Türkçe Sözde Algoritma Kodu için LSTM Tabanlı Kaynak Kod Üretimi

Abstract

Öğrencilerin algoritmik düşünme ve programlama yetenekleri, okullarda ve üniversitelerde teknolojik eğitim programlarında tartışmalı ve popüler bir konudur. İyi bir matematiksel ve analitik geçmişe sahip olmayan öğrenciler, bilgisayar programlamayı öğrenmede zorluklar yaşayabilir. Dahası, programlama öğrenme, bireyin sözde algoritma kodu ile kaynak kodu arasında bağlantı kurması için oldukça zordur. Başka bir problem, bir program kodu yazmak için gereken zamandır. Bu problemi çözmek için öğrencilere sözde kod ile kaynak kod arasındaki ilişkiyi analiz etme ve fark etme konusunda eğitim veren bazı araçlar vardır. Bu çalışmada, Türkçe sözde kodlardan LSTM tabanlı kaynak kod üreteci olan bir derin öğrenme yöntemi önerilmiştir. Bu amaçla, Uzun-Kısa Süreli Bellek (LSTM)'i eğitmek için veri seti olarak meslek yüksek okulundaki programlamaya giriş dersi sınavları kullanılmıştır. Kullanıcılar bir Türkçe sözde algoritma kodunu sorguladığında C # kaynak kodu üretilmektedir. Önerilen sistemin başarısını ölçmek için üretilen kaynak kodu ve öğretim elemanının kaynak kodu metin benzerlik yöntemleri ile analiz edilmiştir. Sonuçlar, önerilen sistemin öğrencilerin temel programlama becerilerini öğrenmeleri için yararlı olduğunu göstermektedir.

Keywords

• Otomatik kod üretimi
• Derin öğrenme
• Uzun kısa süreli bellek

LSTM Based Source Code Generation for Turkish Pseudo Code of Algorithm

Abstract

Algorithmic thinking and programming abilities of students is controversial and popular issue in technological education programs in schools and universities. Students that have not best mathematical and analytical background may have difficulties in learning computer programing. Moreover, learning programming is highly difficult for a single individual to establish connection between discrete pseudo code of algorithm and source code. Another problem is required time to write a piece of program code. In order to solve this problem, there are some tools that tutor students to get analyze and realize relation between pseudo code and source code. In this study, we propose a deep learning method that is Long Short Term-Memory (LSTM) based source code generator from Turkish pseudo codes. For this purpose, we used Introduction to programming course exams in vocational high school as dataset to train LSTM. When users query a Turkish pseudo code of algorithm, C# source code is generated. In order to measure success of proposed system, generated source code and instructor’s source code is analyzed with text similarity methods. Results show that proposed system is useful for students to learn fundamental programming skills.

Keywords

• Automatic code generation
• Deep learning
• Long short term-memory

References

1. [1] B. Robson, “Computers and viral diseases. Preliminary bioinformatics studies on the design of a synthetic vaccine and a preventative peptidomimetic antagonist against the SARS-CoV-2 (2019-nCoV, COVID-19) coronavirus,” Computers in Biology and Medicine, vol. 119, pp. 103670, 2020.
2. [2] B. Drury, M. Roche, “A survey of the applications of text mining for agriculture,” Computers and Electronics in Agriculture, vol. 163, pp. 104864, 2019.
3. [3] R. Warner, S. D. Sowle, and W. Sadler, “Teaching law with computers,” Rutgers Computer & Tech, vol. 24, no. 107, pp. 156-158, 1998.
4. [4] R. P. Feynman, “Simulating physics with computers,” International Journal of Theoretical Physics, vol. 21, pp. 467-488, 1982.
5. [5] M. Duran, T. Aytaç, “Students' opinions on the use of tablet computers in education,” European Journal of Contemporary Education, vol. 15, no. 1, pp. 65-75, 2016.
6. [6] Y. Qian, J. Lehman, “Students’ misconceptions and other difficulties in introductory programming: a literature review,” ACM Transactions on Computing Education (TOCE), vol. 18, no. 1, pp. 1-24, 2017.
7. [7] E. Lahtinen, K. Ala-Mutka and H. M. Järvinen, “A study of the difficulties of novice programmers,” Acm Sigcse Bulletin, vol. 37, no. 13, pp. 14-18, 2005.
8. [8] P. H. Tan, C. Y. Ting, S. W. Ling, “Learning difficulties in programming courses: Undergraduates' perspective and perception,” in International Conference on Computer Technology and Development, 2009, pp. 42-46.

