RNN (Recurrent Neural Network) & Question Answering Systems 2016


Notes:

RNNs can use their internal memory to process arbitrary sequences of inputs.

Wikipedia:

References:

See also:

100 Best Recurrent Neural Network Videos | RNN (Recurrent Neural Network) & Dialog Systems 2016


Generating factoid questions with recurrent neural networks: The 30m factoid question-answer corpus
IV Serban, A García-Durán, C Gulcehre, S Ahn… – arXiv preprint arXiv …, 2016 – arxiv.org
Page 1. Generating Factoid Questions With Recurrent Neural Networks: The 30M Factoid Question-Answer Corpus Iulian Vlad Serban?? Alberto Garc?a-Durán? Caglar Gulcehre? Sungjin Ahn? Sarath Chandar? Aaron Courville? Yoshua Bengio†? Abstract …

Learning recurrent span representations for extractive question answering
K Lee, S Salant, T Kwiatkowski, A Parikh, D Das… – arXiv preprint arXiv …, 2016 – arxiv.org
… The goal of our extractive question answering system is to predict the single best answer span among all candidates from the passage p, denoted as A(p). Therefore, we define a probability distribution … A theoretically grounded application of dropout in recurrent neural networks …

Visual7w: Grounded question answering in images
Y Zhu, O Groth, M Bernstein… – Proceedings of the IEEE …, 2016 – cv-foundation.org
… Traditional question answering system relies on an elabo- rate pipeline of models involving natural language parsing, knowledge base querying, and answer generation [6]. Re- cent neural network models … Figure 5: Diagram of the recurrent neural network model for pointing QA …

Learning Statistical Scripts with LSTM Recurrent Neural Networks.
K Pichotta, RJ Mooney – AAAI, 2016 – aaai.org
… will have a job inter- view,” and perhaps “Smith will find a job.” Inferences of this type are required for robust question-answering systems … We describe a novel statistical script model which uses a recurrent neural network with a Long Short- Term Memory (LSTM) architecture …

Dialog-based language learning
JE Weston – Advances in Neural Information Processing Systems, 2016 – papers.nips.cc
… Imitation Learning This approach involves simply imitating one of the speakers in observed di- alogs, which is essentially a supervised learning objective3. This is the setting that most existing di- alog learning, as well as question answer systems, employ for learning …

Question answering on freebase via relation extraction and textual evidence
K Xu, S Reddy, Y Feng, S Huang, D Zhao – arXiv preprint arXiv …, 2016 – arxiv.org
Page 1. Question Answering on Freebase via Relation Extraction and Textual Evidence Kun Xu1 and Siva Reddy2 and Yansong Feng1,a and Songfang Huang3 and Dongyan Zhao1 1Institute of Computer Science & Technology …

A Deep Architecture for Semantic Matching with Multiple Positional Sentence Representations.
S Wan, Y Lan, J Guo, J Xu, L Pang, X Cheng – AAAI, 2016 – aaai.org
… Long short term memory (LSTM) is an advanced type of Recurrent Neural Network by further using memory cells and gates to learn … Answers which is a community question answering system where some users propose questions to the system and other users will submit their …

Deep recurrent models with fast-forward connections for neural machine translation
J Zhou, Y Cao, X Wang, P Li, W Xu – arXiv preprint arXiv:1606.04199, 2016 – arxiv.org
… Moreover, NMT models can also be easily adapted to other tasks such as dialog systems (Vinyals and Le, 2015), question answering systems (Yu et al … The recurrent neural network (RNN), or its spe- cific form the LSTM, is generally used as the basic unit of the encoding and …

Combining Retrieval, Statistics, and Inference to Answer Elementary Science Questions.
P Clark, O Etzioni, T Khot, A Sabharwal, O Tafjord… – AAAI, 2016 – aaai.org
… The embeddings are learned using the recurrent neural network language model (RNNLM) (Mikolov et al. 2010; 2013) … Brill, E.; Dumais, S.; and Banko, M. 2002. An analysis of the AskMSR question-answering system. In Proceedings of EMNLP, 257–264 …

Sentence pair scoring: Towards unified framework for text comprehension
P Baudiš, J Pichl, T Vysko?il, J Šedivý – arXiv preprint arXiv:1603.06127, 2016 – arxiv.org
… based on an extension of the curatedv2 ques- tion dataset (introduced in (Baudiš and Šedivý, 2015), further denoisified by Mechanical Turkers) with candidate sentences as retrieved by the Yo- daQA question answering system (Baudiš, 2015 … 3.2 Recurrent Neural Networks …

Sequence-based Structured Prediction for Semantic Parsing.
C Xiao, M Dymetman, C Gardent – ACL (1), 2016 – aclweb.org
… tic parsing, has received a lot of attention recently, in particular in the context of building Question- Answering systems (Kwiatkowski et al … Given the recently shown effectiveness of RNNs (Recurrent Neural Networks), in particu- lar Long Short Term Memory (LSTM) networks …

Recurrent Neural Network Encoder with Attention for Community Question Answering
WN Hsu, Y Zhang, J Glass – arXiv preprint arXiv:1603.07044, 2016 – arxiv.org
Page 1. Recurrent Neural Network Encoder with Attention for Community Question Answering Wei-Ning … Glass. 2015. Vectorslu: A continuous word vector approach to answer selection in community question answering systems. In …

