Rule Learning & Dialog Systems


Rule Learning & Dialog Systems
Notes: A “rule learning system” is otherwise known as a learning classifier system (LCS).
See also: CLASSiC (Computational Learning in Adaptive Systems for Spoken Conversation)Learning Classifier & Dialog Systems


Automatically Training a Problematic Dialogue Predictor for a Spoken Dialogue System A Gorin, I Langkilde-Geary, MA Walker, J Wright… – arXiv preprint arXiv: …, 2011 – arxiv.org … ripper is a fast and e cient rule learning system described in more detail in (Cohen, 1995, 1996); we describe it brie y here for completeness. … First, it was important to be able to integrate the results of applying the learner back into the hmihy spoken dialogue system. … Related articles All 3 versions

Automated dialog system and method AL Gorin, IL Geary, MA Walker, JH Wright – US Patent 7,529,667, 2009 – Google Patents … (45) Date of Patent: May 5,2009 (54) AUTOMATED DIALOG SYSTEM AND METHOD … Page 9. US 7,529, ,667 Bl 1 AUTOMATED DIALOG SYSTEM AND METHOD CROSS REFERENCE TO RELATED APPLICATION 5 This is a continuation-in-part of US patent application Ser. … Cited by 5 Related articles All 2 versions

Method and system for predicting understanding errors in a task classification system AL Gorin, IL Geary, MA Walker, JH Wright – US Patent 7,487,088, 2009 – Google Patents … the invention in a telephone customer care system, this invention may be applied to any single mode, or multimodal, dialog system. … 5 The experiments reported here primarily utilize a rule learning program to automatically induce an NLU error clas- sification model from the … Cited by 17 Related articles All 2 versions

Automated dialog system and method AL Gorin, IL Geary, MA Walker, JH Wright – US Patent 7,472,060, 2008 – Google Patents … the invention in a telephone customer care system, this invention may be applied to any single mode, or multimodal, dialog system. … The experiments reported here primarily utilize a rule learning program to automatically induce an NLU error clas- sification model from the … Cited by 15 Related articles All 2 versions

[PDF] from sjtu.edu.cn [PDF] Transformation-based Learning for Semantic parsing F Jurcicek, M Gašic, S Keizer, F Mairesse, B Thomson… – 2009 – bcmi.sjtu.edu.cn … Finally, it learns a compact set of rules that allow it to perform real-time semantic parsing. Note that modern statisti- cal dialogue systems typically exploit multiple ASR hypothe- ses. … Figure 3: Rule learning algorithm. 4. Evaluation … Cited by 4 Related articles All 16 versions

[PDF] from 204.14.132.173 Cognitive niches: An ecological model of strategy selection. JN Marewski, LJ Schooler – Psychological review, 2011 – psycnet.apa.org APA PsycNET Our Apologies! – The following features are not available with your current Browser configuration. – alerts user that their session is about to expire – display, print, save, export, and email selected records – get My … Cited by 15 Related articles All 9 versions

Natural language understanding monitoring system for identifying a task AL Gorin, IL Geary, MA Walker, JH Wright – US Patent 7,933,773, 2011 – Google Patents … the invention in a telephone customer care system, this invention may be 20 applied to any single mode, or multimodal, dialog system. … The experiments reported here primarily utilize a rule learning program to automatically induce an NLU error clas- sification model from the … Cited by 1 Related articles All 3 versions

Method and system for predicting understanding errors in a task classification system AL Gorin, IL Geary, MA Walker, JH Wright – US Patent 8,095,363, 2012 – Google Patents … the invention in a telephone customer care system, this invention may be applied to any single mode, or multimodal, dialog system. … The experiments reported here primarily utilize a rule learning program to automatically induce an NLU error clas- si?cation model from the … Related articles All 2 versions

[PDF] from ntu.edu.tw Towards learned feedback for enhancing trust in information seeking dialogue for radiologists D Sonntag – Proceedings of the 16th international conference on …, 2011 – dl.acm.org … Thereby, the dialogue system also acts as middle- ware between the clients and the backend services [11]. … Association Rules (Data Mining) Association rule learning is a typical data mining technique; established methods can be found in [1] or in [3]. Associa- tion rules are … Related articles All 2 versions

[PDF] from ijcai.org Introspection and adaptable model integration for dialogue-based question answering D Sonntag – Proceedings of the 21st international jont conference …, 2009 – aaai.org … Our training examples are collected from real user interactions with our baseline dialogue system (also explained in the summative evaluation of the SmartWeb system [Mögele and Schiel, 2007]). … Asso- ciation rule learning is a typical data mining technique. … Cited by 6 Related articles All 10 versions

[PDF] from lrec-conf.org [PDF] Witchcraft: A workbench for intelligent exploration of human computer conversations A Schmitt, G Bertrand, T Heinroth… – Proc. of LREC, Valetta, …, 2010 – lrec-conf.org … 1. Introduction The growing task complexity of spoken dialogue systems such as telephone-based speech applications requires new tools to support system designers and speech scientists in analyzing human machine dialogues. … Walker et al. employ a rule-learning al- … Cited by 4 Related articles All 3 versions

[PDF] from mit.edu [PDF] Learning to Adapt in Dialogue Systems: Data-driven Models for Personality Recognition and Generation F Mairesse – 2008 – people.csail.mit.edu Page 1. Learning to Adapt in Dialogue Systems: Data-driven Models for Personality Recognition and Generation François Mairesse … February 2008 Page 2. Abstract Dialogue systems are artefacts that converse with human users in order to achieve some task. … Cited by 2 Related articles All 7 versions

