Notes:
TiMBL is an open-source software package that implements memory-based learning algorithms for classification tasks. It is primarily used in natural language processing as a machine learning classifier component, but its use extends to other domains as well. The algorithms in TiMBL use a memory-based approach, where the training set is stored explicitly in memory and new cases are classified by extrapolation from the most similar stored cases. The decision-tree-based implementation of TiMBL makes it more efficient in classification than a standard k-nearest neighbor algorithm in many cases.
The references below mention that the package could be used for semantic interpretation of text in terms of frames and roles, which could be beneficial for many applications such as question answering, information extraction, semantic dialogue systems, as well as statistical machine translation or automatic text summarization. The text also highlights that the package could be used in practical applications such as dialogue systems development and evaluation, and in linguistics research. They also mention that they are interested in developing more sophisticated spoken dialogue systems by automatically modeling discourse structure.
Resources:
Wikipedia:
See also:
CLASSiC (Computational Learning in Adaptive Systems for Spoken Conversation) | Learning Classifier & Dialog Systems | Rule Learning & Dialog Systems
User simulations for context-sensitive speech recognition in spoken dialogue systems [PDF] from aclweb.org O Lemon – Proceedings of the 12th Conference of the European …, 2009 – dl.acm.org … The results, obtained using TiMBL and an n-gram User Sim- ulation, show a significant relative reduction of Word Error Rate of 5% (this is 44% of the pos- sible WER improvement on this data … Clearly, this improvement would result in better dialogue system performance overall. … Cited by 15 – Related articles – All 12 versions
Combining acoustic and pragmatic features to predict recognition performance in spoken dialogue systems [PDF] from upenn.edu M Gabsdil, O Lemon – Proceedings of the 42nd Annual Meeting on …, 2004 – dl.acm.org … The approach is novel in combining machine learning with n-best processing for spoken dialogue systems using the Information State Update approach. Our best results, obtained using TiMBL with op- timized parameters, show a 25% weighted f-score improvement over a … Cited by 70 – Related articles – All 19 versions
[PDF] Context-based Re-ranking and Grounding Classification of N-Best Hypotheses [PDF] from psu.edu R Jonson – 2008 – Citeseer … To see if it would be possible to automate this task for application in dialogue systems we have used the memory-based learner TiMBL (Daelemans et al., 2001) to inves- tigate the optimization and variability of the ac- curacy of re-ranking and grounding classification of speech … Related articles – View as HTML – All 4 versions
[PDF] TiMBL: Tilburg Memory-Based Learner [PDF] from uvt.nl W Daelemans, J Zavrel, K Van der Sloot… – 2009 – ilk.uvt.nl Page 1. TiMBL: Tilburg Memory-Based Learner version 6.2 Reference Guide ILK Technical Report – ILK 09-01 Walter Daelemans* Jakub Zavrel*† Ko van der Sloot Antal van den Bosch … 11 4.2 Using TiMBL . . . . . … Cited by 14 – Related articles – View as HTML
[PDF] Transformation-based and Memory-based Learning for Detecting Speech Recognition Errors [PDF] from psu.edu G Skantze – 2008 – Citeseer … Since many interpretation modules in dialogue systems are mainly dependent on content words, the performance of these are important for detection. … This gives a baseline of 69.8%. For these words, the scores for the classifiers were 87.7% (µ-TBL) and 87.0% (TiMBL). … Related articles – View as HTML – All 5 versions
Multi-level information and automatic dialog act detection in human-human spoken dialogs [PDF] from cnrs.fr S Rosset, D Tribout… – Speech Communication, 2008 – Elsevier … Most dialog systems exploit the information present at the lexical and semantic levels. … The IB1-IG implementation of Machine Based Learning from the TiMBL software package (Daelemans et al., 2003) was employed using the Manhattan distance, one of the most basic metrics … Cited by 11 – Related articles – All 18 versions
Dialogue context-based re-ranking of ASR hypotheses [PDF] from pitt.edu R Jonson – Spoken Language Technology Workshop, 2006. …, 2006 – ieeexplore.ieee.org … To see if it would be possible to automate this 175 Page 3. task for application in dialogue systems we have used the memory-based learner TiMBL [10] to re-rank ASR hypothe- ses. We have experimented with different features … Cited by 13 – Related articles – All 4 versions
