Regina Barzilay
ABSTRACT
A content selection component determines which information should be conveyed in the output of a natural language generation system. We present an efficient method for automatically learning content selection rules from a corpus and its related database. Our modeling framework treats content selection as a collective classification problem, thus allowing us to capture contextual dependencies between input items. Experiments in a sports domain demonstrate that this approach achieves a substantial improvement over context-agnostic methods.
- J. Besag. 1986. On the statistical analysis of dirty pictures. Journal of the Royal Statistical Society, 48:259--302.Google Scholar
- Y. Boykov, O. Veksler, R. Zabih. 1999. Fast approximate energy minimization via graph cuts. In ICCV, 377--384. Google ScholarDigital Library
- P. A. Duboue, K. R. McKeown. 2003. Statistical acquisition of content selection rules for natural language generation. In Proceedings of the EMNLP, 121--128. Google ScholarDigital Library
- D. Greig, B. Porteous, A. Seheult. 1989. Exact maximum a posteriori estimation for binary images. Journal of the Royal Statistical Society, 51(2):271--279.Google Scholar
- K. Kukich. 1983. Design of a knowledge-based report generator. In Proceedings of the ACL, 145--150. Google ScholarDigital Library
- J. Kupiec, J. O. Pedersen, F. Chen. 1995. A trainable document summarizer. In Proceedings of the SIGIR, 68--73. Google ScholarDigital Library
- K. R. McKeown. 1985. Text Generation: Using Discourse Strategies and Focus Constraints to Generate Natural Language Text. Cambridge University Press. Google ScholarDigital Library
- B. Pang, L. Lee. 2004. A sentimental education: Sentiment analysis using subjectivity summarization based on minimum cuts. In Proceedings of the ACL, 271--278, Barcelona, Spain. Google ScholarDigital Library
- G. Parks. 1990. An intelligent stochastic optimization routine for nuclear fuel cycle design. Nuclear Technology, 89:233--246.Google ScholarCross Ref
- E. Reiter, R. Dale. 2000. Building Natural Language Generation Systems. Cambridge University Press, Cambridge. Google ScholarDigital Library
- J. Robin. 1994. Revision-Based Generation of Natural Language Summaries Providing Historical Background. Ph.D. thesis, Columbia University. Google ScholarDigital Library
- M. Rogati, Y. Yang. 2002. High-performing feature selection for text classification. In Proceedings of the CIKM, 659--661. Google ScholarDigital Library
- R. E. Schapire, Y. Singer. 2000. Boostexter: A boosting-based system for text categorization. Machine Learning, 39(2/3):135--168. Google ScholarDigital Library
- S. G. Sripada, E. Reiter, J. Hunter, J. Yu. 2001. A two-stage model for content determination. In Proceedings of the ACL-ENLG, 3--10. Google ScholarDigital Library
- K. Tanaka-Ishii, K. Hasida, I. Noda. 1998. Reactive content selection in the generation of real-time soccer commentary. In Proceedings of the ACL/COLING, 1282--1288. Google ScholarDigital Library
- B. Taskar, P. Abbeel, D. Koller. 2002. Discriminative probabilistic models for relational data. In Proceedings of the UAI, 485--495. Google ScholarDigital Library
- Collective content selection for concept-to-text generation
Recommendations
Understanding collective content: purposes, characteristics and collaborative practices
C&T '09: Proceedings of the fourth international conference on Communities and technologiesUser-created media content is being increasingly shared with the communities people belong to. The content has a role of a motivator in social interaction within the communities. In fact, the content creation and management can be often seen as a ...
Content-Based Tag Generation for the Grouping of Tags
ELML '09: Proceedings of the 2009 International Conference on Mobile, Hybrid, and On-line LearningA tagging system can encounter too few or too many tags. To solve these problems, we propose a content-based automatic generation of tags. Applied to an e-Learning 2.0 application, the proposal creates tags based on lecture slide contents, generating an ...
Data-to-text generation with content selection and planning
AAAI'19/IAAI'19/EAAI'19: Proceedings of the Thirty-Third AAAI Conference on Artificial Intelligence and Thirty-First Innovative Applications of Artificial Intelligence Conference and Ninth AAAI Symposium on Educational Advances in Artificial IntelligenceRecent advances in data-to-text generation have led to the use of large-scale datasets and neural network models which are trained end-to-end, without explicitly modeling what to say and in what order. In this work, we present a neural network ...
Comments