JDS Journal of Data Science 1683-86021680-743X1680-743X School of Statistics, Renmin University of China JDS1122 10.6339/24-JDS1122 Data Science in Action BIE: Binary Image Encoding for the Classification of Tabular Data https://orcid.org/0000-0003-1804-3263 HalladayJamesjehalladay112@gmail.com1∗ https://orcid.org/0000-0002-0000-8307 CullenDrake1 https://orcid.org/0000-0001-6529-5493 BrinerNathan1 MillerDarrin1 PrimeauRiley1 AvilaAbraham1 WatsonWarin1 https://orcid.org/0000-0001-6864-6893 BasnetRam1 https://orcid.org/0000-0002-1279-689X DoleckTenzin2 Department of Computer Science and Engineering, Colorado Mesa University (CMU), Grand Junction, CO 81501, USA Faculty of Education, Simon Fraser University, Burnaby, BC V5A 1S6, Canada Corresponding author. Email: jehalladay112@gmail.com. 20251942024231109129 Supplementary Material

We provide the code and datasets separately in the supplementary material. Included in the code is also all of the figures included in the paper in svg, pdf, and png format. The code reflects the contents of the github repository used for these experiments at the time of publication (Halladay et al., 2023).

131020231322024 2025 The Author(s). Published by the School of Statistics and the Center for Applied Statistics, Renmin University of China.2025 Open access article under the CC BY license.

There has been remarkable progress in the field of deep learning, particularly in areas such as image classification, object detection, speech recognition, and natural language processing. Convolutional Neural Networks (CNNs) have emerged as a dominant model of computation in this domain, delivering exceptional accuracy in image recognition tasks. Inspired by their success, researchers have explored the application of CNNs to tabular data. However, CNNs trained on structured tabular data often yield subpar results. Hence, there has been a demonstrated gap between the performance of deep learning models and shallow models on tabular data. To that end, Tabular-to-Image (T2I) algorithms have been introduced to convert tabular data into an unstructured image format. T2I algorithms enable the encoding of spatial information into the image, which CNN models can effectively utilize for classification. In this work, we propose two novel T2I algorithms, Binary Image Encoding (BIE) and correlated Binary Image Encoding (cBIE), which preserve complex relationships in the generated image by leveraging the native binary representation of the data. Additionally, cBIE captures more spatial information by reordering columns based on their correlation to a feature. To evaluate the performance of our algorithms, we conducted experiments using four benchmark datasets, employing ResNet-50 as the deep learning model. Our results show that the ResNet-50 models trained with images generated using BIE and cBIE consistently outperformed or matched models trained on images created using the previous State of the Art method, Image Generator for Tabular Data (IGTD).

