Multimodal Deep Learning for Malware Behavior Detection with Integrated Database Storage Optimization
Abstract
With the rapid development of Internet technology, the types of malwares are constantly increasing, which brings significant challenges to network security. Traditional malware detection methods mainly rely on feature extraction and classifier design, but these methods have certain limitations when dealing with complex and changeable malware behaviors. To solve this problem, this study proposes a malware behavior detection method based on multi-modal deep learning combined with database storage optimization technology. This method will extract multi-dimensional malware features and utilize deep learning models for learning and classification to improve the accuracy and efficiency of detection. The experimental data results show that the proposed method in this study is highly accurate and robust in malware behavior detection. By detecting 1000 malware samples, the method in this study can accurately identify the behavioral characteristics of 950 of them and effectively classify them, with a detection accuracy of up to 95%. Compared with other traditional malware detection methods, the detection accuracy of traditional methods is 70% on average, while the method in this study can reach more than 90%. In terms of false reporting rate, the traditional method is about 30%, but the method in this study can be controlled within 5%. In terms of false alarm rate, the traditional method is about 20%, but the method in this study can be reduced to about 3%, which shows apparent advantages. By introducing database storage optimization technology, the method in this study can not only improve the accuracy of malware detection but also effectively reduce the storage pressure of the database, reduce the storage space of the database by about 40% compared with that before optimization, improve the running efficiency of the system, and shorten the overall response time of the system by about 30%.
Full Text:
PDFReferences
Li, M., S. Deng, H. Zhou & Y. Qin, “A path selection scheme for detecting malicious behavior based on deep reinforcement learning in SDN/NFV-Enabled network,” Computer Networks, vol. 236, pp. 110034, 2023.
Maniriho, P., A. N. Mahmood & M. J. M. Chowdhury, “A study on malicious software behaviour analysis and detection techniques: Taxonomy, current trends and challenges,” Future Generation Computer Systems, vol. 130, pp. 1-18, 2022.
Shafi, M., A. H. Lashkari & A. H. Roudsari, “NTLFlowLyzer: Towards generating an intrusion detection dataset and intruders behavior profiling through network and transport layers traffic analysis and pattern extraction,” Computers & Security, vol. 148, pp. 104160, 2025.
Singh, J. & J. Singh, “Detection of malicious software by analyzing the behavioral artifacts using machine learning algorithms,” Information and Software Technology, vol. 121, pp. 106273, 2020.
Valluri, B. P. & N. Sharma, “Trusted head node for Node Behaviour Analysis for malicious node detection in wireless sensor networks,” Measurement: Sensors, vol. 36, pp. 101159, 2024.
Wang, J., B. Zhang, K. Wang, Y. Wang & W. Han, “BFTDiagnosis: An automated security testing framework with malicious behavior injection for BFT protocols,” Computer Networks, vol. 249, pp. 110404, 2024.
Guo, S., N. Sun, Y. Pei, & Q. Li, “3D-UNet-LSTM: A deep learning-based radar echo extrapolation model for convective nowcasting,” Remote Sensing, vol. 15, no. 6, 2023.
Kuriki, D., F. Briffod, T. Shiraiwa & M. Enoki, “Multimodal deep learning framework to predict strain localization of Mg/LPSO two-phase alloys,” Acta Materialia, vol. 281, pp. 120398, 2024.
Peng, L., S. Jian, M. Li, Z. Kan, L. Qiao & D. Li, “A unified multimodal classification framework based on deep metric learning,” Neural Networks, vol. 181, pp. 106747, 2025.
Guzel, M., M. Kalkan, E. Bostanci, K. Acici, & T. Asuroglu, “Cloud type classification using deep learning with cloud images,” Peerj Computer Science, vol. 10,2024.
Hadadi, F., J.H. Dawes, D. Shin, D. Bianculli, & L. Briand, “Systematic evaluation of deep learning models for log-based failure prediction,” Empirical Software Engineering, vol.29, no. 5, 2024.
Hashmi, A., O.M. Barukab, & A.H. Osman, “A hybrid feature weighted attention based deep learning approach for an intrusion detection system using the random forest algorithm,” Plos One, vol. 19, no. 5, 2024.