[PDF] Reordering Attachment Candidates in the CSLI Dialogue System’s DMT [PDF] from stanford.edu F Ratiu – 2009 – cs229.stanford.edu … 4 4. Data Collection Data used for these experiments was collected during several user sessions in which the performance of this dialogue system was evaluated. … The machine learning approach (ML-Reorder) uses classification obtained from the memory based learner TiMBL. … Related articles – View as HTML – All 7 versions
Visinav: A system for visual search and navigation on web data A Harth – Web Semantics: Science, Services and Agents on the …, 2010 – Elsevier … However, there is the open question of how end users should express complex queries over such datasets. A promising approach is to use a menu-based dialogue system in which users construct the query incrementally [34] and [38]. … 2, Information available about timbl:i. … Cited by 9 – Related articles – All 2 versions
Classifying non-sentential utterances in dialogue: A machine learning approach [PDF] from kcl.ac.uk R Fernández, J Ginzburg… – Computational Linguistics, 2007 – dl.acm.org … is to develop a classification model whose output can be fed into a dialogue processing system-be it a full dialogue system or, for … four machine learn- ing systems: the rule induction learner SLIPPER (Cohen and Singer 1999), the memory-based learner TiMBL (Daelemans et … Cited by 14 – Related articles – BL Direct – All 23 versions
[PDF] Dialogue Context-Based Speech Recognition using User Simulation [PDF] from sevas.org.in I Konstas – 2008 – sevas.org.in … alignment with the true transcript. Their approach to the problem is in two steps: first they use TiMBL (Daelemans et al., 2007), a memory based classifier, in order to pre- … < crosstalk. The corpus used was extracted with the WITAS dialogue system (Lem- … Related articles – View as HTML – All 5 versions
[PDF] Cue-Based Dialogue Act Classification [PDF] from shef.ac.uk N Webb – 2010 – nlp.shef.ac.uk … This includes the role das can play in practical applications such as dialogue systems de- velopment and evaluation. … linguistics, recent work, closely linked to the development and deployment of spoken language dialogue systems, has focused on the some of the more … Related articles – View as HTML
Method and apparatus for providing proper or partial proper name recognition F Weng, L Zhao – US Patent 7,865,356, 2011 – Google Patents … Consequently, the spo- ken dialog systems and/or information extraction programs may be improved. BRIEF DESCRIPTION OF THE DRAWINGS FIG. … Page 11. US 7,865,356 B2 is a normalization factor. 3 4 “Timbl”, which was used to perform the experiments described herein. … Related articles – All 5 versions
[PDF] Machine learning techniques in dialogue act recognition [PDF] from rakenduslingvistika.ee M Fišel – Estonian Papers Applied Linguistics, 2007 – rakenduslingvistika.ee … Hagen and Popowich (2000) describe a dialogue system which is based on a grammar of dialogue acts, which is used to determine the … well as a comprehensive description of memory-based learning can be found in the Tilburg Memory-Based Learner (TiMBL) reference … Cited by 6 – Related articles – View as HTML – All 10 versions
Reducing recognition error rate based on context relationships among dialogue turns [PDF] from mit.edu HC Wu… – Eighth Annual Conference of the International …, 2007 – isca-speech.org … The goal is to both provide a confidence score and to reduce recognition error, in a dialogue system in- … Machine learning approaches used include the memory based learner TiMBL [3, 1] and the rule induction learner RIP- PER [4, 1]. Below are two examples of typical learned … Cited by 3 – Related articles – All 10 versions
An information state based dialogue manager for a mobile robot [PDF] from psu.edu M Quinderé, LS Lopes… – Eighth Annual Conference …, 2007 – isca-speech.org … Information State Approach Dialogue Systems can be divided into [2]: Finite State Systems, Frame-Based Systems and Advanced Systems, which … The semantic ex- traction can be shallow – performed by Tilburg Memory Based Learner (TiMBL) or deep – performed by LCFlex … Cited by 3 – Related articles – All 6 versions
[PDF] Classifying Non-Sentential Utterances in Dialogue: A Machine Learning Approach [PDF] from sas.ac.uk S Lappin, R Fernandez… – Computational Linguistics, 2007 – sas-space.sas.ac.uk … is to develop a classification model whose output can be fed into a dialogue processing system-be it a full dialogue system or, for … four machine learn- ing systems: the rule induction learner SLIPPER (Cohen and Singer 1999), the memory-based learner TiMBL (Daelemans et … Cited by 1 – Related articles – View as HTML – All 2 versions
