Realizers In Natural Language Processing

Notes:

Realization, also spelled as “realisation,” is a subtask of Natural Language Generation (NLG) which involves taking a syntactic representation of a message or idea and converting it into an actual text in a human language. Realization is the process of generating natural language text from a structured representation, like a formal language or a database. It’s a step in the NLG process that comes after the generation of the structured representation, which is often in the form of a semantic representation or a logical form.

A realizer, also spelled as “realiser,” is a system or tool that is used to perform the task of realization in Natural Language Generation (NLG). It takes a structured representation of a message or idea, such as a formal language or a database, and converts it into an actual text in a natural language, like English or Japanese. A realizer can be a rule-based system, a statistical model, or a neural network-based model. It can also be a standalone tool or a module within a larger NLG system. The realizer uses various techniques like grammar rules, templates, and language models to generate natural language text that is coherent, grammatically correct and semantically appropriate. It is used in various applications such as text generation, chatbot, automated writing, and many more.

Behavior realizer, also known as a behavior generator, is a system or tool that is used to generate behavior for artificial agents such as robots, virtual characters, and autonomous vehicles. These agents are designed to interact with their environment and perform tasks, and the behavior realizer generates the actions and responses of the agent in response to the environment and user inputs.
Chart realizer, such as OpenCCG (Open-Source Combinatory Categorial Grammar), is a system that generates sentences in natural language from logical forms as input. Logical forms are a way of representing the meaning of a sentence in a formal, structured way, often using predicate logic.
DLUs are units of meaning that capture more information than traditional surface-level words, and in DSyntS, the dependency tree is constructed using DLUs as the nodes rather than surface-level words. This representation allows for more accurate understanding of the sentence structure and meaning.
DSynt stands for “deep syntactic” and it refers to the process of syntactic analysis, which is the process of analyzing the syntactic structure of a sentence. Deep syntactic analysis can be used as a step in the NLG process before the realization, to provide a more accurate and rich representation of the meaning of the text, which can be used as input for the realization process.
DSyntR stands for “deep syntactic representation” which is a method of representing the syntactic structure of a sentence in a way that captures more linguistic information than a traditional surface-level parse tree. Deep syntactic representation can be used as a rich and accurate representation of the meaning of the text, which can be used as input for the realization process.
DSyntS stands for “deep syntactic structure” which is a dependency tree that is labeled with deep lexical units (DLUs). Deep syntactic structure can be used as a rich and accurate representation of the meaning of the text, which can be used as input for the realization process.
Linguistic realizer is a system or tool that generates natural language text from a structured representation of the message or idea. Linguistic realizer takes various forms of input such as logical forms, semantic representation, or formal language and converts it into natural language text that is coherent, grammatically correct, and semantically appropriate. The generated text should be understandable by humans. Linguistic realizers use various techniques such as grammar rules, templates, and language models to produce natural language text. These systems are widely used in natural language generation (NLG) applications such as text generation, chatbots, automated writing, and many more.
Multimodal realizer is a system or tool that generates both verbal and nonverbal behavior for virtual humans, robots or other artificial agents. The realizer uses a multimodal representation of the behavior, which includes verbal and nonverbal cues such as speech, facial expressions, gestures, and body movements. The realizer generates the behavior in real-time, so the virtual human can interact with the environment and users in a natural and believable way.
Sentence realizer is a system or tool that generates sentences in natural language from a structured representation, such as syntax or logical forms, as input. Sentence realizer takes various forms of input such as deep syntactic structure, logical forms, semantic representation, or formal language and converts it into natural language text that is coherent, grammatically correct, and semantically appropriate. The generated text should be understandable by humans. Sentence realizers use various techniques such as grammar rules, templates, and language models to produce natural language text. These systems are widely used in natural language generation (NLG) applications such as text generation, chatbots, automated writing, and many more.
Surface realizer is a system or tool that generates natural language text from abstract, syntactic representations of each sentence. It takes the syntactic structure of a sentence, represented in a formal, abstract way, such as a parse tree or dependency tree, and converts it into a coherent and grammatically correct sentence in a natural language.

Resources:

openccg.sourceforge.net .. an open-source realizer with a number of nice features, including the ability to use statistical language models to make realization decisions.
cogentex.com/technology/realpro .. RealPro is a text generation “engine” that performs syntactic realization — i.e., the transformation of abstract syntactic specifications of natural language sentences (or phrases) into their corresponding surface forms.
simplenlg.googlecode.com .. a document realizing engine with an API which is intended to be simple to learn and use, and focused on limiting scope to only finding the surface area of a document.
kpml .. the oldest realizer, which has been under development in different guises since the 1980s.
cs.bgu.ac.il/surge .. a realiser which was widely used in the 1990s, and is still used in some projects today.

Wikipedia:

Generating natural language texts from business process models H Leopold, J Mendling, A Polyvyanyy – Advanced Information Systems …, 2012 – Springer … Text Planning Sentence Planning Realization RealPro- Realizer Natural Language Text … Page 11. 74 H. Leopold, J. Mendling, and A. Polyvyanyy license costs, generation speed and Java compatibility, we decided to utilize the DSynT-based realizer RealPro from CoGenTex. … Cited by 12 Related articles All 16 versions

Two-stage stochastic natural language generation for email synthesis by modeling sender style and topic structure YN Chen, AI Rudnicky – INLG 2014, 2014 – aclweb.org … 5 Surface Realization Our surface realizer has four elements: building language models, generating text content, scoring email candidates, and filling slots … In Proceedings of the 2009 Conference on Empirical Methods in Natural Language Processing, volume 2, pages 919–928 … Cited by 1

Natural Language Generation as Incremental Planning Under Uncertainty: Adaptive Information Presentation for Statistical Dialogue Systems V Rieser, O Lemon, S Keizer – Audio, Speech, and Language …, 2014 – ieeexplore.ieee.org … B. NLG Realiser We implemented a NLG realizer for the chosen IP structures and attribute choices, in order to realize the wizards’ choices in real time. This generator is based on data from the stochastic sentence planner SPaRKy [39]. … Related articles

