UDC 681.3
IMPROVING TEXT GENERATION THROUGH INTRODUCING COHERENCE METRICS
Abstract. Text-based interaction using mobile devices is now ubiquitous, its main outlets being social networks, messengers, email conversations, virtual assistants, accessibility applications, etc. Its status implies the need to facilitate text input by the user and to devise ways to provide verbal feedback. In this paper, we discuss a method of unique text generation for mobile devices and its evaluation methodology as a solution for both stated challenges. We consider the opportunities given by the use of context (location, weather, scheduled events, etc.), the limitations in terms of computational resources and data usage, and the inherent subjectivity of creative task assessment given the number variety of possibly acceptable outputs. The comparison with other text generation approaches shows that the use of coherence metrics helps to achieve higher quality in terms of human perception. The Spearman correlation between the values of the proposed coherence metric and the human assessment of text readability is 0.86, which indicates the high quality of the metrics and the effectiveness of the method as a whole.
Keywords: natural language processing, automatic natural language text generation, coherency, coherence metrics.
FULL TEXT
REFERENCES
- Ruder S., Ghaffari P., Breslin J.G. Character-level and multi-channel convolutional neural networks for large-scale authorship attribution. 2016. URL: http://arxiv.org/abs/1609.06686.
- Agarwal B., Ramampiaro H., Langseth H., Ruocco M. A deep network model for paraphrase detection in short text messages. Information Processing and Management. 2018. Vol. 54, N 6. P. 922–937.
- Asghar N., Poupart P., Hoey J., Jiang X., Mou L. Affective neural response generation. Advances in Information Retrival. Proc. 40th European Conf. on IR Research, ECIR 2018 (March 26–29, 2018, Grenoble, France). P. 154–166.
- Chen Y.-N., Celikyilmaz A., Hakkani-Tur D. Deep learning for dialogue systems. Proc. 27th Int. Conf. on Computational Linguistics: Tutorial Abstracts. 2018. P. 25–31. URL: https:// www.aclweb.org/anthology/C18-3006.pdf.
- Radford A., Wu J., Child R., Luan D., Amodei D., Sutskever I. Language models are unsupervised multitask learners. OpenAI Blog. URL: https://openai.com/blog/better-language-models/ (Feb 14, 2019).
- Kiros R., Zhu Y., Salakhutdinov R.R., Zemel R., Urtasun R., Torralba A., Fidler S. Skip-thought vectors. Proc. NIPS 2015. (December 7–12, 2015, Montreal, Quebec, Canada). Vol. 2. P. 3294–3302.
- Jain P., Agrawal P., Mishra A., Sukhwani M., Laha A., Sankaranarayanan K. Story generation from sequence of independent short descriptions. 2017. URL: http:// arxiv.org/abs/1707.05501.
- McIntyre N., Lapata M. Learning to tell tales: a data-driven approach to story generation. Proc. 47th Annual Meeting of the ACL and the 4th IJCNLP of the AFNLP (2–7 August 2009, Suntec, Singapore). 2009. P. 217–225.
- Rishes E., Lukib S.M., Elson D.K., Walker M.A. Generating different story tellings from semantic representations of narrative. Proc. 6th ICIDS 2013. (November 6–9, Istanbul, Turkey). P. 192–204.
- Leppnen L., Munezero M., Granroth-Wilding M., Toivonen H. Data-driven news generation for automated journalism. Proc. 10th INLG 2017. (September, 2017, Santiago de Compostela, Spain). P. 188–197.
- Papineni K., Roukos S., Ward T., Zhu W.-J. BLEU: A method for automatic evaluation of machine translation. Proc. 40th Annual Meeting on Association for Computational Linguistics. (July, 2002, Philadelphia, Pennsylvania, USA). 2002. P. 311–318.
- Ji Y., Eisenstein J. Discriminative improvements to distributional sentence similarity. Proc. 2013 Conference on Empirical Methods in Natural Language Processing. (October, 2013, Seattle, Washington, USA). P. 891–896.
- Foltz P. W., Kintsch W., Landauer T.K. The measurement of textual coherence with latent semantic analysis. Discourse Processes. 1998. Vol. 25, N 2–3. P. 285–307.
- Barzilay R., Lapata M. Modeling local coherence: an entity-based approach. Computational Linguistics. 2008. Vol. 34, N 1. P. 1–34.
- Li J., Hovy E. A model of coherence based on distributed sentence representation. Proc. EMNLP 2014. (October 25–29, 2014, Doha, Qatar). P. 2039–2048.
- Li J., Jurafsky D. Neural net models of open-domain discourse coherence. Proc. EMNLP 2017. (September, 2017, Copenhagen, Denmark). P. 198–209.
- Basile V., Condori R.L., Cabrio E. Measuring frame instance relatedness. Proc. 7th Joint Conference on Lexical and Computational Semantics (*SEM) (June 5–6, 2018, New Orleans). 2018. P. 245–254.
- Mesgar M., Strube M. A neural local coherence model for text quality assessment. Proc. 2018 Conference on Empirical Methods in Natural Language Processing (October 31 – November 4, 2018, Brussels, Belgium). P. 4328–4339.
- Wu Z., Palmer M. Verbs semantics and lexical selection. Proc. 32nd Annual Meeting on Association for Computational Linguistics (June 27–30, 1994, Las Cruces, New Mexico). 1994. P. 133–138.
- Mostafazadeh N., Chambers N., He X., Parikh D., Batra D., Vanderwende L., Kohli P., Allen J. A corpus and cloze evaluation for deeper understanding of commonsense stories. Proc. NAACL-HLT 2016. (June 12–17, 2016, San Diego, California). P. 839–849.
- Yao L., Peng N., Weisahedel R., Kbight K., Zhao D., Yan R. Plan-And-Write: Towards better automatic storytelling. 2018. URL: https://arxiv.org/abs/1811.05701.
- Lin T.-Y, Maire M., Belongie S., Hays J., Peroba P., Ramanan D., Dollar P., Zitnicl L. Microsoft COCO: Common objects in context. Proc. ECCV 2014. (September 6–12, 2014, Zurich, Switzerland). P. 740–75.