[PDF] from suendermann.com [PDF] Is it possible to predict task completion in automated troubleshooters A Schmitt, M Scholz, W Minker… – Proc. of the …, 2010 – suendermann.com … For more complex systems, the prediction accuracy significantly decreases the further the call progresses. An early prediction of task completion can have several benefits for the dialog system and the user. … [1]. They employ RIP- PER, a rule-learning algorithm, to implement a … Cited by 11 Related articles All 7 versions

[PDF] from thepieraccinis.com Value-based optimal decision for dialog systems E Levin, R Pieraccini – Spoken Language Technology …, 2006 – ieeexplore.ieee.org … depends on several factors, such as the cost of human agents and the particular commercial agreement between the dialog system developer and … We used the rule-learning system SLIPPER2 based on confi- dence-rated boosting [4] which is an extension of .the widely-used … Cited by 24 Related articles All 17 versions

[PDF] from christopia.net The relative effectiveness of human tutoring, intelligent tutoring systems, and other tutoring systems K VanLehn – Educational Psychologist, 2011 – Taylor & Francis … Netherlands: IOS. (pp. 349–356) View all references; VanLehn, 1999126. VanLehn, K. 1999. Rule learning events in the acquisition of a complex skill: An evaluation of Cascade. Journal of the Learning Sciences , 8: 179–221. [Taylor & … Cited by 11 Related articles All 3 versions

[CITATION] Transformation-Based Learning for Semantic Parsing F Jurcícek, M Gašic, S Keizer, F Mairesse, B Thomson… – Tenth Annual Conference of …, 2009 Related articles

[CITATION] When calls go wrong: How to detect problematic calls based on log-files and emotions? O Herm, A Schmitt, J Liscombe – Ninth Annual Conference of the International …, 2008 Cited by 16 Related articles All 3 versions

[BOOK] Towards Adaptive Spoken Dialog Systems A Schmitt, W Minker – 2012 – books.google.com Page 1. Alexander Schmitt – Wolfgang Minker Towards Adaptive Spoken Dialog Systems a Q Springer Page 2. Towards Adaptive Spoken Dialog Systems Page 3. Alexander Schmitt • Wolfgang Minker Towards Adaptive Spoken Dialog Systems 123 Page 4. … Related articles All 3 versions

Detecting Problematic Dialogs with Automated Agents A Schmitt, C Hank, J Liscombe – … in Multimodal Dialogue Systems, 2008 – Springer … employ RIPPER, a rule-learning algorithm, to implement a Problematic Dialogue Predictor forecasting the call-outcome of calls in the HMIHY (How May I … The application belongs to a new gen- eration of dialog systems that goes beyond classical FAQ IVRs in that it takes similar … Cited by 11 Related articles BL Direct All 2 versions

Text categorization methods for automatic estimation of verbal intelligence F Fernández-Martínez, K Zablotskaya… – Expert Systems with …, 2012 – Elsevier … 2 explains the adaptation process of spoken dialog systems based on verbal intelligence estimation in more detail … regression models, Bayesian probabilistic approaches, Nearest Neighbors approaches, Rocchio algorithm, decision trees, inductive rule learning, neural networks … Related articles All 2 versions

[PDF] from arxiv.org Learning Symbolic Models of Stochastic Domains LP Kaelbling, HM Pasula, LS Zettlemoyer – arXiv preprint arXiv:1110.2211, 2011 – arxiv.org Page 1. Journal of Artificial Intelligence Research 29 (2007) 309-352 Submitted 6/06; published 7/07 Learning Symbolic Models of Stochastic Domains Hanna M. Pasula pasula@csail.mit.edu Luke S. Zettlemoyer lsz@csail.mit.edu Leslie Pack Kaelbling lpk@csail.mit.edu … Related articles All 3 versions

Language Technology Support for Czech G Rehm, H Uszkoreit – The Czech Language in the Digital Age, 2012 – Springer … Some of speech departments are working onmanyprojects inthe speech field, being able to offer simple dialog systems, covering most of voice technology. … For example, mak- ing a rule-based system adaptive by adding a module for rule learning, or, making a statistical MT …

[PDF] from etsmtl.ca Identifying problematic dialogs in a human-computer dialog system HC Truong – 2010 – espace.etsmtl.ca … RIPPER (like other learning programs eg, C5.0 and CART) is a fast and efficient rule learning system described in more detail in (Cohen, 1995) and (Cohen, 1996). The best result … with their system ‘How May I Help You?’ HMIHY is a spoken dialogue system for customer … Related articles All 2 versions

Automated dialog method with first and second thresholds for adapted dialog strategy AL Gorin, IL Geary, MA Walker, JH Wright – US Patent 7,440,893, 2008 – Google Patents … (Continued) Primary Examiner—Martin Lerner (57) ABSTRACT This invention concerns a method and system for monitoring an automated dialog system for the automatic recognition of language understanding errors based on a user’s input com- munications. … Cited by 6 Related articles All 2 versions

PISA: A framework for multiagent classification using argumentation M Wardeh, F Coenen, TB Capon – Data & Knowledge Engineering, 2012 – Elsevier … Classification rules are rules whose consequent is a class label; such rules are used in some types of classifier to categorise “unseen” data. Classification rules can be generated using a number of mechanisms such as decision trees and inductive rule learning. … Related articles All 2 versions

Language Technology Support for French J Mariani, P Paroubek, G Francopoulo, A Max… – The French Language in …, 2012 – Springer … Similarly, oral dialog systems are presently only available for very constrained applications. … of the two paradigms by integrating the good features of each; for example, making a rule-based system adap- tive by adding a module for rule learning, or making a statistical MT system …