9. [9] V. Renumol, S. Jayaprakash, and D. Janakiram, “Classification of cognitive difficulties of students to learn computer programming,” Indian Institute of Technology, vol. 12, pp. 1-12, 2009.
10. [10] M. Egea, C. Dania, “SQL-PL4OCL: An automatic code generator from OCL to SQL procedural language,” Software & Systems Modeling, vol. 18, no. 1, pp. 769-791, 2019.
11. [11] M. Allamanis, D. Tarlow, A. Gordon, and Y. Wei, “Bimodal modelling of source code and natural language,” in International Conference on Machine Learning, 2015, pp. 2123-2132.
12. [12] M. Raghothaman, Y. Wei, and Y. Hamadi, “Swim: synthesizing what i mean-code search and idiomatic snippet synthesis,” in IEEE/ACM International Conference on Software Engineering (ICSE), 2016, pp. 357-367.
13. [13] J. Galenson, P. Reames, R. Bodik, B. Hartmann, and K. Sen, “Codehint: dynamic and interactive synthesis of code snippets,” in International Conference on Software Engineering, 2014, pp. 653-663.
14. [14] T. T. Nguyen, A. T. Nguyen, H. A. Nguyen, and T. N. Nguyen, “A statistical semantic language model for source code,” in Joint Meeting on Foundations of Software Engineering, 2013, pp. 532-542.
15. [15] C. Maddison, and D. Tarlow, “Structured generative models of natural source code,” in International Conference on Machine Learning, 2014, pp. 649-657.
16. [16] E. Parisotto, A. R. Mohamed, R. Singh, L. Li, D. Zhou, and P. Kohli, ”Neuro-symbolic program synthesis,” 2016. [Online]. Available: arXiv:1611.01855.
17. [17] M. Balog, A. L. Gaunt, M. Brockschmidt, S. Nowozin, and D. Tarlow, “Deepcoder: learning to write programs,” 2016. [Online]. Available: arXiv:1611.01989.
18. [18] M. H. Manshadi, D. Gildea, and J. F. Allen, “Integrating programming by example and natural language programming,” in AAAI Conference on Artificial Intelligence, 2013, pp. 661-667.
19. [19] H. Lieberman, Your Wish is my Command: Programming by Example, Burlington, Massachusetts, USA: Morgan Kaufmann Publishers, 2001.
20. [20] S. Gulwani, W. R. Harris, and R. Singh, “Spreadsheet data manipulation using examples,” Communications of the ACM, vol. 55, no. 8, pp. 97-105, 2012.
21. [21] M. Raza, S. Gulwani, and N. Milic-Frayling, “Compositional program synthesis from natural language and examples,” in International Joint Conference on Artificial Intelligence, 2015, pp. 792-800.
22. [22] T. Lei, F. Long, R. Barzilay, and M. Rinard, “From natural language specifications to program input parsers,” in Annual Meeting of the Association for Computational Linguistics, 2013, pp. 1294-1303.
23. [23] Y. Danilchenko, and R. Fox, “Automated code generation using case-based reasoning, routine design and template-based programming,” in Midwest Artificial Intelligence and Cognitive Science Conference, 2012, pp. 119-125.
24. [24] S. Mukherjee, T. Chakrabarti, “Automatic algorithm specification to source code translation,” Indian Journal of Computer Science and Engineering (IJCSE), vol. 2, no. 2, pp. 146-159, 2011.
25. [25] L. Mou, R. Men, G. Li, L. Zhang, and Z. Jin, “On end-to-end program generation from user intention by deep neural networks,” 2015. [Online]. Available: arXiv:1510.07211.
26. [26] X. Chen, C. Liu, and D. Song, “Tree-to-tree neural networks for program translation,” in Advances in Neural Information Processing Systems, 2018, pp. 2547-2557.
27. [27] V. V. Nabiyev, Yapay Zeka, 4. baskı, Ankara, Türkiye: Seçkin Yayıncılık, 2012.
28. [28] M. H. Stefanini, Y. Demazeau, “TALISMAN: A multi-agent system for natural language processing,” in Brazilian Symposium on Artificial Intelligence, 1995, pp. 312-322.
29. [29] S. Sun, C. Luo, and J. Chen, “A review of natural language processing techniques for opinion mining systems,” Information fusion, vol. 36, pp. 10-25, 2017.