[PDF] Results of the French Evalda-Media evaluation campaign for literal understanding [PDF] from brandeis.edu H Bonneau-Maynard, C Ayache, F Bechet… – LREC’ …, 2006 – pages.cs.brandeis.edu … 2003. Timbl: Tilburg memory based learner, v5.0, reference guide. In ILK Technical Report ILK-03-10. A. Denis, M. Quignard, and G. Pittel. 2006. A deep- parsing approach to natural language understanding in dialogue system: Results of a corpus-based evaluation. In LREC. … Cited by 28 – Related articles – View as HTML – All 12 versions
Memory-Based Learning [PDF] from ua.ac.be W Daelemans… – The Handbook of …, 2010 – Wiley Online Library … TiMBL is an open source software package implementing all algorithms and metrics discussed here.1 First, a visual example serves to … dialogue history features (eg, pre- vious dialogue acts), and acoustic features of recognized speech in the context of spoken dialogue systems. … Related articles – All 5 versions
Visinav: Visual web data search and navigation [PDF] from psu.edu A Harth – Database and Expert Systems Applications, 2009 – Springer … A promising approach is to use a menu-based dialogue system in which users incrementally construct the query [17] [19]. … a-Friend, http://foaf-project.org/ 5 Semantically Interlinked Online Communities, http://sioc-project.org/ 6 Dublin Core, http://dublincore.org/ 7 timbl:i expands … Cited by 10 – Related articles – All 8 versions
WALTER DAELEMANS AND AVAN den BOSCH – The Handbook of Computational …, 2010 – books.google.com … TiMBL is an open source software package implementing all algorithms and metrics discussed here. … be a mix of bag-of-word features, dialogue history features (eg, pre- vious dialogue acts), and acoustic features of recognized speech in the context of spoken dialogue systems. …
[PDF] Machine learning for shallow interpretation of user utterances in spoken dialogue systems [PDF] from cam.ac.uk P Lendvai, A van den Bosch… – … on Dialogue Systems: …, 2003 – cl.cam.ac.uk … We hypothesize that with the use of additional features the current perfor- mance would improve, so that eventually it could be integrated and tested in the context of a spoken dialogue system to achieve more … TiMBL: Tilburg memory based learner, version 4.3, reference guide. … Cited by 24 – Related articles – All 15 versions
Unified treatment of data-sparseness and data-overfitting in maximum entropy modeling F Weng… – EP Patent 1,783,744, 2007 – freepatentsonline.com … If the speech recognizer 101 for the exemplary dialog system does not produce class labels for the class-based statistical language models, a … A memory-based learning package that may be used is Timbl, which is discussed, for example, by Daelemans, ”TIMBL: Tilburg Memory … Related articles – Cached – All 2 versions
[PDF] Dialogue Act Recognition Techniques [PDF] from psu.edu M Fishel – GSLT/NGSLT course on dialogue systems: Linkoping …, 2006 – Citeseer … (Fernandez et al., self-defined SLIPPER linguistic and 87% 2005) antecedent TIMBL (same) 87% MaxEnt (same) 87% (Sanchis and Castro, self-defined MLP reduced lexicon 92% 2002) BoW Table 1: DA recognition in dialogue systems 18 Page 19. project DA Cited by 2 – Related articles – View as HTML – All 3 versions
[PDF] Memory-based understanding of user utterances in a spoken dialogue system: Effects of feature selection and co-learning [PDF] from uvt.nl A Van den Bosch – Workshop Proceedings of the 6th International …, 2005 – arno.uvt.nl … Daelemans, W., Zavrel, J., van der Sloot, K., van den Bosch, A.: TiMBL: Tilburg Memory Based Learner, version 5.1, Reference guide. … Lendvai, P., van den Bosch, A., Krahmer, E.: Machine learning for shallow interpre- tation of user utterances in spoken dialogue systems. … Cited by 4 – Related articles – View as HTML – All 7 versions
[PDF] Using dialogue context to improve parsing performance in dialogue systems [PDF] from unam.mx I Meza-Ruiz, O Lemon – International Workshop on …, 2005 – turing.iimas.unam.mx Page 1. Using Dialogue Context to Improve Parsing Performance in Dialogue Systems Ivan Vladimir Meza-Ruiz T H E U NIVER S I T Y … rest of the thesis. 1.1 Dialogue Systems There is a wide range of dialogue systems which have been developed, or which are … Cited by 5 – Related articles – View as HTML – All 2 versions
Classifying ellipsis in dialogue: a machine learning approach [PDF] from upenn.edu R Fernández, J Ginzburg… – Proceedings of the 20th …, 2004 – dl.acm.org … et al., to ap- pear), we implemented G&S’s analysis of di- rect sluices as part of an interpretation module in a dialogue system. … The second, TiMBL, uses a memory-based ma- chine learning procedure to classify a sluice by generalising over similar environments in which the … Cited by 29 – Related articles – All 28 versions