[PDF] from psu.edu [PDF] Transformation-based and Memory-based Learning for Detecting Speech Recognition Errors G Skantze – 2008 – Citeseer … ASR results. However, it is doubtful whether 85.1% (or 87.7% for content words) is good enough to use in a dialogue system. In order … be corrected. In this case, a rule-learning algorithm such as µ-TBL must be used. For example … Related articles All 5 versions

[PDF] from arxiv.org Granularity-adaptive proof presentation M Schiller, C Benzmüller – arXiv preprint arXiv:0903.0314, 2009 – arxiv.org … The development of the dialog system prototype was guided by empirical studies using a mock-up of the DIALOG system [6]. One research challenge … We evaluated rule learning using C5.0 on our sample using 10 fold cross valida- tion, which resulted in a mean percentage of … Cited by 3 Related articles All 9 versions

A new Testbed for Semi-Automatic Usability Evaluation and Optimization of Spoken Dialogue Systems S Möller, KP Engelbrecht, M Pucher, P Fröhlich… – … Speech Dialog Systems, 2008 – Springer … Evaluation and Optimization of Spoken Dialogue Systems … Any model that reduces this error rate can help to predict the recognition performance. Inspired by the work of Hirschberg [14], we adopt the rule-learning pro- gram “RIPPER” from Cohen [17]. … Cited by 2 Related articles

[PDF] from uni-muenchen.de Analyzing collaborative learning processes automatically: Exploiting the advances of computational linguistics in computer-supported collaborative learning C Rosé, YC Wang, Y Cui, J Arguello… – International Journal of …, 2008 – Springer … This keeps the size of the feature space down, which aids in effective rule learning. … 1998, 2001; Page 1968; Page and Petersen 1995; Landauer and Dumais 1997; Foltz et al. 1998; Laham 2000) and tutorial dialogue systems (Graesser et al. 1998; Rosé et al. 2001; Aleven et al. … Cited by 111 Related articles All 26 versions

LEXICAL ANSWER TYPE CONFIDENCE ESTIMATION AND APPLICATION JJ Fan, DA Ferrucci, DC Gondek… – US Patent App. 13/ …, 2011 – Google Patents … Thus, a LAT may be detectable in a question through pattern LAT detection rules. These rules are implemented and are encoded or learned by machines auto- matically through association rule learning. A natural lan- guage understanding model may implement these rules. … All 2 versions

[PDF] from ftw.at [PDF] TIDE: A testbed for interactive spoken dialogue system evaluation S Möller, KP Engelbrecht, M Pucher… – Proc. Intl. Conf. …, 2007 – userver.ftw.at TIDE: A Testbed for Interactive Spoken Dialogue System Evaluation Sebastian Möller 1 , Klaus-Peter Engelbrecht 1 , Michael Pucher 2 , Peter Fröhlich 2 … Inspired by the work of Hirschberg [10], we also adopt the rule-learning program “RIPPER” from Cohen [14] to generate … Cited by 8 Related articles All 2 versions

[PDF] from archives-ouvertes.fr Modeling Tutoring Knowledge J Bourdeau, M Grandbastien – Advances in Intelligent Tutoring Systems, 2010 – Springer … 11). Classifications of this variable include the characteristics of the domain (math, science, language, atti- tude learning), the topics (concept or rule learning), the skills (problem solving, reading, writing), and the competencies. … Cited by 1 Related articles All 6 versions

Method and system for predicting understanding errors in automated dialog systems AL Gorin, IL Geary, MA Walker, JH Wright – US Patent 7,003,459, 2006 – Google Patents … invention in a 5 telephone customer care system, this invention may be applied to any single mode, or multimodal, dialog system. … The experiments reported here primarily utilize a rule learning program to automatically induce an NLU error classification model from the 11,787 … Cited by 17 Related articles All 2 versions

SCORING CANDIDATES USING STRUCTURAL INFORMATION IN SEMI-STRUCTURED DOCUMENTS FOR QUESTION ANSWERING SYSTEMS JJ Fan, DA Ferrucci – US Patent App. 13/244,351, 2011 – Google Patents Page 1. IJS20l20089622A1 (12) Patent Application Publication (10) Pub. No.: US 2012/0089622 A1 (19) United States Fan et al. (43) Pub. Date: Apr. 12, 2012 (54) SCORING CANDIDATES USING STRUCTURAL INFORMATION … All 2 versions

[PDF] from arxiv.org Learning content selection rules for generating object descriptions in dialogue PW Jordan, M Walker – Journal of Artificial Intelligence Research, 2005 – aaai.org … Department of Computer Science, University of Sheffield Regent Court, 211 Portobello Street Sheffield S1 4DP, UK Abstract A fundamental requirement of any task-oriented dialogue system is the ability to gen- erate object descriptions that refer to objects in the task domain. … Cited by 62 Related articles All 23 versions

[BOOK] Logical and relational learning L De Raedt – 2008 – books.google.com … 2006 W. Wahlster (Ed.) SmartKom: Foundations of Multimodal Dialogue Systems XVIII, 644 pages. 2006 B. Goertzel, C. Pennachin (Eds.) Arti?cial General Intelligence XVI, 509 pages. … 161 6.3 Case Study 1: Rule Learning and Foil …..161 6.3.1 Foil’s Problem Setting . . … Cited by 163 Related articles All 11 versions

