural network (RNN) architecture … They first collected a large-scale video QA dataset with 18100 videos and 175076 candidate QA pairs by using a state-of-the-art question generation method [6]. Then the Video QA model takes a video and the …

Distractor Generation with Generative Adversarial Nets for Automatically Creating Fill-in-the-blank Questions
C Liang, X Yang, D Wham, B Pursel, R Passonneau… – 2017 – pdfs.semanticscholar.org
… 2007. An automatic multiple- choice question generation scheme for english adjective understanding. In ICCE. 137–142. [15] Andrew L Maas, Awni Y Hannun, and Andrew Y Ng. 2013. Rectifier nonlinearities improve neural network acoustic models …

Overview of the NLPCC 2017 Shared Task: Open Domain Chinese Question Answering
N Duan, D Tang – National CCF Conference on Natural Language …, 2017 – Springer
… Besides, we also used Automatic Question Generation technique to generate a set faked questions, and mixed them into human labeled questions to form a … Dong, L., Wei, F., Zhou, M., Xu, K.: Question answering over freebase with multi-column convolutional neural networks …

Towards Building an Automated Fact-Checking System
D Jimenez – Proceedings of the 2017 ACM International …, 2017 – dl.acm.org
… ready for deployment in a real-world system [3]. Ultimately, my results have shown that there is still much work to be done in question generation and claim … Go- ing forward we plan to assess the viability of other tech- niques, such as neural networks, in claim to query map- ping …

Generating Natural Language Question-Answer Pairs from a Knowledge Graph Using a RNN Based Question Generation Model
S Reddy, D Raghu, MM Khapra, S Joshi – … of the 15th Conference of the …, 2017 – aclweb.org
… use a Recurrent Neural Network (RNN) (Werbos, 1990; Rumelhart et al., 1988) based model with Long Short Term Mem- ory (LSTM) (Hochreiter and Schmidhuber, 1997) units to generate questions from the given set of keywords. The input to our question generation model is a …

Two-Stage Synthesis Networks for Transfer Learning in Machine Comprehension
D Golub, PS Huang, X He, L Deng – arXiv preprint arXiv:1706.09789, 2017 – arxiv.org
… et al., 2016) has allowed for the train- ing of data-intensive but expressive models such as deep neural networks (Wang et … the process of generating question-answer pairs into two steps, answer generation conditioned on the paragraph, and question generation conditioned on …

Large-Scale Simple Question Generation by Template-Based Seq2seq Learning
T Liu, B Wei, B Chang, Z Sui – National CCF Conference on Natural …, 2017 – Springer
… Furthermore, even more labeled question-answer pairs are required in neural network-based QA systems. Automatic question generation (QG) [4, 6, 15] has become a popular task for solving the data insufficient problems in QA systems …

Question Answering with Character-Level LSTM Encoders and Model-Based Data Augmentation
RZ Wang, CD Zhan, ZH Ling – … Based on Naturally Annotated Big Data, 2017 – Springer
… We have improved the existing approach [9] by employing LSTMs to encode entities and predicates, introducing a new strategy to generate negative samples for model training, and augmenting training set with neural-network-based question generation method …

Neural Joke Generation
H Ren, Q Yang – stanford.edu
… corresponding natural language jokes. The model combines ideas from recent neural network architectures for machine translation, factoid question generation, as well as neural image captioning (NIC). The produced jokes are …

INAUGURAL COURSE PROJECT POSTER FAIR
N Yazdani – cs.washington.edu
… your inquisitive companion might ask you relevant questions like “What are you studying?” or “When are you graduating?” Previous work in question generation has focused … Finally, a deep neural network is trained on these cluster assignments to distinguish the musical styles …

A Unified Query-based Generative Model for Question Generation and Question Answering
L Song, Z Wang, W Hamza – arXiv preprint arXiv:1709.01058, 2017 – arxiv.org
Page 1. A Unified Query-based Generative Model for Question Generation and Question Answering … Abstract We propose a query-based generative model for solving both tasks of question generation (QG) and question an- swering (QA) …

