Skip to main content

Energy-efficient prediction of smartphone unlocking

Abstract

We investigate the predictability of the next unlock event on smartphones, using machine learning and smartphone contextual data. In a 2-week field study with 27 participants, we demonstrate that it is possible to predict when the next unlock event will occur. Additionally, we show how our approach can improve accuracy and energy efficiency by solely relying on software-related contextual data. Based on our findings, smartphone applications and operating systems can improve their energy efficiency by utilising short-term predictions to minimise unnecessary executions, or launch computation-intensive tasks, such as OS updates, in the locked state. For instance, by inferring the next unlock event, smartphones can pre-emptively collect sensor data or prepare timely content to improve the user experience during the subsequent phone usage session.

This is a preview of subscription content, access via your institution.

Fig. 1
Fig. 2
Fig. 3
Fig. 4
Fig. 5
Fig. 6
Fig. 7
Fig. 8
Fig. 9
Fig. 10
Fig. 11
Fig. 12
Fig. 13
Fig. 14
Fig. 15
Fig. 16
Fig. 17
Fig. 18
Fig. 19

References

  1. Abdullah S, Matthews M, Murnane EL, Gay G, Choudhury T (2014) Towards circadian computing: “early to bed and early to rise” makes some of us unhealthy and sleep deprived. In: Proceedings of the 2014 ACM international joint conference on pervasive and ubiquitous computing, UbiComp ’14. ACM, New York, pp 673–684. https://doi.org/10.1145/2632048.2632100

  2. Avrahami D, Hudson SE (2006) Responsiveness in instant messaging: predictive models supporting inter-personal communication. In: Proceedings of the SIGCHI conference on human factors in computing systems. ACM, pp 731–740

  3. Banovic N, Brant C, Mankoff J, Dey A (2014) Proactivetasks: the short of mobile device use sessions. In: Proceedings of the 16th international conference on human-computer interaction with mobile devices & services. ACM, pp 243–252

  4. Ben Abdesslem F, Phillips A, Henderson T (2009) Less is more: energy-efficient mobile sensing with senseless. In: Proceedings of the 1st ACM workshop on networking, systems, and applications for mobile handhelds. ACM, pp 61–62

  5. van Berkel N, Luo C, Anagnostopoulos T, Ferreira D, Goncalves J, Hosio S, Kostakos V (2016) A systematic assessment of smartphone usage gaps. In: Proceedings of the 2016 CHI conference on human factors in computing systems. ACM, pp 4711–4721

  6. Bobadilla J, Ortega F, Hernando A, Bernal J (2012) A collaborative filtering approach to mitigate the new user cold start problem. Knowl-Based Syst 26:225–238

    Article  Google Scholar 

  7. Breiman L (2001) Random forests. Mach Learn 45(1):5–32

    MATH  Article  Google Scholar 

  8. Brown B, McGregor M, McMillan D (2014) 100 days of iphone use: understanding the details of mobile device use. In: Proceedings of the 16th international conference on human-computer interaction with mobile devices & services. ACM, pp 223–232

  9. Carroll A, Heiser G (2010) An analysis of power consumption in a smartphone

  10. Chang CC, Lin CJ (2011) Libsvm: a library for support vector machines. ACM Trans Intell Syst Technol (TIST) 2(3):27

    Google Scholar 

  11. Chon Y, Talipov E, Shin H, Cha H (2011) Mobility prediction-based smartphone energy optimization for everyday location monitoring. In: Proceedings of the 9th ACM conference on embedded networked sensor systems. ACM, pp 82–95

  12. Do TMT, Blom J, Gatica-Perez D (2011) Smartphone usage in the wild: a large-scale analysis of applications and context. In: Proceedings of the 13th international conference on multimodal interfaces. ACM, pp 353–360

  13. Falaki H, Mahajan R, Kandula S, Lymberopoulos D, Govindan R, Estrin D (2010) Diversity in smartphone usage. In: Proceedings of the 8th international conference on mobile systems, applications, and services. ACM, pp 179–194

  14. Ferreira D, Dey A, Kostakos V (2011) Understanding human-smartphone concerns: a study of battery life. Pervas Comput, 19–33

  15. Ferreira D, Ferreira E, Goncalves J, Kostakos V, Dey AK (2013) Revisiting human-battery interaction with an interactive battery interface. In: Proceedings of the 2013 ACM international joint conference on Pervasive and ubiquitous computing. ACM, pp 563–572

  16. Ferreira D, Goncalves J, Kostakos V, Barkhuus L, Dey AK (2014) Contextual experience sampling of mobile application micro-usage. In: Proceedings of the 16th international conference on human-computer interaction with mobile devices & services. ACM, pp 91–100