USING ONTOLOGICAL INFORMATION IN OPEN DOMAIN TYPE COERCION DA Ferrucci, AA Kalyanpur, JW Murdock IV… – US Patent App. 13/ …, 2011 – Google Patents Page 1. US 20l20078873Al (19) United States (12) Patent Application Publication (10) Pub. No.: US 2012/0078873 A1 Ferrucci et al. (43) Pub. Date: Mar. 29, 2012 (54) USING ONTOLOGICAL INFORMATION IN Publication Classi?cation … All 2 versions

[PDF] from aminer.org Language Structure Using Fuzzy Similarity NS Chaudhari, X Wang – Fuzzy Systems, IEEE Transactions on, 2009 – ieeexplore.ieee.org Page 1. This article has been accepted for publication in a future issue of this journal, but has not been fully edited. Content may change prior to final publication. > TFS-2008-0201.R1 < 1 Abstract—Learning of (context free) grammar … Cited by 2 Related articles All 6 versions

[PDF] from upm.es Error-correction detection and response generation in a spoken dialogue system I Bulyko, K Kirchhoff, M Ostendorf, J Goldberg – Speech Communication, 2005 – Elsevier … Error taxonomies are often specific to particular annotation schemes or individual dialogue systems. … A similar accuracy rate was reported for a detector developed by Hirschberg et al. (2001), which was based on the ripper rule learning algorithm. … Cited by 18 Related articles All 3 versions

SYSTEM AND METHOD FOR PROVIDING QUESTION AND ANSWERS WITH DEFERRED TYPE EVALUATION J Fan, D Ferrucci, DC Gondek… – US Patent App. 12/ …, 2008 – Google Patents … LAT detection rules can be encoded manually or learned by machine automatically through association rule learning. In this case the natural language understanding moduel can be limited to implementation the simple rules as described above. … All 2 versions

[BOOK] Errrors and Intelligence in Compter-Assisted Language Learning: Parsers and Pedagogues T Heift, M Schulze – 2007 – books.google.com Page 1. Errors and Intelligence in Computer-Assisted Language Learning Parsers and Pedagogues Trude Heift and Mathias Schulze Routledge Taylor & Francis Croup Page 2. Errors and Intelligence in Computer-Assisted Language Learning Page 3. … Cited by 82 Related articles All 4 versions

[PDF] from washington.edu Learning symbolic models of stochastic domains HM Pasula, LS Zettlemoyer, LP Kaelbling – Journal of Artificial Intelligence …, 2007 – aaai.org Page 1. Journal of Artificial Intelligence Research 29 (2007) 309-352 Submitted 6/06; published 7/07 Learning Symbolic Models of Stochastic Domains Hanna M. Pasula pasula@csail.mit.edu Luke S. Zettlemoyer lsz@csail.mit.edu Leslie Pack Kaelbling lpk@csail.mit.edu … Cited by 80 Related articles All 19 versions

APPARATUS AND METHOD FOR DETERMINING RELEVANCE OF INPUT SPEECH O Kalinli – US Patent 20,120,259,638, 2012 – freepatentsonline.com … 20070118380, Method and device for controlling a speech dialog system, May, 2007, Konig. … 20100100379, VOICE RECOGNITION CORRELATION RULE LEARNING SYSTEM, VOICE RECOGNITION CORRELATION RULE LEARNING PROGRAM, AND VOICE RECOGNITION …

[PDF] from abdn.ac.uk [PDF] Informing Dialogue Strategy through Argumentation-Derived Evidence CD Emele – 2011 – csd.abdn.ac.uk … We explored a number of machine learning algorithms, such as decision trees (using C4.5), rule learning (using sequential covering), instance-based learner (using k-nearest neighbours), and semantic-enriched decision tree learner (using STree). … Related articles All 8 versions

[PDF] from userapi.com [PDF] Acquiring Conceptual Expertise from Modeling: The Case of Elementary Physics K Vanlehn, B van de Sande – Development of Professional …, 2009 – cs5538.userapi.com Page 374. 356 In many domains, the real world is modeled with systems of equations. Such a model uses variables to represent domain properties and equations to rep- resent applications of domain principles. Given a set of … Cited by 1 Related articles All 5 versions

Method and system for predicting understanding errors in a task classification system AL Gorin, IL Geary, MA Walker, JH Wright – US Patent 7,127,395, 2006 – Google Patents … the invention in a telephone customer care system, this invention may be applied to any single mode, or multimodal, dialog system. … 20 The experiments reported here primarily utilize a rule learning program to automatically induce an NLU error classification model from the … Cited by 12 Related articles All 2 versions

[PDF] from uic.edu [PDF] Expert tutoring and natural language feedback in intelligent tutoring systems X Lu – 2007 – cs.uic.edu … system. Like general dialogue systems, dialogue management becomes an intrinsically complex task for designing an ITS. … each state of the network. 2. Form filling: specify the information that the dialogue system must obtain from the user as a set of forms composed of slots. … Cited by 8 Related articles BL Direct All 8 versions

[PDF] from buaa.edu.cn Question classification using head words and their hypernyms Z Huang, M Thint, Z Qin – Proceedings of the Conference on Empirical …, 2008 – dl.acm.org … 1 Introduction An important step in question answering (QA) and other dialog systems is to classify the question to the anticipated type of the answer. … (2002) and Radev et at. (2002), in which language model and Rappier rule learning were employed respectively. … Cited by 33 Related articles All 13 versions

[PDF] from uwindsor.ca [PDF] EXPLORING ADAPTATION IN DIALOG SYSTEMS S STOYANCHEV – 2008 – cs.uwindsor.ca Page 1. EXPLORING ADAPTATION IN DIALOG SYSTEMS BY SVETLANA STOYANCHEV SUBMITTED IN PARTIAL FULFILLMENT … some evidence that people adapt their language use in conversation with computer partners – spoken dialog systems. Spoken dialog sys- … Related articles All 7 versions