Natural Language Generation and Summarization at RALI G Lapalme – ENLG 2013, 2013 – aclweb.org … JSREAL is a work in progress describing a French text realizer in Javascript that can be easily embedded in a web … In Bernadette Sharp and Michael Zock eds., editors, Proceedings of NLPCS 2010 (Natural Language Processing and Cognitive Science), pages 79–89, Funchal … Cited by 1 Related articles All 11 versions

JSREAL: A Text Realizer for Web Programming N Daoust, G Lapalme – Language Production, Cognition, and the Lexicon, 2015 – Springer … Lavoie, B., & Rambow, O. (1997). A fast and portable realizer for text generation systems. In Proceedings of the Fifth Conference on Applied Natural Language Processing (pp. 265–268). Portet, F., Reiter, E., Gatt, A., Hunter, J., Sripada, S., Freer, Y., et al. (2009). …

Deep Natural Language Processing for Italian Sign Language Translation A Mazzei, L Lesmo, C Battaglino, M Vendrame… – AI* IA 2013: Advances in …, 2013 – Springer … the sentence planner is called SentenceDesigner, that builds a tree representing the generic LIS lexical items and some generic syntac- tic relations among them; the realizer is implemented a formal grammar for LIS by Page 5. Deep Natural Language Processing for Italian … Related articles All 2 versions

Natural Language Generation for Nature Conservation: Automating Feedback to Help Volunteers Identify Bumblebee Species. S Blake, A Siddharthan, H Nguyen, N Sharma… – …, 2012 – homepages.abdn.ac.uk … Both the linguistic and structural realisations are performed by using functionalities provided by the SIMPLENLG realiser library (Gatt and Reiter, 2009). … Goldberg, E., Driedger, N., and Kittredge, R. (1994). Using natural-language processing to produce weather forecasts. … Cited by 1 Related articles All 10 versions

SemScribe: Natural Language Generation for Medical Reports. S Varges, H Bieler, M Stede, LC Faulstich, K Irsig… – LREC, 2012 – ling.uni-potsdam.de … We use a sentence realizer called XbarGen (Sawitzky, 2011) that involves the notion of Xbar-Theory and topological fields of German. … In Re- cent Advances in Natural Language Processing (RANLP vol.II), pages 221–243, Amsterdam and Philadelphia. John Benjamins. … Related articles All 3 versions

BEETLE II: Deep Natural Language Understanding and Automatic Feedback Generation for Intelligent Tutoring in Basic Electricity and Electronics M Dzikovska, N Steinhauser, E Farrow, J Moore… – International Journal of …, 2014 – Springer … The Basic Electricity and Electronics Tuto- rial Learning Environment (BEETLE II) advances the state of the art in dynamic adaptive feedback generation and natural language processing (NLP) by extending symbolic NLP techniques to support unrestricted student natural …

Context-Sensitive Natural Language Generation: From Knowledge-Driven to Data-Driven Techniques N Dethlefs – Language and Linguistics Compass, 2014 – Wiley Online Library … The comparison will place a particular focus on the influence that the general shift from knowledge-driven to data-driven methods experienced in Natural Language Processing in the last decades has had on research in context-sensitive NLG. … Related articles All 2 versions

Assessing the influence of personal preferences on the choice of vocabulary for natural language generation R Hervás, V Francisco, P Gervás – Information Processing & Management, 2013 – Elsevier … generation systems is Michael Elhadad’s PhD Thesis (Elhadad, 1992), which addresses this particular problem for a system employing the FUF surface realizer (Elhadad, 1993). … There is a strong tradition of corpus-based evaluation in the field of Natural Language Processing. … Cited by 1 Related articles All 5 versions

Representation of linguistic and domain knowledge for second language learning in virtual worlds A Denis, I Falk, C Gardent… – LREC-The eighth …, 2012 – hal.archives-ouvertes.fr … There has been much debate, both theoretical and practical, on how to link ontologies and lexicons in natural language processing (NLP) applications. … has been selected, sentence generation is carried out using the GenI (Gardent and Kow, 2007) sur- face realiser, a generic … Cited by 1 Related articles All 16 versions

Natural Language Generation in the context of Multimodal Interaction in Portuguese JC Pereira, AJS Teixeira, JS Pinto – Electrónica e Telecomunicações, 2013 – revistas.ua.pt … SimpleNLG operates with three modules [35]: Lexicon/ morphology system – decides what words should be used; Realiser – generates texts … [9] E. Reiter, “Natural Language Generation,” The Handbook of Computational Linguistics and Natural Language Processing, A. Clark … Related articles

Integrating natural language generation and model-based reasoning for explanation generation D Matheson, GM Coghill… – … Intelligence (UKCI), 2012 …, 2012 – ieeexplore.ieee.org … For this reason we have not explored realisation-level tasks in our research and instead used a popular open-source realiser, SimpleNLG[23]. III. … Cambridge University Press, 2000. [23] E. Reiter, “Simplenlg open source realiser,” 2011, http://code.google.com/p/simplenlg/. Related articles

All together now A Hartholt, D Traum, SC Marsella, A Shapiro… – Intelligent Virtual …, 2013 – Springer … Toolkit contains modules that cover audio-visual sensing, nonverbal behavior understanding, speech recognition, natural language processing, nonverbal behavior … van Welbergen, H., Reidsma, D., Ruttkay, ZM, Zwiers, J.: Elckerlyc – A BML Realizer for continuous, multimodal … Cited by 18 Related articles All 5 versions

SPARTIQULATION–Verbalizing SPARQL queries B Ell, D Vrandecic, E Simperl – Proceedings of ILD Workshop, …, 2012 – ceur-ws.org … 1 Introduction Much research has been done to combine the fields of Databases and Natural Language Processing to provide natural language interfaces to database sys- tems [22]. … Document Planner Microplanner Surface Realizer Content determination Document structuring … Cited by 20 Related articles All 8 versions

