Linear Classifiers & Dialog Systems


Linear Classifiers & Dialog Systems
Linear Classifier 


Detection of task-incomplete dialogs based on utterance-and-behavior tag N-gram for spoken dialog systems [PDF] from nagoya-u.ac.jp S Hara, N Kitaoka… – … in Florence, Italy, on August 27 …, 2011 – ir.nul.nagoya-u.ac.jp … The aim of this study is to construct a model to detect “task incomplete” dialogs in spoken dialog systems using real-world data. … value of the classification accuracy, whether COMPLETE or INCOMPLETE, is shown in Figure 2. The re- sult of the SVM-Linear classifier indicated a … Related articles – All 4 versions

[PDF] Very short utterances in conversation [PDF] from uba.ar J Edlund, M Heldner, S Al Moubayed… – Proceedings of …, 2010 – publicaciones.dc.uba.ar … The second contribution of this paper is to propose a method to better deal with turntaking in spoken dialogue systems by continuously detecting VSUS and … K-NN is a non-parametric non-linear classifier which does not build a model for training data, but builds a local model for … Cited by 7 – Related articles – View as HTML – All 7 versions

Multimodal person independent recognition of workload related biosignal patterns [PDF] from kit.edu J Jarvis, F Putze, D Heger… – Proceedings of the 13th …, 2011 – dl.acm.org … has a negative impact on driving performance and therefore seriously affects safety in every day traffic (eg [9]). In-vehicle dialog systems designed to … As a straight forward approach we evaluate a linear classifier on a one-dimensional feature space for each feature and use the … Related articles – All 2 versions

Linear classifier with reject option for the detection of vocal fold paralysis and vocal fold edema [PDF] from hindawi.com C Kotropoulos… – EURASIP Journal on Advances in Signal …, 2009 – dl.acm.org … Linear Classifier with Reject Option for the Detection of … The receiver operating characteristic curve of the linear classifier, that stems from the Bayes classifier when Gaussian class conditional probability density functions with equal covariance matrices are assumed, is derived. … Related articles – All 6 versions

Text categorization methods for automatic estimation of verbal intelligence F Fernández-Martínez, K Zablotskaya… – Expert Systems with …, 2012 – Elsevier … Fig. 2 explains the adaptation process of spoken dialog systems based on verbal intelligence estimation in more detail. When … large). Rocchio and Naive Bayes are linear classifiers whereas kNN is an example of a non-linear one. …

The Automatic Recognition of Emotions in Speech [PDF] from kuleuven.be A Batliner, B Schuller, D Seppi, S Steidl… – Emotion-Oriented …, 2011 – Springer Page 1. The Automatic Recognition of Emotions in Speech Anton Batliner, Björn Schuller, Dino Seppi, Stefan Steidl, Laurence Devillers, Laurence Vidrascu, Thurid Vogt, Vered Aharonson, and Noam Amir Abstract In this chapter … Cited by 2 – Related articles – All 5 versions

[CITATION] Linear Classifier with Reject Option for the Detection of Vocal Fold Paralysis and Vocal Fold Edema K Constantine, A Gonzalo R – … on Advances in …, 2009 – Hindawi Publishing Corporation Related articles – All 3 versions

Modeling and predicting quality in spoken human-computer interaction [PDF] from aclweb.org A Schmitt, B Schatz… – Proceedings of the SIGDIAL 2011 …, 2011 – dl.acm.org … Modeling and Predicting Quality in Spoken Human-Computer Interaction Alexander Schmitt, Benjamin Schatz and Wolfgang Minker Dialogue Systems Research Group Institute for Information Technology Ulm University, Germany … Cited by 1 – Related articles – All 8 versions

[PDF] Applying Multiclass Bandit algorithms to call-type classification [PDF] from univ-mrs.fr L Ralaivola, B Favre, P Gotab… – … of IEEE Automatic …, 2011 – pageperso.lif.univ-mrs.fr … We prove that it is possible to learn linear classifiers in this setting, by estimating adequate expecta- tions inspired by the Multiclass Bandit … The goal of this paper is to apply the multiclass bandit paradigm to Spoken Dialog Systems (SDS) through the online adaptation of Call … Related articles – View as HTML

Emotion Recognition of EMG Based on Improved LM BP Neural Network and SVM [PDF] from academypublisher.com S Yang… – Journal of Software, 2011 – academypublisher.com … A two-class classification problem is assumed for simplification. The problem of finding a linear classifier for given data points with a known class label can be described as a problem of finding a separating hyper plane T xb ? + that satisfies: … Related articles – All 5 versions