SYSTEMS AND METHODS FOR FILTERING DICTATED AND NON-DICTATED SECTIONS OF DOCUMENTS AB Carus, L Lapshina… – US Patent …, 2011 – freepatentsonline.com … 20050055205, Intelligent user adaptation in dialog systems, March, 2005, Jersak et al. … Still other embodiments may employ the IGTree and IB algorithms of the TiMBL machine-learning classifier system to generate accurate and compact classifiers. … Cached
From robust spoken language understanding to knowledge acquisition and management [PDF] from pitt.edu LS Lopes, AJS Teixeira, M Quinderé… – Ninth European …, 2005 – isca-speech.org … Wielenga et al.(eds.), (1990) IOS Press, Amsterdam [4] Daelemans, W., J. Zavrel, A. van den Bosch, and K. van der Sloot.: TiMBL: Tilburg Memory … [12] Seabra Lopes, L., A. Teixeira, M. Quinderé, A Knowl- edge Representation and Reasoning Module for a Dialogue System in a … Cited by 7 – Related articles – All 6 versions
Early error detection on word level [PDF] from psu.edu G Skantze… – COST278 and ISCA Tutorial and Research …, 2004 – isca-speech.org … 2003). TiMBL: Tilburg Memory Based Learner, version 5.0, Reference Guide. ILK Technical Report 03-10. [9] Skantze, G. (2003). Exploring Human Error Handling Strategies: Implications for Spoken Dialogue Systems. In Proceedings … Cited by 22 – Related articles – All 9 versions
[PDF] Using machine learning for non-sentential utterance classification [PDF] from uni-potsdam.de R Fernández, J Ginzburg… – Proceedings of the Sixth …, 2005 – ling.uni-potsdam.de … Punctuation tags (that would correspond to intona- tion patterns in a spoken dialogue system) help to extract the values of these features … im- plement three different learning strategies: SLIP- PER, a rule induction system presented in (Cohen and Singer, 1999); TiMBL, a memory … Cited by 9 – Related articles – View as HTML – All 11 versions
Towards finding and fixing fragments: Using ML to identify non-sentential utterances and their antecedents in multi-party dialogue [PDF] from upenn.edu D Schlangen – Proceedings of the 43rd Annual Meeting on …, 2005 – dl.acm.org … TIMBL (Tilburg Memory-Based Learner), (Daelemans et al., 2003), which implements a memory-based learning algorithm (IB1) which pre- dicts the class of a test data point by looking at its distance to all examples from the training data, us- ing some distance metric. … Cited by 9 – Related articles – BL Direct – All 22 versions
[PDF] About the Usefulness and Learnability of Argument-Diagrams from Real Discussions [PDF] from amiproject.org R Rienks… – Proceedings of MLMI 2006, 2006 – amiproject.org … For the TiMBL results the overlap metric was used. … In addition to the compression we have made use of n-grams of POS-tags which has previously been done in research concerning the creation of backchannels in a spoken dialogue system [Cathcart et al., 2003]. … Cited by 8 – Related articles – View as HTML – All 5 versions
Classifying recognition results for spoken dialog systems [PDF] from upenn.edu M Gabsdil – Proceedings of the 41st Annual Meeting on Association …, 2003 – dl.acm.org … Spotting erroneous utterances and words is a ma- jor task in spoken dialog systems. … We found that by using the machine learners TiMBL and Rip- per we can improve the results in both tasks as com- pared to predicting recognition quality solely on the basis of the acoustic … Cited by 7 – Related articles – All 17 versions
Automatic Bare Sluice Disambiguation in Dialogue* R Fernández… – 2005 – Citeseer … capacity to recognise and interpret sluices-bare wh-phrases that exhibit a sentential meaning-is essential to maintaining cohesive interaction between human users and a machine interlocutor in a dialogue system. … 4, TiMBL: Tilburg Memory Based Learner Reference Guide. … Cited by 3 – Related articles – Cached – All 2 versions
[PDF] Clause Boundary Detection in Transcribed Spoken Language [PDF] from uio.no F Jørgensen – Proceedings from NODALIDA, 2007 – folk.uio.no … syntax (eg Part of Speech tagging), syntax (eg parsing and (semi-)automatic treebank construction) and se- mantics (in dialogue systems, where eg … of features from the context of the conjunctions 3. Apply a machine learning method (memory based learning/TiMBL) 4. Evaluate … Cited by 1 – Related articles – View as HTML – All 4 versions
Automatic detection of dialog acts based on multilevel information [PDF] from 192.44.78.170 S Rosset… – Eighth International Conference on Spoken …, 2004 – isca-speech.org … We are also interested in automatically modeling the discourse structure in order to develop more sophisticated spoken dialog systems. … for natural language processing [2, 3]). We employed the IB1-IG implementation of Machine Based Learning from TiMBL software package … Cited by 7 – Related articles – All 14 versions
