MaxEnt & Dialog Systems 2014


Resources:

Wikipedia:

See also:

100 Best Apache OpenNLP Videos | Apache OpenNLP & Dialog Systems


The second dialog state tracking challenge M Henderson, B Thomson… – 15th Annual Meeting of …, 2014 – anthology.aclweb.org … This consists of all 1117 dialogs with DM-POMDP. Paid Amazon Mechanical Turkers were as- signed tasks and asked to call the dialog systems. … A maximum entropy model (maxent) is used (details in (Metallinou et al., 2013)), which is common practice for stacking classifiers. … Cited by 31 Related articles All 15 versions

Web-style ranking and SLU combination for dialog state tracking JD Williams – 15th Annual Meeting of the Special Interest Group on …, 2014 – aclweb.org … trained di- rect models have been applied, and two studies on dialog data from two publicly deployed dialog systems suggest direct … Maximum entropy (maxent) models have been proposed which provide the strengths of both of these approaches (Metallinou et al., 2013). … Cited by 20 Related articles All 13 versions

The SJTU system for dialog state tracking challenge 2 K Sun, L Chen, S Zhu, K Yu – 15th Annual Meeting of the Special Interest …, 2014 – aclweb.org … state tracking is important because spo- ken dialog systems rely on it to choose proper actions as spoken dialog systems interact with users. … f4 bias (i, r)= 1 In particular, all above feature function are 0 when i? 0. 4.2 Maximum Entropy Model Total 6 MaxEnt models (Bohus … Cited by 8 Related articles All 8 versions

Markovian discriminative modeling for dialog state tracking H Ren, W Xu, Y Yan – 15th Annual Meeting of the Special …, 2014 – anthology.aclweb.org … 2000. Maximum entropy markov mod- els for information extraction and segmentation. In Pat Langley, editor, ICML, pages 591–598. Morgan Kaufmann. … Jason Williams. 2012. A critical analysis of two statistical spoken dialog systems in public use. … Cited by 3 Related articles All 8 versions

Out-of-vocabulary word detection in a speech-to-speech translation system HK Kuo, EE Kislal, L Mangu, H Soltau… – Acoustics, Speech and …, 2014 – ieeexplore.ieee.org … however, such as a dialogue system with confirmation or error cor- rection, especially when playing back audio snippets to the user, ac- curate … Our OOV detector is based on a maximum entropy (maxent) classi- fier similar to that described in [7, 8]. The speech recognizer uses a … Cited by 3 Related articles All 2 versions

Detecting ‘Request Alternatives’ User Dialog Acts from Dialog Context Y Ma, E Fosler-Lussier – … Workshop on Spoken Dialog Systems, 2014 – uni-ulm.de … Table 3 Selected top features with corresponding class labels from maximum entropy classifier top … SRW 2013 p. 91 3. Manning, C., Klein, D.: Optimization, maxent models, and … Proceedings of 5th International Workshop on Spoken Dialog Systems Napa, January 17-20, 2014 … Cited by 1 Related articles All 2 versions

DietTalk: Diet and Health Assistant Based on Spoken Dialog System S Jung, SH SeonghanRyu, GG Lee – uni-ulm.de … [3] C. Lee, S. Jung, S. Kim, and GG Lee, “Example-based dialog modeling for practical multi- domain dialog system,” Speech Communication, 15(5): 466-484, 2009 [4] A. Ratnaparkhi and MP Marcus, “Maximum entropy models for natural language ambiguity resolution,” Ph. … Cited by 1 Related articles All 2 versions

Micro-Counseling Dialog System based on Semantic Content S Han, Y Kim, GG Lee – uni-ulm.de … To detect a statement dialog act, we used the MaxEnt algorithm [2] using … Development of multi-party anticipatory knowledge-intensive natural language dialog system]. … A Maximum Entropy Approach to Natural Language Processing, Association for Computational Linguistics, pp … Cited by 1 Related articles All 2 versions

A discriminative sequence model for dialog state tracking using user goal change detection Y Ma, E Fosler-Lussier – Spoken Language Technology …, 2014 – ieeexplore.ieee.org … Most state-of-the-art spoken dialog systems assume the user goal is fixed during a … With the trained MaxEnt user goal change detector, a binary decision – whether the user changes … CRF) improve performance over a turn-by-turn model (eg, via a Maximum Entropy system), and … Related articles All 2 versions

A Brief History and Recent Trend of Spoken Dialog Systems (SDS) T Kawahara – 2014 – ncmmsc.org 2014/3/28 1 A Brief History and Recent Trend of Spoken Dialog Systems (SDS) Tatsuya Kawahara (Kyoto University, Japan) … [Zue,Glass,Leung,Phillips,Seneff1989] • First real spoken dialog system • Navigation of Cambridge, MA – Similar to current smart phone applications … Related articles All 2 versions

Markovian discriminative modeling for cross-domain dialog state tracking H Ren, W Xu, Y Yan – Spoken Language Technology …, 2014 – ieeexplore.ieee.org … Spoken dialog systems (SDS) are currently widely used in daily lives and industry for various purpose … could been used as f, for example, Maximum En- tropy Model (MaxEnt), Conditional Random … It can also be seen as a gener- alization of the Maximum Entropy Markov Model … Cited by 1 Related articles

An end-to-end dialog system for TV program discovery D Ramachandran, PZ Yeh, W Jarrold… – … (SLT), 2014 IEEE, 2014 – ieeexplore.ieee.org … We start by giving an overview of the architecture of our end-to-end dialog system for TV program discovery, along with a brief description of the main components … The NER is a machine- learned approach that uses the maximum entropy framework similar to [3, 4]. Canonicalizer … Cited by 2 Related articles

A Demonstration of Dialogue Processing in SimSensei Kiosk F Morbini, D DeVault, K Georgila, R Artstein… – 15th Annual Meeting of …, 2014 – aclweb.org … edu Abstract This demonstration highlights the dia- logue processing in SimSensei Kiosk, a virtual human dialogue system that con- ducts interviews related to … This classifier is trained using the Switchboard DAMSL corpus (Jurafsky et al., 1997) using a maximum entropy model. … Related articles All 11 versions

Evaluation of Invalid Input Discrimination Using Bag-of-Words for Speech-Oriented Guidance System H Majima, R Torres, H Kawanami, S Hara… – Natural Interaction with …, 2014 – Springer … Maximum entropy (ME) method [6] provides a general purpose machine learning technique for classification … Saruwatari, H., Shikano, K.: Noise robust real world spoken dialogue system using GMM … cjlin/libsvm (2001) 6. Manning, C., Klein, D.: Optimization, Maxent Models, and … Related articles All 5 versions

Neural Network Models for Lexical Addressee Detection S Ravuri, A Stolcke – Fifteenth Annual Conference of the …, 2014 – mazsola.iit.uni-miskolc.hu … More information about the dialog system itself and its spoken language understanding approach can be found in [7]. The resulting corpus comprises 6.3 hours of recordings over 38 … of the joint training approaches, however, beats the maximum entropy language model baseline … Cited by 1 Related articles All 12 versions