Exploiting ontology lexica for generating natural language texts from RDF data P Cimiano, J Lüker, D Nagel, C Unger – 2013 – pub.uni-bielefeld.de … as selected by the lexicalisation com- ponent. The task of the surface realiser then is to find an appropriate syntactic tree from the parsed corpus that can be used to realise the involved concepts. An example of such a parse … Cited by 5 Related articles All 8 versions

Shakespearean spatial rules C Talbot, GM Youngblood – … of the 2013 international conference on …, 2013 – dl.acm.org … Factors, Languages, Theory Keywords Shakespeare; Hamlet; Spatial Reasoning; Natural Language Processing; Plays; BML Realizer; SmartBody; Speech Act Theory; Agent Reasoning; Agent Planning 1. INTRODUCTION Current … Cited by 5 Related articles All 4 versions

The surface realisation task: Recent developments and future plans A Belz, B Bohnet, S Mille, L Wanner… – Proceedings of the Seventh …, 2012 – dl.acm.org … Bernd Bohnet Institute for Natural Language Processing University of Stuttgart 70174 Stuttgart bohnet@ims.uni … of the surface realiser with the best performance, a surface realiser which is … coverage multilingual deep sen- tence generation with a stochastic multi-level realizer. … Cited by 8 Related articles All 9 versions

[BOOK] Conversational Interfaces O Lemon – 2012 – Springer … ability to combine multiple information sources (calendar, address book, web search, Wolfram Alpha, etc) as from its current natural language processing capabilities. … expressions for different types of users, and the use of crowd-sourced data to train a statistical surface realiser. … Cited by 10 Related articles All 7 versions

Training a statistical surface realiser from automatic slot labelling H Cuayáhuitl, N Dethlefs, H Hastie, X Liu – 2014 – macs.hw.ac.uk … Many natural language processing modules are trained on a set of labelled examples and are then ideally expected to also work on unseen inputs. … A surface realiser within an SDS typically receives a single dialogue act as input [1], which lacks the detailed syntactic and lexical … Cited by 1

Classifiers for data-driven deep sentence generation M Ballesteros, S Mille, L Wanner – Proceedings of the 8th International …, 2014 – taln.upf.edu … 2010. Broad coverage multilingual deep sentence generation with a stochastic multi- level realizer. … In Proceedings of the 2008 Conference on Em- pirical Methods in Natural Language Processing (EMNLP), pages 177–185, Honolulu, Hawaii. … Cited by 2 Related articles All 2 versions

Natural Language Generation in the context of the Semantic Web N Bouayad-Agha, G Casamayor, L Wanner – Semantic Web, 2013 – IOS Press … Finally, the surface realizer is responsible for rendering each sentence plan into a sentence string. It is therefore obvious that it is the document plan- ning module and, to some extent, the sentence plan- ning module that must be able to cope with semantic representations. … Cited by 7 Related articles All 3 versions

Verbalization of 3d scenes based on natural language generation techniques V Golfinopoulos, D Makris, G Bardis, G Miaoulis… – Intelligent Computer …, 2012 – Springer Page 1. D. Plemenos and G. Miaoulis (Eds.): Intelligent Comp. Graphics 2011, SCI 374, pp. 19–39. springerlink.com © Springer-Verlag Berlin Heidelberg 2012 Verbalization of 3D Scenes Based on Natural Language Generation Techniques … Related articles All 4 versions

Spatial cues in hamlet C Talbot, GM Youngblood – Intelligent Virtual Agents, 2012 – Springer … We compared this to the use of just the play-script and some very simplistic natural language processing to block the same act of the same … to play New York [3]. We utilized a 2D simulation for our experiment (top-down view of the stage), however a BML realizer (like SmartBody … Cited by 8 Related articles All 4 versions

Creating a natural language summary from a compressed causal graph C Puente, JA Olivas, E Garrido… – IFSA World Congress …, 2013 – ieeexplore.ieee.org … After this, a factory is created which is responsible for the generation of structures and a realizer that transforms … Proceedings of the 2009 Conference on Empirical Methods in Natural Language Processing, pages 890–899, Singapore, 6-7 August 2009 ACL and AFNLP (2009). … Cited by 1 Related articles

Natural language generation and semantic web technologies N Bouayad-Agha, G Casamayor… – Semantic Web …, 2012 – semantic-web-journal.org … Fi- nally, the surface realizer is responsible for rendering each sentence plan into a sentence string. It is obvi- ously the document planning module and partially also the sentence planning module that must be able to cope with semantic representations. … Cited by 3 Related articles All 2 versions

The C2BDI Agent Architecture for Teamwork Coordination Using Spoken Dialogues between Virtual Agents and Users M Barange, A Kabil, P Chevaillier – Advances in Practical Applications of …, 2014 – Springer … Decision Making Dialogue Management Belief Revision Behavior Realiser Perception … 317 the task in progress, resource allocation etc. through the means of dialogue processing. Natural Language Processing: The ASR recognises spoken phrases with confidence level. … Cited by 1 Related articles All 2 versions

Minimal dependency length in realization ranking M White, R Rajkumar – … Methods in Natural Language Processing and …, 2012 – dl.acm.org Proceedings of the 2012 Joint Conference on Empirical Methods in Natural Language Processing and Computational Natural Language Learning, pages 244–255, Jeju … We provide here a brief overview of CCG and the OpenCCG realizer; for further details, see the works cited … Cited by 6 Related articles All 5 versions

Conditional Random Fields for Responsive Surface Realisation using Global Features. N Dethlefs, HW Hastie, H Cuayáhuitl, O Lemon – ACL (1), 2013 – macs.hw.ac.uk … They train a surface realiser based on a domain-dependent language model and use an overgeneration and ranking ap- proach. … As shown in the architecture diagram in Fig- ure 1, a CRF surface realiser takes a semantic tree as input. … Cited by 9 Related articles All 7 versions

