RBMT (Rule-Based Machine Translation) & Dialog Systems

Notes:

Rule-based machine translation (RBMT) is a type of machine translation that uses a set of linguistic rules and constraints to translate text from one language to another. In contrast to other machine translation approaches, such as statistical machine translation, RBMT relies on the use of explicit rules and constraints to guide the translation process, rather than relying on large collections of translated text to learn and generate translations.

In the context of dialog systems, RBMT can be used to enable the system to understand and generate text in multiple languages. For example, if a dialog system is designed to assist users with tasks that involve text in multiple languages, RBMT can be used to translate text from one language to another, allowing the system to understand and respond to user queries in multiple languages.

In addition to its use in multi-lingual dialog systems, RBMT can also be used to improve the accuracy and fluency of translations generated by other machine translation approaches. For example, RBMT can be used to post-process translations generated by statistical machine translation, applying a set of linguistic rules and constraints to improve the quality of the generated translations.

Rule-based machine translation (RBMT) and knowledge-based machine translation (KBMT) are two different approaches to machine translation that rely on different techniques and technologies to translate text from one language to another. The main difference between RBMT and KBMT is the use of external knowledge sources. While RBMT systems rely solely on predefined rules and patterns, KBMT systems use external knowledge sources to provide additional context and improve the accuracy of the translation.

Knowledge-based machine translation (KBMT) is a type of machine translation that relies on the use of external knowledge sources to improve the accuracy and fluency of the translation. KBMT systems typically use a combination of rule-based and statistical techniques to translate text, but they also draw on external knowledge sources, such as dictionaries, encyclopedias, and specialized databases, to provide context and improve the translation.

One of the main advantages of KBMT is that it can produce more accurate and natural-sounding translations compared to purely rule-based or statistical machine translation systems. This is because KBMT systems are able to use external knowledge sources to better understand the context and meaning of the source text, which can help to disambiguate words and phrases and improve the overall quality of the translation.

KBMT systems are often used in specialized domains, such as legal or technical translation, where the use of specialized terminology and domain-specific knowledge is important for producing accurate translations. They are also used in machine translation systems that are designed to translate text from one language to multiple languages, as the use of external knowledge sources can help to ensure that the translations are consistent and accurate across languages.

Wikipedia:

Rule-based machine translation

Lattice parsing to integrate speech recognition and rule-based machine translation S Köprü, A Yazici – Proceedings of the 12th Conference of the European …, 2009 – dl.acm.org … Lattice Parsing to Integrate Speech Recognition and Rule-Based Machine Translation Selçuk Köprü AppTek, Inc. … Abstract In this paper, we present a novel approach to integrate speech recognition and rule- based machine translation by lattice pars- ing. … Cited by 4 Related articles All 14 versions

Machine Translation L Specia, Y Wilks – The Oxford Handbook of Computational …, 2016 – oxfordhandbooks.com Machine Translation (MT) is and always has been a core application in the field of natural-language processing. It is a very active research area and it has been attracting significant commercial interest, most of which has been driven by the deployment of corpus-based, statistical …

Arabic-Malay Machine Translation Using Rule-Based Approach AJ Alsaket, MJ Ab Aziz – Journal of Computer Science, 2014 – search.proquest.com … Machine Translation (MT) is officially defined as the use of computers to translate messages in the form of text or speech from one natural language (human language) into … Has implemented a machine translation system for Arabic to Malay language for a dialogue system. … Cited by 1 Related articles All 3 versions

Two decades of Lithuanian HLT R Marcinkevi?ien? – 2009 – dspace.ut.ee … The most recent jointly developed tool was a rule-based machine translation system for the translation of English internet texts into Lithua- nian http://www.vertimas.vdu … semantically annotated corpora, – information retrieval and extraction, – dialogue systems, – language learning … Cited by 1 Related articles All 7 versions