Temporal supervised learning for inferring a dialog policy from example conversations L Li, H He, JD Williams – Spoken Language Technology …, 2014 – ieeexplore.ieee.org … improved performance. Finally, we plan to evaluate with a real end-to-end dialog system. 8. ACKNOWLEDGEMENTS Thanks to Dan Bohus for making his maxent software avail- able to us in the experiments. 9. REFERENCES [1 … Related articles All 10 versions

Improving Classification-Based Natural Language Understanding with Non-Expert Annotation F Morbini, E Forbell, K Sagae – 15th Annual Meeting of the Special …, 2014 – aclweb.org … In the next section, we present one ap- plication of this procedure to an existing conversa- tional dialogue system deployed on the web, and … typical interaction with the system is shown in Figure 1. The system and the NLU clas- sifier based on Maximum Entropy models (Berger … All 9 versions

Two-phase reanalysis model for understanding user intention S Kang, J Seo – Pattern Recognition Letters, 2014 – Elsevier … maximum entropy model (Master’s thesis), Sogang University, 1998. [3]; J. Eun, Analysis of speech acts for korean using support vector machine (Master’s thesis), Sogang University, 2002. … Recent approaches to dialog management for spoken dialog systems. J. Comput. Sci. … Related articles All 2 versions

Dialogue Act Recognition Using Probabilistic Networks And Ranking Feature Selection Approaches AA Yahya, AR Ramli – researchgate.net … For instance, the speaker’s utterance “Can you reserve three tickets for me?” to the dialogue system is interpreted as if a user … and regression tree [23], hidden Markov models [24], Naïve Bayes (NB) [8], Static Bayesian Network (SBN) [12, 13], and maximum entropy [16, 17], as … Related articles

Exploiting out-of-vocabulary words for out-of-domain detection in dialog systems S Ryu, D Lee, GG Lee, K Kim… – Big Data and Smart …, 2014 – ieeexplore.ieee.org … K. Komatani, K. Matsuyama, K. Funakoshi, and H. G. Okuno, “A two-stage domain selection framework for extensible multi-domain spoken dialogue systems”, In Proceedings of the SIGdial 2011, 18-29, Portland, Oregon, USA. [9] A. Ratnaparkhi, “Maximum entropy models for … Related articles All 3 versions

Acquisition and use of long-term memory for personalized dialog systems Y Kim, J Bang, J Choi, S Ryu, S Koo… – … Analyses enabling Artificial …, 2014 – Springer … Comput. Speech Lang. 24(2), 150–174 (2010) 5. Lee, C., Jung, S., Kim, S., Lee, GG: Example-based dialog modeling for practical multi- domain dialog system. Speech Commun. … 444–447 16. Ratnaparkhi, A.: A maximum entropy part-of-speech tagger. … Cited by 1 Related articles All 3 versions

Improving domain action classification in goal-oriented dialogues using a mutual retraining method CN Seon, H Lee, H Kim, J Seo – Pattern Recognition Letters, 2014 – Elsevier … To reduce lengthy and rigid interactions of menu-driven navigation and keyword searches, dialogue systems based on a natural language interface have been developed. … The dialogue system should analyze the context of utterance (9) in order to resolve this ambiguity. … Related articles All 2 versions

On a Hybrid NN/HMM Speech Recognition System with a RNN-Based Language Model D Soutner, J Zelinka, L Müller – Speech and Computer, 2014 – Springer … The width of hidden layer was changed from 200 to 700, the number of 4-gram maximum entropy features from 1M to 4M. … Present-day applications of the decoder include com- mercial dictation software, automatic subtitling system, dialogue system and systems for searching in … Related articles All 3 versions

Building A Vocabulary Self-Learning Speech Recognition System L Qin, A Rudnicky – Fifteenth Annual Conference of the …, 2014 – mazsola.iit.uni-miskolc.hu … But in many applications, eg, voice search or spoken dialog systems, OOV words are usually content words such as names and loca- tions which contain … In the hybrid decoding result, we estimate the POS label for OOV words using the Stanford MaxEnt POS tagger [25]. … Cited by 2 Related articles All 5 versions

Leveraging semantic web search and browse sessions for multi-turn spoken dialog systems L Wang, L Heck, D Hakkani-Tur – Acoustics, Speech and Signal …, 2014 – ieeexplore.ieee.org … entities. For Baseline 2 (No Session), we follow [25] to construct a multiclass Maximum Entropy classifier with one vs all strategy. Discussion. Table … log. 4. ENTITY EXTRACTION IN SPOKEN DIALOG SYSTEMS SDS Corpora. To … Cited by 8 Related articles All 9 versions

Chengqing Zong: Statistical natural language processing X Zhang – Machine Translation, 2014 – Springer … And, in the statistical models section, up-to-date adaptive language models such as cache-based, mixture-based and maximum-entropy-based models are introduced … abstracting and information extraction (Chapter 15), and human-computer dialogue systems (Chapter 16). … Related articles All 4 versions

R-cube: a dialogue agent for Restaurant Recommendation and Reservation S Kim, RE Banchs – Asia-Pacific Signal and Information …, 2014 – ieeexplore.ieee.org … and VD Pietra, “A maximum entropy approach to natural language processing,” Computational Linguistics, vol. 22, no. 1, pp. 39–71, 1996. [8] Lee, C., Jung, S., Kim, S., Lee, G., “Example-based dialog modeling for practical multi-domain dialog system,” Speech Communication. … Cited by 2 Related articles All 2 versions

Detecting deletions in ASR output MS Seigel, PC Woodland – Acoustics, Speech and Signal …, 2014 – ieeexplore.ieee.org … may be deleted from information-bearing “slots”, such as in information extraction and dialogue systems. … This result is verified by making use of a non-sequential maximum entropy model in … Baseline LAPR -0.047 0.314 -0.019 0.358 MaxEnt LAPR -0.044 0.312 0.010 0.351 1 … Cited by 1 Related articles All 3 versions

Probabilistic multiparty dialogue management for a game master robot C Kennington, K Funakoshi, Y Takahashi… – Proceedings of the 2014 …, 2014 – dl.acm.org … Furthermore, if a participant becomes confused during the game, the dialogue system (ds) will offer hints so the game can proceed. … When invoked, a Maximum Entropy classifier uses a simple feature set (a somewhat different ap- proach from [5]) to determine if a user receives … Cited by 1 Related articles

A Comprehensive Study on Natural Language Processing and Natural Language Interface to Databases U Shafique, H Qaiser – International Journal of Innovation and …, 2014 – researchgate.net … Maximum entropy approach [10] is also based on statistical analysis that combines diverse pieces of contextual evidence in order to … the period including in Machine and Automatic Translation, Information Retrieval (IR), Information Extraction (IE), Dialogue Systems and Question … Cited by 1 Related articles All 2 versions