[PDF] from upenn.edu Detecting emotion in speech: experiments in three domains J Liscombe – Proceedings of the 2006 Conference of the North …, 2006 – dl.acm.org … This results in non-human-like and even inappropriate behavior on the part of the spoken dialogue system. … to evaluate the predictive power of each fea- ture extracted from the EPSaT utterances, I ran machine learning experiments using RIPPER, a rule-learning al- gorithm. … Cited by 7 Related articles All 28 versions

[PDF] from ucsc.edu [PDF] Automatically training a problematic dialogue predictor for a spoken dialogue system MA Walker, I Langkilde-Geary… – Journal of Artificial …, 2002 – users.soe.ucsc.edu … ripper is a fast and e cient rule learning system described in more detail in (Cohen, 1995, 1996); we describe it brie y here for completeness. … First, it was important to be able to integrate the results of applying the learner back into the hmihy spoken dialogue system. … Cited by 57 Related articles All 27 versions

[PDF] from dtic.mil The Andes physics tutoring system: Lessons learned K Schulze, JA Shapiro, R Shelby, D Treacy… – International Journal of …, 2005 – IOS Press Page 1. International Journal of Artificial Intelligence in Education 15 (2005) 147–204 147 IOS Press 1560-4292/05/$17.00 © 2005 – IOS Press and the authors. All rights reserved The Andes Physics Tutoring System: Lessons Learned … Cited by 253 Related articles All 47 versions

Method and system for predicting understanding errors in a task classification system AL Gorin, IL Geary, MA Walker, JH Wright – US Patent 6,751,591, 2004 – Google Patents … invention in a telephone customer care system, this invention may be 10 applied to any single mode, or multimodal, dialog system. … The experiments reported here primarily utilize a rule learning program to automatically induce an NLU error classification model from the 11,787 … Cited by 37 Related articles All 3 versions

[PDF] from psu.edu [PDF] Refining Tailored Scaffolding for Meta-Cognitive Skills during Analogical Problem Solving K Muldner, C Conati – Workshop on Metacognition and Self-Regulated …, 2007 – Citeseer … References 1. Aleven, V., O. Popescu, C. Torrey, & K. Koedinger (2004) Evaluating the Effectiveness of a Tutorial Dialogue System for Self-Explanation. … 21. VanLehn, K. (1999) Rule-Learning Events in the Acquisition of a Complex Skill: An Evaluation of Cascade. … Related articles All 7 versions

[PDF] from psu.edu Early error detection on word level G Skantze, J Edlund – COST278 and ISCA Tutorial and Research …, 2004 – isca-speech.org … Since many interpretation modules in dialogue systems are mainly dependent on content words, the performance of these are important for … methods for improving the performance of a specific application without collecting more data for models, a rule- learning algorithm such … Cited by 22 Related articles All 12 versions

[PDF] from sri.com [BOOK] Logical particle filtering LS Zettlemoyer, HM Pasula, LP Kaelbling – 2008 – pal.sri.com … This problem is pervasive in AI: a dialogue system has to estimate the belief state of the user; an office-assistant must track the states and … This representation is inspired by a similar one previously used for probabilistic rule learning in the fully observable setting [8]. Let s be the … Cited by 12 Related articles All 19 versions

[PDF] from irit.fr Agents that argue and explain classifications L Amgoud, M Serrurier – Autonomous Agents and Multi-Agent Systems, 2008 – Springer Page 1. Auton Agent Multi-Agent Syst (2008) 16:187–209 DOI 10.1007/s10458-007-9025-6 Agents that argue and explain classifications Leila Amgoud · Mathieu Serrurier Published online: 28 December 2007 Springer Science+Business Media, LLC 2007 … Cited by 5 Related articles BL Direct All 5 versions

[PDF] from uvt.nl [PDF] Memory-based understanding of user utterances in a spoken dialogue system: Effects of feature selection and co-learning A Van den Bosch – Workshop Proceedings of the 6th International …, 2005 – arno.uvt.nl … Similar results have already been attained by applying rule learning to the same data, yielding comparable feature sets and performances for the … As for dialogue systems, we believe our results with the k-nearest neighbor approach using small amounts of features reaffirms the … Cited by 3 Related articles All 6 versions

[PDF] from cmu.edu [PDF] TTS From Zero Building Synthetic Voices for New Languages J Kominek – 2009 – lti.cs.cmu.edu … in great dialogs about dialog systems. As a visiting researcher several years ago, Marelie Davel … 3.1.7 Reweighting with phone transition probabilities …..55 3.1.8 Rule learning architectures: pros and cons…..57 … Cited by 4 Related articles All 7 versions

[PDF] from swrtec.de [PDF] Multi-Modal Task Instructions to Robots by Naive Users JC Wolf – 2008 – swrtec.de Page 1. MULTI-MODAL TASK INSTRUCTIONS TO ROBOTS BY NAIVE USERS by JOERG CHRISTIAN WOLF A thesis submitted to the UNIVERSITY OF PLYMOUTH, UK in partial fulfilment for the degree of DOCTOR OF PHILOSOPHY … Related articles All 9 versions

[PDF] from uvt.nl [PDF] Timbl: Tilburg memory-based learner W Daelemans, J Zavrel, K Van der Sloot… – Version, 2007 – ilk.uvt.nl Page 1. TiMBL: Tilburg Memory-Based Learner version 6.3 Reference Guide ILK Technical Report – ILK 10-01 Walter Daelemans* Jakub Zavrel*† Ko van der Sloot Antal van den Bosch Induction of Linguistic Knowledge Research … Cited by 96 Related articles All 6 versions