  17. Ferreira D, Kostakos V, Dey AK (2015) Aware: mobile context instrumentation framework. Front ICT 2:6

    Article  Google Scholar 

  18. Fischer JE, Greenhalgh C, Benford S (2011) Investigating episodes of mobile phone activity as indicators of opportune moments to deliver notifications. In: Proceedings of the 13th international conference on human computer interaction with mobile devices and services. ACM, pp 181–190

  19. Genuer R, Poggi JM, Tuleau-Malot C (2010) Variable selection using random forests. Pattern Recogn Lett 31(14):2225–2236

    Article  Google Scholar 

  20. Google (2017) Activity recognition api, https://developers.google.com/android/reference/com/google/android/gms/location/ActivityRecognitionApi

  21. Google (2017) Sensor, http://developer.android.com/reference/android/hardware/Sensor.html

  22. Ha JH, Chin B, Park DH, Ryu SH, Yu J (2008) Characteristics of excessive cellular phone use in korean adolescents. CyberPsychol Behav 11(6):783–784

    Article  Google Scholar 

  23. Hall M, Frank E, Holmes G, Pfahringer B, Reutemann P, Witten IH (2009) The weka data mining software: an update. ACM SIGKDD Explor Newslett 11(1):10–18

    Article  Google Scholar 

  24. Joachims T (1998) Text categorization with support vector machines: learning with many relevant features. Machine learning: ECML-98, 137–142

  25. Jones SL, Ferreira D, Hosio S, Goncalves J, Kostakos V (2015) Revisitation analysis of smartphone app use. In: International joint conference on pervasive and ubiquitous computing, UbiComp, pp 1197–1208. https://doi.org/10.1145/2750858.2807542, http://people.eng.unimelb.edu.au/vkostakos/files/papers/ubicomp15.pdf

  26. Khalilia M, Chakraborty S, Popescu M (2011) Predicting disease risks from highly imbalanced data using random forest. BMC Med Inform Decis Making 11(1):51

    Article  Google Scholar 

  27. Khoshgoftaar TM, Golawala M, Van Hulse J (2007) An empirical study of learning from imbalanced data using random forest. In: 19th IEEE International conference on tools with artificial intelligence, 2007. ICTAI 2007, vol 2. IEEE, pp 310–317

  28. Kostakos V, Musolesi M (2017) Avoiding pitfalls when using machine learning in hci studies. Interactions 24(4):34–37

    Article  Google Scholar 

  29. Krstic I (2016) Behind the scenes of ios security. Black Hat

  30. Lane N, Miluzzo E, Lu H, Peebles D, Choudhury T, Campbell A (2010) A survey of mobile phone sensing. IEEE Commun Mag, 48(9)

  31. Lee M, Bak C, Lee JW (2014) A prediction and auto-execution system of smartphone application services based on user context-awareness. J Syst Archit 60(8):702–710

    Article  Google Scholar 

  32. Liu Y, Xu C, Cheung SC (2013) Where has my battery gone? finding sensor related energy black holes in smartphone applications. In: 2013 IEEE International conference on pervasive computing and communications (PerCom). IEEE, pp 2–10

  33. Lu H, Yang J, Liu Z, Lane N, Choudhury T, Campbell A (2010) The jigsaw continuous sensing engine for mobile phone applications. In: Proceedings of the 8th ACM conference on embedded networked sensor systems. ACM, pp 71–84

  34. Oulasvirta A, Rattenbury T, Ma L, Raita E (2012) Habits make smartphone use more pervasive. Pers Ubiquit Comput 16:105–114

    Article  Google Scholar 

  35. Pielot M (2014) Large-scale evaluation of call-availability prediction. In: Proceedings of the 2014 ACM international joint conference on pervasive and ubiquitous computing. ACM, pp 933–937

  36. Pielot M, Dingler T, Pedro JS, Oliver N (2015) When attention is not scarce-detecting boredom from mobile phone usage. In: Proceedings of the 2015 ACM international joint conference on pervasive and ubiquitous computing. ACM, pp 825–836

  37. Pielot M, de Oliveira R, Kwak H, Oliver N (2014) Didn’t you see my message?: predicting attentiveness to mobile instant messages. In: Proceedings of the 32nd annual ACM conference on human factors in computing systems. ACM, pp 3319–3328

  38. Poppinga B, Heuten W, Boll S (2014) Sensor-based identification of opportune moments for triggering notifications. IEEE Pervas Comput 13(1):22–29

