Skip to main content
select distinct n.content as name, r.content as role, t.nameID as nameID, n.unitID as unitID, n.departmentID as departmentID, n.facultyID as facultyID, n.provinceID as provinceID, n.countryID as countryID from transaction as t, Name as n, Role as r where t.itemID='104368' and t.nameID=n.id and t.roleID=r.id
scopus



A Comparative analysis of appliance classifiers for wireless Classification system

Researcher

เครือข่ายคณะผู้วิจัย


เครือข่ายนักวิจัย+ผลงานวิจัย (full screen)

Abstract

This paper presents a Performance comparison of four Classification techniques (i.e., Decision tree, Naïve bayes, Neuron network, and Support vector machine) for Appliance Classification by analyzing each appliance's electricity usage sent via a wireless sensor network. To measure and collect the actual electrical power consumed by each device, we designed sensor circuits, each of which is deployed inside each power outlet. The measured data are sent to a centralized system via a wireless sensor network (which can also be used to deliver control commands to turn on/off each appliance). The system uses the data to classify a type of each appliance connected to each of the outlet. Since this research is to be detecting electrical usage at each outlet (instead of at the main circuit as in previous works), the system can be developed further to help identifying the Abnormal operation of each appliance, and to automatically recognize the device when it is moved to another outlet, making possible automatic appliance on/off control. As a result, it could promote home safety and energy savings without affecting users' normal behaviors. Comparing the accuracies of classifying 40 electric devices using the four techniques, we found that 1) standard deviation of measured electricity usage is one of necessary attributes for accurately classifying appliance operating states, and 2) the Decision tree algorithm (i.e., C4.5) performs best (with the error of 5.73%). Copyright © 2014 ACM.

Appliance classification (1 items found) | Classification technique (39 items found) | Support vector machine (134 items found) | Performance comparison (65 items found) | Abnormal operation (1 items found) | Classification (2214 items found) | Neuron network (5 items found) | Decision tree (105 items found) | Naïve bayes (40 items found) | Wireless sensor networkClassification | Wireless sensor networks | Support vector machines | Decision-tree algorithm | Classification system | Comparative analysis | Centralized systems | Energy conservation | Energy consumption | Energy utilization | Neural networks | Neuron networks | of information | Decision trees | mathematics | Data mining | Equipment | Sodium | Trees |

ต้นฉบับข้อมูล : scopus