Blocking harmful images with a deep learning based next generation firewall

Kenan Baysal; Deniz Taşkin

Blocking harmful images with a deep learning based next generation firewall

Abstract

There are various blocking and filtering algorithms for protection against harmful contents on the Internet. However, it is impossible to classify particularly the visual contents according to their genres and block them through traditional methods. In order to block the harmful visual contents, such as various advertisements and social media posts, we need to review and classify them as per their contents. Deep learning method is today’s most efficient method to review the visual contents. In this study, only the harmful images were blocked without completely blocking the entire website. Alcoholic drinks were selected as the harmful content data set. For this purpose, a training was provided with 4.6 million images by using CNN (Convolutional Neural Net-works) and GoogLeNet architecture. At the end of this training, 97.6469% of accuracy was achieved. F1 score was calculated as 87.75526188% at the end of the test conducted with 154501 images. The images were determined through the network traffic via mitmproxy and classi-fied as harmful or harmless thanks to the trained model, and the filtering process was successfully completed.

Keywords

References

REFERENCES
[1] McCulloch WS, Pitts W. A logical calculus of the ideas immanent in nervous activity. Bull Math Biophys 1943;5:115133. [CrossRef]
[2] Demir I, Karaboğa HA. Modeling mathematics achievement with deep learning methods. Sigma J Eng Nat Sci 2021;39:3340. [CrossRef]
[3] Rajkovic KM, Avramovic JM, Milic PS, Stamenkovic OS. Optimization of ultrasound-assisted base-catalyzed methanolysis of sunflower oil using response surface and artificial neural network methodologies. Chem Eng J 2013;215:8289. [CrossRef]
[4] Zettler AH, Poisel R, Reichl I, Stadler G. Pressure Sensitive Grouting (PSG) using an artificial neural network combined with fuzzy logic. Int J Rock Mech Min Sci 1997;34:358. [CrossRef]
[5] Ma F, Sun T, Liu L, Jing H. Detection and diagnosis of chronic kidney disease using deep learning-based heterogeneous modified artificial neural network. Future Gener Comput Syst 2020;111:1726. [CrossRef]
[6] Sabilla SI, Sarno R, Siswantoro J. Estimating gas concentration using artificial neural network for electronic nose. Procedia Comput Sci 2017;124:181188. [CrossRef]
[7] Esen H, Esen M, Ozsolak O. Modelling and experimental performance analysis of solar-assisted ground source heat pump system. J Exp Theor Artif Intell 2017;29:117. [CrossRef]

[8] Esen H, Inalli M, Sengur A, Esen M. Predicting performance of a ground-source heat pump system using fuzzy weighted pre-processing-based ANFIS. Build Environ 2008;43:21782187. [CrossRef]
[9] Efe E, Alganci U. Determination of land cover change with multi-temporal Sentinel 2 satellite images and machine learning-based algorithms. Geomatik Derg 2023;8:2734. [Turkish] [CrossRef]
[10] We are social. Special report – Digital 2021. Your ultimate guide to the evolving digital world. Available at: https://wearesocial.com/digital-2021. Accessed on Jul 2, 2024.
[11] Yaraş E, Yetkin Özbük RY, Çorlu P. Emmy ödüllü dizilerde alkol ve sigara ürün yerleştirme uygulamalarının içerik analizi yöntemi ile incelenmesi. Kastamonu Univ İktis İdar Bil Fak Derg 2018;20:6784.
[12] İplikçi HG, Batu M. Digital communication and children: A content analysis of advertisements on the websites for children in Turkey. J Akdeniz İletiş 2018;29:242256. Turkish.
[13] Uzun R. The Protection of children from media content and in media content: A study of ethical codes for children in media. J Akdeniz İletiş 2014;22:152167.
[14] Kanbur BN. The effects of visual media and subliminal messages on child health. İstanbul Gelişim Univ Sağ Bil Derg 2020;10:94106. [Turkish] [CrossRef]
[15] Berners-Lee CM. Cybernetics and forecasting. Nature 1968;219:202203. [CrossRef]
[16] Lauriola I, Lavelli A, Aiolli F. An introduction to deep learning in natural language processing: Models, techniques, and tools. Neurocomput 2022;470:443456. [CrossRef]
[17] Gupta A, Anpalagan A, Guan L, Khwaja AS. Deep learning for object detection and scene perception in self-driving cars: Survey, challenges, and open issues. Array 2021;10:100057. [CrossRef]
[18] Fahad S, Ranjan A, Yadav J, Deepak A. A survey of speech emotion recognition in natural environment. Digit Signal Process 2021;110:102951. [CrossRef]
[19] Possemiers A, Lee I. Evaluating deep learned voice compression for use in video games. Expert Syst Appl 2021;181:115180. [CrossRef]
[20] Du X, Cai Y, Wang S, Zhang L. Owerview of deep learning. In proceedings of the 31st Youth Academic Annual Conference of Chinese Association of Automation; 2016 Nov 1113; Wuhan, China. IEEE; 2016. pp. 15964. [CrossRef]
[21] Pathak AR, Pandey M, Rautaray S. Application of deep learning for object detection. Procedia Comput Sci 2018;132:17061717. [CrossRef]
[22] Hu H, Pang L, Shi Z. Image matting in the perception granular deep learning. Knowl Based Syst 2016;102:5163. [CrossRef]
[23] Tiken C. Deep learning applications. Master's thesis. Istanbul: Istanbul Univ; 2015.
[24] Dong S, Wang P, Abbas K. A survey on deep learning and its applications. Comput Sci Rev 2021;40:100379. [CrossRef]
[25] Yiğit ÖE, Alp S, Öz E. Prediction of bist price indices: A comparative study between traditional and deep learning methods. Sigma J Eng Nat Sci 2020;38:16931704.
[26] Lecun Y, Bottou L, Bengio Y, Haffner P. Gradient-based learning applied to document recognition. Proc IEEE 1998;86:22782324. [CrossRef]
[27] Krizhevsky A, Sutskever I, Hinton GE. ImageNet classification with deep convolutional neural networks. Commun ACM. 2017;60:8490. [CrossRef]
[28] Wikipedia. List of alcoholic drinks. Available at: https://en.wikipedia.org/wiki/List_of_alcoholic_drinks. Accessed Jul 2, 2024.
[29] Sarkar N. Mean square error matrix comparison of some estimators in linear regressions with multicollinearity. Stat Probabil Lett 1996;30:133138. [CrossRef]
[30] Taylor L, Nitschke G. Improving deep learning with generic data augmentation. In proceedings of the 2018 IEEE Symposium Series on Computational Intelligence (SSCI); 2017 Nov 1821; Bangalore, India. IEEE; 2018. p. 15422547. [CrossRef]
[31] Shorten C, Khoshgoftaar TM. A survey on image data augmentation for deep learning. J Big Data 2019;6:60. [CrossRef]
[32] Augmentor. Available at: https://augmentor.readthedocs.io/en/master/. Accessed on Jul 2, 2024.
[33] Jia Y, Shelhamer E, Donahue J, Karayev S, Long J, Girshick R, et al. Convolutional architecture for fast feature embedding. In proceedings of the 22nd ACM International Conference on Multimedia; 2015 Jun 1819; California, USA. ACM; 2015. pp. 6758.
[34] Amidi A, Amidi S. Machine Learning tips and tricks cheatsheet. Available at: https://stanford.edu/~shervine/teaching/cs-229/cheatsheet-machine-learning-tips- and-tricks. Accessed on Jul 2, 2024.
[35] Mitmproxy Docs. Mitmproxy. Available at: https://docs.mitmproxy.org/stable/. Accessed Jul 2, 2024.
[36] Wang Y, Xu G, Liu X, Mao W, Si C, Pedrycz W, et al. Identifying vulnerabilities of SSL/TLS certificate verification in Android apps with static and dynamic analysis. J Syst Softw 2020;167:110609. [CrossRef]