Automatic stress detection in emergency (telephone) calls I Lefter, LJM Rothkrantz, DA Van Leeuwen… – International Journal of …, 2011 – Inderscience … used with the help of Wizard of Oz (WoZ) scenarios (Steidl et al., 2008) or computer-based dialogue systems (Zeng et al … Experiments using linear classifiers (Krajewski and Kroger, 2007), Gaussian mixture models (GMM) (Truong and van Leeuwen, 2007), neural networks (NN … Cited by 4 – Related articles – All 6 versions

A Regularized Maximum Figure-of-Merit (rMFoM) Approach to Supervised and Semi-supervised Learning C Ma… – Audio, Speech, and Language Processing, …, 2011 – ieeexplore.ieee.org Page 1. Copyright (c) 2010 IEEE. Personal use is permitted. For any other purposes, Permission must be obtained from the IEEE by emailing pubs-permissions@ieee.org. This article has been accepted for publication in a future issue of this journal, but has not been fully edited. … Related articles – All 2 versions

Survey on speech emotion recognition: Features, classification schemes, and databases M El Ayadi, MS Kamel… – Pattern Recognition, 2011 – Elsevier Cited by 23 – Related articles – All 4 versions

[PDF] A Preliminary Study of Prosody Based Detection of Questions in Arabic Speech Monologues [PDF] from kfupm.edu.sa O Khan, WG Al-Khatib… – Arabian Journal for Science …, 2010 – ajse.kfupm.edu.sa … vector machines (SVM), artificial neural networks such as MLP-NN or RBF, and simpler classifiers such as k-NN, linear classifiers like LDC … This research was carried out on a corpus of spoken tutor dialogues for the development of intelligent tutoring spoken dialogue systems. … Related articles – View as HTML

Recognising realistic emotions and affect in speech: State of the art and lessons learnt from the first challenge [PDF] from kuleuven.be B Schuller, A Batliner, S Steidl… – Speech Communication, 2011 – Elsevier … A natural extension of LDCs are Support Vector Machines (SVM): If the input data have not been previously (implicitly) linearly transformed, which may have increased or decreased the number of features, and if the linear classifier obeys a maximum-margin fitting criterion, then … Cited by 33 – Related articles – All 9 versions

Multimodal fusion for multimedia analysis: a survey [PDF] from nus.edu.sg PK Atrey, MA Hossain, A El Saddik… – Multimedia Systems, 2010 – Springer … However, in general, these rules are domain specific and defining the rules requires proper knowledge of the domain. This fusion method is widely used in the domain of multimodal dialog systems and sports video analysis. 3.2 Classification-based fusion methods … Cited by 28 – Related articles – All 9 versions

Model parameters selection for SVM classification using Particle Swarm Optimization M Hric, M Chmulik… – … ), 2011 21st International …, 2011 – ieeexplore.ieee.org … It is non- probabilistic binary linear classifier. … pdf. [13] JUHÁR, J., ONDÁŠ, S., CIŽMÁR, A., JARINA, R., RUSKO, M., ROZINAJ, G. Development of Slovak GALAXY / VoiceXML based spoken language dialogue system to retrieve information from the internet. … Cited by 1 – Related articles

Improved Malay Vowel Feature Extraction Method Based on First and Second Formants AMY Shahrul, F Siraj, S Yaacob… – Computational …, 2010 – ieeexplore.ieee.org … trained Neural Network (LM), K- Nearest Neighbours (KNN), Multinomial Logistic Regression (MLR) and a linear classifier of Linear … workshop on Perception and Interactive Technologies for Speech- Based Systems: Perception in Multimodal Dialogue Systems Kloster Irsee … Related articles – All 3 versions

Short-term emotion assessment in a recall paradigm [PDF] from im2.ch G Chanel, JJM Kierkels, M Soleymani… – International Journal of …, 2009 – Elsevier Cited by 52 – Related articles – All 9 versions

Inductive and example-based learning for text classification [PDF] from microsoft.com YY Wang, X Li… – Ninth Annual Conference of the …, 2008 – isca-speech.org … Review data from the web can be used to train a sentiment classifier for spoken dialog systems like Voice-Rate [2], which allows callers to … factor of | |, which is also absorbed by K( ). This equation suggests that the vector space model be viewed as a linear classifier, where each … Cited by 1 – Related articles – All 6 versions