Generation of effective referring expressions in situated context K Garoufi, A Koller – Language, Cognition and Neuroscience, 2014 – Taylor & Francis … View all references) present a model of a “rational speaker”, which is based on a maximum entropy learner and generates references optimally with respect to an embedded … Learning to adapt to unknown users: Referring expression generation in spoken dialogue systems. … Cited by 7 Related articles All 2 versions

Dialogue-based Exploration of Graphics for Users with a Visual Disability J Plhak – 2014 – is.muni.cz … 30 4.1.1 DialogueStrategies . . . . . 32 4.2 Types of Dialogue Systems . . . . . 32 4.3 Dialogue Systems Supported by Ontologies . . . . . 34 4.3.1 UK Cancer Referrals and Home Control . . . . . 34 4.3.2 SmartWeb . . . . . … Related articles All 3 versions

Predicate-argument structure analysis with zero-anaphora resolution for dialogue systems K Imamura, R Higashinaka, T Izumi – Proc. COLING, 2014 – anthology.aclweb.org … When we build dialogue systems on PASA, predicate phrases will be identified using part-of-speech patterns that include verbs, adjectives, and copular verbs. … 4.2 Models The models for the selector are based on maximum entropy classification. … Cited by 3 Related articles All 4 versions

Cluster based Chinese Abbreviation Modeling Y Shi, YC Pan, MY Hwang – Fifteenth Annual Conference …, 2014 – mazsola.iit.uni-miskolc.hu … In some practical applications (eg dialogue systems and voice search systems), document level context information is not available. … In this paper, we apply RNN with maximum entropy extension to Chinese abbreviation modeling. … Related articles All 4 versions

Automatic Dialogue Act Recognition with Syntactic P Král, C Cerisara – Citeseer … large applicative systems, such as the VERBMOBIL [14], NE- SPOLE [15] and C-STAR [16] machine translation and dialogue systems that rely … successfully use in [29] for joint DA segmentation and classification hidden-event language models and a maximum entropy classifier. … Related articles All 4 versions

Automatic dialogue act recognition with syntactic features P Král, C Cerisara – Language Resources and Evaluation, 2014 – Springer … 2006 ) and C-STAR (Blanchon and Boitet 2000 ) machine translation and dialogue systems that rely on dialogue act classification. … ( 2006 ) successfully use in for joint DA segmentation and classification hidden-event language models and a maximum entropy classifier. … Cited by 2 Related articles All 5 versions

Application of deep belief networks for natural language understanding R Sarikaya, GE Hinton, A Deoras – Audio, Speech, and …, 2014 – ieeexplore.ieee.org … tuned with backpropagation. We compare a DBN-initialized neural network to three widely used text classification algorithms: support vector machines (SVM), boosting and maximum entropy (MaxEnt). The plain DBN-based … Cited by 16 Related articles All 8 versions

Question answering system: A heuristic approach V Bhoir, MA Potey – Applications of Digital Information and Web …, 2014 – ieeexplore.ieee.org … RELATED WORK Since 1960s, a variety of natural language database frontends, dialog systems and language understanding systems were … semantically richer treebank [5]. IBM’s Statistical Question Answering System [6] is an application of maximum entropy classification for … Related articles All 2 versions

Hierarchical Dirichlet Process Topic Modeling for Large Number of Answer Types Classification in Open domain Question Answering S Park, D Lee, J Choi, S Ryu, Y Kim, S Kown… – Information Retrieval …, 2014 – Springer … We used cluster ID-tagged data to train a Maximum Entropy (ME) model. … Technology] and National Research Founda- tion of Korean (NRF) [NRF-2014R1A2A1A01003041, Development of multi-party anticipatory knowledge-intensive natural language dialog system]. … Related articles All 3 versions

Re-ranking ASR Outputs for Spoken Sentence Retrieval Y Song, H Ahn, H Kim – ceur-ws.org … Error Correction Using Maximum Entropy Language Model. In: Proceedings of the International Speech Communication Association, pp.2137-2140 (2004) 5. Choi, J., Lee, D., Ryu, S., Lee, K., Lee, GG: Engine-Independent ASR Error Management for Dialog Systems. … Related articles All 3 versions

Topic classification of spoken inquiries using transductive support vector machine R Torres, H Kawanami, T Matsui, H Saruwatari… – Natural Interaction with …, 2014 – Springer … and Smartphones: Putting Spoken Dialog Systems into Practice, DOI 10.1007/978-1-4614-8280- 2 23, © Springer Science+Business Media New York 2014 261 Page 2. 262 R. Torres et al. function (RBF) kernel, PrefixSpan boosting (pboost), and maximum entropy (ME) [5]. We … Related articles All 5 versions

Identifying Simple Narrative Structure in Personal Narratives R Swanson, E Rahimtoroghi, T Corcoran, MA Walker – nwnlp2014.cs.sfu.ca … Reid Swanson, Elahe Rahimtoroghi, Thomas Corcoran and Marilyn A. Walker Natural Language and Dialog Systems Lab University of California … Naive Bayes (Witten and Frank, 2005), Confidence Weighted Linear Classifier (Dredze et al., 2008), Maximum Entropy (Witten and … Related articles

Tutor Dialogue Planning with Contextual Information and Discourse Structure R Fisher, R Simmons – cs.cmu.edu … make relation learning tractable, they often rely on rela- tively simplistic classifiers such as linear support vector machines and maximum entropy classifiers [7 … A Contextual Knowledge Base is a set of facts about a user and the environ- ment [9]. In dialogue systems, examples of … Related articles All 2 versions

IBM’s Belief Tracker: Results On Dialog State Tracking Challenge Datasets R Kadlec, J Libovický, J Macek, J Kleindienst – EACL 2014, 2014 – anthology.aclweb.org … The System col- umn shows what dialog system was used to col- lect the dataset. … The overall winner of the DSTC (Lee and Es- kenazi, 2013) used a maximum entropy model, which they claim to be outperformed by bringing more structure to the model by using the Condi- tional … Cited by 1 All 6 versions

Representing Syntactic-Semantic Knowledge from English Texts R Guidry Jr, J Chen – … on the International Conference on Artificial …, 2014 – world-comp.org … 5 Reference [1] James Allen, George Ferguson, Bradford W. Miller, Eric K. Ringger, and Teresa Sikorski Zollo. Dialogue systems: From theory to practice in TRAINS96. … A Maximum Entropy Approach to Natural Language Processing. Computational Linguist., 22(1):39–71, 1996. … Related articles All 2 versions

Semantic language models for Automatic Speech Recognition AO Bayer, G Riccardi – Spoken Language Technology …, 2014 – ieeexplore.ieee.org … [3] R. Lau, R. Rosenfeld, and S. Roukos, “Trigger-based language models: a maximum entropy approach,” in … J. Castro Bieda, M., and R. De-Mori, “Cache neural network language models based on long-distance dependencies for a spoken dialog system,” in Proceedings … Cited by 1 Related articles All 3 versions