Are you asking the right questions? Teaching Machines to Ask Clarification Questions
S Rao – Proceedings of ACL 2017, Student Research …, 2017 – aclweb.org
… framework. 7 Conclusion In our work, we introduce a novel dataset for clar- ification question generation, and build a model that integrates neural network structure with the classic notion of expected value of perfect infor- mation. Our …

Mobile Cognition: Enabling Deep Learning Computing in the Internet of Things Era
A Vega, S Eldridge, A Buyuktosunoglu, P Bose – 2017 – dtic.mil
… level with the aforemen- tioned neural networks implemented in Keras, a high-level deep learning framework for rapid neural network proto- typing … version of the parame- ters, the cloud should be capable of sending updates back at a speed close to the question generation rate …

The Forgettable-Watcher Model for Video Question Answering
H Xue, Z Zhao, D Cai – arXiv preprint arXiv:1705.01253, 2017 – arxiv.org
… [Noh et al., 2016] Hyeonwoo Noh, Paul Hongsuck Seo, and Bohyung Han. Image question answering using convolu- tional neural network with dynamic parameter prediction … The question generation shared task and evaluation chal- lenge. In The University of Memphis …

Hi, how can I help you?: Automating enterprise IT support help desks
S Mani, N Gantayat, R Aralikatte, M Gupta… – arXiv preprint arXiv …, 2017 – arxiv.org
… Extracting simplified statements for factual question generation. In Proceedings of QG2010: The Third Workshop on Ques-tion Generation, 11. [Hofmann 1999] Hofmann, T. 1999 … ACM. [Kim 2014] Kim, Y. 2014. Convolutional neural networks for sentence classification …

Plan, Attend, Generate: Planning for Sequence-to-Sequence Models
F Dutil, C Gulcehre, A Trischler… – arXiv preprint arXiv …, 2017 – papers.nips.cc
… We evaluate our planning models on the recently proposed question-generation task (Yuan et al., 2017), where the goal is to generate a question conditioned on a document and an answer … Both the document and the answer are encoded via recurrent neural networks, and 6 …

Automatic Generation of News Comments Based on Gated Attention Neural Networks
HT Zheng, W Wang, W Chen, JY Chen… – IEEE …, 2017 – ieeexplore.ieee.org
… Neural Networks … JC20140001). ABSTRACT With the development of Recurrent Neural Networks (RNN), various Natural Language Generation (NLG) tasks have boomed in the past few years, such as response generation in conversation and poetry generation …

Plan, attend, generate: Planning for sequence-to-sequence models
C Gulcehre, F Dutil, A Trischler… – Advances in Neural …, 2017 – papers.nips.cc
… We evaluate our planning models on the recently proposed question-generation task (Yuan et al., 2017), where the goal is to generate a question conditioned on a document and an answer … Both the document and the answer are encoded via recurrent neural networks, and 6 …

An Elevator Pitch on Deep Learning
YT Hu, AG Schwing – GetMobile: Mobile Computing and …, 2017 – dl.acm.org
… FUTURE CHALLENGES Deep learning is the beginning of a more data driven era, and applications like visual question answering, question generation and video segmentation will transform … Dermatologist-level classification of skin cancer with deep neural networks. Nature …

Controlling item difficulty for automatic vocabulary question generation
Y Susanti, T Tokunaga… – … and Practice in …, 2017 – telrp.springeropen.com
… 2001) and artificial neural networks (Boldt and Freedle 1996; Perkins et al. 1995). More recently, Loukina et al … Unlike this previous research, the aim of the present study is to control item difficulty in the automatic question generation task …

A REVIEW ON GENERATIVE CONVERSATIONAL MODEL
E Varghese, MTR Pillai – data.conferenceworld.in
… This system has three sections Plain text extraction Question generation and Question and answers … Neural Network (RNN)that canmaintain memory through time or other sequential inputs. VI.CONCLUSION This paper provides a review of generative conversational model …