Valence-arousal evaluation using physiological signals in an emotion recall paradigm G Chanel, K Ansari-Asl… – Systems, Man and …, 2007 – ieeexplore.ieee.org … Two linear classifiers, a Linear Discriminant Analysis (LDA, see discriminant functions for the normal density in [22]) and a Support Vector … and J. Williams, “Emotion Recognition Using Bio-Sensors: First Step Toward an Automatic System,” Affective Dialog Systems: Tutorial And … Cited by 38 – Related articles – All 3 versions

Speech Recognition by Machine, A Review [PDF] from arxiv.org MA Anusuya… – Arxiv preprint arXiv:1001.2267, 2010 – arxiv.org … 6, No. 3, 2009 length data has to be transformed to fixed length vectors before SVMs can be used. It is a generalized linear classifier with maximum-margin fitting functions. This fitting function provides regularization which helps the classifier generalized better. … Cited by 20 – Related articles – All 7 versions

Scaling large margin classifiers for spoken language understanding [PDF] from att.com P Haffner – Speech Communication, 2006 – Elsevier … needed, the robustness of learning convergence, and the quality of generalization make these regularized linear classifiers ideal candidates … for topic classification and spoken language understanding, and are currently deployed in the AT&T VoiceTone (r) spoken dialog system. … Cited by 31 – Related articles – All 4 versions

Named Entity Recognition F Béchet – Spoken Language Understanding, 2011 – Wiley Online Library … In NER from speech, the SLU module of Spoken Dialogue Systems (SDS) often relies on the detection and extraction of application-specific entities such as telephone numbers or customers’ names. … 10.1.3.2 Spoken Dialogue Systems … Cited by 1 – Related articles – All 2 versions

Robust acoustic and semantic modeling in a telephone-based spoken dialog system [PDF] from uni-halle.de KT Mengistu – 2009 – deposit.ddb.de Robust Acoustic and Semantic Modeling in a Telephone-based Spoken Dialog System Kinfe Tadesse Mengistu (M. Sc.) … alog system. We also show the feasibility of building a robust, multi-domain telephone-based spoken dialog system framework that can be seamlessly used … Related articles – Library Search – All 5 versions

To talk or not to talk with a computer [PDF] from uni-erlangen.de A Batliner, C Hacker… – Journal on Multimodal User Interfaces, 2008 – Springer … Nowadays, the dialogue partner does not need to be a hu- man being but can be an automatic dialogue system as well. … 3 Outline For automatic dialogue systems, a good classification per- formance is most important; the way how to achieve this could be treated as a black-box. … Cited by 2 – Related articles – All 3 versions

[PDF] Towards detection of interest during movie scenes [PDF] from lifl.fr JJM Kierkels… – PetaMedia Workshop on Implicit, Human-Centered …, 2008 – lifl.fr … By using features derived from standard physiological signals and using linear classifiers it was first shown that it is possible to … P, and Williams J, “Emotion Recognition Using Bio-sensors: First Steps towards an Automatic System,” in Affective Dialogue Systems Springer Berlin … Cited by 2 – Related articles – View as HTML

[CITATION] Voice Driven Multimedia Interactive System with Ubiquitous Sound Recognition for Digital Home Application HT Kao – 2008 Cited by 1 – Related articles

A Flexible Classifier Design Framework Based on Multiobjective Programming [PDF] from berkeley.edu S Yaman… – Audio, Speech, and Language Processing …, 2008 – ieeexplore.ieee.org … recall, or F1 measure. Linear classifiers are discriminatively trained using the MCE framework for the task of automatic image annotation in [15], and for the task of automatic language identification in [16]. In [17], discriminant … Cited by 5 – Related articles – BL Direct – All 6 versions

Round-robin duel discriminative language models in one-pass decoding with on-the-fly error correction [PDF] from 140.124.72.88 T Oba, T Hori, A Ito… – Acoustics, Speech and …, 2011 – ieeexplore.ieee.org … In online applications such as spoken dialog systems, speech recognizers need to work in real time and with low latency … Shinji Watanabe, Takaaki Hori, and Atsushi Nakamura, “Large vocabu- lary continuous speech recognition using WFST-based linear classifier for structured … Related articles – All 3 versions

Learning to rank for information retrieval and natural language processing H Li – Synthesis Lectures on Human Language Technologies, 2011 – morganclaypool.com … 2010 Page 5. iii 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 Introduction … Cited by 11 – Related articles – Library Search – All 3 versions