Information state based speech recognition [PDF] from gu.se R Jonson – rapport nr.: Gothenburg Monographs in Linguistics 41, 2010 – gupea.ub.gu.se … 65 3.2 The TrindiKit platform for dialogue system development . . . . . 66 3.3 The GoDiS dialogue system . . . . . 67 3.3.1 GoDiS information state . . . . . … 86 3.6.1 TiMBL . . . . . … Related articles – Library Search – All 7 versions
[PDF] Dialogue move prediction from the information state using TiMBL [PDF] from gu.se R Jonson – 2005 – ling.gu.se … rj@ling.gu.se Dec, 2005 Abstract In this paper we explore dialogue move prediction ie predicting what the user of a dialogue system may do in his/her next turn. We have used the machine learner TiMBL [TiMBL] to predict user dialogue moves from information states. … Related articles – View as HTML
[PDF] Automatic Bare Sluice Disambiguation in Dialogue [PDF] from sas.ac.uk S Lappin, R Fernandez… – 2005 – sas-space.sas.ac.uk … If a dialogue system does not assign the correct interpretation to a sluice, it will not respond correctly to the ques- tion. … In section 4 we use a set of features to manually annotate a data set of sluices, and run two machine learning algorithms: SLIPPER and TiMBL, which yield … Related articles – View as HTML – All 8 versions
[PDF] Dialogue act tagging with Memory based learning [PDF] from kth.se A Hjalmarsson – 2005 – speech.kth.se … 6. Daelemans Walter, Jakub Zavrel, Ko van der Sloot, and Antal van den Bosch “TiMBL: Tilburg Memory Based Learner, version 5.1, Reference … Rotaru, M., Dialog Act Tagging using Memory-Based Learning, Term Project in Dialog Systems, University of Pittsburgh, Spring 2002. … Related articles – View as HTML
[PDF] Using Machine Learning for Non-Sentential Utterance Classification [PDF] from sas.ac.uk S Lappin, R Fernandez… – 2005 – sas-space.sas.ac.uk … Punctuation tags (that would correspond to intona- tion patterns in a spoken dialogue system) help to extract the values of these features … im- plement three different learning strategies: SLIP- PER, a rule induction system presented in (Cohen and Singer, 1999); TiMBL, a memory … Related articles – View as HTML – All 2 versions
Educating Lia: The Development of a Linguistically Accurate Memory-Based Lemmatiser for Afrikaans H Groenewald – Intelligent Information Processing III, 2007 – Springer … W. Daelemans, A. Van den Bosch, J. Zavrel and K. Van der Sloot. TiMBL: Tilburg Memory Based Learner, version 5.1, Reference Guide. ILK Technical Report 04-02, 2004. PJ dvi Toit. … The August Spoken Dialogue System. Proceedings of Eurospeech, 1999. R. Hausser. … Related articles – All 6 versions
[PDF] Timbl: Tilburg memory-based learner [PDF] from uvt.nl W Daelemans, J Zavrel, K Van der Sloot… – version, 2002 – 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 … 11 4.2 Using TiMBL . . . . . … Cited by 257 – Related articles – View as HTML – All 27 versions
[PDF] Integrating multiple modalities into SLMs and parsing the output of SLMs [PDF] from eurice.info K Weilhammer, R Jonson, H Burden, J Schatzmann… – 2006 – eurice.info … Out of the scope of task 1.4 we have carried out some work to explore dialogue move prediction ie predicting what the user of a dialogue system may do in his/her next turn. We have used the machine learner TiMBL [DZvdSvdB01] to predict user dialogue moves from … Related articles – View as HTML – All 7 versions
[PDF] Combining acoustic confidences and pragmatic plausibility for classifying spoken chess move instructions [PDF] from upenn.edu M Gabsdil – Proceedings of the 5th SIGdial Workshop on …, 2004 – acl.ldc.upenn.edu … 2002. TIMBL: Tilburg Mem- mory Based Learner, version 4.2, Reference Guide. … Malte Gabsdil and Oliver Lemon. subm. Combining acoustic and pragmatic features to predict recognition performance in spoken dialogue systems. Submitted to ACL-04. Thorsten Joachims. 1999. … Cited by 2 – Related articles – View as HTML – All 16 versions
[PDF] Memory-Based Shallow Parsing for Text Mining [PDF] from pascal-network.org W Daelemans – 2004 – eprints.pascal-network.org … Ontology Extraction and Refinement Question Answering Dialogue Systems Shallow Parsing … Regularities and subregularities / exceptions can be modeled uniformly TiMBL 5.0 • http://ilk.uvt.nl (includes detailed reference guide and guide to related literature) … Cited by 1 – Related articles – View as HTML – All 3 versions
[PDF] Classifying Ellipsis in Dialogue: A Machine Learning Approach [PDF] from sas.ac.uk S Lappin, R Fernandez… – 2004 – sas-space.sas.ac.uk … et al., to ap- pear), we implemented G&S’s analysis of di- rect sluices as part of an interpretation module in a dialogue system. … The second, TiMBL, uses a memory-based ma- chine learning procedure to classify a sluice by generalising over similar environments in which the … Related articles – View as HTML – All 2 versions