Variational Attention for Sequence-to-Sequence Models
H Bahuleyan, L Mou, O Vechtomova… – arXiv preprint arXiv …, 2017 – arxiv.org
… gmail.com Abstract The variational encoder-decoder (VED) encodes source information as a set of random variables using a neural net- work, which in turn is decoded into tar- get data using another neural network. In natural …

A Comparative Evaluation of GMM-Free State Tying Methods for ASR
T Grósz, G Gosztolya, L Tóth – Proc. Interspeech 2017, 2017 – isca-speech.org
… [14] L. Zhu, K. Kilgour, S. Stüker, and A. Waibel, “Gaussian free clus- ter tree construction using Deep Neural Network,” in Proceedings of Interspeech, Dresden, Germany, Sep 2015, pp. 3254–3258 … 3444–3448. [16] K. Beulen and H. Ney, “Automatic question generation for deci …

Joint Learning of Response Ranking and Next Utterance Suggestion in Human-Computer Conversation System
R Yan, D Zhao – Proceedings of the 40th International ACM SIGIR …, 2017 – dl.acm.org
… To this end, we propose a dual recurrent neural network chains with Long-Short Term Memory (LSTM) units for the new conversation task, namely Dual-LSTM Chain Model (Dual-LSTM) … Many matching metrics are proposed for retrieval using deep neural networks …

Co-attending Free-form Regions and Detections with Multi-modal Multiplicative Feature Embedding for Visual Question Answering
P Lu, H Li, W Zhang, J Wang, X Wang – arXiv preprint arXiv:1711.06794, 2017 – arxiv.org
… In this paper, we pro- pose a novel deep neural network for VQA that integrates both attention mechanisms … Great progress has been achieved for different tasks in- cluding image captioning (Karpathy and Fei-Fei 2015), vi- sual question generation (Mostafazadeh et al …

Learning by Asking Questions
I Misra, R Girshick, R Fergus, M Hebert, A Gupta… – arXiv preprint arXiv …, 2017 – arxiv.org
… others [38]. LBA is independent of the backbone VQA model and can be used with any existing architecture. Visual question generation (VQG) was recently pro- posed as an alternative to image captioning [33, 36, 40]. Our work …

High-Order Attention Models for Visual Question Answering
I Schwartz, A Schwing, T Hazan – Advances in Neural Information …, 2017 – papers.nips.cc
… The quest for algorithms which enable cognitive abilities is an important part of machine learning and appears in many facets, eg, in visual question answering tasks [6], image captioning [26], visual question generation [18, 10] and machine comprehension [8]. A common trait in …

Incomplete Follow-up Question Resolution using Retrieval based Sequence to Sequence Learning
V Kumar, S Joshi – Proceedings of the 40th International ACM SIGIR …, 2017 – dl.acm.org
… However, it is not clear to us how this ap- proach can be applied to the task of complete question generation from an incomplete follow-up question … Seq2seq uses a Recurrent Neural Network (RNN) based encoder decoder network …

Plan, Attend, Generate: Character-Level Neural Machine Translation with Planning
C Gulcehre, F Dutil, A Trischler, Y Bengio – Proceedings of the 2nd …, 2017 – aclweb.org
… Its encoder is a recur- rent neural network (RNN) that reads the source (a sequence of byte pairs representing text in some lan- guage … We demonstrated our models’ capabilities through results on character-level machine translation, an algorithmic task, and question generation …

TriviaQA: A Large Scale Distantly Supervised Challenge Dataset for Reading Comprehension
M Joshi, E Choi, DS Weld, L Zettlemoyer – arXiv preprint arXiv:1705.03551, 2017 – arxiv.org
… This decoupling of question generation from evidence collection allows us to control for potential bias in question style or content, while offering organi … on CNN Dailymail and MCTest (Chen et al., 2016; Richardson et al., 2013) and a state-of-the-art neural network baseline (Seo …