Automatic Recognition of Boredom in Video Games Using Novel Biosignal Moment-Based Features [HTML] from computer.org D Giakoumis, D Tzovaras… – Affective Computing, …, 2011 – ieeexplore.ieee.org Page 1. IEEE TRANSACTIONS ON AFFECTIVE COMPUTING, MANUSCRIPT ID 1 Automatic Recognition of Boredom in Video Games using novel Biosignal Moment-based Features Dimitris Giakoumis, Dimitrios Tzovaras, Konstantinos Moustakas, George Hassapis … Cited by 3 – Related articles – All 10 versions

New Churn Prediction Strategies in the Telecom Industry D Ruta, C Adl… – Intelligent data analysis: developing …, 2009 – books.google.com Page 241. Chapter XIII New Churn Prediction Strategies in Telecom Industry Dymitr Ruta BT Group, Research and Venturing, UK Christoph Adl BT Group, Research and Venturing, UK Detlef Nauck BT Group, Research and … All 7 versions

Experiments on decision fusion for driver recognition [PDF] from sabanciuniv.edu H Erdogan, A Ereil… – Advances for In-Vehicle and Mobile Systems, 2007 – Springer … Acoustic Information Research (ClAIR) at Nagoya University has built a multi-modal corpus inside a vehicle, where each driver was required to carry out conversations with three different dialog systems while driving [2 … Fisher combiner (Fisher): A linear classifier that minimizes … Cited by 5 – Related articles – All 5 versions

Review: Personalizing recommendations for tourists [PDF] from 140.134.131.64 K Kabassi – Telematics and Informatics, 2010 – dl.acm.org … International Journal of Man-Machine Studies. v18. 199-214. Rich, 1989. Stereotypes and user modeling. In: Kobsa, A., Wahlster, W. (Eds.), User Models in Dialog Systems, Springer-Verlag, Heidelberg. pp. 199-214. Rich, 1999. … Cited by 25 – Related articles – All 7 versions

Gesture Salience as a Hidden Variable for Coreference Resolution and Keyframe Extraction [PDF] from mit.edu J Eisenstein, R Barzilay… – Journal of Artificial Intelligence …, 2008 – aaai.org … supplement natural language pro- cessing. Much of this research has taken place in the context of spoken language dialogue systems incorporating pen gestures and automatically recognized speech. An early exam- ple of … Cited by 3 – Related articles – All 18 versions

Private emotions versus social interaction: a data-driven approach towards analysing emotion in speech A Batliner, S Steidl, C Hacker… – User Modeling and User-Adapted …, 2008 – Springer Page 1. User Model User-Adap Inter (2008) 18:175-206 DOI 10.1007/s11257-007-9039-4 ORIGINAL PAPER Private emotions versus social interaction: a data-driven approach towards analysing emotion in speech Anton Batliner · Stefan Steidl · Christian Hacker · Elmar Nöth … Cited by 43 – Related articles – BL Direct – All 3 versions

Speech, Text and Braille Conversion Technology R Hoffmann – Assistive Technology for Visually Impaired and Blind …, 2008 – Springer … There are a number of different methods for partitioning the feature space, of which three will now be discussed briefly. • Linear classifier. A linear classifier separates the classes by planes (or lines in the two-dimensional example). … Related articles – Library Search

A vector space modeling approach to spoken language identification [PDF] from technion.ac.il H Li, B Ma… – Audio, Speech, and Language …, 2007 – ieeexplore.ieee.org Page 1. IEEE TRANSACTIONS ON AUDIO, SPEECH, AND LANGUAGE PROCESSING, VOL. 15, NO. 1, JANUARY 2007 271 A Vector Space Modeling Approach to Spoken Language Identification Haizhou Li, Senior Member … Cited by 110 – Related articles – BL Direct – All 12 versions

Contextual recognition of head gestures [PDF] from psu.edu LP Morency, C Sidner, C Lee… – Proceedings of the 7th …, 2005 – dl.acm.org … We use a multi-class SVM for the integrator since experimen- tally it gave us better performance than a linear classifier or simple … J. van Kuppevelt, L. Dybkjaer, and N. Bernsen (eds.), Natural, Intelligent and Effective Interaction with Multimodal Dialogue Systems, chapter Social … Cited by 77 – Related articles – All 20 versions

Spoken Language Understanding: from Spoken Utterances to Semantic Structures [PDF] from unitn.it M Dinarelli – 2010 – eprints-phd.biblio.unitn.it … database. Following the ATIS project, DARPA Communicator project aimed at building a spoken dialog system automatically providing information on flights and travel reservation. … 104 7 SLU in Spoken Dialog Systems 109 7.1 Dialogue System Architecture . . . . . … Cited by 1 – Related articles