 computer vision DeepInsight IGTD native representation ResNet-50 tabular-to-image State of ColoradoThis work was supported by the State of Colorado through funds appropriated for cybersecurity by a piece of legislation dubbed “Cyber Coding Cryptology for State Records.” References Aeberhard S, Forina M (1991). Wine. UCI Machine Learning Repository. https://doi.org/10.24432/C5PC7J Albanese C, Li D, Lobachevskiy E, Meissner G (2013). A comparative analysis of correlation approaches in finance. The Journal of Derivatives, 21(2): 42–66. https://doi.org/10.3905/jod.2013.21.2.042 Alom MZ, Taha TM, Yakopcic C, Westberg S, Sidike P, Nasrin MS, et al. (2018). The history began from alexnet: A comprehensive survey on deep learning approaches. Buturović L, Miljković D (2020). A novel method for classification of tabular data using convolutional neural networks. BioRxiv, 2020–05. https://doi.org/10.1101/2020.05.02.074203 Cawley GC, Talbot NL (2010). On over-fitting in model selection and subsequent selection bias in performance evaluation. Journal of Machine Learning Research, 11: 2079–2107. Choi S, Fang C, Haddad D, Kim M (2022). Predictive modeling of charge levels for battery electric vehicles using cnn efficientnet and igtd algorithm. arXiv preprint: https://arxiv.org/abs/2206.03612 Das HP, Spanos CJ (2022). Improved dequantization and normalization methods for tabular data pre-processing in smart buildings. In: Proceedings of the 9th ACM International Conference on Systems for Energy-Efficient Buildings, Cities, and Transportation, 168–177. Fisher RA (1936). The use of multiple measurements in taxonomic problems. Annals of Eugenics, 7(2): 179–188. https://doi.org/10.1111/j.1469-1809.1936.tb02137.x Fisher RA (1988). Iris. UCI Machine Learning Repository. https://doi.org/10.24432/C56C76 Gokhale M, Mohanty SK, Ojha A (2023). Genevit: Gene vision transformer with improved deepinsight for cancer classification. Computers in Biology and Medicine, 155: 106643. https://doi.org/10.1016/j.compbiomed.2023.106643 Halladay J, Cullen D, Briner N, Warren J, Fye K, Basnet R, et al. (2022). Detection and characterization of ddos attacks using time-based features. IEEE Access, 10: 49794–49807. https://doi.org/10.1109/ACCESS.2022.3173319 Halladay J, Cullen D, Briner N, Watson W, Miller D, Primeau R (2023). Binary image transformation: Github repository. Hand DJ, Christen P, Kirielle N (2021). F*: An interpretable transformation of the f-measure. Machine Learning, 110(3): 451–456. https://doi.org/10.1007/s10994-021-05964-1 He H, Garcia EA (2009). Learning from imbalanced data. IEEE Transactions on Knowledge and Data Engineering, 21(9): 1263–1284. https://doi.org/10.1109/TKDE.2008.239 He K, Zhang X, Ren S, Sun J (2016a). Deep residual learning for image recognition. In: Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, 770–778. He K, Zhang X, Ren S, Sun J (2016b). Identity mappings in deep residual networks. In: Proceedings, Part IV 14. Computer Vision–ECCV 2016: 14th European Conference. Amsterdam, The Netherlands. October 11–14, 2016, 630–645. Springer. Hossin M, Sulaiman MN (2015). A review on evaluation metrics for data classification evaluations. International Journal of Data Mining & Knowledge Management Process, 5(2): 1. https://doi.org/10.5121/ijdkp.2015.5201 Iqbal MI, Mukta MSH, Hasan AR, Islam S (2022). A dynamic weighted tabular method for convolutional neural networks. IEEE Access, 10: 134183–134198. https://doi.org/10.1109/ACCESS.2022.3231102 Koonce B (2021). ResNet 50, 63–72. Apress, Berkeley, CA Krizhevsky A, Sutskever I, Hinton GE (2012). Imagenet classification with deep convolutional neural networks. Advances in neural information processing systems, (F Pereira, CJ Burges, L Bottou, KQ Weinberger, eds.) 25. Krupski J, Graniszewski W, Iwanowski M (2021). Data transformation schemes for cnn-based network traffic analysis: A survey. Electronics, 10(16): 2042. https://doi.org/10.3390/electronics10162042 Ling CX, Huang J, Zhang H (2003). Auc: A better measure than accuracy in comparing learning algorithms. In: Advances in Artificial Intelligence: 16th Conference of the Canadian Society for Computational Studies of Intelligence (Y Xiang, (B Chaib-Draa, eds.), volume 16 of Proceedings, AI 2003, Halifax, Canada, June 11–13, 2003, 329–341. Springer. Moskalenko A, Moskalenko V, Shaiekhov A, Zaretskyi M (2020). Multi-layer model and training method for information-extreme malware traffic detector. In: CMIS, 288–299. Noroozi M, Favaro P (2016). Unsupervised learning of visual representations by solving jigsaw puzzles. In: European Conference on Computer Vision, 69–84. Springer. Rabbah J, Ridouani M, Hassouni L (2022). A new churn prediction model based on deep insight features transformation for convolution neural network architecture and stacknet. International Journal of Web-Based Learning and Teaching Technologies (IJWLTT), 17(1): 1–18. https://doi.org/10.4018/ijwltt.300342 Seger C (2018). An investigation of categorical variable encoding techniques in machine learning: Binary versus one-hot and feature hashing. Sharafaldin I, Lashkari AH, Hakak S, Ghorbani AA (2019). Developing realistic distributed denial of service (ddos) attack dataset and taxonomy. In: 2019 International Carnahan Conference on Security Technology (ICCST), 1–8. IEEE. Sharma A, Vans E, Shigemizu D, Boroevich KA, Tsunoda T (2019). Deepinsight: A methodology to transform a non-image data to an image for convolution neural network architecture. Scientific Reports, 9(1): 11399. https://doi.org/10.1038/s41598-019-47765-6 Simon M, Rodner E, Denzler J (2016). Imagenet pre-trained models with batch normalization. arXiv preprint: https://arxiv.org/abs/1612.01452 Simonyan K, Zisserman A (2014). Very deep convolutional networks for large-scale image recognition. arXiv preprint: https://arxiv.org/abs/1409.1556 Sun B, Yang L, Zhang W, Lin M, Dong P, Young C, et al. (2019). Supertml: Two-dimensional word embedding for the precognition on structured tabular data. In: Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops. 2973–2981. Szegedy C, Vanhoucke V, Ioffe S, Shlens J, Wojna Z (2016). Rethinking the inception architecture for computer vision. In: Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, 2818–2826. Taheri A, Ebrahimnezhad H, Sedaaghi MH (2022). Prediction of the critical temperature of superconducting materials using image regression and ensemble deep learning. Materials Today Communications, 33: 104743. https://doi.org/10.1016/j.mtcomm.2022.104743 Wang W, Zhu M, Zeng X, Ye X, Sheng Y (2017). Malware traffic classification using convolutional neural network for representation learning. In: 2017 International Conference on Information Networking (ICOIN), 712–717. IEEE. Wolberg W, Mangasarian O, Street N, Street W (1995). Breast Cancer Wisconsin (Diagnostic). UCI Machine Learning Repository. https://doi.org/10.24432/C5DW2B Xie S, Girshick R, Dollár P, Tu Z, He K (2017). Aggregated residual transformations for deep neural networks. In: Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, 1492–1500. Zhu Y, Brettin T, Xia F, Partin A, Shukla M, Yoo H, et al. (2021). Converting tabular data into images for deep learning with convolutional neural networks. Scientific Reports, 11(1): 11325. https://doi.org/10.1038/s41598-021-90923-y