智能软件可靠性综合评测指标体系模型的构建

高德鹏

doi:10.70693/cjst.v2i6.2173

Authors

高德鹏南通理工学院

DOI:

https://doi.org/10.70693/cjst.v2i6.2173

Keywords:

智能软件, 深度神经网络, 可靠性, 指标体系, 综合评测

Abstract

人工智能驱动下的智能软件应用规模持续扩大，传统软件的可靠性评测方法因难以适配数据驱动的深度神经网络模型，无法直接用于智能软件的可靠性评估。针对目前智能软件缺少系统化、完备化可靠性评测指标体系的问题，本文从模型、数据、运行平台三个维度解析智能软件可靠性的内涵，面向目标识别、目标检测、智能决策典型业务场景，构建包含通用性能评测指标、智能计算能力评测指标、智能任务评测指标、可靠性评估指标的综合评测指标体系模型。该体系从平台资源消耗、智能计算加速效能、智能任务输出效果、软件运行失效等多个层面实现对智能软件可靠性的多维度度量，可为智能软件的可靠性测试、性能优化与质量评估提供理论参考与评测依据。

References

[1] 高萍萍.大数据驱动下的智能软件开发框架研究[J].信息与电脑,2026,38(9):207-209.

[2] He W, Li B, Song D. Decision Boundary Analysis of Adversarial Examples[C]. International Conference on Learning Representations. 2018,1-15.

[3] Guo J., Bao W., Wang J., et al. A Comprehensive Evaluation Framework for Deep Model Robustness[J]. Pattern Recognition, 2023, 137: 109308.

[4] Zhang D., Dong Y. Adv-BDPM: Adversarial Attack Based on Boundary Diffusion Probability Model[J]. Neural Networks, 2023, 167: 730-740.

[5] Ukita J, Ohki K. Adversarial Attacks and Defenses Using Feature-Space Stochasticity[J]. Neural Networks, 2023.

[6] Pons L., Ozkaya I. Priority Quality Attributes for Engineering AI-enabled Systems[J]. arXiv.1911.02912, 2019.

[7] Goldstein B F, Srinivasan S, Das D, et al. Reliability evaluation of compressed deep learning models[C]//2020 IEEE 11th Latin American Symposium on Circuits & Systems (LASCAS). IEEE, 2020: 1-5.

[8] Kumar R, Kumar S, Tiwari S K. A study of software reliability on big data open source software[J]. International Journal of System Assurance Engineering and Management, 2019, 10(2): 242-250.

[9] Alshemali B, Kalita J. Improving the reliability of deep neural networks in NLP: A review[J]. Knowledge-Based Systems, 2020, 191: 105210.