A detection-based pattern recognition framework and its applications [PDF] from gatech.edu C Ma – 2010 – smartech.gatech.edu Page 1. A DETECTION-BASED PATTERN RECOGNITION FRAMEWORK AND ITS APPLICATIONS A Thesis Presented to The Academic Faculty by Chengyuan Ma In Partial Fulfillment of the Requirements for the Degree Doctor … Cited by 2 – Related articles – Library Search – All 6 versions

The role of context in head gesture recognition [PDF] from pitt.edu L Morency, C Sidner, C Lee… – PROCEEDINGS OF THE …, 2006 – aaai.org … The following section describes how we can automati- cally extract lexical, punctuation and timing features from the dialog system. … We use a multi-class SVM for the integrator since experi- mentally it gave us better performance than a linear classifier or simple thresholding. … Cited by 8 – Related articles – BL Direct – All 11 versions

Medical data mining: Improving information accessibility using online patient drug reviews [PDF] from 18.7.29.232 S Seneff, YA Li – 2011 – 18.7.29.232 … 26 2.2.1 Dialogue Systems….. 26 … and term identification. We also explore the degree to which statistical methods such as co-occurrence mea- sures, linear classifiers, and topic models can be used to extract summary information … Cited by 2 – Related articles – All 3 versions

Automatic Understanding of ATC Speech: Study of Prospectives and Field Experiments for Several Controller Positions [PDF] from upm.es JM Pardo, J Ferreiros, F Fernandez… – Aerospace and …, 2011 – ieeexplore.ieee.org … identification cannot be corrected later in the system. In [38] a full recognizer is also proposed and the recognizer scores are normalized and compared with a linear classifier. Another typical approach seen in the literature is … Cited by 2 – Related articles – All 4 versions

[BOOK] How does real affect affect affect recognition in speech? [PDF] from utwente.nl KP Truong – 2009 – doc.utwente.nl … as being in stress. Interaction with machines, robots or spoken dialog systems in call centers, will feel much more natural and will be much more effective if human emotions can be recognized. Some research communities aim … Cited by 1 – Related articles – Library Search – All 9 versions

Rational kernels: Theory and algorithms [PDF] from psu.edu C Cortes, P Haffner… – The Journal of Machine Learning …, 2004 – dl.acm.org … of general families of PDS rational kernels that are useful in many of these applications and report the result of our experiments illustrating the use of rational kernels in several difficult large-vocabulary spoken-dialog classification tasks based on deployed spoken-dialog systems … Cited by 101 – Related articles – All 29 versions

[PDF] Automatic Language Identification [PDF] from limsi.fr M Adda-Decker – 2008 – limsi.fr Page 1. Automatic Language Identification Martine Adda-Decker July 20, 2008 Page 2. 2 Page 3. 8.1. Introduction When listening to our native language we, speech and hearing enabled humans, immediately identify the language being spoken. … Related articles – View as HTML – All 2 versions

[PDF] Privacy-Preserving Machine Learning for Speech Processing [PDF] from cmu.edu MA Pathak – 2012 – cs.cmu.edu … Bhiksha is truly a visionary, and has made many contributions to multiple research areas: speech and signal processing, machine learning, privacy, optimization, dialog systems, knowledge representation, to name a few – he is symbolic of the interdisciplinary culture that LTI … View as HTML

[PDF] Machine Learning in Speech and Language Processing [PDF] from att.com JA Bilmes… – Proc. ICASSP, 2005 – research.att.com … Input Space f Preprocessing Linear Classifier Avoid preprocessing: Work in input space? Expensive … Making kernels 14 Page 15. Example with Polynomial Kernels Imagine a linear classifier over all pairs of input features: ” d(d + 1)/2 features. ” Dot product in Hilbert space: O( … Cited by 3 – Related articles – View as HTML – All 4 versions

Silence/Speech detection method based on set of decision graphs [PDF] from zcu.cz J Trmal, J Zelinka, J Vanek… – Text, Speech and Dialogue, 2006 – Springer … i=1 a1,i · xi + a0 = 0, (4) where ai, j and a0 are parameters of the hyper-plane. The third sub-classifier is a linear classifier which is the consequence of the equality of the covariance matrices. … In the future, we intend to develop a robust real-time SSD for dialog systems. … Related articles – BL Direct – All 5 versions