30. [30] T. Strzalkowski, F. Lin, J. Wang, and J. Perez-Carballo, “Evaluating natural language processing techniques in information retrieval,” in Natural Language Information Retrieval, Dordrecht: Springer, 1999, pp. 113-145.
31. [31] T. Nasukawa, J. Yi, “Sentiment analysis: Capturing favorability using natural language processing,” in International Conference on Knowledge Capture, 2003, pp. 70-77.
32. [32] Y. Aktaş, E. Y. İnce, and A. Çakır, “Doğal dil işleme kullanarak bilgisayar ağ terimlerinin wordnet ontolojisinde uyarlanması,” Teknik Bilimler Dergisi, vol. 7, no. 2, pp. 1-9, 2017.
33. [33] J. Cushing, R. Hastings, “Introducing computational linguistics with NLTK (natural language toolkit),” Journal of Computing Sciences in Colleges, vol. 25, no. 1, pp. 167-169, 2009.
34. [34] S. Savaş, N. Topaloğlu, “Data analysis through social media according to the classified crime,” Turkish Journal of Electrical Engineering & Computer Sciences, vol. 27, no. 1, pp. 407-420, 2019.
35. [35] E. Y. İnce, “Spell checking and error correcting application for Turkish,” International Journal of Information and Electronics Engineering, vol. 7, no. 2, pp. 68-71, 2017.
36. [36] S. Deerwester, S. T. Dumais, G. W. Furnas, T. K. Landauer, and R. Harshman, “Indexing by latent semantic analysis,” Journal of the American Society for Information Science, vol. 41, no. 6, pp. 391-407, 1990.
37. [37] S. T. Dumais, “Latent semantic analysis,” Annual Review of Information Science and Technology, vol. 38, no. 1, pp. 188-230, 2004.
38. [38] L. Deng, D. Yu, “Deep learning: methods and applications,” Foundations and Trends® in Signal Processing, vol. 7, no. 3, pp. 197-387, 2014.
39. [39] Y. LeCun, Y. Bengio, and G. Hinton, ”Deep learning,” Nature, vol. 521, no. 7553, pp. 436-444, 2015.
40. [40] A. Krizhevsky, I. Sutskever, and G. E. Hinton, “Imagenet classification with deep convolutional neural networks,” in Advances in Neural Information Processing Systems, 2012, pp. 1097-1105.
41. [41] J. Salamon, J. P. Bello, “Deep convolutional neural networks and data augmentation for environmental sound classification,” IEEE Signal Processing Letters, vol. 24, no. 3, pp. 279-283, 2017.
42. [42] Y. Bengio, P. Simard, and P. Frasconi, “Learning long-term dependencies with gradient descent is difficult,” IEEE Transactions on Neural Networks, vol. 5, no. 2, pp. 157-166, 1994.
43. [43] A. Graves, and J. Schmidhuber, “Offline handwriting recognition with multidimensional recurrent neural networks,” in Advances in Neural Information Processing Systems, 2009, pp. 545-552.
44. [44] T. Hughes, and K. Mierle, “Recurrent neural networks for voice activity detection,” in IEEE International Conference on Acoustics, Speech and Signal Processing, 2013, pp. 7378-7382.
45. [45] C. Wang, H. Yang, C. Bartz, and C. Meinel, “Image captioning with deep bidirectional LSTMs,” in ACM International Conference on Multimedia, 2016, pp. 988-997.
46. [46] M. N. Al-Kabi, T. M. Hailat, E. M. Al-Shawakfa, and I. M. Alsmadi, “Evaluating English to Arabic machine translation using BLEU,” International Journal of Advanced Computer Science and Applications, vol. 4, no. 1, 2013.
47. [47] S. Stoll, N. C. Camgoz, S. Hadfield, and R. Bowden, “Text2Sign: Towards sign language production using neural machine translation and generative adversarial networks,” International Journal of Computer Vision, vol. 128, pp. 891–908, 2020.
48. [48] T. Y. Lin, and P. Dollar. (2016, Feb 14) Mscocoapi [Online]. Available: https://github.com/cocodataset/cocoapi.
49. [49] W. B. Langdon, J. Dolado, F. Sarro, and M. Harman, “Exact mean absolute error of baseline predictor MARP0,” Information and Software Technology, vol. 73, pp. 16-18, 2016.
50. [50] E. Arısoy, H. Dutağacı, and L. M. Arslan, “A unified language model for large vocabulary continuous speech recognition of Turkish,” Signal Processing, vol. 86, no. 10, pp. 2844-2862, 2006.