New Insights on Nontechnical Losses Characterization Through Evolutionary-Based Feature Selection

被引:42
作者
Oba Ramos, Caio Cesar [1 ]
de Souza, Andre Nunes [1 ]
Falcao, Alexandre Xavier [2 ]
Papa, Joao Paulo [3 ]
机构
[1] Univ Sao Paulo, Dept Elect Engn, BR-05508970 Sao Paulo, Brazil
[2] Univ Estadual Campinas, Inst Comp, BR-13083852 Sao Paulo, Brazil
[3] UNESP Univ Estadual Paulista, Dept Comp, BR-17033360 Sao Paulo, Brazil
基金
巴西圣保罗研究基金会;
关键词
Feature selection; gravitational search algorithm; harmony search; nontechnical losses; optimum-path forest; particle swarm optimization; pattern recognition;
D O I
10.1109/TPWRD.2011.2170182
中图分类号
TM [电工技术]; TN [电子技术、通信技术];
学科分类号
0808 ; 0809 ;
摘要
Although nontechnical losses automatic identification has been massively studied, the problem of selecting the most representative features in order to boost the identification accuracy and to characterize possible illegal consumers has not attracted much attention in this context. In this paper, we focus on this problem by reviewing three evolutionary-based techniques for feature selection, and we also introduce one of them in this context. The results demonstrated that selecting the most representative features can improve a lot of the classification accuracy of possible frauds in datasets composed by industrial and commercial profiles.
引用
收藏
页码:140 / 146
页数:7
相关论文
共 26 条
[1]  
Al-Ani A., 2005, INT J COMPUTATIONAL, V2, P53
[2]  
[Anonymous], 2001, SWARM INTELL-US
[3]  
[Anonymous], COMPUT ELECT ENG
[4]  
[Anonymous], 2011, ACM T INTEL SYST TEC
[5]  
[Anonymous], 2003, IMPLEMENTATION FAST
[6]  
[Anonymous], LECT NOTES COMPUT SC
[7]  
[Anonymous], 2003, Genetic programming IV: routine human-competitive machine intelligence
[8]   Ant system: Optimization by a colony of cooperating agents [J].
Dorigo, M ;
Maniezzo, V ;
Colorni, A .
IEEE TRANSACTIONS ON SYSTEMS MAN AND CYBERNETICS PART B-CYBERNETICS, 1996, 26 (01) :29-41
[9]  
Fan RE, 2008, J MACH LEARN RES, V9, P1871
[10]  
Firpi H. A., 2004, Proceedings. 33rd Applied Imagery Pattern Recognition Workshop, P112