A Two-Step Approach for Efficient Domain Selection in Multi-Domain Dialog Systems I Lee, S Kim, K Kim, D Lee, J Choi, S Ryu… – Natural Interaction with …, 2014 – Springer … The goal of our proposed method (Fig. 12.1) is to select the domain expert, which can generate the system response to the user intention in multi-domain dialog systems. … We used the maximum entropy (ME) classifier as a preselection model to compute the domain score. … Cited by 1 Related articles All 3 versions

Language Resources and Technology in Latvia (2010–2014) I Skadi?a, I Auzi?a, G B?rzdi?š… – … : Proceedings of the …, 2014 – books.google.com … 4.8. Multimedia Novel research has been started to apply language technologies to dialog systems for smartphones. Two … Pinnis, M., Goba, K.(2011). Maximum Entropy Model for Disambiguation of Rich Morphological Tags. In … Related articles All 3 versions

Identifying narrative clause types in personal stories R Swanson, E Rahimtoroghi… – 15th Annual Meeting …, 2014 – anthology.aclweb.org … Thomas Corcoran and Marilyn A. Walker Natural Language and Dialog Systems Lab University … Confidence Weighted Linear Classifier (CWLC)(Dredze et al., 2008), Maximum Entropy (ME)(Witten … 750 1000 # Training Clauses F? Score Model q CRF CWLC MaxEnt NB Figure 2 … Cited by 2 Related articles All 8 versions

Inter-Annotator Agreement on Spontaneous Czech Language T Valenta, L Šmídl, J Švec, D Soutner – Text, Speech and Dialogue, 2014 – Springer … To record the corpus, a simple dialogue system was developed. … The vocabulary of RNN LM was shortlisted to 40 k most frequent words, the size of the hidden layer was chosen h = 400 and the model was trained together with 8 M maximum entropy features. … Cited by 1 Related articles All 4 versions

Heterogeneous networks and their applications: Scientometrics, name disambiguation, and topic modeling B King, R Jha, DR Radev – Transactions of the …, 2014 – tacl2013.cs.columbia.edu … 1 A Maximum Entropy Model for … NLP support vector machines errors space classification correcting word parsing detecting MaxEnt models entropy maximum approach based attachment model models phrase prepositional disambiguation Dialogue systems dialogue spoken … Cited by 2 Related articles All 13 versions

Towards Modeling Collaborative Task Oriented Multimodal Human-human Dialogues L Chen – 2014 – indigo.uic.edu … tial multimodal dialogue systems. SmartKom was a mixed-initiative dialogue system with full symmetric multimodality: the combined input of speech, gesture, and facial expression … sequences. (Sridhar et al., 2009) used a Maximum Entropy based model with prosody fea- … All 3 versions

QAS SD Joshi – ijaiem.org … Since 1960s, a variety of natural language database frontends, dialog systems and language understanding systems were created. … using semantically richer tree bank [5]. IBM’s Statistical Question Answering System [6] is an application of maximum entropy classification for … Related articles

TSVD as a Statistical Estimator in the Latent Semantic Analysis Paradigm G Pilato, G Vassallo – ieeexplore.ieee.org … applications, such as natural language understanding, cognitive modeling, speech recog- nition, smart indexing, anti-spam filters, dialogue systems and other … This consideration is related to the principles of maximum entropy [4]: the model should be the simplest possible, given … Cited by 1 Related articles All 5 versions

SemEval-2014 Task 6: Supervised Semantic Parsing of Robotic Spatial Commands K Dukes – SemEval 2014, 2014 – aclweb.org … of applica- tions, including question answering (Kwiat- kowski et al., 2013; Krishnamurthy and Mitchell, 2012), dialog systems (Artzi and … Statistical Statistical maximum entropy parser 87.35 60.84-26.51 RoBox Evang and Bos Statistical CCG parser+ structured perceptron 86.80 … Related articles All 9 versions

Domain Specific Named Entity Recognition (DSNER) from Web Documents P Kumar, RK Goel, PS Sharma – International Journal of …, 2014 – search.proquest.com … representations of sentences like in the case of Information Extraction systems [2][3] and Human-Machine Dialogue systems or merely … Models (HMM) [15], Vector Machines (SVM) [14], Decision Trees, Conditional Random Fields (CRF) [17] and Maximum Entropy Models [16]. … Related articles All 2 versions

Conditional Random Field In Segmentation And Noun Phrase Inclination Tasks For Russian AA Romanenko, P II – dialog-21.ru … Temporal expressions extraction is important for natural language under- standing modules of spoken dialog systems. … L-CRF is a discriminative model and in this aspect it resembles the popular Maximum entropy Markov model (MEMM). … Related articles

Boosting methods in machine learning (I) C Shen – 2014 – itee.uq.edu.au Page 1. Boosting methods in machine learning (I) Chunhua Shen Te University of Adelaide July 2014 NB: Some slides are adapted from R. Schapire’s NIPS tutorial Page 2. Example: “How May I Help You?” [Gorin et al.] • goal … Related articles

An Approach for Intention Perception Based on Knowledge Network H Li, G Sun, B Xu – … , Knowledge and Grids (SKG), 2014 10th …, 2014 – ieeexplore.ieee.org … The Maximum entropy classifier is adopted to classify those entries into large mount of categories using structural … Since the real conversation system like a catch ball game, we will devote this model to build an interactive dialogue system. … Related articles All 2 versions

Language Learning via Unsupervised Corpus Analysis B Goertzel, C Pennachin, N Geisweiller – Engineering General Intelligence …, 2014 – Springer … phenomena. We also believe that dependency grammars have a more natural fit with maximum-entropy ideas, where a depen- dency relationship can be literally interpreted as the mutual information between word-pairs [Yur98]. …

Sequential labeling for tracking dynamic dialog states S Kim, RE Banchs – 15th Annual Meeting of the Special Interest Group …, 2014 – aclweb.org … 1 Introduction A dialog manager is one of the key components of a dialog system, which aims at determining the system actions to generate … one is based on CRFs for our pro- posed sequential labeling approach; and the other is a baseline using maximum entropy (ME) that … Cited by 5 Related articles All 8 versions

Automated transcription of conversational Call Center speech–with respect to non-verbal acoustic events G Sárosi, B Tarján, T Fegyó… – Intelligent Decision …, 2014 – speechlab.tmit.bme.hu … First simple pattern matching on the ASR transcript to answer a set of standard quality control questions and then maximum entropy ranking, where their goal was to estimate the probability of a call being bad based on fea- tures extracted from the automatic transcription. … Related articles All 4 versions

Engine-independent asr error management for dialog systems J Choi, D Lee, S Ryu, K Lee, K Kim, H Noh… – … Dialogue Systems …, 2014 – uni-ulm.de … Proceedings of 5th International Workshop on Spoken Dialog Systems Napa, January 17-20, 2014 25 Page 11. 1. M. Jeong, S. Jung, and GG Lee, “Speech recognition error correction using maximum entropy language model”, In Proceedings of the International Speech … Cited by 4 Related articles