Sequence to backward and forward sequences: A content-introducing approach to generative short-text conversation
L Mou, Y Song, R Yan, G Li, L Zhang, Z Jin – arXiv preprint arXiv …, 2016 – arxiv.org
… Sutskever et al. (2014) propose seq2seq for machine translation; the idea is to encode a source sen- tence as a vector by a recurrent neural network (RNN) and to decode the vector to a target sentence by another RNN … In a dialogue-like question-answering system, Yin et al …

Question answering on linked data: Challenges and future directions
S Shekarpour, KM Endris, A Jaya Kumar… – Proceedings of the 25th …, 2016 – dl.acm.org
… [17] A. Graves and N. Jaitly. “Towards end-to-end speech recognition with recurrent neural networks”. In: 31st International Conference on Machine Learning (ICML-14) … [20] G. Hu et al. “SpeechQoogle: An Open-Domain Question Answering System with Speech Interface” …

Strategy and Policy Learning for Non-Task-Oriented Conversational Systems.
Z Yu, Z Xu, AW Black, AI Rudnicky – SIGDIAL Conference, 2016 – aclweb.org
… systems, such as machine translation (Ritter et al., 2011), retrieval-based response se- lection (Banchs and Li, 2012), and sequence-to- sequence recurrent neural network (Vinyals and Le … Reinforcement learning is also used in question-answering systems (Misu et al., 2012) …

Survey of named entity recognition systems with respect to Indian and foreign languages
N Patil, AS Patil, BV Pawar – International Journal of Computer …, 2016 – search.proquest.com
… Question answering systems pinpoint relevant information by expressing question in natural language whose answers are extracted by … CRF), Hidden Morkov Models (HMM), Maximum Entropy (ME), Memory-Based Learning (MBL), Recurrent Neural Networks (RNN), Support …

Text Analytics: the convergence of Big Data and Artificial Intelligence.
A Moreno, T Redondo – IJIMAI, 2016 – researchgate.net
… OpenEphyra [22] was an open-source question answering system, originally derived from Ephyra, which was developed by Nico Schlaefer and … Deep Learning is a very broad field and most promising work is moving around Recurrent Neural Networks (RNN) and Convolutional …

Learning semantic relatedness in community question answering using neural models
H Nassif, M Mohtarami, J Glass – ACL 2016, 2016 – aclweb.org
… 2013. Joint language and transla- tion modeling with recurrent neural networks … 2015. Vectorslu: A continuous word vector approach to answer selection in com- munity question answering systems. SemEval-2015, page 282. 144 Page 159 …

Learning executable semantic parsers for natural language understanding
P Liang – Communications of the ACM, 2016 – dl.acm.org
… Even question answering systems relied less on understanding and more on a shallower analysis coupled with a large collection of unstructured text … Due to their empirical success, there has been a recent surge of interest in using recurrent neural networks and their extensions …

Match-srnn: Modeling the recursive matching structure with spatial rnn
S Wan, Y Lan, J Xu, J Guo, L Pang, X Cheng – arXiv preprint arXiv …, 2016 – arxiv.org
… et al., 2014] utilize convolutional neural network (CNN), while LSTM- RNN [Palangi et al., 2015] adopts recurrent neural network with long … Answers, a commu- nity question answering system where some users propose questions to the system and other users will submit their an …

SelQA: A New Benchmark for Selection-based Question Answering
T Jurczyk, M Zhai, JD Choi – Tools with Artificial Intelligence …, 2016 – ieeexplore.ieee.org
… to mitigate these limitations to allow for a more thorough reading comprehension evaluation of open-domain question answering systems … convolutional and recurrent neural networks are implemented to analyze this corpus and to provide strong baseline measures for future …

RDI_Team at SemEval-2016 Task 3: RDI Unsupervised Framework for Text Ranking
A Magooda, A Gomaa, A Mahgoub, H Ahmed… – Proceedings of the 10th …, 2016 – aclweb.org
… The community question ranking problem is dif- ferent than ordinary question answering system that aims to generate a satisfactory answer for a … Using the previous generated samples a recurrent neural network for language modeling (Mikolov et al., 2011) with the following …

A vision enriched intelligent agent with image description generation
L Zhang, B Fielding, P Kinghorn, K Mistry – Proceedings of the 2016 …, 2016 – dl.acm.org
… We have extended the agent’s vocabulary monumentally through the implementation of a question-answering system using Wikipedia as a data source … as the Fast R-CNN, and pairing it with a more efficient sentence generation method such as Recurrent Neural Networks [15] …

Overview of NTCIR-13
MP Kato, Y Liu, C Gurrin, H Joho… – Proceedings of the …, 2016 – research.nii.ac.jp
… We used recurrent neural networks (RNNs) to extract the answer … We have been developing question answering systems for the world history multiple-choice questions in the National Center Test for University Admissions …

Seeing is believing: the quest for multimodal knowledge by Gerard de Melo and Niket Tandon, with Martin Vesely as coordinator
G de Melo, N Tandon – ACM SIGWEB Newsletter, 2016 – dl.acm.org
… For instance, humans can no longer keep up with IBM’s question answering system Watson, not even all-time Jeopardy … While sufficient for traditional statistical NLP methods of the past, it turns out that deep recurrent neural networks need even more training data [Vinyals et al …

” A Distorted Skull Lies in the Bottom Center…” Identifying Paintings from Text Descriptions
A Guha, M Iyyer, J Boyd-Graber – … of the Workshop on Human-Computer …, 2016 – aclweb.org
… Abstract Most question answering systems use symbolic or text information … 2015. Conditional random fields as recurrent neural networks. In Interna- tional Conference on Computer Vision. Yuke Zhu, Oliver Groth, Michael Bernstein, and Li Fei- Fei. 2015 …

