We address the problem that different users have different lexical knowledge about problem domains, so that automated dialogue systems need to adapt their generation choices online to the users' domain knowledge as it encounters them. We approach this problem using policy learning in Markov Decision Processes (MDP). In contrast to related work we propose a new statistical user model which incorporates the lexical knowledge of different users. We evaluate this user model by showing that it allows us to learn dialogue policies that automatically adapt their choice of referring expressions online to different users, and that these policies are significantly better than adaptive hand-coded policies for this problem. The learned policies are consistently between 2 and 8 turns shorter than a range of different hand-coded but adaptive baseline lexical alignment policies.
{"title":"Learning Lexical Alignment Policies for Generating Referring Expressions for Spoken Dialogue Systems","authors":"S. Janarthanam, Oliver Lemon","doi":"10.3115/1610195.1610206","DOIUrl":"https://doi.org/10.3115/1610195.1610206","url":null,"abstract":"We address the problem that different users have different lexical knowledge about problem domains, so that automated dialogue systems need to adapt their generation choices online to the users' domain knowledge as it encounters them. We approach this problem using policy learning in Markov Decision Processes (MDP). In contrast to related work we propose a new statistical user model which incorporates the lexical knowledge of different users. We evaluate this user model by showing that it allows us to learn dialogue policies that automatically adapt their choice of referring expressions online to different users, and that these policies are significantly better than adaptive hand-coded policies for this problem. The learned policies are consistently between 2 and 8 turns shorter than a range of different hand-coded but adaptive baseline lexical alignment policies.","PeriodicalId":307841,"journal":{"name":"European Workshop on Natural Language Generation","volume":"13 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2009-03-30","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"130927697","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
This paper argues for a game-theoretic approach to content determination that uses text-type specific strategies in order to determine the optimal content for various user types. By means of content determination for the description of numerical data the benefits of a game-theoretic treatment of content determination are outlined.
{"title":"Towards a Game-Theoretic Approach to Content Determination","authors":"R. Klabunde","doi":"10.3115/1610195.1610211","DOIUrl":"https://doi.org/10.3115/1610195.1610211","url":null,"abstract":"This paper argues for a game-theoretic approach to content determination that uses text-type specific strategies in order to determine the optimal content for various user types. By means of content determination for the description of numerical data the benefits of a game-theoretic treatment of content determination are outlined.","PeriodicalId":307841,"journal":{"name":"European Workshop on Natural Language Generation","volume":"106 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2009-03-30","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"121318936","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
This paper introduces a class-based approach to ordering prenominal modifiers. Modifiers are grouped into broad classes based on where they tend to occur prenominally, and a framework is developed to order sets of modifiers based on their classes. This system is developed to generate several orderings for modifiers with more flexible positional constraints, and lends itself to bootstrapping for the classification of previously unseen modifiers.
{"title":"Class-Based Ordering of Prenominal Modifiers","authors":"Margaret Mitchell","doi":"10.3115/1610195.1610203","DOIUrl":"https://doi.org/10.3115/1610195.1610203","url":null,"abstract":"This paper introduces a class-based approach to ordering prenominal modifiers. Modifiers are grouped into broad classes based on where they tend to occur prenominally, and a framework is developed to order sets of modifiers based on their classes. This system is developed to generate several orderings for modifiers with more flexible positional constraints, and lends itself to bootstrapping for the classification of previously unseen modifiers.","PeriodicalId":307841,"journal":{"name":"European Workshop on Natural Language Generation","volume":"80 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2009-03-30","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"123654412","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Colin Kelly, Ann A. Copestake, Nikiforos Karamanis
The content selection component of a natural language generation system decides which information should be communicated in its output. We use information from reports on the game of cricket. We first describe a simple factoid-to-text alignment algorithm then treat content selection as a collective classification problem and demonstrate that simple 'grouping' of statistics at various levels of granularity yields substantially improved results over a probabilistic baseline. We additionally show that holding back of specific types of input data, and linking database structures with commonality further increase performance.
{"title":"Investigating Content Selection for Language Generation using Machine Learning","authors":"Colin Kelly, Ann A. Copestake, Nikiforos Karamanis","doi":"10.3115/1610195.1610218","DOIUrl":"https://doi.org/10.3115/1610195.1610218","url":null,"abstract":"The content selection component of a natural language generation system decides which information should be communicated in its output. We use information from reports on the game of cricket. We first describe a simple factoid-to-text alignment algorithm then treat content selection as a collective classification problem and demonstrate that simple 'grouping' of statistics at various levels of granularity yields substantially improved results over a probabilistic baseline. We additionally show that holding back of specific types of input data, and linking database structures with commonality further increase performance.","PeriodicalId":307841,"journal":{"name":"European Workshop on Natural Language Generation","volume":"14 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2009-03-30","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"125706673","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
This paper gives an overview of ongoing work on a system for the generation of NL descriptions of classes defined in OWL ontologies. We present a general structuring approach for such descriptions. Since OWL ontologies do not by default contain the information necessary for lexicalization, lexical information has to be added to the data via annotations. A rule-based mechanism for automatically deriving these annotations is presented.