Part-of-speech tagging in written slang V Korolainen – 2014 – jyx.jyu.fi … FIGURE 1: Architecture pipeline for a Spoken Dialogue System….. … Intelligence API Application Programming Interface CRF Conditional Random Fields HCI Human-Computer Interaction HMM Hidden Markov Model MEHMM Maximum-Entropy Hidden Markov Model … Related articles

ASR Independent Hybrid Recurrent Neural Network Based Error Correction for Dialog System Applications J Choi, S Ryu, K Lee, Y Kim, S Koo, J Bang… – … enabling Artificial Agents …, 2014 – Springer … In: Intenational Workshop Series on Spoken Dialogue Systems Technology (IWSDS) (2014). 4. Evermann, G., Woodland, P.: Posterior probability … ITS)MathSciNet. 7. Jeong, M., Jung, S., Lee, GG: Speech recognition error correction using maximum entropy language model. … Related articles All 2 versions

Combining multiple parallel streams for improved speech processing JTR de Sousa Miranda – 2014 – l2f.inesc-id.pt … it enables downstream processing methods (machine translation, spoken document retrieval, dialog systems, etc.) to recover from ASR errors, if different information sources are available … (CRF), Maximum Entropy, and HMM-based [33]. Hidden Event Language Models (HELM) … Related articles

Multi-Task CRF Model for Predicting Issue Resolution Status in Social Media based Customer Care A Agarwal, S Kataria – umiacs.umd.edu Page 1. Multi-Task CRF Model for Predicting Issue Resolution Status in Social Media based Customer Care Arvind Agarwal Palo Alto Research Center Webster, New York, USA arvind.agarwal@xerox.com Saurabh Kataria Palo …

On-line and Off-line Chinese-Portuguese Translation Service for Mobile Applications RE Banchs, A Gelbukh – Computación y Sistemas, 2014 – scielo.org.mx … 7. Och, FJ & Ney, H. (2002). Discriminative Training and Maximum Entropy Models for Statistical Machine Translation. In Proc. … His recent areas of research include Machine Translation, Information Retrieval, Cross-language Information Retrieval and Dialogue Systems. … Related articles All 17 versions

Korean anaphora recognition system to develop healthcare dialogue-type agent J Yang, Y Lee – Healthcare informatics research, 2014 – synapse.koreamed.org … Several algorithms, such as naïve Bayesian, sequential minimal optimization, and maximum entropy, were utilized to assume the parameters, and the Java … medical free-text to extract information and knowledge in an automatic manner or to design a dialogue system which can … Cited by 1 Related articles All 9 versions

Adapting dependency parsing to spontaneous speech for open domain spoken language understanding F Bechet, A Nasr, B Favre – Fifteenth Annual Conference …, 2014 – mazsola.iit.uni-miskolc.hu … work was fol- lowed by a large number of studies, with variants using other kinds of classifiers such as maximum entropy [6] or SVM [7]. /storage/raid2 Recently, in [8], a FrameNet parser was used to process spontaneous speech for the development of a Spoken Dialog System. … Related articles All 7 versions

Hypotheses ranking for robust domain classification and tracking in dialogue systems JP Robichaud, PA Crook, P Xu… – Proc. of …, 2014 – mazsola.iit.uni-miskolc.hu … Index Terms: dialogue systems, natural language understand- ing, hypothesis ranking, re-ranking, contextual domain classifi- cation, lambda rank, gradient boosted … Each hypothesis is then classi- fied using a maximum entropy classifier for its category (in- tent within a domain). … Cited by 3 Related articles All 10 versions

Using lexical, syntactic and semantic features for non-terminal grammar rule induction in spoken dialogue systems G Athanasopoulou, I Klasinas… – … (SLT), 2014 IEEE, 2014 – ieeexplore.ieee.org … 6. CONCLUSIONS We proposed a lightly supervised corpus-based algorithm for the automatic induction of non-terminal grammar rules for spoken dialogue systems. … 29, no. 4, pp. 589–637, 2003. [9] E. Charniak, “A maximum-entropy-inspired parser,” in Proceedings of 1st … Cited by 1 Related articles

Context Awareness and Personalization in Dialogue Planning RWH Fisher – 2014 – cs.cmu.edu … ing, with much of the recent work utilizing comparatively simple models such as linear support vector machines and maximum entropy classifiers [20 … it may be preferable to use inverse reinforcement learning (also called imitation learning) to train a dialogue system using human … Related articles All 2 versions

A generalized rule based tracker for dialogue state tracking K Sun, L Chen, S Zhu, K Yu – Spoken Language Technology …, 2014 – ieeexplore.ieee.org … the results of [1]. In contrast, dis- criminative state tracking models were successfully used for spoken dialogue systems [6]. The … Up to now, many discriminative statistical approaches includ- ing Maximum Entropy [12], Conditional Random Field [13], Deep Neural Network (DNN … Cited by 6 Related articles All 2 versions

Inverse reinforcement learning for micro-turn management D Kim, C Breslin, P Tsiakoulis, M Gašic… – Proceedings of the …, 2014 – 193.6.4.39 … [18] BD Ziebart, A. Maas, JA Bagnell, and AK Dey, “Maximum entropy inverse reinforcement learning,” in Proc. … [20] B. Thomson and S. Young, “Bayesian update of dialogue state: A POMDP framework for spoken dialogue systems,” Computer Speech and Language, vol. 24, no. … Cited by 3 Related articles All 11 versions

[BOOK] Natural Interaction with Robots, Knowbots and Smartphones: Putting Spoken Dialog Systems Into Practice J Mariani, S Rosset, M Garnier-Rizet, L Devillers – 2014 – books.google.com … University, and the Department of Statistical Modeling at the Institute of Statistical Mathematics (Japan), present a method for detecting invalid inputs for a spoken dialog system. … They compare two different methods, one based on SVM and the other on maximum entropy. … Cited by 1 All 3 versions

Incorporating Weak Statistics for Low-Resource Language Modeling S Novotney – 2014 – jscholarship.library.jhu.edu Page 1. Incorporating Weak Statistics for Low-Resource Language Modeling by Scott Novotney A dissertation submitted to The Johns Hopkins University in conformity with the requirements for the degree of Doctor of Philosophy. Baltimore, Maryland February, 2014 … Related articles All 2 versions

Robust dialog state tracking using delexicalised recurrent neural networks and unsupervised adaptation M Henderson, B Thomson… – … Workshop (SLT), 2014 …, 2014 – ieeexplore.ieee.org … Before being de- ployed, most dialog systems are trained for well-defined and static domains such as this … Several successful approaches for dialog state tracking have been entered and evaluated in the first two DSTCs, in- cluding maximum entropy models [3], web-style ranking … Cited by 10 Related articles All 7 versions

Modelling User Experience in Human-Robot Interactions K Jokinen, G Wilcock – Multimodal Analyses enabling Artificial Agents in …, 2014 – Springer … In spoken dialogue systems [12] this has been an important design principle discussed under the concepts of grounding and … We experimented with Weka’s logistic regression (maximum entropy model) and Support Vector Machine algorithms, but the differences between the … Related articles All 3 versions