Open-domain Factoid Question Answering via Knowledge Graph Search
A Aghaebrahimian, F Jurc?cek – Proceedings of the …, 2016 – pdfs.semanticscholar.org
… 27 Page 7. 9 Conclusion We introduced a question answering system with no dependence on external lexicons or any other tool … 2011. Rnnlm – recurrent neural network language modeling toolkit. V. Punyakanok, D. Roth, and W. Yih. 2008 …

Knowledge base question answering based on deep learning models
Z Xie, Z Zeng, G Zhou, T He – International Conference on Computer …, 2016 – Springer
… Automatic question answering systems are aimed at returning the direct and exact answers to natural language questions … Long short-Term Memory (LSTM) [20] is one of the popular variations of Recurrent Neural Networks (RNN) which is widely used to deal with variable-length …

Lyapunov filtering of objectivity for Spanish sentiment model
I Chaturvedi, E Cambria… – Neural Networks (IJCNN) …, 2016 – ieeexplore.ieee.org
… lower level features are learned using convolution and the higher level features are learned using recurrent neural networks (RNNs) guided by … sentiment flow, in [19], the authors used recurrent CNN to model the dynamics in dialogue tracking and question answering systems …

Learning for Biomedical Information Extraction: Methodological Review of Recent Advances
F Liu, J Chen, A Jagannatha, H Yu – arXiv preprint arXiv:1606.07993, 2016 – arxiv.org
… deep belief networks (DBNs)[138]; (2) supervised/discriminative, eg, deep neural networks (DNNs)[139], convolutional neural networks (CNNs)[140] and recurrent neural networks(RNNs)[ 141 … AskHERMES: An online question answering system for complex clinical questions …

Towards a framework for closed-domain question answering in Italian
E Damiano, R Spinelli, M Esposito… – … -Image Technology & …, 2016 – ieeexplore.ieee.org
… 172–176. [3] F. Ture and O. Jojic, “Simple and effective question an- swering with recurrent neural networks,” arXiv, 2016. [4] J. Straková, M. Straka, and J. Hajic, Neural Networks for Featureless Named Entity Recognition in Czech …

Imisound: An unsupervised system for sound query by vocal imitation
Y Zhang, Z Duan – Acoustics, Speech and Signal Processing …, 2016 – ieeexplore.ieee.org
… We also would like to adopt more advanced deep neural networks such as Recurrent Neural Networks (RNN) to model the temporal evolution of vocal … [20] Dragomir R. Radev, Hong Qi, Harris Wu, and Weiguo Fan, “Evalutating web-based question answering systems,” in Proc …

Full-Time Supervision based Bidirectional RNN for Factoid Question Answering
D Xu, WJ Li – arXiv preprint arXiv:1606.05854, 2016 – arxiv.org
… neural network for document modeling. EMNLP, 2015. [11] Jeffrey Pennington, Richard Socher, and Christopher D. Manning. Glove: Global vectors for word representation. EMNLP, 2014. [12] Ellen Riloff and Michael Thelen. A rule-based question answering system for reading …

Context-aware Natural Language Generation for Spoken Dialogue Systems.
H Zhou, M Huang, X Zhu – COLING, 2016 – aclweb.org
… 1 Introduction Natural language generation (NLG), the task of generating natural language from a knowledge base or a logical form representation, is an important component of dialogue or question answering system … Speech recognition with deep recurrent neural networks …

Hierarchical attention model for improved machine comprehension of spoken content
W Fang, JY Hsu, H Lee, LS Lee – … Technology Workshop (SLT) …, 2016 – ieeexplore.ieee.org
… based Multi-hop Recurrent Neural Network (AMRNN) mentioned above in the previous work [1] uti- lizes the attention mechanism with recurrent neural networks (RNN) [19] to … [6] PR Comas, J. Turmo, and L. M`arquez, “Sibyl, a fac- toid question-answering system for spoken …

Towards Machine Comprehension of Spoken Content: Initial TOEFL Listening Comprehension Test by Machine
BH Tseng, SS Shen, HY Lee, LS Lee – arXiv preprint arXiv:1608.06378, 2016 – arxiv.org
… graphs for question answering through conversational dialog.” [5] PR Comas, J. Turmo, and L. M`arquez, “Sibyl, a factoid question-answering system for spoken … [11] A. Graves, A.-r. Mohamed, and G. Hinton, “Speech recognition with deep recurrent neural networks,” in Acoustics …

A deep neural network for chinese zero pronoun resolution
Y Qingyu, Z Weinan, Z Yu… – arXiv preprint arXiv …, 2016 – pdfs.semanticscholar.org
… As the inputs of both parts of ZPSNN are se- quences of words, one way to encode the in- put word sequences is via a recurrent neural net- work (Elman, 1991). Recurrent neural networks (RNN) have been widely exploited to deal with variable-length sequence input …

Structural Sentence Similarity Estimation for Short Texts.
W Ma, T Suel – FLAIRS Conference, 2016 – aaai.org
… For example sentimental analysis tasks focus more on critical words while question answering systems rely more on syn- tactic features … 2015; Zhang, Zhao, and LeCun 2015; Hu et al. 2014) and Recurrent Neural Networks (RNNs)(Cho et al. 2014; Hill et al. 2015; Chung et al …