[PDF] Classifying Ellipsis in Dialogue: A Machine Learning Approach [PDF] from kcl.ac.uk J Ginzburg, R Fernandez… – 2004 – calcium.dcs.kcl.ac.uk … et al., to ap- pear), we implemented G&S’s analysis of di- rect sluices as part of an interpretation module in a dialogue system. … The second, TiMBL, uses a memory-based ma- chine learning procedure to classify a sluice by generalising over similar environments in which the … Related articles – View as HTML
Context dependent speech recognition [PDF] from ed.ac.uk S Andersson – 2006 – era.lib.ed.ac.uk … Page 14. 6 CHAPTER 1. CONTEXT SENSITIVE ASR IN DIALOGUE SYSTEMS 1.2.3 Learning Methods … RIPPER was also used in Gabsdil and Lemon (2004) but was discarded because a memory based learner, TiMBL (Daelemans et al., 2002) per- formed better. … Cited by 1 – Related articles – All 5 versions
The dual of denial: Two uses of disconfirmations in dialogue and their prosodic correlates [PDF] from uvt.nl E Krahmer, M Swerts, M Theune… – Speech Communication, 2002 – Elsevier … Using a corpus of interactions with two Dutch spoken dialogue systems, prosodic correlates of users’ disconfirmations were investigated. … Author Keywords: Spoken dialogue systems; Prosody; Error detection; Information grounding; Perception. … Cited by 25 – Related articles – All 17 versions
[PDF] Memory-based robust interpretation of recognised speech [PDF] from psu.edu P Lendvai, A Van den Bosch, E Krahmer… – … of SPECOM’04, 9th …, 2004 – Citeseer … Most if not all spoken dialogue systems (SDSs) contain an understanding module that extracts semantic and prag- matic information from the word … We employ the IB1 algorithm, implemented in the TiMBL software package (version 5.0) [6]. The IB1 algorithm is an example of a … Cited by 5 – Related articles – All 15 versions
[PDF] Dialog act tagging using memory-based learning [PDF] from pitt.edu M Rotaru – Term project, University of Pittsburgh, 2002 – cs.pitt.edu … 1.Introduction A better understanding of the semantics of user utterances in a spoken dialog system (SDS) will lead to more efficient and robust SDS. … For our experiments, we employed the IB1-IG implementation of MBL from TiMBL software package (Daelemans et all 2001). … Cited by 7 – Related articles – View as HTML – All 5 versions
[PDF] TiMBL: Tilburg memory based learner [PDF] from ua.ac.be W Daelemans, J Zavrel, K van der Sloot… – ILK 03, 2003 – clips.ua.ac.be Page 1. TiMBL: Tilburg Memory-Based Learner version 5.0 Reference Guide ILK Technical Report – ILK 03-10 Walter Daelemans* Jakub Zavrel*† Ko van der Sloot Antal van den Bosch … 9 4.2 Using TiMBL . . . . . … Cited by 32 – Related articles – View as HTML – All 12 versions
Unified treatment of data-sparseness and data-overfitting in maximum entropy modeling F Weng… – US Patent App. 11/266,867, 2005 – Google Patents … Partial Proper Name Recognition [0040] If the speech recognizer 101 for the exemplary dialog system does not produce class labels for the … A memory-based learning package that may be used is Timbl, which is discussed, for example, by Daelemans, “TIMBL: Tilburg Memory …
Book Review [DOC] from naba.org.uk S McRoy – Computational Linguistics, 2004 – MIT Press … TiMBL: Tilburg Memory Based Learner, version 2.0, Reference Guide. … sys- tems for tutoring and training; achieving robust human-machine communication; detecting and repairing communication errors; developing tools for real-time intelligent dialog systems; and investigating … Cited by 1 – Related articles – BL Direct – All 4 versions
Exceptionality and natural language learning [PDF] from upenn.edu M Rotaru, DJ Litman – Proceedings of the seventh conference on …, 2003 – dl.acm.org … come from the area of spoken dialog systems and have smaller datasets and more features (with many of the features being numeric, in contrast … Our memory-based learner is called IB1-IG and is part of TiMBL, a software package developed by the ILK Research Group, Tilburg … Cited by 4 – Related articles – All 48 versions
[PDF] D4. 1: Integration of Learning and Adaptivity with the ISU approach [PDF] from shef.ac.uk O Lemon, K Georgila, J Henderson, M Gabsdil… – 2005 – nlp.shef.ac.uk … 7 2 Classifier Machine Learning methods for ISU dialogue systems 9 2.1 Memory-based learning . … ISU dialogue systems This chapter explains how we have explored the use of several different machine learning techniques with Information State Update dialogue management. … Cited by 15 – Related articles – All 6 versions
[PDF] Improving machine-learned detection of miscommunications in human-machine dialogues through informed data splitting [PDF] from psu.edu P Lendvai, A Van den Bosch, E Krahmer… – Proceedings of the …, 2002 – Citeseer … 11 Page 12. RIPPER significantly better than TiMBL? Global data Split data … 5 Discussion In this paper we have studied the use of two machine learning techniques, namely RIPPER and IB1-GR, for error detection in spoken dialogue systems. … Cited by 11 – Related articles – View as HTML – All 17 versions
