Book contents
- Frontmatter
- Contents
- List of contributors
- 1 Introduction
- Part I Joint construction
- Part II Reference
- Part III Handling uncertainty
- Part IV Engagement
- 9 Data-driven methods for linguistic style control
- 10 Integration of cultural factors into the behavioral models of virtual characters
- 11 Natural language generation for augmentative and assistive technologies
- Part V Evaluation and shared tasks
- Author index
- Subject index
- References
9 - Data-driven methods for linguistic style control
from Part IV - Engagement
Published online by Cambridge University Press: 05 July 2014
Book contents
- Frontmatter
- Contents
- List of contributors
- 1 Introduction
- Part I Joint construction
- Part II Reference
- Part III Handling uncertainty
- Part IV Engagement
- 9 Data-driven methods for linguistic style control
- 10 Integration of cultural factors into the behavioral models of virtual characters
- 11 Natural language generation for augmentative and assistive technologies
- Part V Evaluation and shared tasks
- Author index
- Subject index
- References
Summary
Introduction
Modern spoken language interfaces typically ignore a fundamental component of human communication: human speakers tailor their speech and language based on their audience, their communicative goal, and their overall personality (Scherer, 1979; Brennan and Clark, 1996; Pickering and Garrod, 2004). They control their linguistic style for many reasons, including social (e.g., to communicate social distance to the hearer), rhetorical (e.g., for persuasiveness), or task-based (e.g., to facilitate the assimilation of new material). As a result, a close acquaintance, a politician, or a teacher are expected to communicate differently, even if they were to convey the same underlying meaning. In contrast, the style of most human–computer interfaces is chosen once for all at development time, typically resulting in cold, repetitive language, or machinese. This chapter focuses on methods that provide an alternative to machinese by learning to control the linguistic style of computer interfaces from data.
Natural Language Generation (NLG) differs from other areas of natural language processing in that it is an under-constrained problem. Whereas the natural language understanding task requires learning a mapping from linguistic forms to the corresponding meaning representations, NLG systems must learn the reverse one-to-many mapping and choose among all possible realizations of a given input meaning. The criterion to optimize is often unclear, and largely dependent on the target application. Hence it is important to identify relevant control dimensions – i.e., linguistic styles – to optimize the generation process based on the application context and the user.
- Type
- Chapter
- Information
- Natural Language Generation in Interactive Systems , pp. 207 - 226Publisher: Cambridge University PressPrint publication year: 2014
References
(2009). Prodigy-METEO: Pre-alpha release notes. Technical Report NLTG-09-01, Natural Language Technology Group, CMIS, University of Brighton.Google Scholar
and (2003). Factored language models and generalized parallel backoff. In Proceedings of the Human Language Technology Conference of the North American Chapter of the Association for Computational Linguistics (HLT-NAACL), pages 4-6, Edmonton, Canada. Association for Computational Linguistics.Google Scholar
and (1996). Conceptual pacts and lexical choice in conversation. Journal of Experimental Psychology, 22(6):1482-1493.Google ScholarPubMed
, , , and (2004). Evaluating cognitive load in spoken language interfaces using a dual-task paradigm. In Proceedings of the International Conference on Spoken Language Processing (INTERSPEECH), pages 1721-1724, Jeju Island, Korea. International Speech Communication Association.Google Scholar
(1981). The MRC psycholinguistic database. Quarterly Journal of Experimental Psychology, 33A:497-505.Google Scholar
and (1992). NEO PI-R Professional Manual. Psychological Assessment Resources, Odessa, FL.Google Scholar
and (1993). A computational theory of goal-directed style in syntax. Computational Linguistics, 19(3):451-499.Google Scholar
(1990). Language and personality. In and , editors, Handbook of Language and Social Psychology, pages 73-95. Wiley, New York, NY.Google Scholar
and (1996). Mean and variance adaptation within the MLLR framework. Computer Speech & Language, 10(4):249-264.CrossRefGoogle Scholar
and (2002). Taking care of the linguistic features of extraversion. In Proceedings of the Annual Conference of the Cognitive Science Society (CogSci), pages 363-368, Fairfax, Virginia. Cognitive Science Society.Google Scholar
(1990). An alternative “description of personality”: The Big-Five factor structure. Journal of Personality and Social Psychology, 59(6):1216-1229.CrossRefGoogle ScholarPubMed
, , and (2003). A very brief measure of the Big-Five personality domains. Journal of Research in Personality, 37(6):504-528.CrossRefGoogle Scholar
and (1996). Stylistic decision-making in natural language generation. In and , editors, Trends in Natural Language Generation: An Artificial Intelligence Perspective, volume 1036, pages 125-143. Springer LNCS, Berlin, Germany.Google Scholar
(1988). Generating Natural Language under Pragmatic Constraints. Lawrence Erlbaum Associates, Hillsdale, NJ.Google Scholar
, , and (2006). Individuality and alignment in generated dialogues. In Proceedings of the International Conference on Natural Language Generation (INLG), pages 25-32, Sydney, Australia. Association for Computational Linguistics.Google Scholar