Discourse Sense Classification from Scratch using Focused RNNs.
G Weiss, M Bajec – CoNLL Shared Task, 2016 – aclweb.org
… Any recurrent neural network (RNN) can be used as its building block, but we decided to use the GRU layer (Chung et al., 2014 … They are designed to only work with question-answering systems, use a weighted com- bination of all input states, and can focus on only one aspect …

An empirical evaluation of various deep learning architectures for bi-sequence classification tasks
A Laha, V Raykar – arXiv preprint arXiv:1607.04853, 2016 – arxiv.org
… Another class of problems originated from question-answering systems also known as answer selection, where given a question, a candidate … For many single sequence classification tasks, the state-of- the-art approaches are based on recurrent neural networks (RNN variants …

Deep LSTM based Feature Mapping for Query Classification
Y Shi, K Yao, L Tian, D Jiang – Proceedings of the 2016 Conference of …, 2016 – aclweb.org
… For query classifications, recurrent neural networks (RNNs) and convolutional neural networks … The question type classification task is to clas- sify a question into a specific type, which is a very important step in question answering system …

Social question answering: Textual, user, and network features for best answer prediction
P Molino, LM Aiello, P Lops – ACM Transactions on Information Systems …, 2016 – dl.acm.org
… 2013] have also been used, showing potential in addressing the retrieval of synonyms and hypernyms. Recurrent neural network language models [Yih et al. 2013] have been studied as well, confirming that lexical semantics is suitable to tackle the problem. 2.2 …

Learning sentiment and semantic relatedness in user generated content using neural models
HM Nassif – 2016 – dspace.mit.edu
… The Community Question Answering system is concerned with automatically find- ing the related questions in an existing set of questions, and finding the relevant answers to a new question … 33 2.2 Recurrent Neural Networks . . . . . 34 …

Neural Reasoning for Legal Text Understanding.
KJ Adebayo, G Boella, L Di Caro – JURIX, 2016 – books.google.com
… We implemented a Memory Network-based Question Answering system which test a Machine’s un- derstanding of legal text and identifies whether … Recurrent Neural Networks (RNNs)[6] have connections that have loops, adding feedback and memory to the networks over time …

Negation Scope Detection with Recurrent Neural Networks Models in Review Texts
L Lazib, Y Zhao, B Qin, T Liu – International Conference of Young …, 2016 – Springer
… of negation by tackling it as a sequence labelling problem, and use a variety of recurrent neural networks (RNNs) models … that are sensitive to polarity in the information extraction and natural language processing domain, like sentiment analysis or question answering systems …

Community-Based Question Answering via Heterogeneous Social Network Learning
M Ester – 2016 – aaai.org
… we choose long-short term memory (LSTM) (Hochreiter and Schmid- huber 1997) instead of traditional recurrent neural network to learn the … is built up by collecting the crowd- sourced data from a popular high-quality community-based question answering system, Quora, and …

Neural Clinical Paraphrase Generation with Attention
SA Hasan, B Liu, J Liu, A Qadir, K Lee, V Datla… – ClinicalNLP …, 2016 – aclweb.org
… We propose an end-to-end neural network built on an attention-based bidirectional Recurrent Neural Network (RNN) architecture … generating query variants or pattern alternatives for information retrieval, information extraction or question answering systems, creating reference …

A Chinese text paraphrase detection method based on dependency tree
Y Jiang, Y Hao, X Zhu – … , and Control (ICNSC), 2016 IEEE 13th …, 2016 – ieeexplore.ieee.org
… In question answering systems, we want to know whether a query can be mapped to a candidate question … They use a deep network to train word vectors and sentence vectors such as recurrent neural network(RNN) and so on …

Training IBM Watson using Automatically Generated Question-Answer Pairs
J Lee, G Kim, J Yoo, C Jung, M Kim, S Yoon – arXiv preprint arXiv …, 2016 – arxiv.org
… answering in natural languages. It is believed that IBM Watson can understand large corpora and answer relevant questions more effectively than any other question-answering system currently available. To unleash the full …

History question classification and representation for Chinese Gaokao
K Yu, Q Liu, Y Zheng, T Zhao… – … Processing (IALP), 2016 …, 2016 – ieeexplore.ieee.org
… A question representation with high accuracy is an essential part of a KB-based question answering system and this paper gives a promising method on both entity … [8] T. Mikolov, M. Karafiat, 1. Burget, J. Cernocky, and S. Khudanpur, “Recurrent neural network based language …

Using Recurrent Neural Networks in Trajectory Prediction
J Moore, B Ribeiro – 2016 – cs.purdue.edu
… Recurrent Neural networks are a class of Neural networks that have been used extensively in sequence prediction problems … have started to become start of the art in speech translation [1], machine translation [7], image captioning [8], and question answering systems [9]. All of …

QA-It: Classifying Non-Referential It for Question Answer Pairs
T Lee, A Lutz, JD Choi – ACL 2016, 2016 – anthology.aclweb.org
… need to perform coref- erence resolution for question answering systems. In the future, we will double the size of our an- notation so we can train a better model and have a more meaningful evaluation. We are also planning on developing a recurrent neural network model for …

FastHybrid: A Hybrid Model for Efficient Answer Selection.
L Wang, M Tan, J Han – COLING, 2016 – aclweb.org
… Abstract Answer selection is a core component in any question-answering systems. It aims to select cor- rect answer sentences for a given question from a pool of candidate sentences … 2013. Speech recognition with deep recurrent neural networks …