The learning vector quantization algorithm applied to automatic text classification tasks MT Martín-Valdivia, LA Urena-López… – Neural Networks, 2007 – Elsevier Cited by 11 – Related articles – All 7 versions
Blueprint for a high performance NLP Infrastructure [PDF] from upenn.edu JR Curran – Proceedings of the HLT-NAACL 2003 workshop on …, 2003 – dl.acm.org … There have already been several attempts to develop distributed NLP systems for dialogue systems (Bayer et al., 2001) and speech recognition (Ha- cioglu and Pellom, 2003). … 2002. TiMBL: Tilburg Memory-Based Learner reference guide. … Cited by 14 – Related articles – All 47 versions
[PDF] Nordic Graduate School of Language Technology (NGSLT) [PDF] from 130.241.54.13 R Cooper – 2007 – 130.241.54.13 … Spring Term Dialogue Systems Lars Ahrenberg, Linköpings universitet, Sweden, GSLT Java Development for HLT Lars Degerstedt, Linköpings universitet, Sweden, GSLT Lexical Semantics Åke Viberg, Uppsala University, Swe- den, GSLT Machine Learning Joakim Nivre … Related articles – View as HTML – All 4 versions
Ontological Engineering and the Semantic Web J Gómez-Pérez… – Advanced Techniques in Web Intelligence-I, 2010 – Springer Page 1. Chapter 8 Ontological Engineering and the Semantic Web José Manuel Gómez-Pérez and Carlos Ruiz Abstract. This chapter focuses on ontologies as means of representing knowledge and reason across the various domains. … Cited by 2 – Related articles – All 2 versions
[BOOK] Finite state morphology [HTML] from mendeley.com KR Beesley… – 2003 – mendeley.com … TiMBL: Tilburg Memory Based Learner, version 2.0, Reference Guide. … sys- tems for tutoring and training; achieving robust human-machine communication; detecting and repairing communication errors; developing tools for real-time intelligent dialog systems; and investigating … Cited by 520 – Related articles – Cached – Library Search – All 12 versions
Mining Diagnostic Text Reports by Learning to Annotate Knowledge Roles [PDF] from cuc.edu.cn E Mustafaraj, M Hoof… – … Language Processing and …, 2007 – books.google.com … The authors envision that the semantic interpretation of text in terms of frames and roles would contribute to many applications, like question answering, information extraction, semantic dialogue systems, as well as statistical machine translation or automatic text summarization … Cited by 6 – Related articles – All 13 versions
[PDF] D2. 1: Integration of ontological knowledge with the ISU approach [PDF] from eurice.info M Gabsdil, S Larsson, O Lemon, P Manchon… – 2005 – eurice.info … Memory-based learning (implemented using TiMBL)is used over Information State Update representations of dialogue context. This resulted in over 50% reduction in error rates for speech recognition in a task-based dialogue system. … Related articles – All 10 versions
[PDF] Content assessment in intelligent computer-aided language learning: Meaning error diagnosis for English as a second language [PDF] from psu.edu SM Bailey – 2008 – Citeseer … 203 6.9 Breakdown of Data Sets by Judgment Labels . . . . . 205 6.10 Hypothetical Performance Metric Example Data . . . . . 206 6.11 TiMBL Distance Metrics . . . . . 212 6.12 Development Set Performance with(out) AA Pairs . . . . . … Cited by 6 – Related articles – View as HTML – Library Search – All 8 versions
Unhappy bedfellows: the relationship of AI and IR [PDF] from shef.ac.uk Y Wilks – Charting a New Course: Natural Language Processing …, 2005 – Springer Page 1. 255 John I. Tait (ed.), Charting a New Course: Natural Language Processing and Information Retrieval. Essays in Honour of Karen Spärck Jones. 255-282 (c) 2005 Springer. Printed in the Netherlands. YORICK A. WILKS … Cited by 3 – Related articles – All 9 versions
Word sense disambiguation [PDF] from 212.3.125.93 M Stevenson… – The Oxford Handbook of Comp. …, 2003 – books.google.com … Examples of’final’tasks are MT, information extraction (IE), and dialogue systems, all of which are described in Part III of this volume. … 1998. TiMBL: Tilburg memory based learner version 1. o. ILK Technical Report 98-03. Gale, W., K. Church. and D. Yarowsky. … Cited by 88 – Related articles – Library Search – All 7 versions
System and method for tokenization of text J Carrier, AB Carus, WF Cote… – US Patent App. …, 2004 – freepatentsonline.com … 20070083375, METHOD AND SYSTEM FOR TEMPLATE INQUIRY DIALOGUE SYSTEM, April, 2007, Lee et al. … In one embodiment of the invention, the tokenizer 15 may be configured to call up a package known as TiMBL (Tilburg Memory Based Learner). … Cached