TURKOISE: a Mechanical Turk-based Tailor-made Metric for Spoken Language Translation Systems in the Medical Domain M Starlander – 2014 – archive-ouverte.unige.ch … Popescu-Belis et al, 2004) that automatic metrics like BLEU are often not well suited for rule-based machine translation (RBMT) systems … have been involved in tasks ranging from labelling to transcription and have recently moved to spoken dialogue systems evaluation (Jurcícek … Related articles All 7 versions

Measuring User Acceptability of Machine Translations to Diagnose System Errors B Hui – Citeseer … Towards Developing General Models of Usability with PARADISE. Natural Language Engineering: Special Issue on Best Practice in Spoken Dialogue Systems, 2000. ? Y. Wilks. Systran: It obviously works but how much can it be improved? J. Newton (ed.), … Related articles All 2 versions

ASR Normalization for Machine Translation H Huang, C Feng, J Wang… – … Human-Machine Systems …, 2010 – ieeexplore.ieee.org … That is especially for RBMT (rule-based machine translation), since RBMT needs to parse the source language. … Reference [2] described a SDS (Spoken Dialogue Systems) toolkit which deals with Chinese dialogue and makes it easy for not only experts, but also lay people to … Cited by 2 Related articles All 5 versions

A Survey on Decoding Sanskrit Language into English: Using Morphological Analysis T Somwanshi, A Shaikh, P Shete, S Shelar, P DYPIET – ijarcet.org … Syntactic analysis can be utilized for various purpose like, developing a punctuation resolver, dialogue systems with a natural language … Rule Based Machine Translation from English to Malayam Rajan, R; Sivan, R; Ravindran, R; Sornan, KP; Advances in Computing, Control … Related articles

Consistent annotation of eurowordnet with the top concept ontology J Alvez, J Atserias, J Carrera, S Climent… – Proceedings of Fourth …, 2008 – adimen.si.ehu.es … Some NLP systems, such as knowledge-based Machine Translation systems usually include some kind of decision making (eg transfer module, PP … to implement selectional restrictions to limit the search space when parsing and to perform type-coercion in a dialogue system. … Cited by 23 Related articles All 4 versions

Lexical functions in actual NLP-applications JD Apresjan, IM Boguslavsky, LL Iomdin… – Selected lexical and …, 2007 – books.google.com Page 222. Lexical Functions in Actual NLP-Applications Jury D. Apresjan, Igor M. Boguslavsky, Leonid L. lomdin & Leonid L. Tsinman Russian Academy of Sciences, Moscow For Igor Mel’cuk, from his heirs and disciples, with … Cited by 19 Related articles

Bootstrapping a statistical speech translator from a rule-based one M Rayner, P Estrella, P Bouillon – … Rule-Based Machine …, 2011 – journals.linguisticsociety.org Page 1. Linguistic Issues in Language Technology – LiLT Submitted, February 2012 Bootstrapping a Statistical Speech Translator From a Rule-Based One Manny Rayner Pierrette Bouillon Paula Estrella Yukie Nakao Gwen Christian … Cited by 2 Related articles All 9 versions

Design of a hybrid high quality machine translation system K Eberle, J Geiß, M Ginestí-Rosell, B Babych… – Proceedings of the …, 2012 – dl.acm.org … has been around for about 20 years, and for roughly half of this time SMT and the ‘traditional’ Rule-based Machine Translation (RBMT) have … by humans and can be exploited for a lot of applications outside of translation (dictionaries, text understanding, dialogue systems, etc.). … Cited by 5 Related articles All 17 versions

On whose shoulders? Y Wilks – Computational Linguistics, 2008 – MIT Press … At Stanford as a post-doc, I was on the same corridor as Winograd, just arrived from MIT; Schank, then starting to build his Conceptual Dependency empire; and Colby and his large team building the PARRY dialogue system, which included Larry Tesler, later the Apple software … Cited by 5 Related articles All 13 versions

