International Journal of Engineering Research in Computer Science and Engineering (IJERCSE)

LSG: A Regression Based Approach to Automate LaTeX Slides Generation for Technical Articles

Author : Biju P Dais ¹ Smitha C S ²

Date of Publication :7th February 2016

Abstract: A data mining approach for automating the generation of presentation slides from an academic article is presented in this paper. Initially, the system is trained using a large dataset to learn the intricacies involved in how humans do the task of slide generation. The input to the proposed scheme is a technical article. The proposed method operates in 2 stages - Scoring and Selection. During the phase of scoring, the sentences in the input are extracted and their importance is analyzed by calculating a relevance score for each, by using a trained Support Vector Regression model. During the phase of selection, an Integer Linear Programming model with a robust objective function and well defined constraints selects important key phrases and the sentences which best summarizes them from the document. The proposed system can include graphical elements as well to the slides. The resultant slides are output in either TeX or PPT editable formats based on user preference. The sentences are also compressed optionally by the system so as to resemble humanly generated slides to a much higher level.

Reference :

1. M. Utiyama and K. Hasida, ”Automatic slide presentation from semantically annotated documents”, in Proc. ACLWorkshop Conf. Its Appl., 1999, pp. 25- 30.
2. Y. Yasumura, M. Takeichi, and K. Nitta, ”A support system for making presentation slides”, Trans. Japanese Soc. Artif. Intell., vol. 18, pp. 212-220, 2003.
3. M. Sravanthi, C. R. Chowdary, and P. S. Kumar, ”SlidesGen: Automatic generation of presentation slides for a technical paper using summarization”, in Proc. 22nd Int. FLAIRS Conf., 2009, pp. 284-289.
4. M. Sravanthi, C. R. Chowdary, and P. S. Kumar, ”QueSTS: A query specific text summarization approach”, in Proc. 21st Int. FLAIRS Conf., 2008, pp. 219-224.
5. T. Shibata and S. Kurohashi, ”Automatic slide generation based on discourse structure analysis”, in Proc. Int. Joint Conf. Natural Lang. Process., 2005, pp. 754-766
6. Gokul Prasad, K., Mathivanan, H., Jayaprakasam, M., and Geetha, T. V., ”Document summarization and information extraction for generation of presentation slides”, Advances in Recent Technologies in Communication and Computing, 2009. ARTCom’09. International Conference on. IEEE, 2009.
7. Sariki, Tulasi Prasad, Bharadwaja Kumar, and Ramesh Ragala. ”Effective Classroom Presentation Generation Using Text Summarization”
8. S. M. A. Masum, M. Ishizuka, and M. T. Islam, ”Autopresentation: A multi-agent system for building automatic multi-modal presentation of a topic from world wide web information”, in Proc. IEEE/WIC/ACMInt. Conf. Intell. Agent Technol., 2005, pp. 246-249
9. S. M. A. Masum and M. Ishizuka, ”Making topic specific report and multimodal presentation automatically by mining the web resources”, in Proc. IEEE/WIC/ACM Int. Conf. Web Intell., 2006, pp. 240- 246.
10. Hu, Yue, and Xiaojun Wan. ”Ppsgen: learning to generate resentation slides for academic papers”, Proceedings of the Twenty-Third international joint conference on Artificial Inelligence. AAAI Press, 2013
11. V. Vapnik, Statistical Learning Theory. Hoboken, NJ, USA: Wiley, 1998.
12. C. C. Chang and C. J. Lin. (2001), LIBSVM: A library for support vector machines, [Online]. Available http://www.csie.ntu.edu. tw/ cjlin/libsvm
13. Minh-Thang Luong, Thuy Dung Nguyen and Min-Yen Kan (2010) Logical Structure Recovery in Scholarly Articles with Rich Document Features. International Journal of Digital Library Systems (IJDLS), 1(4), 1-23.