SMART HOME AUTOMATION SYSTEM FOR ENERGY CONSUMPTION USING TENSORFLOW-BASED DEEP ENSEMBLE LEARNING
Abstract
Over the past decades, the evolution of new wireless technology has led to increased attention toward Smart Home Automation Systems (SHAS). In the smart home, numerous smart devices are interconnected with the proliferation of the Internet of Things (IoT) technology to provide users with a more comfortable lifestyle. Prior research on the smart home system has enacted machine learning and deep learning techniques to forecast the consecutive activities in the smart home. This research paper aims to enhance the future decision-making of energy consumption with the assistance of environmental factors and home appliances by exploiting the Tensorflow-based deep ensemble learning technique. The enhancement of future decision-making in smart home automation systems primarily involves the classification of energy consumption levels from the knowledge of external environmental factors and energy consumption levels of home appliances through the phases of data preprocessing, feature selection, fuzzy logic-based data labeling, and finally, classification of energy consumption using TensorFlow based deep ensemble learning technique. The data obtained from the effective feature selection technique is subjected to labeling via the fuzzy logic system to classify the energy consumption of smart home appliances. Finally, this work classifies the level of energy consumption based on the labeled knowledge of smart home data using a tensorflow-based deep ensemble learning model. The experimental model implements the proposed deep ensemble learning model in the tensorflow framework, which improves the decision-making performance of energy utilization in the smart home system. Experimental results illustrate that the proposed deep ensemble learning model yields superior classification performance than the other baseline classifiers, such as Artificial Neural Network (ANN), Convolutional Neural Network (CNN), and Long Short-Term Memory (LSTM) on the smart home dataset.

Authors
S. Umamageswari, M. Kannan
Sri Chandrasekharendra Saraswathi Viswa Mahavidyalaya, India

Keywords
Feature Selection Techniques, Fuzzy Logic, Tensorflow Framework, Ensemble Deep Learning Model
Yearly Full Views
JanuaryFebruaryMarchAprilMayJuneJulyAugustSeptemberOctoberNovemberDecember
041514210400000
Published By :
ICTACT
Published In :
ICTACT Journal on Soft Computing
( Volume: 14 , Issue: 3 , Pages: 3282 - 3292 )
Date of Publication :
January 2024
Page Views :
210
Full Text Views :
49

Creative Commons License
This work is licensed under a Creative Commons Attribution-NonCommercial-ShareAlike 4.0 International License.