Speech-to-speech translation to support medical interviews JASG Rodrigues – 2013 – repositorio.ul.pt Page 1. UNIVERSIDADE DE LISBOA Faculdade de Ciências Departamento de Informática SPEECH-TO-SPEECH TRANSLATION TO SUPPORT MEDICAL INTERVIEWS Jo˜ao Ant´onio Santos Gomes Rodrigues PROJETO MESTRADO EM ENGENHARIA INFORM´ATICA … Cited by 2 Related articles All 3 versions

[BOOK] Building a Slovenian-English Language Pair Speech-to-Speech Translation System JŽ Gros, F Miheli?, M Žganec – 2006 – eurasip.org … Galaxy Communicator is a distributed, message-based, hub- and-spoke infrastructure optimized for constructing spoken dialogue systems. … SMT (statistical machine translation), EBMT (example-based machine translation) and RBMT (rule-based machine translation) methods. … Related articles

ACL Lifetime Achievement Award OW Shoulders – Computational Linguistics, 2008 – Citeseer … At Stanford as a post-doc, I was on the same corridor as Winograd, just arrived from MIT; Schank, then starting to build his Conceptual Dependency empire; and Colby and his large team building the PARRY dialogue system, which included Larry Tesler, later the Apple software … Related articles All 2 versions

Pretreatment for speech machine translation X Zhang, C Feng, H Huang – International Conference on Computational …, 2010 – Springer … That is especially for RBMT (rule-based machine translation), since RBMT needs to parse the source language. … Reference [2] described a SDS (Spoken Dialogue Systems) toolkit which deals with Chinese dialogue and makes it easy for not only experts, but also lay people to … Related articles All 4 versions

Language report for Catalan (English version) MA Moreno Bilbao, N Bel, E Revilla Espí, E Garcia… – 2011 – upcommons.upc.edu … Language applications, voice-based user interfaces and dialogue systems are traditionally found in highly specialised domains, and they often … Some of the leading rule-based machine translation systems have been under constant development for more than twenty years. … Related articles

Evolutionary algorithms in natural language processing L Bungum, B Gambäck – Norwegian artificial intelligence symposium (NAIS), 2010 – sics.se … Major application areas are spelling and grammar checking, machine translation, text summarization, ques- tion answering systems, and dialogue systems. … 3.2.3. Rule-Based Machine Translation Rule-based translation (RBT) systems have an important role in the history of MT … Cited by 6 Related articles

[BOOK] Routledge Encyclopedia of Translation Technology S Chan – 2014 – books.google.com … Mikel L. Forcada 9 Pragmatics-based machine translation David Farwell and Stephen Helmreich Rule-based machine translation Yu Shiwen … Dr Ahrenberg’s research spans several areas of NLP including dialogue systems, grammar formalisms and syntactic analysis, corpus … Cited by 8 Related articles All 2 versions

Statistical post-editing and quality estimation for machine translation systems H Bechara – 2014 – doras.dcu.ie … 7 2 Machine Translation, Post-Editing, and Quality Estimation 8 2.1 Rule-Based Machine Translation . . . . . … quality 8 Page 18. 2.1 Rule-Based Machine Translation Rule-based systems construct translations by using usually hand-crafted linguistic … Cited by 1 Related articles All 2 versions

Machine translation and type theory A Ranta – Epistemology versus Ontology, 2012 – Springer Cited by 1 Related articles All 7 versions

[BOOK] Natural language processing of semitic languages I Zitouni – 2014 – Springer … 2 Rule-Based Machine Translation….. 202 6.2. … Prior to IBM, Imed was a researcher at Bell Laboratories, Lucent Technologies, for almost half dozen years working on language modeling, speech recognition, spoken dialog systems, and speech understanding. … Cited by 10 Related articles All 6 versions

A brief state of the art of CNLs for ontology authoring H Safwat, B Davis – International Workshop on Controlled Natural …, 2014 – Springer … the core advantage of GF is its multilingualism in that its primary task is domain specific knowledge based Machine Translation (MT) of … GF applications range from mathematical proofing, dialog systems, patent translation [19], multilin- gual wikis and multilingual generation in … Cited by 7 Related articles All 10 versions