Conversational Exploratory Search via Interactive Storytelling
S Vakulenko, I Markov, M de Rijke – arXiv preprint arXiv:1709.05298, 2017 – arxiv.org
… RQ4.2. Question generation: How to verify user understanding, satisfaction and preferences … Martin et al. [12] generate stories in natural language using two sequence-to-sequence recurrent neural networks (RNNs): (1) event representations are extracted from text using …

An Efficient Approach for Keyphrase Extraction from English Document
IH Emu, AU Ahmed, MM Islam, MS Al Mamun… – 2017 – mecs-press.org
… “Automatic keyphrases extraction from document using neural network”, In ICMLC, pages 633– 641, 2005. [21] Claude Pasquier … [23] Pabitha, P., Suganthi, S. and Ram, R.,. “Key Phrase Extraction Using Naive Bayes? in Question Generation System”, Asian Journal of …

Keyword-based Query Comprehending via Multiple Optimized-Demand Augmentation
B Pan, H Li, Z Zhao, D Cai, X He – arXiv preprint arXiv:1711.00179, 2017 – arxiv.org
… In this paper, we propose a novel neural network system that consists a Demand Optimization Model based on a passage-a ention neural machine translation and a Reader Model that can nd the answer given the optimized question …

Incorporating External Knowledge to Answer Open-Domain Visual Questions with Dynamic Memory Networks
G Li, H Su, W Zhu – arXiv preprint arXiv:1712.00733, 2017 – arxiv.org
… (11) where e(I), e(Q) and e(A) are embedded features for image, question and answer, respectively; h is the fused feature in this common space; and, W(I), W(Q) and W(A) are corresponding to the parameters in neural networks … 4.1.2 Open-domain Question Generation …

Robustness Analysis of Visual Question Answering Models by Basic Questions
JH Huang – 2017 – repository.kaust.edu.sa
… such as machine translation [21, 22], LSTM is a successful application. In [16], they exploit RNN and Convolutional Neural Network (CNN) to build a question generation algorithm, but the generated question sometimes has invalid grammar. The input in [23, 24] is the …

The Achievement of Higher Flexibility in Multiple Choice-based Tests Using Image Classification Techniques
M Afifi, KF Hussain – arXiv preprint arXiv:1711.00972, 2017 – arxiv.org
… 4.3.3 CNN Ordinary neural networks consist of multiple layers of a set of neurons, each which applies a linear-function followed by a non-linear function to the input vector x ? RD … [4] M. Liu, V. Rus, and L. Liu, “Automatic chinese multiple choice question generation using mixed …

Plan, Attend, Generate: Character-level Neural Machine Translation with Planning in the Decoder
C Gulcehre, F Dutil, A Trischler, Y Bengio – arXiv preprint arXiv …, 2017 – arxiv.org
… Our model’s encoder is a recurrent neural network (RNN) that ?Equal Contribution … We demonstrated our models’ capabilities through results on character-level machine translation, an algorithmic task, and question generation …

Scaffolding Networks for Teaching and Learning to Comprehend
A Celikyilmaz, L Deng, L Li, C Wang – arXiv preprint arXiv:1702.08653, 2017 – arxiv.org
… Especially the neural-network-based approaches learn to selectively attend to different parts of the input text relevant to a target question … answer pairs are provided in corpus (ie, at the end of the text), the provided question follows through the question-generation path (1-6) in …

Learning Semantic Patterns for Question Generation and Question Answering
HP Rodrigues – 2017 – l2f.inesc-id.pt
… creating Question/Answer pairs given an information source, through Question Answering (QA) and Question Generation (QG) … Mechanical Turk), and extrinsically, as a supporting data source for real systems in their tasks (for instance, to train neural network systems). Page 5 …