[PDF] Experiences from the spoken Dutch corpus project [PDF] from kuleuven.be N Oostdijk, W Goedertier, F Van Eynde… – Proceedings of …, 2002 – yum.ccl.kuleuven.be … MBT 86.5 89.4 91.2 92.0 94.3 95.6 maxent 83.6 89.4 90.1 92.6 95.2 – Brill 83.3 86.3 87.9 89.9 – – Timbl combiner 94.2 94.3 94.3 95.6 96.2 96.6 … Unraveling the prosodic mechanisms is also of great importance for the further development of human-machine dialog systems. … Cited by 88 – Related articles – View as HTML – All 25 versions
[PDF] Learning Computational Grammars TMR Project Nr. ERBFMRXCT980237 Final Report [PDF] from ua.ac.be J Nerbonne – 2002 – cnts.ua.ac.be … Tiubingen colleague SandraKiubler visited Antwerp in December 1999 for an introduction to the memory-based learning software TiMBL. 11 … RobKoeling worked previously on the grammar of a natural language processing module of a spoken dialogue system.HisPhD research … Related articles – View as HTML – All 6 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 … 47 3.2 The Oslo Corpus of Tagged Norwegian Texts . . . . . 48 3.3 The Oslo-Bergen tagger . . . . . 51 3.4 TiMBL . . . . . 53 3.5 Zhang Le’s maximum entropy modelling toolkit . . . . . 54 3.6 SVMlight . . … Cited by 1 – Related articles – View as HTML – All 4 versions
Combining machine learning and rule-based approaches in Spanish syntactic generation [PDF] from tdx.cat MT Melero Nogués – 2006 – tdx.cat Page 1. 1 Barcelona, 2006 Ph.D. Dissertation To obtain the Ph.D. degree in the Universitat Pompeu Fabra Directed by: Antoni Badia i Cardús Combining Machine Learning and Rule-Based Approaches in Spanish Syntactic Generation Maria Teresa Melero Nogués … Cited by 4 – Related articles – All 11 versions
[PDF] Domain action classification and argument parsing for interlingua-based spoken language translation [PDF] from cmu.edu CT Langley – 2003 – lti.cs.cmu.edu … 94 3.1 Preparation of Training Data ….. 95 3.2 Testing TiMBL Parameter Settings ….. … 121 4.2.2.1 TiMBL ….. … Cited by 4 – Related articles – View as HTML – All 8 versions
Knowledge-lean approaches to metonymy [PDF] from ed.ac.uk Y Peirsman – 2005 – era.lib.ed.ac.uk … It is clear that metonymy is an extremely frequent phenomenon in everyday language, and many NLP tasks such as machine translation or dialogue systems have to be well- equipped to handle it correctly. They have to be able to spot a metonymy, and often also to interpret it. … Related articles – All 5 versions
[PDF] Universiteit van Tilburg Faculteit der Letteren juni 2003 [PDF] from uvt.nl C Wieme – 2003 – arno.uvt.nl … In deze scriptie worden een aantal evaluatiemethoden onderzocht: 1.) Een evaluatiemethode om twee systemen te vergelijken 2.) PARADISE, een generiek evaluatie-model 3.) TiMBL, een alternatieve methode om user satisfaction te voorspellen. … 30 4.6 Berekeningen TiMBL . … Related articles – View as HTML
[PDF] Corpus-based semantic categorisation for anaphora resolution [PDF] from uib.no U Eiken – 2005 – hf.uib.no … 73 4.3 Step III: Using concept groups in TiMBL 74 … is called anaphora resolution and its computer implementation is relevant in many NLP applications, such as machine translation, automatic abstracting, dialogue systems, question answering and information extraction. … Cited by 2 – Related articles – View as HTML – All 3 versions
[PDF] GRADUATE COMMITTEE APPROVAL [PDF] from byu.edu M Emonts – 2002 – contentdm.lib.byu.edu Page 1. MEMORY-BASED TONE RECOGNITION OF CANTONESE SYLLABLES by Michael Emonts A thesis submitted to the faculty of Brigham Young University in partial fulfillment of the requirements for the degree of Master of Arts Department of Linguistics … Related articles – View as HTML – All 4 versions
[PDF] Language Technology for the Lazy [PDF] from kth.se J SJÖBERGH – 2006 – f.kth.se … NLP is when a computer does something with a language normally used by humans, such as English or Swedish. NLP is of course a very large research area, including such diverse things as speech recognition, machine translation, text categorization and dialogue systems. … Related articles – View as HTML – All 18 versions
[BOOK] Memory-based parsing S Kübler – 2004 – books.google.com … Such a parser is an indispensable prerequisite for the successful design of dialogue systems. 1.2 Machine Learning Machine learning (ML) is a term which is used in two different scientific disciplines: computer science and cognitive science. … Cited by 6 – Related articles – Library Search – All 4 versions
Samisk sprakteknologi T Trosterud – Nordisk sprogteknologi-Arbog for Nordisk …, 2003 – books.google.com … V: Multilinguality and general applications: • Semantically annotated corpora • Information retrieval and extraction • Machine translation systems; translation of NPs and simple sentences • Dialog systems • Multilingual lexical-semantic knowledge base • Language learning … Cited by 1 – Related articles