Identificando el humor de tuits utilizando el aprendizaje de preferencias basado en procesos gaussianos

• Autores: Tristan Miller, Iryna Gurevych, Edwin Simpson, Erik Lân Do Dinh
• Localización: Procesamiento del lenguaje natural, ISSN 1135-5948, Nº. 64, 2020, págs. 37-44
• Idioma: español
• Títulos paralelos:
  • Predicting the Humorousness of Tweets Using Gaussian Process Preference Learning
• Enlaces
  • Texto completo
• Resumen
  • español

    Actualmente la mayoría de los sistemas de procesamiento de humor hacen, en el mejor de los casos, distinciones discretas y granulares entre lo cómico y lo convencional. Sin embargo, dichos conceptos se conciben mejor en un espectro más amplio. Este artículo presenta un método probabilístico, un modo de preferencias de aprendizaje basadas en un proceso gaussiano (GPPL), que aprende a clasificar y calificar el humor de textos cortos explotando juicios de preferencia humana y anotaciones lingüísticas generadas en forma automática. Nuestro sistema es similar a uno que previamente había demostrado un buen desempeño en frases en inglés anotadas con anotaciones humorısticas por pares y lo aplicamos a la colección de datos en español de la campaña de evaluación HAHA@IberLEF2019. En este trabajo reportamos el desempeño del sistema para dos subtareas de la campaña: detección de humor y predicción de puntaje de diversión. También presentamos algunos problemas que surgen de la conversión entre los puntajes numéricos utilizados en los datos HAHA@IberLEF2019 y las anotaciones de juicio de pares de documentos requeridas para nuestro método.

  • English

    Most humour processing systems to date make at best discrete, coarsegrained distinctions between the comical and the conventional, yet such notions are better conceptualized as a broad spectrum. In this paper, we present a probabilistic approach, a variant of Gaussian process preference learning (GPPL), that learns to rank and rate the humorousness of short texts by exploiting human preference judgments and automatically sourced linguistic annotations. We apply our system, which is similar to one that had previously shown good performance on English-language one-liners annotated with pairwise humorousness annotations, to the Spanish-language data set of the HAHA@IberLEF2019 evaluation campaign. We report system performance for the campaign's two subtasks, humour detection and funniness score prediction, and discuss some issues arising from the conversion between the numeric scores used in the HAHA@IberLEF2019 data and the pairwise judgment annotations required for our method.