Social and Ethical Impact of Artificial Intelligence on Public-A Case Study of University Students
FF Quraishi, SA Wajid, P Dhiman – 2017 – ijsrset.com
… 2011), accessed August 1, 2016,http://www.cs.cmu.edu/~ark/mheilman/questions /papers/heilman-question-generation-dissertation.pdf … Le Lu, Ziyue Xu, Isabella Nogues,Jianhua Yao, Daniel Mollura, and Ronald M. Summers, “Deep Convolutional Neural Networks for …

Dataset for the First Evaluation on Chinese Machine Reading Comprehension
Y Cui, T Liu, Z Chen, W Ma, S Wang, G Hu – arXiv preprint arXiv …, 2017 – arxiv.org
… 3.4.2. User Query RC Unlike the cloze dataset, we have no automatic question generation procedure in this type … Cui, Y., Liu, T., Chen, Z., Wang, S., and Hu, G. (2016). Consensus attention-based neural networks for chinese reading comprehension …

Product review summarization through question retrieval and diversification
M Liu, Y Fang, AG Choulos, DH Park, X Hu – Information Retrieval Journal, 2017 – Springer
… It usually takes 50 minutes to finish question generation and examination for a single review, which is a very time-consuming process since … 4.2.3, we introduce the Recurrent Neural Network (RNN) Encoder–Decoder, which is used to measure the answerability of a review to a …

Unifying the Video and Question Attentions for Open-Ended Video Question Answering
H Xue, Z Zhao, D Cai – IEEE Transactions on Image Processing, 2017 – ieeexplore.ieee.org
… the questions (around 94%) have answers of length less than or equal to 6. The question generation approach from Heilman and Smith [8] can also generate … The LSTM is an improved recurrent neural network model which can keep long-term information of the sequences …

Role of Premises in Visual Question Answering
A Mahendru – 2017 – vtechworks.lib.vt.edu
… 15 6.1 Question Generation … Simple models like representing questions as bags of words (BoW+I [5]), or encoding the question using a recurrent neural network and train a simple classifier on the encoded question and image (Deeper LSTM [5]) have shown promise …

Year of Publication: 2012
K Rajamani, V Kathiravan – pdfs.semanticscholar.org
… Dynamic question generation is proposed which uses the novel approach of Particle Swarm Optimization … Kennedy, J. , & Eberhart, RC (1995). Particle swarm optimization. In Proceedings of the IEEE international conference on neural networks 4, pp. 1942–1948. Index Terms …

Automatic Generation of Lexical Exercises?
A Fenogenova, E Kuzmenko – ceur-ws.org
… the research on exercise generation involves other sources such as ontologies and thesauruses [5] [10], especially for question generation tasks and … The model was build with Word2vec, a tool based on neural networks, which computes vector representations of words from a …

L2 Learners’ Proficiency Evaluation Using Statistics Based on Relationship Among CEFR Rating Scales
H Tsubaki – International Conference on Data Mining and Big Data, 2017 – Springer
… By using neural network and multiple regression models, the learners’ English proficiency scores were estimated by the 24 statistics based on … 2. Wang, H., Waple, CJ: Kawahara, T.: Computer assisted language learning system based on dynamic question generation and error …

Patient’s experience in Pediatric Primary immunodeficiency Disorders: computerized classification of Questionnaires
U Mücke, C Klemann, U Baumann… – Frontiers in …, 2017 – ncbi.nlm.nih.gov
… questionnaire. For study purpose, we added a step for question generation (see Figure ?Figure11) … Most of these methods are variations of main statistical concepts like vector space methods or artificial neural networks. The …

Proceedings of the 15th Conference of the European Chapter of the Association for Computational Linguistics: Volume 1, Long Papers
M Lapata, P Blunsom, A Koller – Proceedings of the 15th Conference of …, 2017 – aclweb.org
… Sentence Segmentation in Narrative Transcripts from Neuropsychological Tests using Recurrent Convo- lutional Neural Networks Marcos Treviso, Christopher … Language Question-Answer Pairs from a Knowledge Graph Using a RNN Based Question Generation Model Sathish …