Mapping distributional to model-theoretic semantic spaces: a baseline
F Dernoncourt – arXiv preprint arXiv:1607.02802, 2016 – arxiv.org
… Abstract Word embeddings have been shown to be use- ful across state-of-the-art systems in many natural language processing tasks, ranging from question answering systems to depen- dency parsing … 2016. De-identification of patient notes with recurrent neural networks …

A Study of Deep Learning for Legal Question Answering Systems
DV Tran – 2016 – dspace.jaist.ac.jp
… iii, v, vii, 17, 37, 40, 44 QA Question Answering. iii, v, 1, 2, 4, 5 RNN Recurrent Neural Network. 2 … First, can we apply deep learning for legal question answering systems? We compare several developed deep learning models on designated legal question answering corpora …

GRU-RNN Based Question Answering Over Knowledge Base
S Chen, J Wen, R Zhang – … : Semantic, Knowledge, and Linked Big Data, 2016 – Springer
… This architecture consists of two-column independent recurrent neural network (RNN) with Gated Recurrent Unit (GRU) cell [10] … 6 Conclusion. In this paper, we propose our knowledge graph based question answering system …

Learning to Extract Conditional Knowledge for Question Answering using Dialogue
P Wang, L Ji, J Yan, L Jin, WY Ma – … of the 25th ACM International on …, 2016 – dl.acm.org
… Table 3 gives several examples of the CKB elements. Each line represents a knowledge record. The object is different based on d- (a) The CKB QA system. (b) FAQ retrieval system. Figure 1: Question answering systems examples ifferent conditions …

Deep Learning for Natural Language Processing-Research at Noah’s Ark Lab
H Li – 2016 – hangli-hl.com
Page 1. Deep Learning for Natural Language Processing – Research at Noah’s Ark Lab Hang Li Noah’s Ark Lab Huawei Technologies Institute of Software, CAS Beijing Jan 15, 2016 Page 2. DL for NLP @Noah Lab Zhengdong Lu Lifeng Shang Lin Ma Zhaopeng Tu Xin Jiang …

Regression Based Approaches for Detecting and Measuring Textual Similarity
S Sarkar, P Pakray, D Das, A Gelbukh – International Conference on …, 2016 – Springer
… Finding Semantic similarity is an important component in various fields such as information retrieval, question-answering system, machine translation and … our system trained using two neural networks (i) multilayer feed forward network; and (ii) layered recurrent neural network …

LSTM-based Deep Learning Models for Answer Ranking
Z Li, J Huang, Z Zhou, H Zhang… – Data Science in …, 2016 – ieeexplore.ieee.org
… We aim to solve the answer ranking problem in practical question answering system with deep learning approaches … Tang et al., 2015)[ 2 ], machine translation (Bahdanau et al., 2015)[3] and text summarization (Rush et al., 2015)[ 4 ]. Recurrent neural network (RNN), especially …

Speech Intent Recognition for Robots
B Shen, D Inkpen – … and Computers in Sciences and in Industry …, 2016 – ieeexplore.ieee.org
… However, the system in their work is limited to provide information delivery services, similar to a question answering system, but deployed on a … neighbor method with a distance function [7]. Recent research also proposed the idea of using recurrent neural networks for sequence …

Are Deep Learning Approaches Suitable for Natural Language Processing?
S Alshahrani, E Kapetanios – … on Applications of Natural Language to …, 2016 – Springer
… have been recommended for processing sequences [10], while Recursive Neural Networks are collections of recurrent networks that can address trees [6]. Another application uses Recurrent Neural Networks for question answering systems about paragraphs [15], and a Neural …

A neural network approach for knowledge-driven response generation
P Vougiouklis, J Hare, E Simperl – 2016 – eprints.soton.ac.uk
… It should also be beneficial in the development of Question- Answering systems, by enhancing their ability to generate human-like … recent systems that are employed in the automatic response generation domain are based on Recurrent Neural Networks (RNNs) (Sordoni et al …

Predicting answer types for question-answering
I Bogatyy – Retrieved on, 2016 – pdfs.semanticscholar.org
… Many of the question-answering systems capable of answering a broad range of world knowledge on a production scale tend to have … Regarding the third part, model selection, the core choice has been between an architecture based on a recurrent neural network, like [2] or [3 …

Multi-level Gated Recurrent Neural Network for dialog act classification.
W Li, Y Wu – COLING, 2016 – pdfs.semanticscholar.org
… Henceforth, we apply a novel multi-level gated recurrent neural network (GRNN) with non-textual information to predict the DA tag … for example, the DA of the current sentence provides very important information for answer generation in an automatic question answering system …

CS381V Final Project Report Visual Question Answering using Natural Language Object Retrieval and Saliency Cues
A Padmakumar, A Saran – pdfs.semanticscholar.org
… The goal of this project is to examine the effect of ex- plicitly providing two additional sources of information to a Visual Question Answering system apart from the orig … DPPNet is also similar in goal but uses a dif- ferent type of recurrent neural network – a GRU instead of an LSTM …

Visual Question Answering Using Various Methods
S Qu – cs224d.stanford.edu
… In general, the visual question answer system are required to answer all kinds of questions people might ask relate or not related with the image … They align the convolutional neural network with bidirectional recurrent neural network …

Neural Paraphrase Generation with Stacked Residual LSTM Networks
A Prakash, SA Hasan, K Lee, V Datla, A Qadir… – arXiv preprint arXiv …, 2016 – arxiv.org
… in several NLP applications, for example, by generating query variants or pattern alternatives for information retrieval, information extraction or question answering systems, by creating reference … Most of the deep learning models for NLP use Recurrent Neural Networks (RNNs) …