• Referencias bibliográficas
  • Attardo, S. 1994. Linguistic Theories of Humor. Mouton de Gruyter, Berlin.
  • Beck, D., T. Cohn, and L. Specia. 2014. Joint emotion analysis via multi-task Gaussian processes. In Proc. 2014 Conference on Empirical Methods...
  • Bell, N. D. 2017. Failed humor. In S. Attardo, editor, The Routledge Handbook of Language and Humor. Routledge, New York, pages 356–370.
  • Bradley, R. A. and M. E. Terry. 1952. Rank analysis of incomplete block designs: I. The method of paired comparisons. Biometrika, 39(3/4):324–345,...
  • Carretero-Dios, H., C. P´erez, and G. BuelaCasal. 2010. Assessing the appreciation of the content and structure of humor: Construction of...
  • Castro, S., L. Chiruzzo, and A. Rosá. 2018. Overview of the HAHA task: Humor analysis based on human annotation at IberEval 2018. In Proc....
  • Tristan Miller, Erik-Lân Do Dinh, Edwin Simpson, Iryna Gurevych 42 Castro, S., L. Chiruzzo, A. Rosá, D. Garat, and G. Moncecchi. 2018. A crowdannotated...
  • Chiruzzo, L., S. Castro, M. Etcheverry, D. Garat, J. J. Prada, and A. Rosá. 2019. Overview of HAHA at IberLEF 2019: Humor analysis based on...
  • Chu, W. and Z. Ghahramani. 2005. Preference learning with Gaussian processes. In Proc. 22nd International Conference on Machine Learning,...
  • Cohn, T. and L. Specia. 2013. Modelling annotator bias with multi-task Gaussian processes: An application to machine translation quality estimation....
  • Deriu, J., A. Lucchi, V. De Luca, A. Severyn, S. Müller, M. Cieliebak, T. Hoffmann, and M. Jaggi. 2017. Leveraging large amounts of weakly...
  • Felt, P., E. K. Ringger, and K. D. Seppi. 2016. Semantic annotation aggregation with conditional crowdsourcing models and word embeddings....
  • Finn, A. and J. J. Louviere. 1992. Determining the appropriate response to evidence of public concern: The case of food safety. J. Public...
  • Flynn, T. N. and A. A. J. Marley. 2014. Best– worst scaling: Theory and methods. In S. Hess and A. Daly, editors, Handbook of Choice Modelling....
  • Gonzalez-Agirre, A., E. Laparra, and G. Rigau. 2012. Multilingual Central Repository version 3.0. In Proc. 8th International Conference on...
  • Hempelmann, C. F. 2003. Paronomasic Puns: Target Recoverability Towards Automatic Generation. Ph.D. thesis, Purdue University, West Lafayette,...
  • Hempelmann, C. F. 2008. Computational humor: Beyond the pun? In V. Raskin, editor, The Primer of Humor Research. Mouton de Gruyter, Berlin,...
  • Hoffman, M. D., D. M. Blei, C. Wang, and J. W. Paisley. 2013. Stochastic variational inference. J. Mach. Learn. Res., 14:1303– 1347, May.
  • Holton, A. E. and S. C. Lewis. 2011. Journalists, social media, and the use of humor on Twitter. Electron. J. Commun., 21(1&2).
  • Hossain, N., J. Krumm, M. Gamon, and H. Kautz. 2020. SemEval-2020 Task 7: Assessing the funniness of edited news headlines. In Proc. 14th...
  • Ismailov, A. 2019. Humor analysis based on human annotation challenge at IberLEF 2019: First-place solution. In Proc. Iberian Languages Evaluation...
  • Kido, H. and K. Okamoto. 2017. A Bayesian approach to argument-based reasoning for attack estimation. In Proc. 26th International Joint Conference...
  • Kiritchenko, S. and S. M. Mohammad. 2016. Capturing reliable fine-grained sentiment associations by crowdsourcing and best– worst scaling....
  • Lampos, V., N. Aletras, D. Preotiuc-Pietro, and T. Cohn. 2014. Predicting and characterising user impact on Twitter. In Proc. 14th Conference...
  • Lippman, L. G. and M. L. Dunn. 2000. Contextual connections within puns: Effects on perceived humor and memory. J. Gen. Psychol., 127(2):185–197,...
  • Luce, R. D. 1959. On the possible psychophysical laws. Psychol. Rev., 66(2):81– 95.
  • Marley, A. A. J. and J. J. Louviere. 2005. Some probabilistic models of best, worst, and best–worst choices. J. Math. Psychol., 49(6):464–480.
  • Mosteller, F. 1951. Remarks on the method of paired comparisons: I. The least squares solution assuming equal standard deviations and equal...
  • Nickisch, H. and C. E. Rasmussen. 2008. Approximations for binary Gaussian process classification. J. of Mach. Learn. Res., 9:2035–2078, October.
  • Ortega-Bueno, R., C. E. Muñiz Cuza, J. E. Medina Pagola, and P. Rosso. 2018. UO UPV: Deep linguistic humor detection in Spanish social media....
  • Plackett, R. L. 1975. The analysis of permutations. J. Royal Stat. Soc. Ser. C (Appl. Stat.), 24(2):193–202.
  • Rasmussen, C. E. and C. K. I. Williams. 2006. Gaussian Processes for Machine Learning. Adaptive Computation and Machine Learning. MIT Press,...
  • Shahaf, D., E. Horvitz, and R. Mankoff. 2015. Inside jokes: Identifying humorous cartoon captions. In Proc. 21th ACM SIGKDD International...
  • Shifman, L. 2013. Memes in Digital Culture. MIT Press, Cambridge, MA, USA.
  • Simpson, E., E.-L. Do Dinh, T. Miller, and I. Gurevych. 2019. Predicting humorousness and metaphor novelty with Gaussian process preference...
  • Simpson, E. and I. Gurevych. 2018. Finding convincing arguments using scalable Bayesian preference learning. Trans. Assoc. Comput. Ling.,...
  • Simpson, E. D., M. Venanzi, S. Reece, P. Kohli, J. Guiver, S. J. Roberts, and N. R. Jennings. 2015. Language understanding in the wild: Combining...
  • Thurstone, L. L. 1927. A law of comparative judgment. Psychol. Rev., 34(4):273–286.
  • Titov, I. and A. Klementiev. 2012. A Bayesian approach to unsupervised semantic role induction. In Proc. 13th Conference of the European Chapter...
  • Titov, I. and A. Klementiev. Proc. ACL 2015 Workshop on Noisy User-generated Text. Association for Computational Linguistics.
  • Xiong, H. Y., Y. Barash, and B. J. Frey. 2011. Bayesian prediction of tissue-regulated splicing using RNA sequence and cellular context. Bioinform.,...