Intelligent Math Tutor: Problem-Based Approach to Create Cognizance
M Gupta, N Gantayat, R Sindhgatta – Proceedings of the Fourth (2017) …, 2017 – dl.acm.org
… Good Question! Statistical Ranking for Question Generation. In HLT-NAACL 2010, pages 609–617 6. Danqi Chen and Christopher Manning. 2014. A Fast and Accurate Dependency Parser Using Neural Networks. In Proceedings of EMNLP 2014 …

Neural Wikipedian: Generating Textual Summaries from Knowledge Base Triples
P Vougiouklis, H Elsahar, LA Kaffee, C Gravier… – arXiv preprint arXiv …, 2017 – arxiv.org
… Conventional systems based on neural networks when employed on NLG tasks, such as Machine Translation [6] or Question Generation [11] are incapable of learning high quality vector repre- sentation for the infrequent tokens (ie either words or entities) in their training dataset …

Big Media Data Analysis
A Iosifidis, A Tefas, I Pitas, M Gabbouj – 2017 – Elsevier
… Indeed, the GPU processing allowed for training very deep neural networks for solving various learning tasks with very large-scale datasets … Existing datasets target the problems of generic object and scene analysis [[80]; [81] ; [82]], visual question generation and answering [[83 …

Diverse and Accurate Image Description Using a Variational Auto-Encoder with an Additive Gaussian Encoding Space
L Wang, A Schwing, S Lazebnik – Advances in Neural Information …, 2017 – papers.nips.cc
… given the images. At test time, the decoder is seeded with an image feature and different z samples, so that multiple z’s result in multiple questions. While Jain et al. [14] obtained promising question generation performance with the above CVAE model equipped with a fixed …

Proceedings of the 55th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers)
R Barzilay, MY Kan – Proceedings of the 55th Annual Meeting of the …, 2017 – aclweb.org
… Yasuhide Miura, Motoki Taniguchi, Tomoki Taniguchi and Tomoko Ohkuma. Unifying Text, Metadata, and User Network Representations with a Neural Network for Geolocation Pre- diction. • Ramakanth Pasunuru and Mohit Bansal …

The Art of Deep Connection-Towards Natural and Pragmatic Conversational Agent Interactions
A Ray – 2017 – vtechworks.lib.vt.edu
… For visual vs. non-visual question detection, we use a Long Short-Term Memory (LSTM) recurrent neural network (RNN) trained on part of speech (POS) tags to capture visual-specific linguistic structure. For true vs … Question generation …

Improving sentiment analysis with document-level semantic relationships from rhetoric discourse structures
J Märkle-Huß, S Feuerriegel… – … of the 50th …, 2017 – hl-128-171-57-22.library.manoa …
… For instance, [5] uses syntax and lexical information to obtain sentence- but not document- level semantic structures. Similarly, [6] studies the internal composition of sentences by considering the syntax tree structure with the help of recursive deep neural networks …

Predicting Recall Probability to Adaptively Prioritize Study
S Mooney, K Sun, E Bomgardner – 2017 – teaching-machines.cc
… we need to move beyond two dimensional cue-target KC encoding and implement more comprehensive KC modeling and question generation … promising techniques such as Bayesian Networks in which knowledge states are latent variables, and neural networks which are …

An Interactive Tour Guide for a Heritage Site
S Chelaramani, V Muthireddy… – Proceedings of the …, 2017 – openaccess.thecvf.com
… We adopt this method as one of our retrieval methods, since it outperforms other frameworks. Recently, several approaches also generate captions of images using neural networks [16, 19, 44, 48] … 3.4. Question Generation …