[PDF] from pitt.edu Ontological technologies for user modelling S Sosnovsky, D Dicheva – International Journal of Metadata, …, 2010 – Inderscience … current information task. The short-term model is discarded when the session ends. 2.1.3 Goals, plans, tasks and needs Modelling users’ goals and plans has been widely exploited in intelligent dialog systems. Knowledge of what … Cited by 11 Related articles All 7 versions

[PDF] from mit.edu Characterizing phonetic transformations and fine-grained acoustic differences across dialects NFY Chen – 2011 – dspace.mit.edu … tying) trained on 5-Dialect Arabic Corpus. . . . . 148 9-1 Example of rule learning limitation in current system setup. . . . . 151 9-2 Limitation shown in Figure 9-1 can be elegantly dealt with simply by reversing the direction of all state transition arcs. . . . . 152 … Related articles All 3 versions

[PDF] from upenn.edu A hybrid text classification approach for analysis of student essays CP Rosé, A Roque, D Bhembe, K Vanlehn – Proceedings of the HLT- …, 2003 – dl.acm.org … We also plan to experiment with other rule learning approaches, such as RIPPER (Cohen, 1995). 7 Acknowledgments … Can we help students with high initial competency? In Pro- ceedings of the ITS Workshop on Empirical Methods for Tutorial Dialogue Systems. 74 Page 8. … Cited by 45 Related articles All 56 versions

[PDF] from pdx.edu Hahoe kaist robot theatre: learning rules of interactive robot behavior as a multiple-valued logic synthesis problem M Perkowski, T Sasao, JH Kim, M Lukac… – … -Valued Logic, 2005. …, 2005 – ieeexplore.ieee.org … 4.1. Constructive induction While commercial dialog systems are boring with their repeating “I do not understand. Please repeat the last sentence” behaviors, our robots are rarely “confused”. They always do something, and in most cases their action is slightly unexpected. … Cited by 6 Related articles All 7 versions

[PDF] from unr.edu Efficient and non-parametric reasoning over user preferences C Domshlak, T Joachims – User Modeling and User-Adapted Interaction, 2007 – Springer Page 1. User Model User-Adap Inter (2007) 17:41–69 DOI 10.1007/s11257-006- 9022-5 ORIGINAL PAPER Efficient and non-parametric reasoning over user preferences Carmel Domshlak · Thorsten Joachims Received: 31 … Cited by 21 Related articles BL Direct All 5 versions

[PDF] from cmu.edu An evaluation of a hybrid language understanding approach for robust selection of tutoring goals CP Rosé, K VanLehn – International Journal of AI in Education, 2005 – IOS Press … Halloun & Hestenes, 1985). In our previous work on the Atlas-Andes system, we have demonstrated one way an implemented tutorial dialogue system can address this problem (Rosé et al., 2001a). We evaluated two versions … Cited by 14 Related articles All 18 versions

[PDF] from ubc.ca [PDF] From example studying to problem solving via tailored computer-based meta-cognitive scaffolding: hypotheses and design C Conati, K Muldner, G Carenini – Journal of Technology, …, 2006 – people.cs.ubc.ca Page 1. From Example Studying to Problem Solving via Tailored Computer-Based Meta-Cognitive Scaffolding: Hypotheses and Design CRISTINA CONATI*, KASIA MULDNER AND GIUSEPPE CARENINI Department of Computer … Cited by 8 Related articles All 14 versions

[PDF] from psu.edu Multi-dimensional relational sequence mining F Esposito, N Di Mauro, TMA Basile, S Ferilli – Fundamenta Informaticae, 2008 – IOS Press … Indeed, some domains have a stronger need for user Page 3. F. Esposito et al. / Multi-Dimensional Relational Sequence Mining 25 models to tune the interface characteristics and behaviours: among the most classical ones, Computer Aided Instruction and Dialog Systems. … Cited by 15 Related articles All 7 versions

[PDF] from psu.edu [PDF] MULTI-MODAL TASK INSTRUCTIONS TO ROBOTS BY NAIVE USERS MPHIL/PHD TRANSFER REPORT JC WOLF, S TEAM – 2006 – Citeseer … However sequence learning alone is not enough. For more advanced tasks rule learning is necessary. Page 8. – 8 – … The dialogue system is based on an investigation of a corpus of human-human and simulated human-machine dialogs. … All 3 versions

[PDF] from shef.ac.uk [PDF] Toward Portable Information Extraction MV Tablan – 2009 – nlp.shef.ac.uk Page 1. MV Tablan Toward Portable Information Extraction Submitted for the degree of Doctor of Philosophy 2009 Page 2. Page 3. Toward Portable Information Extraction Mihai Valentin Tablan Submitted for the degree of Doctor of Philosophy … Related articles All 2 versions

[PDF] from ajou.ac.kr Reflections on Gordon Pask’s adaptive teaching concepts and their relevance to modern knowledge systems G Mallen – Proceedings of the 5th conference on Creativity & …, 2005 – dl.acm.org … 87 Page 3. principles to simple rule learning. … Our specific goal is to be able to relate the subjective knowledge state to the objective or expert knowledge which has been used to build the database and use the dialogue systems to bridge the gap. … Cited by 2 Related articles All 3 versions