[PDF] The design and development of a system for controlling automotive functions using speech recognition [PDF] from utm.my A Ahmad – 2006 – eprints.utm.my … 22 2.5 Classic example of a linear classifier 22 2.6 Basic idea behind SVM 23 … order to 1000-3000 words, which are typical sizes for current research-based spoken dialogue systems. Some applications can make every restrictive assumption possible. For instance, … Related articles – View as HTML

A Study of recent classification algorithms and a novel approach for biosignal data classification [PDF] from rit.edu E Cinar – 2010 – ritdml.rit.edu Page 1. A STUDY OF RECENT CLASSIFICATION ALGORITHMS AND A NOVEL APPROACH FOR BIOSIGNAL DATA CLASSIFICATION By EYUP CINAR Thesis Submitted to the Faculty of Rochester Institute of Technology in partial fulfillment of the … Related articles – Library Search

Towards real life applications in emotion recognition D Küstner, R Tato, T Kemp… – Affective Dialogue Systems, 2004 – Springer … Different classifiers have been tested: linear classifiers, neural networks and support vector machines (C-SVM with RBF-kernel). … non-normalized energy for increas- ing intensity (AIBO corpus, clean speech, three classes (activation dimension)), linear classifier … Cited by 10 – Related articles – All 4 versions

[PDF] Gesture in automatic discourse processing [PDF] from psu.edu J Eisenstein – 2008 – Citeseer … gesture or other non-verbal features. Section 2.2 describes multimodal interfaces and dialogue systems in which human users interact with computers using gesture and sometimes speech. Section 2.3 describes the application of prosody to natural … Cited by 6 – Related articles – View as HTML – All 14 versions

Gesture in automatic discourse processing [PDF] from mit.edu R Barzilay, R Davis… – 2008 – dspace.mit.edu … gesture or other non-verbal features. Section 2.2 describes multimodal interfaces and dialogue systems in which human users interact with computers using gesture and sometimes speech. Section 2.3 describes the application of prosody to natural … Related articles – All 2 versions

Using viewing time to infer user preference in recommender systems [PDF] from lancs.ac.uk J Parsons, P Ralph… – 2004 – eprints.lancs.ac.uk Page 1. Using Viewing Time to Infer User Preference in Recommender Systems Jeffrey Parsons, Paul Ralph, Katherine Gallagher Faculty of Business Administration Memorial University of Newfoundland St. John’s, NL, Canada … Cited by 33 – Related articles – All 10 versions

Gesture in Automatic Discourse Processing [PDF] from 18.7.29.232 R Davis, J Eisenstein – 2008 – 18.7.29.232 … gesture or other non-verbal features. Section 2.2 describes multimodal interfaces and dialogue systems in which human users interact with computers using gesture and sometimes speech. Section 2.3 describes the application of prosody to natural … Related articles – All 2 versions

One Approach to Statistical Speech Recognition M Blagojevic, Z Durovic… – Computer as a Tool, …, 2005 – ieeexplore.ieee.org … The short overview of theoretical background of different phases of the algorithm including basic signals filtering, start-point and endpoint detection, dimension reduction and linear classifiers design are included in the paper. … Related articles

Semantic role labeling [PDF] from rug.nl M Palmer, D Gildea… – Synthesis Lectures on Human …, 2010 – morganclaypool.com … Semantic Role Labeling Martha Palmer, Daniel Gildea, and Nianwen Xue 2010 Spoken Dialogue Systems Kristiina Jokinen and Michael McTear 2010 Introduction to Chinese Natural Language Processing Kam-Fai Wong, Wenji Li, Ruifeng Xu, and Zheng-sheng Zhang 2009 … Cited by 12 – Related articles – Library Search – All 12 versions

[PDF] Improving the Performance of LVCSR Using Ensemble of Acoustic Models [PDF] from cmu.edu R Zhang – 2006 – cs.cmu.edu Page 1. Improving the Performance of LVCSR Using Ensemble of Acoustic Models Thesis Proposal By Rong Zhang Language Technologies Institute School of Computer Science Carnegie Mellon University May, 2006 Committee: Alexander Rudnicky Tanja Schultz … Related articles – View as HTML

[PDF] Verification of Multi-Class Recognition Decision: A Classification Approach [PDF] from psu.edu T Matsui, FK Soong… – IEICE transactions on information and …, 2005 – Citeseer … The automatic speech recognition (ASR) module in a human-machine dialogue system needs a high word recog- nition accuracy to make it useful for … eg, phone and word-levels), use of different features (eg, the state du- ration[5]-[7]), use of non-linear classifiers, integration of … Related articles – View as HTML – BL Direct – All 9 versions