He, Y., Y. Ma, K. Huang, L. Wang, & J. Zhang, “Abnormal data detection and recovery of sensors network based on spatiotemporal deep learning methodology,” Measurement, vol. 228, 2024.
Hnamte, V., H. Nhung-Nguyen, J. Hussain, & Y. Hwa-Kim, “A novel two-stage deep learning model for network intrusion detection: LSTM-AE,” IEEE Access, vol. 11, pp. 37131-37148, 2023.
Huang, K., K. Wei, F. Li, C. Yang, & W. Gui, “LSTM-MPC: A deep learning based predictive control method for multimode process control,” IEEE Transactions on Industrial Electronics, vol. 70, no. 11, pp. 11544-11554, 2023.
Chang, Z.H., Y. Zhang, and H.C. Chen, “Dynamic LSTM for 5G signal power amplifier behavioral model,” IEEE Microwave and Wireless Technology Letters, vol. 34, no. 6, pp. 683-686, 2024.
Liu, B., & J. Cai, “Comparative analysis of nonlinear behavioral models for GaN HEMTs based on machine learning techniques,” International Journal of Numerical Modelling-Electronic Networks Devices and Fields, vol. 37, no. 2, 2024.
Wang, R., R. Toth, P. J. W. Koelewijn, & I. R. Manchester, “Virtual control contraction metrics:Convex nonlinear feedback design via behavioral embedding,” International Journal of Robust and Nonlinear Control, vol. 34, no. 12, pp. 7698-7721, 2024.
Zong, L., & J. Fang, “Deep visual computing of behavioral characteristics in complex scenarios and embedded object recognition applications,” Sensors, vol. 24, no. 14, 2024.
Bhushan, K., & B. B. Gupta, “Network flow analysis for detection and mitigation of Fraudulent Resource Consumption(FRC)attacks in multimedia cloud computing,” Multimedia Tools and Applications, vol. 78, no. 4, pp. 4267-4298, 2019.
Shang, Y. & T. Fu, “Multimodal fusion: A study on speech-text emotion recognition with the integration of deep learning,” Intelligent Systems with Applications, vol. 24, pp. 200436, 2024.
Chauhan, V. K. & A. Kumar, “Cascaded capsule twin attentional dilated convolutional network for malicious URL detection,” Expert Systems with Applications, vol. 262, pp. 125507, 2025.
Jeon, S.-E., Y.-S. Oh, Y.-J. Lee & I.-G. Lee, “Suboptimal feature selection techniques for effective malicious traffic detection on lightweight devices,” CMES - Computer Modeling in Engineering and Sciences, vol. 140, no. 2, pp. 1669-1687, 2024.
Kalaria, R., A. S. M. Kayes, W. Rahayu, E. Pardede & A. Salehi S, “IoTPredictor: A security framework for predicting IoT device behaviours and detecting malicious devices against cyber attacks,” Computers & Security, vol. 146, pp. 104037, 2024.
Khashan, O. A., “Dual-stage machine learning approach for advanced malicious node detection in WSNs,” Ad Hoc Networks, vol. 166, pp. 103672, 2025.
Kolasa, D., K. Pilch & W. Mazurczyk, “Federated learning secure model: A framework for malicious clients detection,” SoftwareX, vol. 27, pp. 101765, 2024.
Wang, Y., R. Xiao, J. Sun & S. Jin, “MC-Det: Multi-channel representation fusion for malicious domain name detection,” Computer Networks, vol. 255, pp. 110847, 2024.
Yevsikov, A., T. Muralidharan, T. Panker & N. Nissim, “CADefender: Detection of unknown malicious AutoLISP computer-aided design files using designated feature extraction and machine learning methods,” Engineering Applications of Artificial Intelligence, vol. 138, pp. 109414, 2024.
Afify, G. M., A. El Bastawissy & O. M. Hegazy, “A hybrid filtering approach for storage optimization in main-memory cloud database,” Egyptian Informatics Journal, vol. 16, no. 3, pp. 329-337, 2015.
Rafi, S. M., R. Yogesh & M. Sriram, “Optimized dual access control for cloud-based data storage and distribution using global-context residual recurrent neural network,” Computers & Security, vol., pp. 104183, 2024.
DOI: https://doi.org/10.31449/inf.v49i10.8876
This work is licensed under a Creative Commons Attribution 3.0 License.