Requirement Specifications Using Natural Language T Bures, P Hnetynka, P Kroha, V Simko – 2012 – d3s.mff.cuni.cz … 5 2 Natural Language Processing 10 2.1 Tokens . . . . . … be expensive. The cost to fix them can be up to a third of the total production cost [5]. Moreover, many system failures are attributed to poor requirements analysis [52]. 9 Page 17. Chapter 2 Natural Language Processing … Cited by 3 Related articles All 2 versions

Better Surface Realization through Psycholinguistics R Rajkumar, M White – Language and Linguistics Compass, 2014 – Wiley Online Library … Finally, the realizer determines the exact surface form of the text, tackling the interrelated tasks of ordering constituents, inflecting … Going further, compared to other natural language processing (NLP) tasks, NLG systems have to deal with deciding between competing choices at …

Creating a Cognitive Agent in a Virtual World: Planning, Navigation, and Natural Language Generation W Hewlett – 2013 – escholarship.org … generated. The vast majority of work in natural language processing and natural lan- guage generation is focused on enabling computer-human discourse; communica- … 2 Page 18. language processing, such as interpreting the wide range of utterances that a hu- man can make. … Related articles All 4 versions

Creating and Updating Personalized and Verbalized Business Process Descriptions J Kolb, H Leopold, J Mendling, M Reichert – The Practice of Enterprise …, 2013 – Springer … Words). Computational Linguistics 13(3-4), 261–275 (1987) 24. Lavoie, B., Rambow, O.: A Fast and Portable Realizer for Text Generation Systems. In: Applied Natural Language Processing, ACL, pp. 265–268 (1997) 25. Weber … Cited by 2 Related articles All 11 versions

Portuguese text generation using factored language models EM de Novais, I Paraboni – Journal of the Brazilian Computer Society, 2013 – Springer … Abstract As in many other natural language processing (NLP) fields, the use of statistical methods is now part of mainstream natural language … the systems described in [14] (using a graph-based parsing algorithm) and [15] (a grammar-based chart realiser), focused on … Cited by 9 Related articles All 4 versions

Designing Appropriate Feedback for Virtual Agents and Robots. M Lohse, H van Welbergen – Position paper at RO-MAN 2012 …, 2012 – pub.uni-bielefeld.de … disciplinary effort, re- quiring expertise in many research areas, including computer animation, perception, cognitive modeling, emotions and personality, natural language processing, speech recognition … The Realizer executes behavior specified in BML onto a (virtual) agent. … Cited by 4 Related articles All 5 versions

Natural Language Generation and Semantic Technologies K Staykova – Cybernetics and Information Technologies, 2014 – degruyter.com … of NLG is the necessity for considering much more varieties of meaning types than those traditionally addressed in Natural Language Processing. … The third module, called Surface Realiser, transforms the lexico-grammatical specification of each sentence to a sequence of …

Generating Elliptic Coordination C Gardent, S Narayan – Proceedings of the 14th …, 2013 – hal.archives-ouvertes.fr … Finally, we evaluate an existing surface realiser on the resulting dataset. We show that, after rewriting, the generator achieves a cover- age of 76% and a BLEU score of 0.74 on the elliptical data. 1 Introduction … 5 Generating Elliptic Coordination 5.1 The Surface Realiser … Related articles All 10 versions

Fully abstractive approach to guided summarization PE Genest, G Lapalme – Proceedings of the 50th Annual Meeting of the …, 2012 – dl.acm.org … They are implemented us- ing the SimpleNLG realizer (Gatt and Reiter, 2009), which takes a sentence structure and words in their root form as input and gives a sentence with re- solved agreements and sentence markers as … The Stanford Natural Language Processing Group. … Cited by 9 Related articles All 11 versions

Gennext: A consolidated domain adaptable nlg system F Schilder, B Howald, R Kondadadi – ENLG 2013, 2013 – aclweb.org … In par- ticular, GenNext reduces complexity by avoiding the necessity of having a separate document plan- ner, surface realizer, etc., and extensive … In Proceedings of the 2010 Confer- ence on Empirical Methods for Natural Language Processing (EMNLP 2010), pages 502–512 … Cited by 1 Related articles All 5 versions

Leveraging High Level Visual Information for Matching Images and Captions F Yan, K Mikolajczyk – Lecture Notes in Computer Science, 2014 – epubs.surrey.ac.uk … Owing to recent advances in computer vision (CV), natural language processing (NLP) and machine learning (ML), in- tegrated modelling of vision and language is finding more … A surface realiser is then employed to produce captions as constrained by the lexicon and grammar. …

Creating an artificially intelligent director (aid) for theatre and virtual environments C Talbot – Proceedings of the 2013 international conference on …, 2013 – dl.acm.org … Keywords Theatre; Director; Artificial Intelligence; Spatial Reason- ing; Natural Language Processing; Machine Learning; Force- Directed Graphs; Rules Engine; Shakespeare; Hamlet; BML Realizer; Agent Reasoning; Agent Planning … Related articles All 4 versions

Generation for grammar engineering C Gardent, G Kruszewski – Proceedings of the Seventh International …, 2012 – dl.acm.org … As with parsing, these approaches permit detecting under-generation in that an input for which the surface realiser fails to produce a sentence points to shortcomings either in the sur- face realisation algorithm or in the grammar/lexicon. … Cited by 2 Related articles All 15 versions

One step further towards stochastic semantic sentence generation B Bohnet, S Mille, L Wanner – … Dependency Theory. Frontiers in …, 2013 – books.google.com … We address both tasks—the amendment of an existing annotation with the purpose to make it more adequate for generation and the development of a semantic stochastic realizer. … In this paper, we present a deep sentence realizer that achieves this goal. … Cited by 2 Related articles All 3 versions

A Semi-Supervised Clustering Approach for Semantic Slot Labelling H Cuayáhuitl, N Dethlefs, H Hastie – 2014 – macs.hw.ac.uk … Natural language processing modules are often trained from labelled examples and therefore rely on the existence of human-annotated corpora, potentially ignoring … [7] H. Cuayáhuitl, N. Dethlefs, H. Hastie, and X. Liu, “Training a statistical surface realiser from automatic slot … Cited by 1