Multilingual speech recognition with language identification B Ma1, C Guan, H Li… – Seventh International Conference …, 2002 – isca-speech.org … Spoken dialogue systems have received great interests in recent years by helping business reduce operational cost and upgrade customer services … utterances further based on the acoustic scores coming from Equation (8). In our experiments, a simple linear classifier is trained … Cited by 18 – Related articles

Confidence scoring for ANN-based spoken language understanding [PDF] from titech.ac.jp C Wutiwiwatchai… – … . ASRU’03. 2003 IEEE Workshop on, 2003 – ieeexplore.ieee.org … The SLU system was integrated in our spoken dialogue system, namely Thai Interactive Hotel Reservation Agent (TRA … The linear features can be classified by a simple linear classifier such as the Fisher linear discriminant analysis, which uses only one threshold for classification … Cited by 4 – Related articles – All 7 versions

[PDF] Weighted Finite-State Transducer Algorithms. An Overview [PDF] from snu.ac.kr M Mohri – STUDIES IN FUZZINESS AND SOFT COMPUTING, 2004 – bi.snu.ac.kr … They have been used with weighted transducers of several hundred million states and transitions to create large-vocabulary speech recognition and complex spoken-dialog systems. … Page 16. Cortes, Haffner, and Mohri deployed spoken-dialog systems. … Cited by 15 – Related articles – View as HTML – BL Direct – All 12 versions

[BOOK] Progress in pattern recognition, image analysis and applications: 9th Iberoamerican Congress on Pattern Recognition, CIARP 2004, Puebla, Mexico, October … [PDF] from unileon.es A Sanfeliu, JFM Trinidad… – 2004 – books.google.com … Classes Random Neural Network Model 336 Jose Aguilar New Bounds and Approximations for the Error of Linear Classifiers 342 Luis … Gil-Garcia, Jose M. Badia-Contelles, and Aurora Pons-Porrata Hidden Markov Models for Understanding in a Dialogue System 410 Fernando … Cited by 1 – Related articles – Library Search – All 12 versions

[PDF] Confidence estimation for statistical machine translation [PDF] from 128.220.117.40 J Blatz, E Fitzgerald, G Foster… – Johns Hopkins …, 2003 – 128.220.117.40 … know when you are wrong, particularly when users are involved! • extensive previous work in SR, eg spoken dialog systems • almost no work outside SR. . . … 22 Page 23. Multi Layer Perceptrons A discriminative model generalising linear classifiers: ^P(c|x) = s(v c .h(W.x)) … Cited by 6 – Related articles – View as HTML – All 7 versions

[BOOK] Affect in text and speech [PDF] from illinois.edu ECO Alm – 2008 – books.google.com … see Wilson, Wiebe, and Hoffman (2005) for a discussion within the context of sentiment analysis), or considering affective meaning and its instantiations within machine translation, literary analysis, language learning, or educational or service-related dialogue systems, and so … Cited by 29 – Related articles – All 10 versions

[PDF] Bimodal Affect Recognition Recognizing aggression using vision and sound [PDF] from uva.nl M Metternich – 2006 – staff.science.uva.nl … It is used to find the linear combination of features which best separates two or more classes of object or event. The resulting combinations may be used as a linear classifier, or more commonly in dimensionality reduction before later classification. … Related articles – View as HTML

Using graphone models in automatic speech recognition [PDF] from mit.edu JR Glass, IL Hetherington, SX Wang – 2009 – dspace.mit.edu … This often results in user frustration and reduced user confidence in speech dialogue systems in general. … Multiple confidence measures can also be combined to construct confidence vectors, which are then passed through a linear classifier to determine if the corresponding … Cited by 7 – Related articles – Library Search – All 3 versions

[PDF] Exploring Biologically-Inspired Interactive Networks for Object Recognition [PDF] from diva-portal.org M Saifullah – 2011 – liu.diva-portal.org Page 1. Linköping Studies in Science and Technology Thesis No. 1466 Exploring Biologically-Inspired Interactive Networks for Object Recognition by Mohammad Saifullah Submitted to Linköping Institute of Technology at Linköping … Cited by 1 – Related articles – View as HTML – All 2 versions