Details

Primary Language

English

Subjects

Clinical Chemistry

Journal Section

Research Article

Authors

Kenan Baysal ^*
0000-0002-2205-7185
Türkiye

Deniz Taşkin This is me
0000-0001-7374-8165
Türkiye

Publication Date

August 1, 2024

Submission Date

January 6, 2023

Acceptance Date

March 12, 2023

Published in Issue

Year 2024 Volume: 42 Number: 4

IZ

https://izlik.org/JA98YY98UA

Cite

RIS / Bibtex

APA

Baysal, K., & Taşkin, D. (2024). Blocking harmful images with a deep learning based next generation firewall. Sigma Journal of Engineering and Natural Sciences, 42(4), 1133-1147. https://izlik.org/JA98YY98UA

AMA

1.Baysal K, Taşkin D. Blocking harmful images with a deep learning based next generation firewall. SIGMA. 2024;42(4):1133-1147. https://izlik.org/JA98YY98UA

Chicago

Baysal, Kenan, and Deniz Taşkin. 2024. “Blocking Harmful Images With a Deep Learning Based Next Generation Firewall”. Sigma Journal of Engineering and Natural Sciences 42 (4): 1133-47. https://izlik.org/JA98YY98UA.

EndNote

Baysal K, Taşkin D (August 1, 2024) Blocking harmful images with a deep learning based next generation firewall. Sigma Journal of Engineering and Natural Sciences 42 4 1133–1147.

IEEE

[1]K. Baysal and D. Taşkin, “Blocking harmful images with a deep learning based next generation firewall”, SIGMA, vol. 42, no. 4, pp. 1133–1147, Aug. 2024, [Online]. Available: https://izlik.org/JA98YY98UA

ISNAD

Baysal, Kenan - Taşkin, Deniz. “Blocking Harmful Images With a Deep Learning Based Next Generation Firewall”. Sigma Journal of Engineering and Natural Sciences 42/4 (August 1, 2024): 1133-1147. https://izlik.org/JA98YY98UA.

JAMA

1.Baysal K, Taşkin D. Blocking harmful images with a deep learning based next generation firewall. SIGMA. 2024;42:1133–1147.

MLA

Baysal, Kenan, and Deniz Taşkin. “Blocking Harmful Images With a Deep Learning Based Next Generation Firewall”. Sigma Journal of Engineering and Natural Sciences, vol. 42, no. 4, Aug. 2024, pp. 1133-47, https://izlik.org/JA98YY98UA.

Vancouver

1.Kenan Baysal, Deniz Taşkin. Blocking harmful images with a deep learning based next generation firewall. SIGMA [Internet]. 2024 Aug. 1;42(4):1133-47. Available from: https://izlik.org/JA98YY98UA