FORECASTING THE NUMBER OF INCIDENTS USING A NEURAL NETWORK BASED ON THE KERAS FRAMEWORK

UDC 004.032.26

 

Mushchuk Artur Nikolaevich – Master's degree student, the Department of Software Engineering. Belarusian State Technological University (13a, Sverdlova str., 220006, Minsk, Republic of Belarus). E-mail: mushuk-artur@mail.ru

DOI: https://doi.org/10.52065/2520-6141-2024-278-9

 

Key words: neural network, machine learning, Keras, long short-term memory.

For citation: Mushchuk A. N. Forecasting the number of incidents using a neural network based on the Keras framework. Proceedings of BSTU, issue 3, Physics and Mathematics Informatics, 2024, no. 1 (278), pp. 58–63 (In Russian). DOI: 10.52065/2520-6141-2024-278-9.

Abstract

The article presents the process of developing a neural network architecture for predicting the number of incidents based on socio-economic indicators in Python using the Keras framework. The entire design cycle of such architectures is considered. The methodology for constructing an architecture includes several stages: collecting, grouping and processing data using appropriate methods, forming a training and test set, selecting a suitable network architecture taking into account the specific requirements of the forecasting problem being solved. Next, the neural network is implemented, trained, and tested on an independent data set to evaluate performance and accuracy. A comparative analysis of the proposed architecture with the known ones is carried out according to the criterion of mean absolute error.

Download

References

  1. Tufail Sh., Riggs H., Tariq M., Sarwat A. I. Advancements and Challenges in Machine Learning: A Comprehensive Review of Models, Libraries, Applications, and Algorithms Electronics, 2023, vol. 12, no. 8, p. 1789. DOI: 10.3390/electronics12081789.
  2. Paulenka D. A., Kovalev V. A., Snezhko E. V., Liauchuk V. A., Pechkovsky E. I. Recognition of underlying surface using a convolutional neural network on a single-board computer. Informatika [Informatics], 2020, vol. 17, no. 3, pp. 36–43. DOI: 10.37661/1816-0301-2020-17-3-36-43 (In Russian).
  3. Sorokina V. V, Ablameyko S. V. Detection of human body parts on the image using the neural networks and the attention model. Zhurnal Belorusskogo gosudarstvennogo universiteta. Matematika. Informatika [Journal of the Belarusian State University. Mathematics and Informatics], 2022, vol. 2, pp. 94–106. DOI: 10.33581/2520-6508-2022-2-94-106. (In Russian).
  4. Portugal I., Alencar P., Cowan D. The use of machine learning algorithms in recommender systems: A systematic review. Expert Systems with Application, 2018, vol. 97, pp. 205–227. DOI: 10.1016/j.eswa. 2017.12.020.
  5. Roslyakova S., Bragina T., Zemlyanova E., Korotkova D., Merkulova Р., Laushkina A., Filippov I. On the possibilities of applying machine learning in lighting control systems. Nauchnyy rezul'tat. Ser.: Informatsionnyye tekhnologii [Research result. Information technologies], 2021, vol. 6, no. 4, pp. 48–63 (In Russian). DOI: 10.18413/2518-1092-2021-6-4-0-7.
  6. Almalki A., Gokaraju B., Acquaah Y., Turlapaty A. Regression Analysis for COVID-19 Infections and Deaths Based on Food Access and Health Issues. Healthcare, 2022, vol. 10, no. 2, p. 324. DOI: 10.3390/healthcare10020324
  7. Bali V. [et al.]. Life Expectancy: Prediction & Analysis using ML. ICRITO: 9th International Conference on Reliability, Infocom Technologies and Optimization (Trends and Future Directions), 2021. P. 1–8. DOI: 10.1109/ICRITO51393.2021.9596123.
  8. Rajendran S., Chamundeswari S., Sinha A. Predicting the academic performance of middle-and highschool students using machine learning algorithms. Social Sciences & Humanities Open, 2022, vol. 6, no. 1, p. 100357. DOI: 10.1016/j.ssaho.2022.100357.
  9. Aydemir B. [et al.]. Predicting the Income Groups and Number of Immigrants by Using Machine Learning (ML). International Journal of Multidisciplinary Studies and Innovative Technologies, 2022, vol. 6, no. 2, pp. 162–168. DOI: 10.36287/ijmsit.6.2.162.
  10. Shah N., Bhagat N., Shah M. Crime forecasting: a machine learning and computer vision approach to crime prediction and prevention. Visual Computing for Industry, Biomedicine, and Art, 2021, vol. 4, pp. 1–14. DOI: 10.1186/s42492-021-00075-z.
  11. Sarker I. Machine learning: Algorithms, real-world applications and research directions. SN computer science, 2021, vol. 2, no. 3, p. 160. DOI: 10.1007/s42979-021-00592-x.
  12. Chollet F. Glubokoye obucheniye na Python [Deep Learning with Python]. St. Petersburg, Piter Publ., 2018. 400 p. (In Russian).
  13. McKinney W. In Python for data analysis. O’Reilly Media, 2013, issue 2, pp. 207–209.
  14. Aurelien J. Prikladnoye mashinnoye obucheniye s pomoshch'yu Scikit-Learn i TensorFlow [Applied machine learning with Scikit-Learn and TensorFlow]. St. Petersburg, Al'fa-kniga Publ., 2018. 684 p. (In Russian).
  15. Goodfellow I., Bengio Y., Courville A. Deep Learning. Massachusetts, Massachusetts Institute of Technology Publ., 2017. 804 p.

09.02.2024