O slideshow foi denunciado.
Utilizamos seu perfil e dados de atividades no LinkedIn para personalizar e exibir anúncios mais relevantes. Altere suas preferências de anúncios quando desejar.

Battery log data mining — Ramon Oliveira (Datart) @PAPIs Connect — São Paulo 2017

234 visualizações

Publicada em

Battery life is critical for smart devices, but optimizing it requires cooperation from the entire software ecosystem. Wasteful software affects user perception about devices’ battery quality. Therefore, a large team within a producer of those smart devices is focused on identifying and correcting energy consumption bugs. Since the software ecosystem grows fast, that team faces a lot of suspect issues, from which only a small fraction turns out to be genuine. Our project aims to streamline energy-related bug processing in devices of the company and its partners, by automatically identifying anomalous behaviors related to battery drain using data mining and machine learning.

Publicada em: Tecnologia
  • Seja o primeiro a comentar

Battery log data mining — Ramon Oliveira (Datart) @PAPIs Connect — São Paulo 2017

  1. 1. ● ○ ○ ○ ● ○ ■ ● ○
  2. 2. Motorola Employees CustomersNew Product
  3. 3. New Product Generate logs Filter suspicious behaviours Manual analysis and triage Fix
  4. 4. ● ● ● ●
  5. 5. ● ○ ○ ● ○
  6. 6. ● ○ ○ ○ ○ ○ ○ ● ○ ○ ○ ○
  7. 7. ● ● ● ● Train Predict Time TimeWeek 1 Train Predict Week 2 Train Predict Week 3
  8. 8. ● ○ ○ ○ ○ ○ ● ○ ● ○
  9. 9. ● ○ ○ ● ○ ○ ○ ● ○ ○ ○
  10. 10. ● ○ ○ Bug thresholdNon-bug threshold
  11. 11. ● ○ ○ ■ ■ ● ○ ○ ○ Not Bug Bug Not Bug 7,961 (90.86%) 801 (9.14%) Bug 106 (4.35%) 2,329 (95.65%) Predicted Actual
  12. 12. 10% FP threshold 5% FN thresholdFN FP
  13. 13. ● ● ○ ○ ● ○ ○
  14. 14. ● ○ ● ○ ● Keyword AUC tethering 92% signal 89% radio 88% pkey 88% duplicate 86% wakelock 83% bluetooth 83% screen 81% kernel 81% streaming 80% gps 80%
  15. 15. ● ○ ● ○ ● ○

×