Answer Quality Prediction Joint Textual and Non-Textual Features
H Liu, J Huang, C An, X Fu – Web Information Systems and …, 2016 – ieeexplore.ieee.org
… In summary, we developed a basic model for question similarity based on recurrent neural network … [15] Moschitti A, Quarteroni S. Linguistic kernels for answer re-ranking in question answering systems[J]. Information Processing & Management, 2011, 47(6): 825-842 …

Question Answering Using Deep Learning
E Stroh, P Mathur – pdfs.semanticscholar.org
… GRU and LSTM units allow recurrent neural networks (RNNs) to handle the longer texts required for QA … of problems encountered by real-world QA systems, they also involve a significant amount of information retrieval engineering in addition to the question-answering system …

A Chinese Question Answering Approach Integrating Count-based and Embedding-based
B Wang, J Niu, L Ma, Y Zhang, L Zhang, J Li, P Zhang… – tcci.ccf.org.cn
… Thus, we explore a Question Answering system which is char- acterized in Chinese for the QA task of NLPCC … a shallow convolutional neural network (CNN) which combines the ordered overlapped information into the hidden layer [8]. Recurrent neural network (RNN) and the …

A Chinese Question Answering Approach Integrating Count-Based and Embedding-Based Features
B Wang, J Niu, L Ma, Y Zhang, L Zhang, J Li… – … on Computer Processing …, 2016 – Springer
… convolutional neural network (CNN) which combines the ordered overlapped information into the hidden layer [8]. Recurrent neural network (RNN) and the … to-end system which is specifically applicable for Chinese can also be the trend for Chinese Question-Answering system …

Neural Architectures for Fine-grained Entity Type Classification
S Shimaoka, P Stenetorp, K Inui, S Riedel – arXiv preprint arXiv …, 2016 – arxiv.org
… Their end goal was to use the resulting types in a question answering system and they developed a conditional random field model that they … They defined 22 types and used a two-part neural classifier that used a recurrent neural network to obtain a vector representation of …

Represent, Aggregate, and Constrain: A Novel Architecture for Machine Reading from Noisy Sources
J Naradowsky, S Riedel – arXiv preprint arXiv:1610.09722, 2016 – arxiv.org
… Additionally, while many question answering systems are designed to extract a single answer to a single query, a user may wish to under- stand many aspects of a single entity or event. In machine reading, this is akin to pairing each text passage with multiple queries …

Natural Language Understanding and Intelligent Applications: 5th CCF Conference on Natural Language Processing and Chinese Computing, NLPCC 2016 …
CY Lin, N Xue, D Zhao, X Huang, Y Feng – 2016 – books.google.com
… and Hengliang Luo Learning to Recognize Protected Health Information in Electronic Health Records with Recurrent Neural Network … Qingrong Xia, Zhenghua Li, Jiayuan Chao, and Min Zhang Open Domain Question Answering System Based on Knowledge …

User authority ranking models for community question answering1
Y Rao, H Xie, X Liu, Q Li, FL Wang… – Journal of Intelligent & …, 2016 – content.iospress.com
… such as imposing minimal hurdles to their users, retrieving the desired data, and developing and evaluating question answering systems … this problem by combining the Convolutional Neural Networks (CNN) for visual recognition and Recurrent Neural Networks (RNN) for …

An Enhanced Intelligent Agent with Image Description Generation
B Fielding, P Kinghorn, K Mistry, L Zhang – International Conference on …, 2016 – Springer
… We have extended ALICE’s vocabulary through the implementation of a question-answering system, using Wikipedia as a data source … The sentence generation functionality could also be altered to use machine learned methods such as Recurrent Neural Networks that have …

Proceedings of the 17th Annual Meeting of the Special Interest Group on Discourse and Dialogue
R Fernandez, W Minker, G Carenini… – Proceedings of the 17th …, 2016 – aclweb.org
… 11 Joint Online Spoken Language Understanding and Language Modeling With Recurrent Neural Networks Bing Liu and Ian Lane … Language Portability for Dialogue Systems: Translating a Question-Answering System from English into Tamil Satheesh Ravi and Ron Artstein …

The Fraunhofer IAIS Audio Mining System: Current State and Future Directions
C Schmidt, M Stadtschnitzer… – … Communication; 12. ITG …, 2016 – ieeexplore.ieee.org
… the analysis result can be inte- grated into larger analytics frameworks such as ontology- based frameworks or question answering systems (see Sec … Recently we also trained acoustical models based on recurrent neural networks (RNN) using the Eesen toolkit [8]. In Table 1, the …

Deep Recurrent Models with Fast-Forward Connections for Neural Machine Translation
JZYCX Wang, PLW Xu – pdfs.semanticscholar.org
… NMT models can also be easily adapted to other tasks such as dialog systems (Vinyals and Le, 2015), question answering systems (Yu et al … Recurrent neural network (RNN) or its specific form LSTM is generally employed as the basic unit of the encoding and decoding function …

Named Entity Recognition in Swedish Health Records with Character-Based Deep Bidirectional LSTMs
S Almgren, S Pavlov, O Mogren – BioTxtM 2016, 2016 – aclweb.org
… 4.1 Model layout We used a deep bidirectional recurrent neural network with LSTM cells … They have been successfully been applied to sentiment analysis (Tang et al., 2015), question answering systems (Hagstedt P Suorra and Mogren, 2016), and machine translation …