Conversation Intention Perception Based on Knowledge Base YZ Chen, HK Li, Y Liu – Trends and Applications in Knowledge Discovery …, 2014 – Springer … The Maximum entropy classifier is adopted to classify those entries into large mount of categories using structural information since the BE pages are … Since the real conversation system likes a catch ball game, we will devote this model to build an interactive dialogue system. … Related articles All 2 versions

Better Surface Realization through Psycholinguistics R Rajkumar, M White – Language and Linguistics Compass, 2014 – Wiley Online Library … al. 2013), thereby making them more suitable for integration into real-time incremental dialog systems (Schlangen and Skantze 2011; Dethlefs et al. 2013). … 2007; Guo et al. 2008), maximum entropy models (Nakanishi et al. 2005 … Cited by 2 Related articles

Learning strategies in table tennis using inverse reinforcement learning K Muelling, A Boularias, B Mohler, B Schölkopf… – Biological …, 2014 – Springer … Ziebart et al. (2008) sug- gested an algorithm where the principle of maximum entropy was exploited. Ramachandran and Amir (2007) modeled the uncertainties involved as probabilities where the demonstra- tions are treated as evidence of the unknown reward function. … Cited by 2 Related articles All 15 versions

Semi-Supervised Learning of Statistical Models for Natural Language Understanding D Zhou, Y He – The Scientific World Journal, 2014 – hindawi.com … Linear-chain CRFs, as a discriminative probabilistic model over sequences of feature vectors and label sequences, have been widely used to model sequential data. This model is analogous to maximum entropy models for structured outputs. … Related articles All 10 versions

Automated grammatical error detection for language learners C Leacock, M Chodorow, M Gamon… – Synthesis lectures on …, 2014 – morganclaypool.com … Semantic Role Labeling Martha Palmer, Daniel Gildea, and Nianwen Xue 2010 Spoken Dialogue Systems Kristiina Jokinen and Michael McTear 2009 Introduction to Chinese Natural Language Processing Kam-Fai Wong, Wenjie Li, Ruifeng Xu, and Zheng-sheng Zhang 2009 … Cited by 19 Related articles All 4 versions

Linguistic Individuality Transformation for Spoken Language M Mizukami, G Neubig, S Sakti, T Toda, S Nakamura – isw3.naist.jp … 13. Franz Josef Och and Hermann Ney. Discriminative training and maximum entropy models for statistical machine translation. In Proc. … Con- struction and analysis of a persuasive dialogue corpus. In 5th International Workshop on Spoken Dialog Systems (IWSDS), 2014. 26. … Related articles All 4 versions

An impact analysis of features in a classification approach to irony detection in product reviews K Buschmeier, P Cimiano, R Klinger – ACL 2014, 2014 – anthology.aclweb.org … 2006.“yeah right”: Sarcasm recogni- tion for spoken dialogue systems. In Proceedings of InterSpeech, pages 1838–1841, Pittsburgh, PA, September. … Dual coordinate descent methods for logistic regression and maximum entropy. Machine Learn- ing, 85 (1–2): 41–75, October. … Cited by 4 Related articles All 11 versions

Introduction to Emotion Recognition A Konar, A Halder… – Emotion Recognition: A …, 2014 – media.johnwiley.com.au … the affective speech. They used multiple classifiers like GMM, SVM, MLP, (Multilayer Perceptron), MDT (Meta Decision Tree), and Maximum Entropy Model (MaxEnt) and got an overall accuracy of 85.79%. Kim et al. [81] have … Related articles All 3 versions

Extraction of non-taxonomic relations from texts to enrich a basic ontology M Ribeiro – 2014 – fenix.tecnico.ulisboa.pt … ern systems usually use machine learning techniques including Hidden Markov Models, Maximum Entropy, Decision Tree, Support Vector Machines, and Conditional Random Fields. … ond phase. The technique used is maximum entropy. The system is trained on the BBN …

Knowledge-based Dialog State Tracking R Kadlec, M Vodolan, J Libovicky… – … (SLT), 2014 IEEE, 2014 – ieeexplore.ieee.org … Even though the dialog state tracking is only a sub-component of a complete dialog system, it was shown that … knowledge-based, whereas the term discriminative models usually means machine-learned discriminative models, most frequently the Maximum Entropy (ME) models … Cited by 4 Related articles

Situated Incremental Natural Language Understanding using a Multimodal, Linguistically-driven Update Model C Kennington, S Kousidis, D Schlangen – Proceedings of CoLing, 2014 – aclweb.org … The final, bolded NLU frame in Figure 1 shows the addressee (in this case, the dialogue system) as the recip- ient of the request, the … Kennington et al., (2013), we use the model represented formally in Equation 2, where P(R|U) is realised using a maximum entropy classifier (ME … Cited by 2 Related articles All 6 versions

Language Technology for eGovernment–Business Cases M Henkel, E Perjons, E Sneiders, J Karlgren… – New Perspectives in …, 2014 – Springer … 2000) Acomb, K., Bloom, J., Dayanidhi, K., Hunter, P., Krogh, P., Levin, E., Pieraccini, R.: Technical support dialog systems: Issues, problems … Ramabhadran, B., Povey, D., Mangu, L., Kingsbury, B.: Automated quality monitoring in the call center with asr and maximum entropy. … Cited by 1 Related articles All 5 versions

Parsing, Expansion And Merging Of Dependency Treebank P Kosaraju – 2014 – web2py.iiit.ac.in … application to several natural language processing applications like machine translation, word sense disambiguation, information retrieval, dialogue systems etc. … categories With the annotated data we built a semantic tagger using Maximum Entropy Model.(A.Ratnaparkhi, … Related articles

LRC Best Thesis Award Winner 2014 JG Rubio – localisation.ie … 16 1.3.3 Maximum Entropy Model … The goal of NLP is to accomplish human-like language processing for a broad range of tasks or ap- plications, for instance information retrieval, information extraction, question answering, summarization, machine translation, dialog systems, … Related articles

On the effective deployment of current machine translation technology J González Rubio – 2014 – riunet.upv.es … 16 1.3.3 Maximum Entropy Model … The goal of NLP is to accomplish human-like language processing for a broad range of tasks or ap- plications, for instance information retrieval, information extraction, question answering, summarization, machine translation, dialog systems, … Cited by 1 Related articles All 2 versions

On the Effective Deployment of Current Machine Translation Technology JG Rubio – prhlt.upv.es … 16 1.3.3 Maximum Entropy Model … The goal of NLP is to accomplish human-like language processing for a broad range of tasks or ap- plications, for instance information retrieval, information extraction, question answering, summarization, machine translation, dialog systems, … Related articles