PROVIDING QUESTION AND ANSWERS WITH DEFERRED TYPE EVALUATION USING TEXT WITH LIMITED STRUCTURE PA Duboue, JJ Fan, DA Ferrucci… – US Patent App. 13/ …, 2011 – Google Patents Page 1. US 20120078902A1 United States (19) (12) Patent Application Publication (10) Pub. No.: US 2012/0078902 A1 Duboue et al. (43) Pub. Date: Mar. 29, 2012 (54) PROVIDING QUESTION AND ANSWERS Publication … All 2 versions

[PDF] from diva-portal.org Question Classification in Question Answering Systems H Sundblad – 2007 – liu.diva-portal.org … Page 23. Question classification 15 2.4.1 Decision rule learning with set-valued features Radev et al. (2002) experiments with machine learning for question classi- fication using decision rule learning with set-valued features (Cohen 1996). … Cited by 1 Related articles All 12 versions

[PDF] from upenn.edu [PDF] A Little Goes a Long Way: Quick Authoring of Semantic Knowledge Sources for Interpretation CP Rosé, B Hall – Proceedings of SCa-NaLu, 2004 – acl.ldc.upenn.edu … Previ- ously, tutorial dialogue systems such as AUTO-TUTOR (Wiemer-Hastings et al., 1998) and Research Methods Tutor (Malatesta et al … A hybrid rule learning approach to classification involving both statistical and symbolic features has been shown to perform better than LSA … Cited by 8 Related articles All 34 versions

[PDF] from teithe.gr A very fast and efficient linear classification algorithm KI Diamantaras, I Michailidis… – Machine Learning for …, 2005 – ieeexplore.ieee.org … This task is very useful for creat- ing semantic representation of sentences like in the case of Information Extraction systems [13] and Human-Machine Dialogue systems, or simply for indexing texts [14]. See related conferences-competitions of NER systems [15, 12]. … Cited by 2 Related articles All 8 versions

[PDF] from cmu.edu [PDF] Automatic improvement of machine translation systems AF Llitjós – 2007 – cs.cmu.edu Page 1. Automatic Improvement of Machine Translation Systems Ariadna Font Llitjós July 2007 CMU-LTI-07-008 Language Technologies Institute School of Computer Science Carnegie Mellon University Pittsburgh, Pennsylvania 15213 … Cited by 4 Related articles All 13 versions

[PDF] from utwente.nl [PDF] A survey of reinforcement learning in relational domains M Van Otterlo – 2005 – eprints.eemcs.utwente.nl Page 1. A Survey of RL in Relational Domains (Van Otterlo 2005) A Survey of Reinforcement Learning in Relational Domains Martijn van Otterlo otterlo@cs.utwente. nl Human Media Interaction (HMI) Department of Computer … Cited by 43 Related articles All 12 versions

[PDF] from uni-saarland.de [PDF] Intrinsic and Extrinsic Approaches to Recognizing Textual Entailment DP der Philosophischen – 2011 – coli.uni-saarland.de … Each utterance in the interpreted version is actually implied or entailed by the utterances in the original conversation. Con- sequently, if we want to build a dialogue system, dealing with this kind of implication or entailment is one of the key challenges. Let alone there … Related articles All 2 versions

[PDF] from psu.edu An efficient incremental architecture for robust interpretation CP Rosé, A Roque, D Bhembe – Proceedings of the second …, 2002 – dl.acm.org … Additionally, for CarmelTC one of the features for each training text that is made available to the rule learning algorithm is the classification obtained using the Rainbow naive bayes clas … Pedagogical content knowledge in a tutorial dialogue system to support self-explanation. … Cited by 29 Related articles All 15 versions

[PDF] from upenn.edu Training a dialogue act tagger for human-human and human-computer travel dialogues R Prasad, M Walker – Proceedings of the 3rd SIGdial workshop on …, 2002 – dl.acm.org … et al., 2000), identifying important parts of a dialogue (Finke et al., 1998), evaluating and comparing spo- ken dialogue systems (Walker et al … Our experiments apply the rule learning program RIPPER (Cohen, 1996) to train a DATE dialogue act tagger for the utterances of the … Cited by 19 Related articles All 81 versions

[PDF] from mdh.se [PDF] An Intelligent FAQ Answering System Using a Combination of Statistic and Semantic IR Techniques L Jianan, P Sävström – 2006 – idt.mdh.se … Originally well known dialogue systems like SHRDLU ( Winograd, 1972) and GUS (Bobrow, 1977) were all built as research systems to help researchers understand the issues involved in interactive advisory system and modeling human dialogue , rather than real commercial … Related articles

[PDF] from stuba.sk [PDF] Information Sciences and Technologies Bulletin of the ACM Slovakia A Andrejko, R Semanèík, J Jakubík, M Lekavý… – 2009 – acmbulletin.fiit.stuba.sk … The main problem of feedback based solutions is that nec- essary questions are mostly hard coded for a particular application. However, there are some solutions that use generation of the questions, such as dialog systems. … Related articles All 6 versions

[PDF] from washington.edu The impact of speech recognition on speech synthesis M Ostendorf, I Bulyko – … Proceedings of 2002 IEEE Workshop on, 2002 – ieeexplore.ieee.org … Second, data-driven learning techniques could be applied to structure and rule learning, not simply param- eter learning, and in that sense they … One important area is in spo- ken dialog systems, where prosody can be useful for dialog act recognition as well as in parsing [5]. A … Cited by 33 Related articles All 11 versions