From image annotation to image description A Gupta, P Mannem – Neural information processing, 2012 – Springer … Keywords: Natural Language Processing, Knowledge-based Informa- tion Systems, Information Retrieval, Natural Language Generation. … We use SimpleNLG [14] for this purpose, which is a surface realizer with significant coverage of English syntax and morphology. … Cited by 9 Related articles All 4 versions

Joint Morphological Generation and Syntactic Linearization L Song, Y Zhang, K Song, Q Liu – Twenty-Eighth AAAI …, 2014 – people.sutd.edu.sg … Recent years have witnessed a great success of joint meth- ods on natural language processing problems, such as joint chunking and part-of-speech tagging (Sutton, McCallum, and Rohanimanesh 2007), joint Chinese word segmenta- tion and part-of-speech tagging (Zhang …

Fuzzy linguistic summaries: Where are we, where can we go? B Bouchon-Meunier, G Moyse – Computational Intelligence for …, 2012 – ieeexplore.ieee.org … the selected values. The “tactical component” which consists of micro/macro planner and a surface realizer (like the realization step in SumTime and BT-45 systems, but more detailed), is based on a G- TAG [27] system. It is … Cited by 4 Related articles All 4 versions

Managing Lexical Ambiguity in the Generation of Referring Expressions IH Khan, M Haleem – International Journal of Intelligent Systems and …, 2013 – mecs-press.org … The problem could arise, however, when a surface realiser would express this formula as the old president. … His areas of research are Natural Language Processing (NLP), particularly Natural Language Generation (NLG), and Evolutionary Computation. … Related articles All 3 versions

Towards Efficient HPSG Generation for German, a Non-Configurational Language. B Crysmann, W Packard – COLING, 2012 – kde.cs.tut.ac.jp Page 1. Towards efficient HPSG generation for German, a non-configurational language Berthold CRYSMANN1 Woodley PACKARD2 (1) LLF (UMR 7110), CNRS & U Paris–Diderot, 5 rue Thomas Mann, F-75205 Paris Cedex … Cited by 1 Related articles All 5 versions

Natural Language Descriptions for Video Streams MUG Khan – 2012 – etheses.whiterose.ac.uk … video retrieval. 1http://trecvid.nist.gov/ 2 Page 14. Machine generated descriptions for videos sequences require investigation in two disciplines, namely image processing and natural language processing. Image processing … Related articles All 7 versions

Structure-driven lexicalist generation S Narayan, C Gardent – … of the 24th International Conference in …, 2012 – hal.inria.fr … For HPSG, (Carroll and Oepen, 2005) present a bottom-up, lexicalist, surface realiser which uses a chart based strategy, subsumption-based local ambiguity factoring and a procedure to selectively unpack the generation forest according to a probability distribution given by a … Cited by 3 Related articles All 10 versions

Optimising incremental dialogue decisions using information density for interactive systems N Dethlefs, H Hastie, V Rieser, O Lemon – … Natural Language Processing …, 2012 – dl.acm.org Page 1. Proceedings of the 2012 Joint Conference on Empirical Methods in Natural Language Processing and Computational Natural Language Learning, pages 82–93, Jeju Island, Korea, 12–14 July 2012. cO2012 Association for Computational Linguistics … Cited by 14 Related articles All 7 versions

Syntax and Data-to-Text Generation C Gardent – Statistical Language and Speech Processing, 2014 – Springer … Variable names have been abbreviated and the kbgen tuple notation converted to terms so as to fit the input format expected by our surface realiser. Fig. … In: Proceedings of the 2010 Conference on Empirical Methods in Natural Language Processing, pp. 502–512. …

Planning for Social Interaction in a Robot Bartender Domain. RPA Petrick, ME Foster – ICAPS, 2013 – aaai.org … as the main input modality, many of the planner’s actions will correspond to speech acts, which introduces a link to natural language processing. … tion module then translates the RST structure into OpenCCG logical forms, which are sent to the OpenCCG realiser which outputs … Cited by 15 Related articles All 5 versions

Question Answering of Informative Web Pages: How Summarisation Technology Helps J De Belder, D de Kok, G van Noord, F Nauze… – Essential Speech and …, 2013 – Springer … P. Spyns and J. Odijk (eds.), Essential Speech and Language Technology for Dutch, Theory and Applications of Natural Language Processing, DOI 10.1007/978 … In order to reorganise the ordering of the words, we use a sentence realiser that, given the dependencies required in … Cited by 1 Related articles All 3 versions

Syntax-based word ordering incorporating a large-scale language model Y Zhang, G Blackwood, S Clark – Proceedings of the 13th Conference of …, 2012 – dl.acm.org Page 1. Proceedings of the 13th Conference of the European Chapter of the Association for Computational Linguistics, pages 736–746, Avignon, France, April 23 – 27 2012. cO2012 Association for Computational Linguistics … Cited by 3 Related articles All 13 versions

A model for synthesizing a combined verbal and nonverbal behavior based on personality traits in human-robot interaction A Aly, A Tapus – Proceedings of the 8th ACM/IEEE international …, 2013 – dl.acm.org … 978-1-4673-3101-2/13/$31.00 © 2013 IEEE 331 HRI 2013 Proceedings Page 8. [37] B. Lavoie and O. Rambow, “A fast and portable realizer for text generation systems,” in Proceedings of the 3rd Conference on Applied Natural Language Processing, 1997, pp. 265–268. … Cited by 10 Related articles All 2 versions

Generating non-projective word order in statistical linearization B Bohnet, A Björkelund, J Kuhn, W Seeker… – … Language Processing …, 2012 – dl.acm.org Proceedings of the 2012 Joint Conference on Empirical Methods in Natural Language Processing and Computational Natural Language Learning, pages 928–939, Jeju Island … They use machine-learning techniques to lift edges in a pre- processing step to a surface realizer. … Cited by 1 Related articles All 5 versions