[PDF] Elements for Learning and Optimizing Expensive Fuctions [PDF] from lri.fr P Rolet – 2010 – lri.fr … This is an illustrative example; MDPs are of course suited to model many real-world processes, such as for instance: Dialogue systems An automated machine must interact with a user to acquire information and perform actions (eg interactive voice response for a … Cited by 1 – Related articles – View as HTML

[PDF] Kernels for Structured Data in Natural Language Processing [PDF] from aist-nara.ac.jp J Suzuki – Doctor Thesis, Nara Institute of Science and …, 2005 – cl.aist-nara.ac.jp Page 1. NAIST-IS-DD0361207 Doctoral Dissertation Kernels for Structured Data in Natural Language Processing Jun Suzuki March 24, 2005 Department of Information Processing Graduate School of Information Science Nara Institute of Science and Technology Page 2. … Cited by 2 – Related articles – View as HTML – All 4 versions

[PDF] A Machine Learning Approach to Anaphora Resolution Including Named Entity Recognition, PP Attachment Disambiguation, and Animacy Detection [PDF] from psu.edu A Nøklestad – 2009 – Citeseer Page 1. A Machine Learning Approach to Anaphora Resolution Including Named Entity Recognition, PP Attachment Disambiguation, and Animacy Detection Anders Nøklestad May 7, 2009 Page 2. 2 For my parents, Randi and Hans Olaf Page 3. Contents 1 Introduction 13 … Cited by 2 – Related articles – View as HTML – All 4 versions

Improvement of the Post-processing Method for Isolated Word OOV Rejection [PDF] from colips.org Y Zhu, C Li… – International Symposium on Chinese Spoken …, 2002 – isca-speech.org … First of all, by using a linear classifier, we combine several promising features presented by others to obtain a confidence measure … speech recognition applications, the rejection of out-of-vocabulary (OOV) words is an important issue, especially in dialogue systems, where acting … Related articles – All 4 versions

[PDF] Department oF Signal and Image Processing 46, rue Barrault Paris 75634 C’edex 13 France [PDF] from archives-ouvertes.fr P Verlinde – 2004 – hal.archives-ouvertes.fr … Neighbor NSA National Security Agency (USA) PE Profile Expert PIN Personal Identification Number PLC Piece-wise Linear Classifier QC Quadratic … shopping, accessing the safe room of your bank, tele-banking, accessing the services of interactive dialogue systems [175], or … Related articles – View as HTML

[BOOK] Code breaking for automatic speech recognition [PDF] from psu.edu V Venkataramani… – 2005 – Citeseer Page 1. Code Breaking for Automatic Speech Recognition Veera Venkataramani A dissertation submitted to the Johns Hopkins University in conformity with the requirements for the degree of Doctor of Philosophy. Baltimore, Maryland 2005 … Cited by 1 – Related articles – View as HTML – Library Search – All 8 versions

[PDF] Determining user state and mental task demand from electroencephalographic data [PDF] from uniklinik-freiburg.org M Honal – University of Karlsruhe, 2005 – uniklinik-freiburg.org Page 1. Determining User State and Mental Task Demand From Electroencephalographic Data Diplomarbeit Matthias Honal University of Karlsruhe Supervisors: Dr. Tanja Schultz Prof. Dr. Alexander Waibel November 2005 Page 2. Page 3. … Cited by 5 – Related articles – View as HTML – All 20 versions

[PDF] Mining Speech Sounds [PDF] from psu.edu G SALVI – KTH: Stockholm, 2006 – Citeseer … However, it is worth noting that many aspects of human-to-human interaction concern the way the speaker’s and listener’s roles are exchanged, and this picture would be misleading if the focus was on dialogue and dialogue systems research. … Related articles – View as HTML – All 9 versions

Computational Terminology: Exploring Bilingual and Monolingual Term Extraction [PDF] from diva-portal.org J Foo – 2012 – liu.diva-portal.org Page 1. Linköping Studies in Science and Technology. Thesis, No. 1523 Computational Terminology: Exploring Bilingual and Monolingual Term Extraction by Jody Foo Submitted to Linköping Institute of Technology at Linköping …

[PDF] Automatic voicemail summarisation for mobile messaging [PDF] from quid5.net K Koumpis – 2002 – server.quid5.net … PARADISE paradigm for dialogue system evaluation … 96 7.6 The ROC curves produced for the binary decision summarisation task using as inputs to a linear classifier the individual features of Table 7.2 with respect to four training sets containing different amount of data. . . . … Cited by 4 – Related articles – Library Search – All 5 versions