Version 1
: Received: 12 October 2024 / Approved: 14 October 2024 / Online: 14 October 2024 (14:22:45 CEST)
How to cite:
Bkheet, S. A.; Agbinya, J. I.; Khamis, G. S. M. Advanced Deep Learning Approach for Smart Home Appliance Identification Using Recurrent Neural Networks with LSTM. Preprints2024, 2024101085. https://doi.org/10.20944/preprints202410.1085.v1
Bkheet, S. A.; Agbinya, J. I.; Khamis, G. S. M. Advanced Deep Learning Approach for Smart Home Appliance Identification Using Recurrent Neural Networks with LSTM. Preprints 2024, 2024101085. https://doi.org/10.20944/preprints202410.1085.v1
Bkheet, S. A.; Agbinya, J. I.; Khamis, G. S. M. Advanced Deep Learning Approach for Smart Home Appliance Identification Using Recurrent Neural Networks with LSTM. Preprints2024, 2024101085. https://doi.org/10.20944/preprints202410.1085.v1
APA Style
Bkheet, S. A., Agbinya, J. I., & Khamis, G. S. M. (2024). Advanced Deep Learning Approach for Smart Home Appliance Identification Using Recurrent Neural Networks with LSTM. Preprints. https://doi.org/10.20944/preprints202410.1085.v1
Chicago/Turabian Style
Bkheet, S. A., Johnson I. Agbinya and Gamal Saad Mohamed Khamis. 2024 "Advanced Deep Learning Approach for Smart Home Appliance Identification Using Recurrent Neural Networks with LSTM" Preprints. https://doi.org/10.20944/preprints202410.1085.v1
Abstract
This study presents the development of a Recurrent Neural Network (RNN) model for classifying smart home device data. Utilizing a dataset sourced from Kaggle, the study explored the processes of data gathering, loading, normalization, and model construction. The RNN, equipped with Long Short-Term Memory (LSTM), was trained and evaluated, demonstrating significant improvements in training and validation accuracy over 10 epochs, culminating in a test accuracy of 83.25%, and only 35.4% loss. The evaluation of the model on the test set provides a test accuracy result, accompanied by an in-depth analysis of ROC curves and Area Under the Curve (AUC) scores for multi-class classification, along with a confusion matrix. The AUC score of 0.9896 indicates outstanding performance in accurately classifying IoT device categories. These findings indicate that the RNN with LSTM exhibits superior learning efficiency and generalization capabilities, making it more suitable for IoT device classification tasks.
This article emphasizes the concept of IoT and reviews recent studies on the application of deep learning models across various IoT domains, including smart homes, industrial systems, and healthcare. Future research could focus on improving real-time processing capabilities, and scalability, and incorporating diverse IoT data types to enhance and broaden the model's practical applications.
Keywords
The Internet of Things (IoT); Smart objects; Recurrent Neural Network (RNN); Long Short-Term Memory (LSTM)
Subject
Computer Science and Mathematics, Artificial Intelligence and Machine Learning
Copyright:
This is an open access article distributed under the Creative Commons Attribution License which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.