Towards cloud-based speech interfaces for open-domain CogInfoCom systems G Wilcock, K Jokinen – … (CogInfoCom), 2013 IEEE 4th …, 2013 – ieeexplore.ieee.org … Text is generated and pitch accents and phrasal melodies are placed on generated text so that the realizer is able to synthesis the correct surface … B. Named Entity Recognition In text-based natural language processing, named entity recognition is a long-standing research topic … Cited by 1 Related articles All 2 versions

A Real-Time Architecture for Embodied Conversational Agents: Beyond Turn-Taking B Nooraei, C Rich, C Sidner – ACHI 2014, The Seventh International …, 2014 – thinkmind.org … Finally, an important real-time issue that our architecture does not directly address is incrementality, especially with respect to natural language processing. … Each behavior that survives the arbitration process results in instantiation of a realizer process(the loops in the figure that … Cited by 1 Related articles All 5 versions

Does Personality Matter? Expressive Generation for Dialogue Interaction MA Walker, J Sawyer, G Lin, S Wing – Natural Interaction with Robots, …, 2014 – Springer … Secondly, we have introduced finer grained distinctions in these modules in order to support expressive generation. Below we will discuss in detail how our PERSONAGE Realizer Sentence Planner Text Planner Dialog Manager Natural Language Generation Prosody Assigner … Related articles All 3 versions

Robust Multilingual Statistical Morphological Generation Models O Dušek, F Jurcícek – ACL 2013, 2013 – aclweb.org … 17–24. B. Lavoie and O. Rambow. 1997. A fast and portable realizer for text generation systems. In Proceedings of the fifth conference on Applied natural language processing, page 265–268. VI Levenshtein. 1966. Binary codes … Cited by 1 Related articles All 3 versions

Generating different story tellings from semantic representations of narrative E Rishes, SM Lukin, DK Elson, MA Walker – Interactive Storytelling, 2013 – Springer … MIT Press (1998) Page 13. 204 E. Rishes et al. 19. Lavoie, B., Rambow, O.: A fast and portable realizer for text generation systems. In: Proceedings of the Third Conference on Applied Natural Language Processing, ANLP 1997, pp. 265–268 (1997) 20. … Cited by 9 Related articles All 5 versions

SFS-TUE: Compound Paraphrasing with a Language Model and Discriminative Reranking Y Versley – Atlanta, Georgia, USA, 2013 – aclweb.org … In a similar fashion, we used a list of verbs and the morphg morphological realizer (Minnen et al., 2001) to extract all occurrences of the patterns “N PREP N”,“Nprep (DET) N” for noun-preposition- noun combinations, and “N that VBZ … Natural Language Processing with Python. … Related articles All 5 versions

Explaining the reactions of a smart environment S Bader – Explanation-aware Computing ExaCt 2012, 2012 – lirmm.fr … addModifier (createPrepositionPhrase (” by”, GS1)); return realiser. realise … 2009. [17] E. Reiter and R. Dale, Building Natural Language Generation Systems, Studies in natural language processing, Cambridge Press, 2000. … Cited by 2 Related articles All 2 versions

A Hybrid Approach to Multi-document Summarization of Opinions in Reviews G Di Fabbrizio, AJ Stent, R Gaizauskas – INLG 2014, 2014 – aclweb.org … And, finally, the surface realizer must select the proper domain lexemes to express positive and negative opinions. Figure 1: STARLET-H hybrid review summarizer architecture Figure 1 shows the architecture we adopted for our STARLET-H hybrid review summarizer. … Cited by 1

Data-to-text summarisation of patient records: Using computer-generated summaries to access patient histories D Scott, C Hallett, R Fettiplace – Patient education and counseling, 2013 – Elsevier … Natural language processing technology can provide useful tools to support clinical decision-making [16] and [17]. … The system design of the Report Generator follows a classical NLG pipeline architecture, with a Content Selector, MicroPlanner and Syntactic Realiser [24]. … Cited by 4 Related articles All 8 versions

Stochastic Language Generation in Dialogue using Factored Language Models F Mairesse, S Young – 2014 – MIT Press Page 1. Stochastic Language Generation in Dialogue using Factored Language Models François Mairesse ? University of Cambridge Steve Young ?? University of Cambridge Most previous work on trainable language generation … Related articles

sArAmsha-A Kannada abstractive summarizer VR Embar, SR Deshpande… – Advances in …, 2013 – ieeexplore.ieee.org … Due to this, automatic text summarization in Natural Language Processing has been explored thoroughly. … Theoretical systems like SimpleNLG Realizer [18] take sentence structure and words in root form as input and give a syntactically correct sentence as output. … Related articles

Communication Aspects: How Landmarks Enrich the Communication Between Human and Machine KF Richter, S Winter – Landmarks, 2014 – Springer … Finally, the surface realizer phase maps the semantic specifications into actual sentences, ie, into (grammatically correct) natural language. … Understanding verbal references to landmarks is first and foremost a matter of natural language processing (NLP) . … Related articles

Natural Language-based Visualization and Modeling for Updatable Process Views W Wipp – 2013 – dbis.eprints.uni-ulm.de … the surface realization step is done by the class SurfaceRealizer. For the step of surface realization, the publicly available realizer named RealPro from … First, with help of the RealPro realizer, the given XML-based DSynT is transformed into a natural language sentence. … Related articles

Modeling Comma Placement in Chinese Text for Better Readability using Linguistic Features and Gaze Information T Hara, C Chen, Y Kano, A Aizawa – ACL 2013, 2013 – aclweb.org … Guo et al.(2010) implemented a three-tier max- imum entropy model incorporating linguistically motivated features for generating commonly used Chinese punctuation marks in unpunctuated sen- tences output by a surface realizer. … Related articles All 2 versions

To what extent does sentence-internal realisation reflect discourse context?: a study on word order S Zarrieß, J Kuhn, A Cahill – Proceedings of the 13th Conference of the …, 2012 – dl.acm.org Page 1. Proceedings of the 13th Conference of the European Chapter of the Association for Computational Linguistics, pages 767–776, Avignon, France, April 23 – 27 2012. cO2012 Association for Computational Linguistics … Cited by 2 Related articles All 8 versions

