Downloads: 0 | Views: 149

Research Paper | Computer Science | India | Volume 13 Issue 7, July 2024 | Popularity: 4.8 / 10

Rate This Article!      

Enhanced Daily Activity Recognition in Smart Homes: Leveraging Feature Selection, Neural Networks, and Appliance Load Signatures

Bina Ajaykumar Shah, Dr. Nagesh Salimath

Abstract: Accurate daily activity identification in smart home environments is becoming more and more necessary as the use of these devices increases. Robust activity identification improves energy economy, guarantees home security, and makes tailored services possible. This research study integrates feature selection approaches, neural networks, and appliance load signatures to propose a complete method for enhancing everyday activity detection in smart homes. Our approach seeks to reduce computing overhead while optimizing the precision and efficacy of activity identification systems. In order to extract the most pertinent characteristics from the sensor data gathered in smart homes, we look into a number of feature selection techniques. Next, we create and put into practice a neural network architecture that is tailored to tasks involving activity recognition. We also include appliance load signatures to the recognition procedure to improve the resilience and efficiency of the model. The outcomes of our experiments show that our method is effective in correctly identifying a wide variety of everyday actions in actual smart home settings. With potential applications in energy management, home automation, and healthcare monitoring, this research advances activity identification systems in smart homes.

Keywords: Smart Homes, Daily Activity Recognition, Feature Selection, Neural Networks, Appliance Load Signatures, Energy Efficiency, Home Automation

Edition: Volume 13 Issue 7, July 2024

Pages: 495 - 498

DOI: https://www.doi.org/10.21275/MR24709161339

Make Sure to Disable the Pop-Up Blocker of Web Browser

Click to Cite this Article

Similar Articles