Tweety-a comprehensive collection of java libraries for logical aspects of artificial intelligence and knowledge representation M Thimm – Proceedings of the 14th International Conference on …, 2014 – colonyofmalice.de … Besides a naive implementation of probabilistic reasoning based on the principle of maximum entropy (Paris 1994) this library also contains several … On top of this implementation of the actual dialogue system a sim- ulation framework was implemented that allows the (ran- dom … Cited by 10 Related articles All 3 versions

[BOOK] Speech and language processing D Jurafsky, JH Martin – 2014 – cs.colorado.edu … A similar spoken dialogue system has been deployed by as- tronauts on the International Space Station. … Techniques such as support vector machines (Boser et al., 1992; Vapnik, 1995), maximum entropy techniques and their equiva- lent formulation as multinomial logistic … Cited by 25 Related articles All 20 versions

SimSensei Kiosk: A virtual human interviewer for healthcare decision support D DeVault, R Artstein, G Benn, T Dey, E Fast… – Proceedings of the …, 2014 – dl.acm.org … Keywords virtual humans; dialogue systems; nonverbal behavior … This maximum entropy clas- sifier is trained using face-to-face and Wizard-of-Oz data to detect various forms of positive responses that serve to assert closeness (a domain-specific dialogue act). … Cited by 23 Related articles All 9 versions

Personal knowledge graph population from user utterances in conversational understanding X Li, G Tur, D Hakkani-Tur, Q Li – … Technology Workshop (SLT), …, 2014 – ieeexplore.ieee.org … User Speech Fig. 2. Framework of Personal Knowledge Graph Construction multi-turn spoken dialog systems. … Page 4. the first order Markov constraint on the model topology. Similar to maximum entropy models, in this model, the conditional probability, p(Y |X) is defined as [51]: … Cited by 1 Related articles All 6 versions

On the use of a fuzzy classifier to speed up the Sp ToBI labeling of the Glissando Spanish corpus D Escudero-Mancebo, L Aguilar-Cuevas… – lrec-conf.org … can help resolve lexical disambigua- tion. In Dialog Systems, the identification of focalized or highlighted items can be crucial in interpreting the mes- sage from a semantic or pragmatic perspective. In Text to Speech, the correspondence … Related articles All 3 versions

A pragmatic tutorial dialogue system: design, implementation and evaluation in a health sciences domain JA McDonald – 2014 – ourarchive.otago.ac.nz … gives a response. Such systems, which employ natural language as their interface, are called dialogue-based ITS or tutorial dialogue systems; they … In this thesis, a ratio- nale is presented for revisiting tutorial dialogue systems in the context of large-class teaching. … Related articles All 2 versions

Understanding questions and finding answers: semantic relation annotation to compute the Expected Answer Type V Petukhova – Proceedings 10th Joint ISO-ACL SIGSEM Workshop on …, 2014 – lrec-conf.org … The system has all components that any traditional dialogue system has: Automatic Speech Recog- nition (ASR) and Speech Generation (eg TTS) modules … The reported in (Huang et al., 2008) the accuracies of SVM and Maximum Entropy (ME) classifiers were 89.2% and 89.0 … Related articles All 3 versions

KIT-Conferences PI Lichtblau – 2014 – isl.anthropomatik.kit.edu … Proceedings of the International Workshop for Spoken Language Translation (IWSLT 2013). Heidelberg, December 5-6, 2013. Maximum Entropy Language Modeling for Russian ASR, Evgeniy Shin, Sebastian Stüker, Kevin Kilgour, Christian Fügen, Alex Waibel. … All 2 versions

Conversational Implicatures L Benotti, P Blackburn – compute.dtu.dk … The characteristics and functions of clarifica- tion subdialogues have been deeply studied by dialogue system researchers [Gabs- dil, 2003; Purver … A similar approach is taken by [DeVault and Stone, 2009] who instead of POMDPs use Maximum Entropy models over abductive … Related articles

Structured Adaptive Regularization of Weight Vectors for a Robust Grapheme-to-Phoneme Conversion Model S SAKTI, G NEUBIG, T Tomoki… – … on Information and …, 2014 – search.ieice.org … classification tasks for natural lan- guage processing (NLP), it has been shown that because rare features are often very informative in NLP, online learning algorithms that have that property have been shown to out- perform batch learning of maximum entropy classifiers and … Related articles All 6 versions

Context and Implicature L Benotti, P Blackburn – Context in Computing, 2014 – Springer … The characteristics and func- tions of clarification subdialogues have been deeply studied by dialogue system researchers (Gabsdil 2003; Purver 2004 … A similar approach is taken by DeVault and Stone (2009) who instead of POMDPs use Maximum Entropy models over … Related articles All 3 versions

Modeling language with structured penalties AK Nelakanti – 2014 – tel.archives-ouvertes.fr … It has applications spanning across various domains, such as dialogue systems, text generation and machine translation among … generative models (Dirichlet, Pitman-Yor and hierarchical Pitman- Yor processes), discriminative models (maximum entropy, conditional random … Related articles All 6 versions

Maximum likelihood inverse reinforcement learning MC Vroman – 2014 – rucore.libraries.rutgers.edu Page 1. MAXIMUM LIKELIHOOD INVERSE REINFORCEMENT LEARNING by MONICA C. VROMAN A dissertation submitted to the Graduate School—New Brunswick Rutgers, The State University of New Jersey In partial fulfillment of the requirements For the degree of … Related articles All 4 versions

Translation rescoring through recurrent neural network language models Á PERIS ABRIL – 2014 – riunet.upv.es … such as automatic summarization, discourse analysis, machine translation, morphological segmentation, natural language generation and understanding, speech recognition, topic segmentation and recognition, information retrieval, dialog systems, question answering, etc. … Related articles All 2 versions

Theoretical analysis of diversity in an ensemble of automatic speech recognition systems K Audhkhasi, AM Zavou, PG Georgiou… – Audio, Speech, and …, 2014 – ieeexplore.ieee.org Page 1. 2329-9290 (c) 2013 IEEE. Personal use is permitted, but republication/ redistribution requires IEEE permission. See http://www.ieee.org/ publications_standards/publications/rights/index.html for more information. This … Cited by 9 Related articles All 4 versions

Integration of complex language models in ASR and LU systems R Justo, MI Torres – Pattern Analysis and Applications, 2014 – Springer Page 1. THEORETICAL ADVANCES Integration of complex language models in ASR and LU systems Raquel Justo • M. Inés Torres Received: 22 June 2012 / Accepted: 16 November 2014 © Springer-Verlag London 2014 Abstract … Related articles

Statistical dialog management for health interventions U Yasavur – 2014 – digitalcommons.fiu.edu … Research endeavors on spoken dialogue systems in the 1990s and 2000s have led to the deployment of commercial spoken dialogue systems (SDS) in microdomains … based on the brief intervention counseling style via spoken dialogue systems. … Related articles All 4 versions

Language independent search in MediaEval’s Spoken Web Search task F Metze, X Anguera, E Barnard, M Davel… – Computer Speech & …, 2014 – Elsevier … cases. These include searching large archives of audio-visual material, dialog systems for access to personal information and (mobile) web search, as well as applications in language learning and pronunciation training. A … Cited by 12 Related articles All 12 versions