CeBIT 2009 TF SerCHo – 2008 – dfki.de Page 1. NEWSLETTER GERMAN RESEARCH CENTER FOR ARTIFICIAL INTELLIGENCE © 2008 DFKI I ISSN – 1615 – 5769 I 22th edition RESEARCH LABS DFKI – 20 years of innovation New research department: Augmented Vision THESEUS MEDICO 2/2008 … All 3 versions

Investigating the role of machine translated text in ASR domain adaptation: Unsupervised and semi-supervised methods H Cucu, L Besacier, C Burileanu… – … Speech Recognition and …, 2011 – ieeexplore.ieee.org … Several approaches for language portability (French to Italian) of dialogue systems are being investigated in [5]. In a previous work [6 … The methodology used in [7] is fundamentally different than ours, because rule-based machine translation techniques are utilized for language … Cited by 12 Related articles All 6 versions

Hindi generation from interlingua S Singh, M Dalal, V Vachani… – Proc. Machine …, 2007 – mt-archive.info … NLG research in recent times is witnessing a flurry of activities in Dialogue Systems in which the generation component addresses … The reasons for traversing a trodden path are- (i) the gradual re-emergence of knowledge based machine translation that needs generating target … Cited by 21 Related articles All 10 versions

Translation rescoring through recurrent neural network language models Á PERIS ABRIL – 2014 – riunet.upv.es … language generation and understanding, speech recognition, topic segmentation and recognition, information retrieval, dialog systems, question answering … Rule-Based Machine Translation (RBMT) was one of the firsts approaches used in machine translation (the first RBMT … Related articles

Advanced Series on Artificial Intelligence: Volume NG BOURBAKIS – World Scientific … AAAI-S6, 1986. Bobrow. DG, Kaplan, RM, Kay, M., Norman, DA, Thompson, H. and Winograd, T., “GUS, a frame-driven dialog system”, >lrli/. Intell. 8 (1977) 155-173. Bobrow, DG and Winograd, T., “An overview of KRL; A knowledge representation language”, Cognitive Sci. …

Chinese in the Grammatical Framework: Grammar, Translation, and Other Applications A Ranta, Y Tian, Q Haiyan – ACL-IJCNLP 2015, 2015 – aclweb.org … (Zimina, 2012). Adapting a GF grammar to a new setting, such as a dialogue system or a domain- specific translator, can be accomplished in a few days (Perera and Ranta, 2007; Ranta et al., 2012). … Apertium: a free/open-source platform for rule- based machine translation. … Related articles All 7 versions

[BOOK] Speech Technology F Chen, K Jokinen – 2010 – Springer … 2 Platform Enterprise Edition JAWS Job access with speech JSP Java Server Pages KBMT Knowledge-Based Machine Translation K-NN K … OAA Open Agent Architecture OCR Optical Character Recognition Software PARADISE PARAdigm for DIalogue System Evaluation PCA … Cited by 6 Related articles All 4 versions

Computational linguistic tools and machine translation system for Kannada language PJ Antony – 2012 – nlp.amrita.edu … 107 3.1.5.5 MT 107 3.1.5.6 Dialogue Systems 107 3.1.5.7 Sentence Understanding 107 … 243 8.1 Challenges for English – Kannada Machine Translation 244 8.2 Proposed Rule- Based Machine Translation System 245 8.2.1 Creation of English-Kannada bilingual dictionary 246 … Related articles All 5 versions

Language Technology Support for Polish G Rehm, H Uszkoreit – The Polish Language in the Digital Age, 2012 – Springer … Recognition 5: Speech-based dialogue system stricting the range of possible user utterances to a lim- ited set of keywords, or manually creating language models that cover a large range of natural language ut- terances. … machine translation began to grow. … All 2 versions