Attentive listening system with backchanneling, response generation and flexible turn-taking
D Lala, P Milhorat, K Inoue, M Ishida… – Proceedings of the 18th …, 2017 – aclweb.org
… We implement a dialogue act tagger which clas- sifies an utterance into questions, statements or others such as greetings. This is currently based on a support vector machine and is moving to a recurrent neural network. Questions …

Learning Algorithms for Broad-Coverage Semantic Parsing
S Swayamdipta – 2017 – cs.cmu.edu
… The specific advance we employ is the stack LSTM (Dyer et al., 2015), a neural network that continuously summarizes the contents of the stack data structures in which a transition-based parser’s state is conventionally encoded …

Deep reinforcement learning: An overview
Y Li – arXiv preprint arXiv:1701.07274, 2017 – arxiv.org
… The integration of reinforcement learning and neural networks dated back to 1990s (Tesauro, 1994; Bertsekas and Tsitsiklis, 1996; Schmidhuber, 2015) … A recurrent neural network (RNN) is often used to process sequential inputs like speech and language, element …

Abstractive Text Summarization Using Deep Learning
CX Tran – 2017 – dspace.jaist.ac.jp
… First of all, we introduce a widely used deep learning model for abstractive summariza- tion, the encoder-decoder model using recurrent neural network (RNN). In this model, the document is simply viewed as a sequence of words … 5 2.1.2 Recurrent Neural Network …

A Big Data architecture for knowledge discovery in PubMed articles
F Gargiulo, S Silvestri, M Ciampi – … and Communications (ISCC) …, 2017 – ieeexplore.ieee.org
… Its aim is to provide a search tool aid based on automatic question generation suggestions, obtained from basic search terms of the … The proposed similarity search engine is based on word2vec alghoritm, a WEs model [6] based on an unsuper- vised Neural Network (NN) that …

Building and Querying an Enterprise Knowledge Graph
D Song, F Schilder, S Hertz, G Saltini… – IEEE Transactions …, 2017 – ieeexplore.ieee.org
… In this example, cases (ie, legal cases) will be the first segment. As the question generation proceeds, our system finds sug- gestions based on the discovered question segments, and produces the following sequence of segments: cases, filed by …

A novel X-FEM based fast computational method for crack propagation
Z Cheng, H Wang, PMB Vitanyi, N Chater, M Barzegari… – arxiv.org
… physics.data-an). arXiv:1708.01648 [pdf, other] Title: 3D-PRNN: Generating Shape Primitives with Recurrent Neural Networks … arXiv:1708.01666 [pdf] Title: An Effective Training Method For Deep Convolutional Neural Network …

A Closer Look at Pure-Text Human-Interaction Proofs
HE Kiziloz, K Bicakci – IEEE Transactions on Human-Machine …, 2017 – ieeexplore.ieee.org
… [13] shows that the hardest category of visual reCAPTCHA HIPs could be solved with 99.8% accuracy using a deep convolutional neural network … Some tests may be unsolvable by a search en- gine at the time of question generation but may become solvable later …

Survey of the State of the Art in Natural Language Generation: Core tasks, applications and evaluation
A Gatt, E Krahmer – arXiv preprint arXiv:1703.09902, 2017 – arxiv.org
… a single survey, and many of these are covered in other individual surveys, including Mani (2001) and Nenkova and McKeown (2011) for summarisation; Androutsopoulos and Malakasiotis (2010) for paraphrasing; and Piwek and Boyer (2012) for automatic question generation …

A cognitive system for business and technical support: A case study
P Dhoolia, P Chugh, P Costa… – IBM Journal of …, 2017 – ieeexplore.ieee.org
… In a constrained case of SAP domain ticket clusters, we perform an additional step of matching a standard error code or message with the problem description, and if a match is found, then we generate a question based on a standard question generation template …

Motivational and Intervention Systems and Monitoring with mHealth Tools
GMT Ahsan – 2017 – epublications.marquette.edu
Page 1. Marquette University e-Publications@Marquette Dissertations (2009 -) Dissertations, Theses, and Professional Projects Motivational and Intervention Systems and Monitoring with mHealth Tools Golam Mushih Tanimul Ahsan Marquette University …