Citation handling: Processing citation texts in scientific documents MA Whidby – 2012 – drum.lib.umd.edu … that depends on generic natural language processing components, such as parsers … 1.1 The parse tree for the citation sentence “To get an estimate of how our realiser compares with existing published results, we revisited the test cases discussed in [Carroll et al, 1999] and … Cited by 5 Related articles All 2 versions

Lexical syntax for Arabic SMT H Hassan – Challenges for Arabic Machine Translation, 2012 – books.google.com … The proposed approach deploys three modules in a cascade:(1) a statistical Supertagger,(2) a statistical Operator tagger, and (3) a deterministic Parsing State Realizer. … Given a supertag sti and an operator oi, the state S. is calculated deter- ministically by the State-Realizer. … Related articles All 2 versions

XMG: eXtensible MetaGrammar B Crabbé, D Duchier, C Gardent, J Le Roux… – Computational …, 2013 – MIT Press Cited by 9 Related articles All 16 versions

Joint Video and Text Parsing for Understanding Events and Answering Queries K Tu, M Meng, MW Lee, TE Choe, SC Zhu – MultiMedia, IEEE, 2014 – ieeexplore.ieee.org … causal parsing detects the causal relations between events and fluent changes.30–32 Our work also involves semantic parsing of text descrip- tions, which is related to a variety of semantic parsing approaches studied in the natural language processing community, including … Cited by 1 Related articles All 3 versions

Automatic caption generation for news images Y Feng, M Lapata – Pattern Analysis and Machine Intelligence, …, 2013 – ieeexplore.ieee.org … A multisentence description is generated using a document planner and a surface realizer. … Within natural language processing, most previous efforts have focused on generating captions to accompany complex graphical presentations such as pie charts and bars (eg, [28], [29 … Cited by 16 Related articles All 11 versions

How emotions affect relations between interlocutors in bad news conversations B van Straalen, D Heylen – Communications Control and …, 2012 – ieeexplore.ieee.org … This holds for the Graphical User Interface (GUI) component and the Behavior realizer component. … a myriad of inter- pretation processes performed by the listener (eg processes that are studied in the fields of Automatic Speech Recogni- tion and Natural language processing). … Related articles

SongRecommend: From summarization to recommendation S Tata, B Di Eugenio – Natural Language Engineering, 2014 – Cambridge Univ Press … the heterogeneity of those non-NP features, the inherent limitations of the current state-of-the-art Natural Language Processing techniques and … We use (McRoy, Channarukul and Ali 2003), a sentence realizer that uses templates augmented with rules to express text structures … Cited by 2 Related articles All 2 versions

On Applying Controlled Natural Languages for Ontology Authoring and Semantic Annotation BP Davis – 2013 – vmserver14.nuigalway.ie … annotation tools attempt to ease this knowledge acquisition barrier, their development often requires access to specialists in Natural Language Processing (NLP). This chal- … Dr Paul Buitelaar for being an veritable treasure chest of natural language processing experience. … Cited by 1 Related articles All 2 versions

Joint Video and Text Parsing for Understanding Events and Answering Queries K Tu, M Meng, MW Lee, TE Choe, SC Zhu – arXiv preprint arXiv: …, 2013 – arxiv.org Page 1. 1 Joint Video and Text Parsing for Understanding Events and Answering Queries Kewei Tu, Meng Meng, Mun Wai Lee, Tae Eun Choe, Song-Chun Zhu Abstract—We propose a framework for parsing video and text jointly … Related articles All 6 versions

AnCora-UPF: A Multi-Level Annotation of Spanish SMABL Wanner – DepLing 2013, 2013 – aclweb.org … c 2013 Charles University in Prague, Matfyzpress, Prague, Czech Republic AnCora-UPF: A Multi-Level Annotation of Spanish Simon Mille1 Alicia Burga1 Leo Wanner1, 2 1 Natural Language Processing Group, Pompeu Fabra University, Barcelona, Spain 2 Institució Catalana … Related articles All 7 versions

Automatic abstractive summarization of meeting conversations T Oya – 2014 – elk.library.ubc.ca … Usually, systems first interpret the concept of the source text using natural language processing (NLP) techniques, such as … Finally, those selected aspects are sent to a language generation module, the SimpleNLG realizer [14], to generate the summary. …

Computation and modeling of information structure M Stede – The Expression of Information Structure, 2012 – books.google.com Page 372. Computation and modeling of information structure Manfred Stede 1. Introduction and overview Generally speaking, research in Computational Linguistics strives to devise models of language processing that are …

Computational generation of referring expressions: A survey E Krahmer, K Van Deemter – Computational Linguistics, 2012 – MIT Press … 2.1 First Beginnings REG can be traced back to the earliest days of Natural Language Processing; Winograd 1972 (Section 8.3.3, Naming Objects and Events), for example, sketches a primitive “incremental” REG algorithm, used in his SHRDLU program. … Cited by 73 Related articles All 12 versions

A real-time architecture for conversational agents B Nooraei – 2012 – wpi.edu … 9 Page 17. From a natural language processing point of view, simultaneity requires an agent to incrementally process the speech input so that it can a) react to whatever has … Arbitrator E n v iro n m e n t Primitive Realizer 1 Primitive Realizer 2 Res 1 Res 2 … Compound Realizer … Cited by 1 Related articles All 2 versions

Perspective-oriented generation of football match summaries: Old tasks, new challenges N Bouayad-Agha, G Casamayor, S Mille… – ACM Transactions on …, 2012 – dl.acm.org … Categories and Subject Descriptors: I.2.7 [Artificial Intelligence]: Natural Language Processing— Language generation General Terms: Human Factors Languages Additional Key Words and Phrases: OWL, content selection, discourse structuring, information structuring … Cited by 7 Related articles

An OpenCCG-Based Approach to Question Generation from Concepts MM Berg, A Isard, JD Moore – Natural Language Processing and …, 2013 – Springer … OpenCCG1 is a collection of natural language processing tools, which provide parsing and realisation support based on the CCG formalism. … The result of this step is the representation in Figure 3, which is – at the same time – the input for the OpenCCG realiser. … Related articles All 4 versions