    Article  Google Scholar 

  39. Shin C, Hong JH, Dey AK (2012) Understanding and prediction of mobile application usage for smart phones. In: Proceedings of the 2012 ACM conference on ubiquitous computing. ACM, pp 173–182

  40. Song J, Sörös G, Pece F, Fanello SR, Izadi S, Keskin C, Hilliges O (2014) In-air gestures around unmodified mobile devices. In: Proceedings of the 27th annual ACM symposium on user interface software and technology. ACM, pp 319–329

  41. Strobl C, Boulesteix AL, Zeileis A, Hothorn T (2007) Bias in random forest variable importance measures: illustrations, sources and a solution. BMC Bioinform 8(1):25

    Article  Google Scholar 

  42. Svetnik V, Liaw A, Tong C, Culberson JC, Sheridan RP, Feuston BP (2003) Random forest: a classification and regression tool for compound classification and qsar modeling. J Chem Inf Comput Sci 43 (6):1947–1958

    Article  Google Scholar 

  43. Truong KN, Shihipar T, Wigdor DJ (2014) Slide to x: unlocking the potential of smartphone unlocking. In: Proceedings of the 32nd annual ACM conference on human factors in computing systems. ACM, pp 3635–3644

  44. Verkasalo H (2009) Contextual patterns in mobile service usage. Pers Ubiquit Comput 13(5):331–342

    Article  Google Scholar 

  45. Xu Y, Lin M, Lu H, Cardone G, Lane N, Chen Z, Campbell A, Choudhury T (2013) Preference, context and communities: a multi-faceted approach to predicting smartphone app usage patterns. In: Proceedings of the 2013 international symposium on wearable computers. ACM, pp 69–76

  46. Yan T, Chu D, Ganesan D, Kansal A, Liu J (2012) Fast app launching for mobile devices using predictive user context. In: Proceedings of the 10th international conference on mobile systems, applications, and services. ACM, pp 113–126

  47. Yuan Y, Raubal M, Liu Y (2012) Correlating mobile phone usage and travel behavior–a case study of harbin, china. Comput Environ Urban Syst 36(2):118–130

    Article  Google Scholar 

  48. Zhuang Z, Kim KH, Singh JP (2010) Improving energy efficiency of location sensing on smartphones. In: Proceedings of the 8th international conference on mobile systems, applications, and services. ACM, pp 315–330

Download references

Author information

Authors and Affiliations

Authors

Corresponding author

Correspondence to Chu Luo.

Additional information

Publisher’s Note

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Appendix: Parameter tuning

Appendix: Parameter tuning

To improve the performance of SVM, we first tried the linear kernel. However, the performance of SVM with linear kernel degraded (ROCArea = 0.523 at 10 min) compared to the RBF kernel (ROCArea = 0.571 at 10 min). Hence, we attempted to apply parameter tuning methods on the RBF kernel. As illustrated in previous work investigating parameter tuning for SVM with the Radial Basis Function kernel [10, 24] , we repeated the 10-fold cross-validation on the data of 10 min time window with replacing the default setting \(\gamma = \frac {1}{m + 1}\) with a wide variety of γ values.

Figure 20 shows the performance of SVM with the RBF kernel having different γ values. Among all γ assignments, γ = 100 achieved the best performance: ROCArea = 0.813, class “no” precision 0.864, class “no” recall 0.945, class “yes” precision 0.852, class “yes” recall 0.682.

Fig. 20
figure 20

SVM with the Radial Basis Function kernel

The results indicate that, given a suitable γ, SVM with the RBF kernel can also achieve high performance in the classification to predict unlocking events. However, due to the large number of possible γ values and our limited computing resources, we could not refine our finding, since parameter tuning is highly time-consuming and computation-intensive (in our case, running each γ value took about 10 days for a normal PC). With γ = 100, although SVM with the RBF kernel achieved considerably good results, the performance of Random Forests was still better. Hence, we focused on Random Forests in further analysis. Future work may conduct deeper investigation about the employment of SVM with the RBF kernel in phone unlocking prediction.

Rights and permissions

Reprints and Permissions

About this article

Verify currency and authenticity via CrossMark

Cite this article

Luo, C., Visuri, A., Klakegg, S. et al. Energy-efficient prediction of smartphone unlocking. Pers Ubiquit Comput 23, 159–177 (2019). https://doi.org/10.1007/s00779-018-01190-0

Download citation

  • Received:

  • Accepted:

  • Published:

  • Issue Date:

  • DOI: https://doi.org/10.1007/s00779-018-01190-0

Keywords

  • Smartphones
  • Machine learning
  • Sensors
  • Context-awareness