Domain Cartridge: Unsupervised Framework for Shallow Domain Ontology Construction from Corpus S Mukherjee, J Ajmera, S Joshi – Proceedings of the 23rd ACM …, 2014 – dl.acm.org … The discovered relations can be used for query expan- sion (eg by considering Synonyms along with the original query), interactive dialogue systems (eg for the user query “the battery of my device depletes very fast”, a DRD inference that ‘battery’ is a Feature-Of ‘phone’ as well … Related articles All 5 versions

A unified framework for translation and understanding allowing discriminative joint decoding for multilingual speech semantic interpretation B Jabaian, F Lefèvre, L Besacier – Computer Speech & Language, 2014 – Elsevier … The framework can be generalized to other components of a dialogue system. Abstract. … It has been shown that using lattices as SLU input decreased the classification error rate in the case of the AT&T’s “How May I Help You” dialogue system. In parallel Servan et al. … Cited by 1 Related articles All 5 versions

Large vocabulary Russian speech recognition using syntactico-statistical language modeling A Karpov, K Markov, I Kipyatkova, D Vazhenina… – Speech …, 2014 – Elsevier … models sure as the Maximum Entropy model and Conditional Random Fields are used. In (Bechet and Nasr, 2009), a syntactic parser for spontaneous speech recognition outputs is used for identification of verbal sub-categorization frames for dialogue systems and spoken … Cited by 19 Related articles All 6 versions

Dialogue POMDP components (Part II): learning the reward function H Chinaei, B Chaib-Draa – International Journal of Speech Technology, 2014 – Springer … This frame- work has been extensively used to model the uncertainty of SDSs (spoken dialogue systems) (Roy et al. 2000; Zhang et al. 2001a,b; Williams and Young 2007; Thomson 2009; Gašic 2011; Pinault and Lefèvre 2011). … Related articles All 5 versions

Realistic Dialogue Engine for Video Games CM Rose – 2014 – ir.lib.uwo.ca … after a while due to the highly deterministic nature of their responses. The types of dialogue systems used in video games could be classified into three main types: cutscenes, branching dialogue trees, and simple natural language processing. Each … Cited by 1 Related articles

Automatic scoring for answers to Arabic test questions WH Gomaa, AA Fahmy – Computer Speech & Language, 2014 – Elsevier … It has been used in the tutorial dialog system Why2-Atlas (VanLehn et al., 2002 … C-Rater, the student answers are parsed, to extract a predicate argument structure that is then categorized as absent, present, or negated for each concept, using a maximum entropy-based matching … Cited by 4 Related articles All 4 versions

Dialog-Based Wayfinding Using Intrinsic and Extrinsic Attributes of Landmarks A Khan – 2014 – gautam5.cse.iitk.ac.in … Page 22. 7 Furthermore, one major advantage of a dialog system is that context-identification can be implemented by studying the response of the user to the instructions/questions. … to resolve ambiguities along the path. Owing to the benefits of a dialog system to …

Stochastic language generation in dialogue using factored language models F Mairesse, S Young – Computational Linguistics, 2014 – MIT Press … HALogen is thus domain-independent, and it was successfully ported to a specific dialogue system domain (Chambers and Allen 2004). … Bagel aims to produce natural utterances within a large dialogue system domain while minimizing the overall development effort. … Cited by 7 Related articles All 14 versions

Integration of Multiple Cues for Speech Activity Detection and Word Segmentation P Mikias – honda-ri.de Page 1. Diploma Thesis Integration of Multiple Cues for Speech Activity Detection and Word Segmentation Paschalis Mikias Advised by: Dr.-Ing. Martin Heckmann* Prof. Dr.-Ing. Dorothea Kolossa Dr.-Ing. Steffen Zeiler * Honda Research Institute Europe … Related articles

Three-dimensional knowledge representation using extended structure graph grammars C Du Plessis – 2014 – 152.106.6.200 … The Beilstein database is searchable with STN International and the Dialog system.[ALL89, BEI92]. The Beilstein database also provides offline structure … The Heilbron Dictionary of organic chemistry (DOC) and related databases is searchable on the Dialog system. Page 27. … Related articles All 4 versions

Algorithm for Analysis and Translation of Sentence Phrases R Lacko – is.muni.cz … Voice XML is a World Wide Web Consortium (W3C) standard for specifying dialogue systems, usually for automated customer services. … to the Stanford NLP Group, OpenNLP also contains tools like tokenizer, POS tagger and parser, but it uses maximum entropy and perceptron … Related articles

Deep stochastic sentence generation S Mille – taln.upf.edu Page 1. Deep stochastic sentence generation Resources and strategies Simon Mille TESI DOCTORAL UPF / 2014 Director de la tesi Prof. Leo Wanner Department of Information and Communication Technologies Page 2. By … Related articles

Deep stochastic sentence generation: resources and strategies S Mille – 2014 – tdx.cat Page 1. Deep stochastic sentence generation Resources and strategies Simon Mille TESI DOCTORAL UPF / 2014 Director de la tesi Prof. Leo Wanner Department of Information and Communication Technologies Page 2. By … Cited by 1 Related articles All 5 versions

CASAM: collaborative human-machine annotation of multimedia RJ Hendley, R Beale, CP Bowers… – Multimedia tools and …, 2014 – Springer Page 1. CASAM: collaborative human-machine annotation of multimedia Robert J. Hendley & Russell Beale & Chris P. Bowers & Christos Georgousopoulos & Charalampos Vassiliou & Petridis Sergios & Ralf Moeller & Eric Karstens & Dimitris Spiliotopoulos … Cited by 2 Related articles All 10 versions

Referring Expression Generation Towards Mediating Shared Perceptual Basis In Situated Dialogue R Fang – 2014 – cse.msu.edu Page 1. REFERRING EXPRESSION GENERATION TOWARDS MEDIATING SHARED PERCEPTUAL BASIS IN SITUATED DIALOGUE By Rui Fang A DISSERTATION Submitted to Michigan State University in partial fulfillment of the requirements for the degree of …

A System for Recognizing Natural Spelling of English Words L Czech, A Waibel, S Stüker, DIT Köhler – Links, 2014 – isl.anthropomatik.kit.edu … There are plenty of use cases for such a system, including out-of-vocabulary learning, error recovery and dialog systems, for example in … for the extraction of named entities are presented, with a focus on conversational telephone speech by using maximum entropy clustering on … Related articles All 2 versions

Foundations and Trends in Signal Processing L Deng, Y Dong – Signal Processing, 2014 – research.microsoft.com Page 1. the essence of knowledge FnT SIG 7:3-4 Deep Learning; Methods and Applications Li Deng and Dong Y u Foundations and Trends® in Signal Processing 7:3-4 Deep Learning Methods and Applications Li Deng and Dong Yu now now Page 2. 7.1. … Cited by 2 Related articles All 12 versions