Summarizing Information Graphics Textually S Demir, S Carberry, KF McCoy – Computational Linguistics, 2012 – MIT Press Cited by 2 Related articles All 6 versions

Virtual agents for the production of linear animations R Damiano, V Lombardo, F Nunnari – Entertainment Computing, 2013 – Elsevier … It allows the description of animated gestures through the composition of both data-driven and procedural animations. It was used to implement a BML realizer, providing all the features needed to accomplish the conversational animation gestures. … Related articles

System Personality and Persuasion in Human-Computer Dialogue PY Andrews – ACM Transactions on Interactive Intelligent Systems ( …, 2012 – dl.acm.org … and Principles]: User/Machine Systems—Human factors; software psychology; I.2.7 [Artificial Intelligence]: Natural Language Processing—Discourse; language … 0, 1]. The PERSONAGE generation framework generates distinct surface forms with the realpro realizer [Lavoie and … Cited by 4 Related articles All 2 versions

Generating Descriptions for Images A Gupta – 2013 – web2py.iiit.ac.in … Prof. Rajeev Sangal has a unique ability to explain even complex things in a very easy and efficient way. I have learnt not only the fundamentals and various applications of Natural Language Processing from him, but also about research and various aspects of life. … Cited by 1 Related articles

Linguistically Motivated Features for CCG Realization Ranking RP Rajkumar – 2012 – rave.ohiolink.edu … In Proceedings of the Workshop on Software Engineering, Testing, and Quality Assurance for Natural Language Processing (SETQA-NLP 2009), pages 45- 46, Boulder, Colorado, June 2009. … 2. Inflection and agreement: In surface realization, the input to the surface realizer … Related articles All 2 versions

Deep stochastic sentence generation: resources and strategies S Mille – 2014 – tdx.cat … annotation schemes for stochastic sentence realization. This decade saw a significant increase of interest in corpus-based (ie sta- tistical, or stochastic) Natural Language Processing (NLP). These tenden- cies have been reflected …

Synthesis of listener vocalizations: towards interactive speech synthesis SC Pammi – 2012 – scidok.sulb.uni-saarland.de Page 1. Synthesis of Listener Vocalizations Towards Interactive Speech Synthesis Dissertation zur Erlangung des Grades des Doktors der Ingenieurwissenschaften der Naturwissenschaftlich- Technischen Fakultäten der Universität des Saarlandes vorgelegt von Sathish Pammi … Cited by 1 Related articles All 2 versions

Conversational Informatics T Nishida, A Nakazawa, Y Ohmoto, Y Mohammad – 2014 – Springer Page 1. Conversational Informatics Toyoaki Nishida Atsushi Nakazawa Yoshimasa Ohmoto Yasser Mohammad A Data-Intensive Approach with Emphasis on Nonverbal Communication Page 2. Conversational Informatics Page 3. Toyoaki Nishida • Atsushi Nakazawa … Cited by 1

Context in content composition N Asher – Philosophy of Linguistics, 2012 – books.google.com Page 251. CONTEXT IN CONTENT COMPOSITION Nicholas Asher 1 INTRODUCTION For a long while, lexical semantics developed independently from formal semantics. Apart from a few, daring forays into the formal world … Related articles All 2 versions

Joint models for concept-to-text generation I Konstas – 2014 – era.lib.ed.ac.uk … However, most conventional search en- gines and natural language processing tools (eg, summarisers) can only handle textual input. … However, most conventional search en- gines and natural language processing tools (eg, summarisers) can only handle textual input. … Related articles

Generating Route Instructions with Android Devices based on the GRUVE Challenge B Migot – 2013 – macs.hw.ac.uk … Content determination creates objects that will be later used by micro planning and a surface realizer. This is more or less to … With this, the structure can be provided and the surface realizer will be able to generate the sentence. …

[BOOK] Multilingual text generation from structured formal representations D Dannélls – 2013 – gupea.ub.gu.se … In the light of the substantial growth of digital content availability in large structured Web ontology standards, and today’s increasingly wid- espread use of smart phones and small electronic devices, there is a growing need for new natural language processing (NLP) technolo … Related articles All 2 versions

Access to multimodal articles for individuals with sight impairments S Carberry, S Elzer Schwartz, K Mccoy… – ACM Transactions on …, 2012 – dl.acm.org … Systems]: Information Interfaces and Presentation— User Interfaces—Natural language; Screen design; I.2 [Computing Methodologies]: Artificial Intelli- gence—Natural language interfaces; I.2.7 [Computing Methodologies]: Natural Language Processing— Language generation … Cited by 3 Related articles All 3 versions

Efficient accurate syntactic direct translation models: one tree at a time H Hassan, K Sima’an, A Way – Machine translation, 2012 – Springer … next. Following (Bangalore and Joshi 1999), only an impoverished deterministic procedure (called a ‘State-Realizer’) is needed in order to compose a state Si with the previous states S0 … Si … 4.3 Parse-state realizer After assigning … Related articles All 8 versions

User behavior fusion in dialog management with multi-modal history cues M Yang, J Tao, L Chao, H Li, D Zhang, H Che… – Multimedia Tools and …, 2014 – Springer Page 1. User behavior fusion in dialog management with multi-modal history cues Minghao Yang & Jianhua Tao & Linlin Chao & Hao Li & Dawei Zhang & Hao Che & Tingli Gao & Bin Liu Received: 20 January 2014 /Revised …

Space Efficient Data Structures in the Word-RAM and Bitprobe Models PK Nicholson – 2013 – uwspace.uwaterloo.ca Page 1. Space Efficient Data Structures in the Word-RAM and Bitprobe Models by Patrick Kevin Nicholson A thesis presented to the University of Waterloo in fulfillment of the thesis requirement for the degree of Doctor of Philosophy in Computer Science … Cited by 1 Related articles All 3 versions