���ڲ���ݻ��㷨�Ĵֲڼ�����Լ��
����1 ,�պ���2
1. ��ɳ�?��ѧ
Attribute reduction of rough sets based on differential evolutionary algorithm
ժҪ ����Լ���Ǵֲڼ�(RS)���۵ĺ�������֮һ��Ӧ�ò���ݻ�(DE)�㷨�����С����Լ����һ���µķ��Բ���ݻ��㷨�����˸Ľ�����һ���µ���Ӧֵ����Ķ�����ʽ�����ڴ˻�������˻��ڲ���ݻ��㷨������Լ���㷨��������ö�����ݶԸ��㷨�����˷���ʵ�飬���������㷨�����˱ȽϷ�����ʵ����������㷨����Ч��,�ܿ��ٵؽ�������Լ��
�ؼ�� ��
�ֲڼ� ,
����Լ�� ,
����ݻ��㷨 ,
��Ⱥ
Abstract ��The attribute reduction is one of the cores of Rough Set (RS) theory. To solve minimum attribute reduction by Differential Evolution (DE) algorithm is a new direction. In this paper, an improved differential evolution algorithm and a new definition form of fitness function were proposed. And on this basis, an attribute reduction algorithm based on the improved differential evolutionary algorithm was put forward. Finally, the simulation experiments and a comparative analysis with an existing algorithm were carried out for the algorithm with multiple sets of data. The experimental results show that the algorithm is effective and fast.
Key words ��
Rough Set (RS)
attribute reduction
Differential Evolution (DE) algorithm
population
�ո�����: 2010-03-12
��������: 2010-09-03
ͨѶ����:
����
E-mail: gaoyi19851003@163.com
[1]
�˸շ� ������ ���պ� ��־�� ����. ���ڸĽ����Ŵ��㷨����������֫�˶������˶�ѧ��ⷽ��
[2]
���Ƿ� ������ �ﾰ�� ����� ���� ����. ����ģ��ֲڼ���D-S֤�����۵Ķ�Դ�����Ϣ�ںϷ���
[3]
���� ������ ��ޱޱ ���� ����ɽ. ���ڴֲڼ��Ļ���ͼ������ѡ����ʶ��
[4]
�Ż�ΰ ������. ��Ⱥ��ģ�ɱ�����߶�ģ̬�����Ż�
[5]
���� ��ǿ ������ ����ׯ. �������Һ���Ľ��PSO�㷨�������
[6]
�»� ����Ȩ �Թ�ΰ. ���ڻ������еĶ���Ⱥ�����Ӳ��㷨
[7]
��ѩ�� ������. ����ȫ�����ƹ�ϵ�IJ��걸ģ����Ϣϵͳ��������Լ��
[8]
�߲� ���ǫ ��־�� �ź���. �ֲڼ��Ŷ�һ������Լ��
[9]
���� ������ ������. ��������Ȩ�غʹֲڼ������������㷨
[10]
��ҫ�� ���� ������. ���ƹ�ϵ�µļ�ֵ��ֵ��Ϣϵͳ
[11]
���� ��. ���걸��Ϣϵͳ�л��������ݲ��ϵ������Լ��
[12]
�¼ҿ� ���ر� �컪��. ���ڶ�߶ȴֲڼ�ģ�͵ľ������Ż��㷨
[13]
�麣Բ ������ ��־ǿ �ܾ� ׯ����. ������Ⱥ��̬���鲥ӵ����ƻ���
[14]
���� ʩ�B ���翥 ���. ����ȫ������ڵ���Ⱥ�����㷨
[15]
�ƹ��� ������ �ű�. ������չʱ�����Petri��Ĵֲ����繥��ģ��
2010, Vol. 30 