Learning to answer questions from semi-structured knowledge sources
A Morales – 2016 – dspace.mit.edu
… Chapter 1 Introduction Question answering systems aim to provide high-precision access to large amounts of information … of information are some of the reasons why the task is so challenging. Question answering systems like Apple’s Siri or Amazon’s Echo make an intelligent …

IMAI RESEARCH GROUP COUNCIL
EDDJS Carrión, RG Crespo, JP Mestras, A Rocha… – academia.edu
… OpenEphyra [22] was an open-source question answering system, originally derived from Ephyra, which was developed by Nico Schlaefer and … Deep Learning is a very broad field and most promising work is moving around Recurrent Neural Networks (RNN) and Convolutional …

Parsing with Recurrent Neural Networks
JH Cross III – 2016 – ir.library.oregonstate.edu
… James Henry Cross III for the degree of Doctor of Philosophy in Computer Science presented on December 8, 2016. Title: Parsing with Recurrent Neural Networks Abstract approved: Liang Huang … Page 5. Parsing with Recurrent Neural Networks by James Henry Cross III …

RNN-Enhanced Deep Residual Neural Networks for Web Page Classification
Y Lin – 2016 – theses.ucalgary.ca
… RecurrentNet or RNN Recurrent Neural Network RecursiveNet Recursive Neural Network … We propose a top RNN classifier, which can work along with aforementioned ResNet. Recurrent neural network (RNN) is good at dealing with variant length inputs, and …

Combining Lexical and Semantic-based Features for Answer Sentence Selection
J Shia, J Xua, Y Yaoa, S Zhenga, B Xua – OKBQA 2016, 2016 – aclweb.org
… TREC- QA, WikiQA (Wang et al., 2007; Yang et al., 2015), allowing researchers to build effective question answering systems (Voorhees and … we infer with a bit radicalness that the Recurrent Neural Networks (RNN), which focus on sequential information, maybe not a proper …

Odpovídání na otázky pomocí dotaz? do strukturovaných databází
J Pichl – 2016 – dspace.cvut.cz
… ix Page 10. Page 11. Contents Introduction 1 Goals 3 1 Question answering system 5 1.1 Types of the questions . . . . . 5 1.2 IBM Watson … 35 3.6.2 Recurrent neural network . . . . . 35 3.6.3 Other models …

Chinese Sentence Similarity Calculation Based on Multi-Features Fusion
H Wang, Y Wen, Y Xian, R Li – Journal of Residuals Science & …, 2016 – dpi-journals.com
… Kunming, China Abstract: Sentence similarity computation plays an increasingly important role in text mining, Web page retrieval, Machine Translation, Speech Recognition and Question-answering systems. Existing methods …

A Study on Image Semantic Analysis Algorithm for Natural Language Understanding
J LUO, HUAJUN WANG, YANMEI LI… – Journal of Residuals …, 2016 – dpi-journals.com
… For example, text summarization, translation, information retrieval, question answering system and so on, but this kind of recognition algorithm is lagging behind, which is not conducive to the research and innovation of new … Second, sort words by recurrent neural network …

Representation learning for natural language processing
C Zhou? ??? – HKU Theses Online (HKUTO), 2016 – hub.hku.hk
… bining a convolutional neural network (CNN) and a long short-term memory (LSTM) recurrent neural network. The input to the LSTM network is sequential … TREC Text REtrieval Conference RNN Recurrent Neural Network LSTM Long Short-Term Memory …

Spoken Language Understanding
M McTear, Z Callejas, D Griol – The Conversational Interface, 2016 – Springer
… Google Now). Utterance 3 has the form of a request but its purpose is to obtain information, so it is more like a question that would be sent to a question answering system or to an online encyclopedia such as Wikipedia. Finally …

Overview of NTCIR-12.
K Kishida, MP Kato – NTCIR, 2016 – research.nii.ac.jp
… Our system achieved 76 points in Benesse mock exam Jun 2015 (Pattern 1) of Phase 2. SML Question-Answering System for World History Essay and Multiple-choice Exams at NTCIR-12 QA@Lab-2 Takuma Takada, Takuya Imagawa, Takuya Matsuzaki and Satoshi Sato [Pdf …

A hybrid approach to domain-independent taxonomy learning
E Lefever – Applied Ontology, 2016 – content.iospress.com
… (2016) propose a hybrid system, where dependency paths (ie extension of Hearts’s patterns) are encoded using a recurrent neural network, and show that the combination of … (2013) was used, which implements recurrent neural networks to learn the word vector representations …

A comparative study of segment representation for biomedical named entity recognition
HL Shashirekha, HA Nayel – Advances in Computing …, 2016 – ieeexplore.ieee.org
… IE systems for various further tasks such as relation extraction between entities, diagnosis classification and biomedical question answering systems … D. Song, and D. Huang, “Biomedical named entity recognition based on extended recurrent neural networks,” in Bioinformatics …

Optimising spoken dialogue systems using Gaussian process reinforcement learning for a large action set
TFW Nicholson, M Gaši? – pdfs.semanticscholar.org
… full belief space. However this work still used the summary action space. Recent work in [48] examined how to user recurrent neural networks (RNNs) to perform policy op- timisation in master action space. They propose a Policy …

Knowledge Questions from Knowledge Graphs
D Seyler, M Yahya, K Berberich – arXiv preprint arXiv:1610.09935, 2016 – arxiv.org
Page 1. Knowledge Questions from Knowledge Graphs Dominic Seyler‡ ‡University of Illinois at Urbana-Champaign Urbana, Illinois dseyler2@illinois.edu Mohamed Yahya† †Max Planck Institute for Informatics Saarland Informatics Campus myahya@mpi-inf.mpg.de …