Computational Natural Language Inference: Robust and Interpretable Question Answering
RR Sharp – 2017 – search.proquest.com
… The second reduces pre-processing costs by limiting itself to a single sentence and using a neural network to learn a latent representation of the background knowledge … 30. 1.3.4 Contribution 4: Using neural networks to rank justications. for interpretable question answering …

Towards the Implementation of an Intelligent Software Agent for the Elderly
AHF Dinevari – 2017 – era.library.ualberta.ca
Page 1. Towards the Implementation of an Intelligent Software Agent for the Elderly by Amir Hossein Faghih Dinevari A thesis submitted in partial fulfillment of the requirements for the degree of Master of Science Department of Computing Science University of Alberta …

DDoS attacks in cloud computing: issues, taxonomy, and future directions
G Somani, MS Gaur, D Sanghi, M Conti… – Computer …, 2017 – Elsevier
Security issues related to the cloud computing are relevant to various stakeholders for an informed cloud adoption decision. Apart from data breaches, the cyber.

Data Science Solution for User Authentication
A Ibrahim – 2017 – ir.lib.uwo.ca
Page 1. Western University Scholarship@Western Electronic Thesis and Dissertation Repository September 2017 Data Science Solution for User Authentication Anas Ibrahim The University of Western Ontario Supervisor Dr. Abdelkader Ouda The University of Western Ontario …

Security challenges in cloud computing: state-of-art
K Bhushan, BB Gupta – International Journal of Big Data …, 2017 – inderscienceonline.com
Page 1. Int. J. Big Data Intelligence, Vol. 4, No. 2, 2017 81 Copyright © 2017 Inderscience Enterprises Ltd. Security challenges in cloud computing: state-of-art Kriti Bhushan and BB Gupta* Department of Computer Engineering …

Crowdsourcing Based Evaluation of Ranking Approaches for Spatial Keyword Querying
J Chen, H Lu, I Keles, CS Jensen – Mobile Data Management …, 2017 – ieeexplore.ieee.org
… B. Binary Question Generation We generate a sufficiently small number of binary questions in four steps. First, we use a learn-to-rank method to construct a list l12 of k most important spatial objects from the two top- k lists l1 and l2 …

Routledge Revivals: Science as a Questioning Process (1996)
N Sanitt – 2017 – books.google.com
Page 1. ROULEDGEREV | V ALS Science as a Questioning Process Nigel Sanitt Page 2. Routledge Revivals Science as a Questioning Process First published in 1996, Science as a Questioning Process evaluates scientific …

Teaching Law Students How to Become Metacognitive Thinkers: Helping Students Develop Their Mental Apps
ES Fruehwald – 2017 – papers.ssrn.com
Page 1. Electronic copy available at: https://ssrn.com/abstract=3048990 Teaching Law Students How to Become Metacognitive Thinkers: Helping Students Develop Their Mental Apps by E. Scott Fruehwald Successful students take charge of their own learning.1 …

Autonomous Cars: An Adaptable Feedback Mechanism For Customised Ethics
S Jaiswal – 2017 – dalspace.library.dal.ca
… 62 Figure 14: The selection sorting algorithm [55] …. 63 Figure 15: The flowchart of question generation phase ….. 64 Figure 16: The flowchart of the decision making phase ….. 67 …

From Discourse Structure to Text Specificity: Studies of Coherence Preferences
JJ Li – 2017 – search.proquest.com
From Discourse Structure to Text Specificity: Studies of Coherence Preferences. Abstract. To successfully communicate through text, a writer needs to organize information into an understandable and well-structured discourse for the targeted audience …

List of Recently Published Quality Research Papers
MN Ansari – Imperial Journal of Interdisciplinary Research, 2017 – imperialjournals.com
List of Recently Published Quality Research Papers.

(Visited 175 times, 1 visits today)