{"title":"Generating Natural Language Descriptions of Ontology Concepts","authors":"Niels Schütte","doi":"10.3115/1610195.1610212","DOIUrl":"https://doi.org/10.3115/1610195.1610212","url":null,"abstract":"This paper gives an overview of ongoing work on a system for the generation of NL descriptions of classes defined in OWL ontologies. We present a general structuring approach for such descriptions. Since OWL ontologies do not by default contain the information necessary for lexicalization, lexical information has to be added to the data via annotations. A rule-based mechanism for automatically deriving these annotations is presented.","PeriodicalId":307841,"journal":{"name":"European Workshop on Natural Language Generation","volume":"37 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2009-03-30","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"121661926","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
This informal position paper brings together some recent developments in formal semantics and pragmatics to argue that the discipline of Game Theory is well placed to become the theoretical backbone of Natural Language Generation. To demonstrate some of the strengths and weaknesses of the Game-Theoretical approach, we focus on the utility of vague expressions. More specifically, we ask what light Game Theory can shed on the question when an NLG system should generate vague language.
{"title":"What Game Theory Can Do for NLG: The Case of Vague Language (Invited Talk)","authors":"Kees van Deemter","doi":"10.3115/1610195.1610221","DOIUrl":"https://doi.org/10.3115/1610195.1610221","url":null,"abstract":"This informal position paper brings together some recent developments in formal semantics and pragmatics to argue that the discipline of Game Theory is well placed to become the theoretical backbone of Natural Language Generation. To demonstrate some of the strengths and weaknesses of the Game-Theoretical approach, we focus on the utility of vague expressions. More specifically, we ask what light Game Theory can shed on the question when an NLG system should generate vague language.","PeriodicalId":307841,"journal":{"name":"European Workshop on Natural Language Generation","volume":"313 2 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2009-03-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"121176254","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
D. Dionne, Salvador de la Puente, C. León, Pablo Gervás, Raquel Hervás
This paper shows a model of automatic instruction giving for guiding human users in virtual 3D environments. A multilevel model for choosing what instruction to give in every state is presented, and so are the different modules that compose the whole generation system. How 3D information in the virtual world is used is explained, and the final order generation is detailed. This model has been implemented as a solution for the GIVE Challenge, an instruction generation challenge.
{"title":"A Model for Human Readable Instruction Generation Using Level-Based Discourse Planning and Dynamic Inference of Attributes","authors":"D. Dionne, Salvador de la Puente, C. León, Pablo Gervás, Raquel Hervás","doi":"10.3115/1610195.1610205","DOIUrl":"https://doi.org/10.3115/1610195.1610205","url":null,"abstract":"This paper shows a model of automatic instruction giving for guiding human users in virtual 3D environments. A multilevel model for choosing what instruction to give in every state is presented, and so are the different modules that compose the whole generation system. How 3D information in the virtual world is used is explained, and the final order generation is detailed. This model has been implemented as a solution for the GIVE Challenge, an instruction generation challenge.","PeriodicalId":307841,"journal":{"name":"European Workshop on Natural Language Generation","volume":"17 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2009-03-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"127214328","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Politeness is an integral part of human language variation, e.g. consider the difference in the pragmatic effect of realizing the same communicative goal with either "Get me a glass of water mate!" or "I wonder if I could possibly have some water please?" This paper presents POLLy (Politeness for Language Learning), a system which combines a natural language generator with an AI Planner to model Brown and Levinson's theory of politeness (BL (2) our indirect strategies which should be the politest forms, are seen as the rudest; and (3) English and Indian native speakers of English have different perceptions of the level of politeness needed to mitigate particular face threats.
{"title":"Generating Politeness in Task Based Interaction: An Evaluation of the Effect of Linguistic Form and Culture","authors":"S. Gupta, M. Walker, D. Romano","doi":"10.3115/1610163.1610173","DOIUrl":"https://doi.org/10.3115/1610163.1610173","url":null,"abstract":"Politeness is an integral part of human language variation, e.g. consider the difference in the pragmatic effect of realizing the same communicative goal with either \"Get me a glass of water mate!\" or \"I wonder if I could possibly have some water please?\" This paper presents POLLy (Politeness for Language Learning), a system which combines a natural language generator with an AI Planner to model Brown and Levinson's theory of politeness (BL (2) our indirect strategies which should be the politest forms, are seen as the rudest; and (3) English and Indian native speakers of English have different perceptions of the level of politeness needed to mitigate particular face threats.","PeriodicalId":307841,"journal":{"name":"European Workshop on Natural Language Generation","volume":"63 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2007-06-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"133741077","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 1993-04-28DOI: 10.1007/3-540-60800-1_35
Bernd Abb, Carsten Günther, Michael Herweg, K. Lebeth, Claudia Maienborn, A. Schopp
{"title":"Incremental Grammatical Encoding - An Outline of the Synphonics Formulator","authors":"Bernd Abb, Carsten Günther, Michael Herweg, K. Lebeth, Claudia Maienborn, A. Schopp","doi":"10.1007/3-540-60800-1_35","DOIUrl":"https://doi.org/10.1007/3-540-60800-1_35","url":null,"abstract":"","PeriodicalId":307841,"journal":{"name":"European Workshop on Natural Language Generation","volume":"6 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1993-04-28","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"115407748","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 1993-04-28DOI: 10.1007/3-540-60800-1_31
Elsa Pascual
{"title":"Integrating Text Formatting and Text Generation","authors":"Elsa Pascual","doi":"10.1007/3-540-60800-1_31","DOIUrl":"https://doi.org/10.1007/3-540-60800-1_31","url":null,"abstract":"","PeriodicalId":307841,"journal":{"name":"European Workshop on Natural Language Generation","volume":"3 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1993-04-28","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"117103708","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
京公网安备 11010802042870号
京ICP备2023020795号-1
扫码关注我们
求助内容：
应助结果提醒方式：