An improved branch and bound algorithm
Liu Zheng-lian · IEEE Software · 2011
How to select optimal features is one of the priority research areas.Currently,BAB and BAB+ algorithms,which are used to select optimal features,are commonly applied.But it takes too long to use convinently.Then theory and algorithm of an improved branch and bound are explained particularly.The essence of this algorithm is to search for an optimal solution in a solution tree.Partial route and Father route are introduced in this paper.The solution tree is got by cutting those unnecessary branches which will not contain the optimal solution.The experimental results demonstrate the effectiveness of the new algorithm.