[BOOK] Toward computational processing of less resourced languages: Primarily experiments for Moroccan Amazigh language FA Allah, S Boulaknadel – 2012 – cdn.intechopen.com … The rule-based machine translation (RBMT) approach is described as interlingual or transfer-based machine translation. … The Human-machine dialogue systems can be represented as a four process: Speech recognition process to transcribe sentences spoken into written text … Cited by 5 Related articles All 4 versions

The limits of machine translation MW Madsen – Center for Language Technology, Univ. of …, 2009 – vantage-siam.com Page 1. The Limits of Machine Translation Mathias Winther Madsen Thesis submitted for the degree of Master in Information Technology and Cognition Department of Scandinavian Studies and Linguistics, Faculty of Humanities, University of Copenhagen … Cited by 17 Related articles All 2 versions

Dynamic QA Systems Using Knowledgebase S Hu – ru.nl … systems. 1.3. Interactive Dialogue System Another famous question-answering system is human-machine dialogue system. In 1950, AlanTuring put forward an idea to treat Q&A as the test toward machine intelligence. He assumed … Related articles

Using Ontology-Based Context in the Portuguese-English Translation of Homographs in Textual Dialogues D Moussallem, R Choren – arXiv preprint arXiv:1510.01886, 2015 – arxiv.org … This paper introduces a novel approach to tackle the existing gap on message translations in dialogue systems. Currently, submitted messages to the dialogue systems are considered as isolated sentences. … Machine Translation, Semantic Web, Ontologies, Dialogue Systems … Related articles All 6 versions

Multi-word unit processing in machine translation J Monti – 2015 – elea.unisa.it … TT Page 18. XVIII Translate, Edit, Publish TEP Translation Memory TM Regular Expression RegEx Rule-based Machine Translation RBMT Statistical Machine Translation SMT Example-based Machine Translation EBMT Phrased-based Statistical Machine Translation PB-SMT … Related articles All 2 versions

Task-based Teaching of Computer-aided Translation in a Progressive Manner JX Liu – Conference Chairs and Editors of the Proceedings – academia.edu … quality checking, 77, 81, 83. quality estimation, 7, 10, 133, 163, 221, 227. questionnaires, 7, 8, 10, 110, 177, 179. rule-based machine translation, 25, 76, 100. search tool, 57, 250. semantic similarity, 86, 87, 88, 89. Sketch Engine, 7, 10, 129, 130, 131, 132, 254. … Related articles All 4 versions

Human Translation Evaluation and its Coverage by Automatic Scores M Vela, AK Schumann, A Wurm – Automatic and Manual Metrics for …, 2014 – academia.edu Page 25. 19 Full Papers Page 26. 20 Human Translation Evaluation and its Coverage by Automatic Scores Mihaela Vela, Anne-Kathrin Schumann, Andrea Wurm Department of Applied Linguistics, Translation and Interpreting … Cited by 3 Related articles All 6 versions

Reuse of free resources in machine translation between nynorsk and bokmål K Unhammer, T Trosterud – 2009 – rua.ua.es … JA Pérez-Ortiz, F. Sánchez-Mart?nez, FM Tyers (eds.) Proceedings of the First International Workshop on Free/Open-Source Rule-Based Machine Translation, p. 35–42 Alacant, Spain … The Fyn- tour Multilingual Weather and Sea Dialogue System–Description and Assessment. … Cited by 6 Related articles All 10 versions

Review of research on speech technology: main contributions from Spanish research groups R San Segundo Hernández… – Journal of Speech …, 2011 – oa.upm.es … This section includes spoken language understanding and translation, spoken dialogue systems, voice-activated question answering (QA … approaches have achieved performance levéis comparable to those achieved by knowledge-based Machine Translation (MT) algorithms … Cited by 2 Related articles All 5 versions

Toward Semantic Machine Translation J Andreas – 2012 – core.ac.uk … design. One notable success in rule-based machine translation is SYSTRAN, which began … who learn a Markov logic network in order to induce semantic forms [PD09]; Artzi et al.’s semantic parser bootstrapped from transcripts of interactions with a dialog system [AZ11]; … Cited by 1 Related articles All 3 versions