, , and (1991). The Big Five inventory-versions 4a and 54. Technical report, Berkeley: University of California, Institute of Personality and Social Research.Google Scholar
and (1998). Generation that exploits corpus-based statistical knowledge. In Proceedings of the Annual Meeting of the Association for Computational Linguistics and the International Conference on Computational Linguistics (COLING-ACL), pages 704-710, Montreal, Canada. Association for Computational Linguistics.Google Scholar
(2002). An empirical verification of coverage and correctness for a generalpurpose sentence generator. In Proceedings of the International Conference on Natural Language Generation (INLG), pages 17-24, Arden Conference Center, NY. Association for Computational Linguistics.Google Scholar
and (1997). A fast and portable realizer for text generation systems. In Proceedings of the Applied Natural Language Processing Conference (ANLP), pages 1-7, Washington, DC. Association for Computational Linguistics.Google Scholar
(2008). Learning to Adapt in Dialogue Systems: Data-driven Models for Personality Recognition and Generation. PhD thesis, Department of Computer Science, University of Sheffield.Google Scholar
, , , , , , and (2010). Phrase-based statistical language generation using graphical models and active learning. In Proceedings of the Annual Meeting of the Association for Computational Linguistics (ACL), pages 1552-1561, Uppsala, Sweden. Association for Computational Linguistics.Google Scholar
and (2007). PERSONAGE: Personality generation for dialogue. In Proceedings of the Annual Meeting of the Association for Computational Linguistics (ACL), pages 496-503, Prague, Czech Republic. Association for Computational Linguistics.Google Scholar
and (2010). Towards personality-based user adaptation: Psychologically informed stylistic language generation. User Modeling and User-Adapted Interaction, 20(3):227-278.CrossRefGoogle Scholar
and (1988). Rhetorical structure theory: Toward a functional theory of text organization. Text, 8(3):243-281.CrossRefGoogle Scholar
, , and (2006). Personality in its natural habitat: Manifestations and implicit folk theories of personality in daily life. Journal of Personality and Social Psychology, 90(5):862-877.CrossRefGoogle ScholarPubMed
(1963). Toward an adequate taxonomy of personality attributes: Replicated factor structure in peer nomination personality rating. Journal ofAbnormal and Social Psychology, 66(6):574-583.Google Scholar
and (2005). Empirically-based control of natural language generation. In Proceedings of the Annual Meeting of the Association for Computational Linguistics (ACL), pages 58-65, Ann Arbor, MI. Association for Computational Linguistics.Google Scholar
, , and (2001). Linguistic inquiry and word count. Available from http://www.liwc.net/. Accessed on 11/24/2013.Google Scholar
and (1999). Linguistic styles: Language use as an individual difference. Journal of Personality and Social Psychology, 77(6):1296-1312.CrossRefGoogle ScholarPubMed
and (2004). Toward a mechanistic psychology of dialogue. Behavioral and Brain Sciences, 27(2):169-226.CrossRefGoogle Scholar
and (2004). Modelling politeness in natural language generation. In Proceedings of the International Conference on Natural Language Generation (INLG), pages 141-150, Brockenhurst, UK. Springer.Google Scholar
(1989). Tutorial on hidden Markov models and selected applications in speech recognition. Proceedings of the IEEE, 77(2):257-286.CrossRefGoogle Scholar
and (2000). Building Natural Language Generation Systems. Cambridge University Press, Cambridge, UK.CrossRefGoogle Scholar
(1979). Personality markers in speech. In and , editors, Social Markers in Speech, pages 147-209. Cambridge University Press, Cambridge, UK.Google Scholar
, , and (2004). Trainable sentence planning for complex information presentation in spoken dialog systems. In Proceedings of the Annual Meeting of the Associationfor Computational Linguistics (ACL), pages 79-86, Barcelona, Spain. Association for Computational Linguistics.Google Scholar
, , , and (2002). User-tailored generation for spoken dialogue: An experiment. In Proceedings of the International Conference on Spoken Language Processing (INTERSPEECH), pages 1281-1284, Denver, CO. International Speech Communication Association.Google Scholar
, , and (2003). NL generation for virtual humans in a complex social environment. In Working Papers of the AAAI Spring Symposium on Natural Language Generation in Spoken and Written Dialogue, pages 151-158, Stanford, CA. AAAI Press.Google Scholar
, , , and (2004). Support vector learning for interdependent and structured output spaces. In Proceedings of the International Conference on Machine Learning (ICML), Banff, Canada. The International Machine Learning Society, Association for Computing Machinery.Google Scholar
, , , , , , and (2002). Speech-Plans: Generating evaluative responses in spoken dialogue. In Proceedings of the International Conference on Natural Language Generation (INLG). Association for Computational Linguistics.Google Scholar
, , , , , , and (2004). Generation and evaluation of user tailored responses in multimodal dialogue. Cognitive Science, 28(5):811-840.CrossRefGoogle Scholar
, , , and (2007). Individual and domain adaptation in sentence planning for dialogue. Journal of Artificial Intelligence Research, 30(1):413-456.Google Scholar
, , , , , and (2008). The politeness effect: Pedagogical agents and learning outcomes. International Journal of Human-Computer Studies, 66(2):98-112.CrossRefGoogle Scholar