Gradient Descent Learns Linear Dynamical Systems
M Hardt, T Ma, B Recht – arXiv preprint arXiv:1609.05191, 2016 – arxiv.org
… Text translation, speech recognition, time series prediction, video captioning and question answering systems, to name a few, are all sequence to sequence learning problems … Recurrent neural networks form an expressive class of non-linear sequence models …

MSc Dissertation Generating Stories from Images
D Wilmot – project-archive.inf.ed.ac.uk
… natural question answering systems. Based on SIND, Sort story (Agrawal et al., 2016) uses an LSTM (Hochreiter & … A joint model that combines a DCNN (Deep Convolutional Neural Network) with an RNN (Recurrent Neural Network, specifically LSTM decoder (Vinyals et al …

Machine Translation Evaluation: A Survey
ALF Han, DF Wong – arXiv preprint arXiv:1605.04515, 2016 – arxiv.org
Page 1. arXiv:1605.04515v6 [cs.CL] 19 Jun 2016 Machine Translation Evaluation: A Survey Aaron Li-Feng Han ILLC Faculty of Science University of Amsterdam l.han@uva.nl Derek F. Wong , Lidia S. Chao NLP2CT Lab Faculty …

USE OF LANGUAGE TECHNOLOGY TO IMPROVE
R Gupta – 2016 – wlv.openrepository.com
… Networks . . . . . 109 vii Page 8. 4.2.1 Recurrent Neural Networks . . . . . 111 … 74 3.4 EditingisInProgress . . . . . 74 4.1 An Unfolded Recurrent Neural Network . . . . . 112 4.2 AnUnfoldedLSTMnetwork …

Effects of an e-Prescribing interface redesign on rates of generic drug prescribing: exploiting default options
S Malhotra, AD Cheriff, JT Gossey… – Journal of the …, 2016 – academic.oup.com
We use cookies to enhance your experience on our website. By continuing to use our website, you are agreeing to our use of cookies. You can change your cookie settings at any time. Find out more …

Learning Open Domain Knowledge From Text
GG Angeli – 2016 – nlp.stanford.edu
… In fact, many downstream NLP applications do query large knowledge bases. Prominent exam- ples include question answering systems (Voorhees, 2001), and semantic parsers (Zelle and Mooney, 1996; Zettlemoyer and Collins, 2007; Kwiatkowski et al., 2013; Berant and …

Computational Autism
B Galitsky – 2016 – Springer
Page 1. Human–Computer Interaction Series Boris Galitsky Computational Autism Page 2. Human–Computer Interaction Series Editors-in-chief Desney Tan, Microsoft Research, USA Jean Vanderdonckt, Université catholique de Louvain, Belgium Page 3 …

Textual Inference for Machine Comprehension
M Gleize – 2016 – theses.fr
Page 1. Le cas échéant, logo de l’établissement co-délivrant le doctorat en cotutelle internationale de thèse , sinon mettre le logo de l’établissement de préparation de la thèse (UPSud, HEC, UVSQ, UEVE, ENS Cachan, Polytechnique, IOGS, …) NNT : 2016SACLS004 …

Health Information Extraction from Social Media
A Nikfarjam – 2016 – search.proquest.com
Health Information Extraction from Social Media. Abstract. Social media is becoming increasingly popular as a platform for sharing personal health-related information. This information can be utilized for public health monitoring …

Paraphrase identification using knowledge-lean techniques
A Eyecioglu Ozmutlu – 2016 – sro.sussex.ac.uk
… PR PAN: Plagiarism Detection Corpus PE: Paraphrase Extraction PG: Paraphrase Generation PI: Paraphrase Identification PoS: Part-of-Speech QA: Question Answering RBF: Radial Basis Function RNNM: Recurrent Neural Network Models …

Understanding and Describing Tennis Videos
MK Sukhwani – 2016 – cvit.iiit.ac.in
… Large scale deployments of combi- nation of Convolutional Neural Networks (CNNs) and Recurrent Neural Networks (RNNs) have … that access videos and their textual descriptions simultaneously – video search, browsing, question-answer system, commercial applications …

Exploiting Semantic and Topic Context to Improve Recognition of Proper Names in Diachronic Audio Documents
I Sheikh – 2016 – tel.archives-ouvertes.fr
Page 1. Exploiting Semantic and Topic Context to Improve Recognition of Proper Names in Diachronic Audio Documents Imran Sheikh To cite this version: Imran Sheikh. Exploiting Semantic and Topic Context to Improve Recognition …

A Stochastic Petri Net based NLU Scheme for Technical Documents Understanding
A Psarologou – 2016 – corescholar.libraries.wright.edu
Page 1. Wright State University CORE Scholar Browse all Theses and Dissertations Theses and Dissertations 2016 A Stochastic Petri Net Based NLU Scheme for Technical Documents Understanding Adamantia Psarologou Wright State University …

3.35 Open Problems and State-of-Art of Session Types
N Yoshida – Compositional Verification Methods for Next …, 2016 – pdfs.semanticscholar.org
Page 24. 22 15191–Compositional Verification Methods for Next-Generation Concurrency 3.35 Open Problems and State-of-Art of Session Types Nobuko Yoshida (Imperial College London, GB) License Creative Commons …

(Visited 54 times, 2 visits today)