05522544

a

05522544 Yan, Luo Changrui, Yu A new hybrid algorithm for feature selection and its application to customer recognition. Dress, Andreas (ed.) et al., Combinatorial optimization and applications. First international conference, COCOA 2007, Xi'an, China, August 14--16, 2007. Proceedings. Berlin: Springer (ISBN 978-3-540-73555-7/pbk). Lecture Notes in Computer Science 4616, 102-111 (2007). 2007 Berlin: Springer EN

doi:10.1007/978-3-540-73556-4_13

Summary: This paper proposes a novel hybrid algorithm for feature selection. This algorithm combines a global optimization algorithm called the simulated annealing algorithm based nested partitions (NP/SA). The resulting hybrid algorithm NP/SA retains the global perspective of the nested partitions algorithm and the local search capabilities of the simulated annealing method. We also present a detailed application of the new algorithm to a customer feature selection problem in customer recognition of a life insurance company and it is found to have great computational efficiency and convergence speed.