[PDF] from wwu.edu [PDF] A Chinese Language Expert System Using Bayesian Learning YY Wu, JJ Zhang – Proc of the Eighth World Multiconference on …, 2004 – cs.wwu.edu … Paek [25, 26] has constructed a Bayesian network model by observing interactions between a user and a spoken dialog system. … Rule-learning in Classification of Email”, University of Texas at Austin, Artificial Intelligence Lab, Technical Report, AI-TR-99-284, 1999. … Cited by 1 Related articles All 3 versions

[PDF] from u2u.net [PDF] Relational sequence learning and user modelling B Demoen, N Lavrac, N JACOBS – 2004 – u2u.net … Nowadays, the focus is more on supporting the student rather than guiding the student through a learning path, but the importance of user models is still present.1 A second historical domain is that of dialog systems, a sub-domain of natural language processing (NLP), which is … Cited by 10 Related articles All 8 versions

[PDF] from ntu.edu.sg [PDF] Semantic relations in information science CS Khoo, JC Na – Annual Review of Information Science and …, 2006 – ntu.edu.sg … filled by terms/concepts extracted from the text are often used as the intermediate representation or interlingua in natural language understanding systems (eg Chan & Franklin, 2003; Minker, Bennacef & Gauvain, 1996), question answering and dialogue systems (eg Takemura … Cited by 42 Related articles BL Direct All 5 versions

[PDF] from upc.edu Ontology-based Information Extraction C Vicient Monllaró – 2011 – upcommons.upc.edu … McCallum 1999; Skounakis, Craven et al. 2003), self-supervised methods (Etzioni, Cafarella et al. 2005), rule learning (Soderland 1999), and conditional random fields (McCallum 2003). These techniques learn a language model … Related articles All 3 versions

[PDF] from uci.edu User modeling for personalized city tours J Fink, A Kobsa – Artificial intelligence review, 2002 – Springer Page 1. Artificial Intelligence Review 18: 33–74, 2002. © 2002 Kluwer Academic Publishers. Printed in the Netherlands. 33 User Modeling for Personalized City Tours1 JOSEF FINK1 and ALFRED KOBSA2 1humanIT–Human … Cited by 157 Related articles BL Direct All 16 versions

[PDF] from tudelft.nl [PDF] Topics in speech recognition DALK Fa – 2006 – kbs.twi.tudelft.nl … For certain spoken telephone dialog systems it is desirable that these English names are properly recognized. … Chapter 2 Theory 2.1 Introduction The purpose of speech dialog systems is the exchange of information in a more interactive manner using speech to communicate. … Related articles All 2 versions

FEMTI, 43 French, 139 M Translation – Machine Translation, 2002 – Springer … dialogue systems, 165, 185, 213 doctor–patient dialogues, 213 elicitation, 245, 271 embedded MT, 77, 99, 139, 165, 213, 245, 271 evaluation, 43 Expedition, 271 … quality models, 43 rapid ramp-up, 271 resources, 1, 99 rule-based MT, 19 rule learning, 245 …

[PDF] from uni-dortmund.de [PDF] Automatically Training a Problematic Dialogue Predictor for a Spoken Dialogue System Marilyn A. Walker walker@ research. att. com Irene Langkilde-Geary … HW Hastie, J Wright, A Gorin – Journal of …, 2002 – www-ai.informatik.uni-dortmund.de … RIPPeR is a fast and efficient rule learning system described in more detail in (Cohen, 1995, 1996); we describe it brie fl y here for completeness. … First, it was important to be able to integrate the results of applying the learner back into the HmIHY spoken dialogue system. … Related articles All 12 versions

[TXT] from gmu.edu Abstraction of reasoning for problem solving and tutoring assistants V Le – 2008 – digilib.gmu.edu … A learning component for acquiring and refining the knowledge of the agent, allowing a wide range of operations, from ontology import and user definition of knowledge base elements (through the use of editors and browsers), to ontology learning and rule learning. … Cited by 1 Related articles All 7 versions

[PDF] from psu.edu [PDF] Long-answer question answering and rhetorical-semantic relations SJ Blair-Goldensohn – 2007 – Citeseer Page 1. Long-Answer Question Answering and Rhetorical-Semantic Relations Sasha J. Blair-Goldensohn Submitted in partial fulfillment of the Requirements for the degree of Doctor of Philosophy in the Graduate School of Arts and Sciences COLUMBIA UNIVERSITY 2007 … Cited by 5 Related articles All 6 versions

[PDF] from naist.jp GMM-based voice conversion applied to emotional speech synthesis H Kawanami, Y Iwami, T Toda, H Saruwatari… – 2003 – library.naist.jp … syn?esis has become an im portant subject due to requests to realize more familiar human interface of a spoken dialogue system, more advanced … To realize it, in vestigation of further modeling of an acoustic space?d con version rule learning system should be considered. … Cited by 22 Related articles All 8 versions

[BOOK] Intelligent information integration in B2B electronic commerce M Brown, D Fensel, G Dabiri, B Omelayenko, Y Ding… – 2002 – books.google.com Page 1. INTELLIGENT INFORMATION NTEGRATION IN B2B ELECTRONIC COMMERCE Dieter Fensel Borys Omelayenko Ying Ding Michel Klein Alan Flett Ellen S elm I ten Guy Botquin Mike Brown Gloria Dabiri Kluwer Academic Publishers Page 2. Page 3. … Cited by 15 Related articles BL Direct All 5 versions

[TXT] from brunel.ac.uk A framework for knowledge discovery within business intelligence for decision support R Basra – 2008 – bura.brunel.ac.uk A Framework for Knowledge Discovery within Business Intelligence for Decision Support A thesis submitted for the degree of Doctor of Philosophy. By Rajveer Singh Basra Brunel Business School Brunel University I Abstract … All 4 versions