A Parallel Corpus Based Translation Using Sentence Similarity SW Ruoro – 2014 – erepository.uonbi.ac.ke … vii ACRONYMS AND ABBREVIATIONS MT Machine translation RBMT Rule-based machine translation SMT Statistical machine translation EBMT Example-based machine translation … applications in areas such as text mining, Web page retrieval, and dialogue systems. Existing …

Spoken Language Translation F Ehsani, R Frederking, M Rayner, P Bouillon – Speech Technology, 2010 – Springer … Interlingua-based MT is one category of a class of MT known as knowledge-based machine translation (KBMT), so called because it utilizes significant linguistic and domain-dependent knowledge sources. … Speaking Multilingual Interactive Natural Dialog System. References. … Cited by 1 Related articles All 7 versions

CNLs for the semantic web: a state of the art H Safwat, B Davis – Language Resources and Evaluation, 2016 – Springer Page 1. ORIGINAL PAPER CNLs for the semantic web: a state of the art Hazem Safwat1 • Brian Davis1 © Springer Science+Business Media Dordrecht 2016 Abstract One of the core challenges for building the semantic web … Cited by 1 Related articles

Combining multiple parallel streams for improved speech processing JTR de Sousa Miranda – 2015 – l2f.inesc-id.pt … RASTA Relative Spectral Transform RBMT Rule-Based Machine Translation RTF Real-time Factor … retrieval engine to locate the most relevant documents. • Spoken dialog systems [82, 54] may collaborate with a user in spoken language to com- plete a certain task. … Related articles All 2 versions

Semantic transfer in Verbmobil A Copestake – 2011 – scidok.sulb.uni-saarland.de … This is also discussed in more detail below. Knowledge-based machine translation (eg Nirenburg and Raskin, 1988). This approach em- phasises reliance on domain knowledge in carrying out translation. … This is particularly true of spoken dialogue systems. … Cited by 16 Related articles All 5 versions

Pgf: A portable run-time format for type-theoretical grammars K Angelov, B Bringert, A Ranta – Journal of Logic, Language and …, 2010 – Springer Page 1. J Log Lang Inf (2010) 19:201–228 DOI 10.1007/s10849-009-9112-y PGF: A Portable Run-time Format for Type-theoretical Grammars Krasimir Angelov · Björn Bringert · Aarne Ranta Published online: 15 December 2009 © Springer Science+Business Media BV 2009 … Cited by 23 Related articles All 13 versions

Annex I-“Description of Work” UPC Spain – molto-project.eu … 2007), and mathematical exercises (Caprotti 2006, in the European project WebALT16). Other projects start with semantic modelling work to build meaning representations from scratch, most notably ones for dialogue systems (Ranta and Cooper 2004, Bringert & al. … Related articles All 2 versions

[BOOK] GF Runtime System K Angelov – 2009 – Citeseer Page 1. THESIS FOR THE DEGREE OF LICENTIATE OF ENGINEERING GF Runtime System Krasimir Angelov Department of Computer Science and Engineering Chalmers University of Technology and Göteborg University SE-412 96 Göteborg Sweden Göteborg, 2009 Page 2 … Cited by 1 Related articles All 6 versions

Human tutorial instruction in the raw Y Gil – ACM Transactions on Interactive Intelligent Systems ( …, 2015 – dl.acm.org … new words. Other corpus analyses reveal the intrinsic ambiguity of natural language in expressing the same kind of information, particularly in analyses done to develop speech recognition and dialogue systems. Despite the … Cited by 3 Related articles All 6 versions

On applying controlled natural languages for ontology authoring and semantic annotation BP Davis – 2013 – aran.library.nuigalway.ie Page 1. Provided by the author(s) and NUI Galway in accordance with publisher policies. Please cite the published version when available. Downloaded 2016-05-27T17:09:24Z Some rights reserved. For more information, please see the item record link above. … Cited by 3 Related articles