University of Twente Student Theses
Resource Profiling for Smart Home Machine Learning Applications
Florian, P. (2025) Resource Profiling for Smart Home Machine Learning Applications.
|
PDF
2MB |
| Abstract: | As smart home devices increasingly incorporate machine learning applications, their resource constraints pose significant challenges for deployment. This research investigates resource profiling to better understand the requirements of machine learning applications in smart home devices under varying conditions. By analyzing the effects of input size, model complexity, workload conditions, and hardware platforms, we aim to understand how these factors influence resource utilization and system performance. Controlled experiments will focus on face recognition and speech recognition. These applications were chosen for their relevance as representative machine learning tasks in smart homes. They cover key areas like security and accessibility, making them ideal for studying resource demands. To capture performance across a range of hardware capabilities, we will conduct testing on two hardware platforms, namely the Raspberry Pi and a Desktop PC. The results will identify critical resource requirements and offer insights for more efficient implementation of machine learning in smart home devices, enhancing their capabilities and reliability. |
| Item Type: | Essay (Bachelor) |
| Faculty: | EEMCS: Electrical Engineering, Mathematics and Computer Science |
| Subject: | 54 computer science |
| Programme: | Computer Science BSc (56964) |
| Link to this item: | https://purl.utwente.nl/essays/105015 |
| Export this item as: | BibTeX EndNote HTML Citation Reference Manager |
Show download statistics for this publication
Show download statistics for this publicationRepository Staff Only: item control page