Page 262 - 《软件学报》2021年第10期

3234 Journal of Software 软件学报 Vol.32, No.10, October 2021

预测模型.此外,本文首次引入了生存分析领域中的 Concordance Index 作为评价指标,从而解决了不可见样本对
评价结果造成的偏差.
实验结果表明:本文提出的模型能够有效地应用于实际场景,显著提升电池续航时间预测的效果.数据显
示,查询时特征、会话特征和历史特征能够不同程度地帮助续航时间预测.其中,查询时特征的效果相对最弱;会
话特征作为用户的短期使用行为数据,效果最好,尤其与电池耗电速度直接相关的特征最为有效;历史特征作为
长期用户行为数据,也能够发挥一定的效果.最终,本文提出的模型可以将预测准确度提升 40 分钟.通过讨论,本
文论证了模型得到的效果在实际场景下具有足够的意义和价值.
下一步,本文计划从两方面作进一步探索:其一,本文希望能够尝试更多维度的特征,例如更多的传感器读
数和更加细粒度的应用使用行为数据;其二,本文希望进一步尝试更加前沿的机器学习模型,尤其是希望能够设
计一种生存分析模型,将不可见样本纳入到训练过程当中,以进一步提升预测的效果.

References:
[1] Balasubramanian N, Balasubramanian A, Venkataramani A. Energy consumption in mobile phones: A measurement study and
implications for network applications. In: Proc. of the 9th ACM SIGCOMM Conf. on Internet Measurement. ACM, 2009. 280293.
[2] Puustinen I, Nurminen JK. The effect of unwanted Internet traffic on cellular phone energy consumption. In: Proc. of the 4th IFIP
Int’l Conf. on New Technologies, Mobility and Security. IEEE, 2011. 15.
[3] Rosen S, Nikravesh A, Guo Y, et al. Revisiting network energy efficiency of mobile apps: Performance in the wild. In: Proc. of the
2015 Internet Measurement Conf. ACM, 2015. 339345.
[4] Shen H, Balasubramanian A, LaMarca A, et al. Enhancing mobile apps to use sensor hubs without programmer effort. In: Proc. of
the 2015 ACM Int’l Joint Conf. on Pervasive and Ubiquitous Computing. ACM, 2015. 227238.
[5] He S, Liu Y, Zhou H. Optimizing smartphone power consumption through dynamic resolution scaling. In: Proc. of the 21st Annual
Int’l Conf. on Mobile Computing and Networking. ACM, 2015. 2739.
[6] Chen X, Jindal A, Ding N, et al. Smartphone background activities in the wild: Origin, energy drain, and optimization. In: Proc. of
the 21st Annual Int’l Conf. on Mobile Computing and Networking. ACM, 2015. 4052.
[7] Draa I C, Tayeb J, Niar S, et al. Application sequence prediction for energy consumption reduction in mobile systems. In: Proc. of
the 2015 IEEE Int’l Conf. on Computer and Information Technology; Ubiquitous Computing and Communications; Dependable,
Autonomic and Secure Computing; Pervasive Intelligence and Computing. IEEE, 2015. 2330.
[8] Mirsky Y, Shabtai A, Rokach L, et al. Sherlock vs. moriarty: A smartphone dataset for cybersecurity research. In: Proc. of the 2016
ACM Workshop on Artificial Intelligence and Security. ACM, 2016. 112.
[9] Shye A, Scholbrock B, Memik G. Into the wild: Studying real user activity patterns to guide power optimizations for mobile
architectures. In: Proc. of the 42nd Annual IEEE/ACM Int’l Symp. on Microarchitecture. ACM, 2009. 168178.
[10] Dong M, Zhong L. Self-constructive high-rate system energy modeling for battery-powered mobile systems. In: Proc. of the 9th
Int’l Conf. on Mobile Systems, Applications, and Services. ACM, 2011. 335348.
[11] Zhang L, Tiwana B, Qian Z, et al. Accurate online power estimation and automatic battery behavior based power model generation
for smartphones. In: Proc. of the 8th IEEE/ACM/IFIP Int’l Conf. on Hardware/Software Codesign and System Synthesis. ACM,
2010. 105114.
[12] Min C, Yoo C, Hwang I, et al. Sandra helps you learn: The more you walk, the more battery your phone drains. In: Proc. of the
2015 ACM Int’l Joint Conf. on Pervasive and Ubiquitous Computing. ACM, 2015. 421432.
[13] Pathak A, Hu YC, Zhang M, et al. Fine-grained power modeling for smartphones using system call tracing. In: Proc. of the 6th
Conf. on Computer Systems. ACM, 2011. 153168.
[14] Mittal R, Kansal A, Chandra R. Empowering developers to estimate app energy consumption. In: Proc. of the 18th Annual Int’l
Conf. on Mobile Computing and Networking. ACM, 2012. 317328.
[15] Li D, Tran AH, Halfond WGJ. Making Web applications more energy efficient for OLED smartphones. In: Proc. of the 36th Int’l
Conf. on Software Engineering. ACM, 2014. 527538.