The Research On Attribute Reduction Based On Genetic Algorithm And Ant Colony Algorithm

Rough set theory has been proved to be an excellent mathematical tool dealing with vague and uncertain description of objects,without any prior knowledge or additional in data mining and knowledge reduction.Attribute reduction,one of the key problems in rough set theory,is able to reduct the redundant atrributes from the databases, to simplify the knowledge presentations ,to improve the efficiency of systems and to support the decision making.And now a more effective attribute reduction algorithm which can reduct the time comliexity and get the better results is still the main research topics.Firstly,the thesis reviews the theories and methods of rough set ,and analyzes the algorithms of attribute reduction based on discernibility matrix,attribute significance,dependability ,genetic algorithm and ant colony algorithm.Then,based on the characteristic of rough set theory ,through analysis the existing primary algorithm of attribute reduction based on genetic algorithm and ant colony algorithm ,two new algorithm of attribute reductions are presented in this thesis.The first one is attribute reduction algorithm based on new ant colony algorithm,the main advantage of the algorithm is that define the fitness function in the algorithm by the dependability of attribute,and through genetic algorithm optimise the the ant colony algorithm ,not only optimize the initial moment of ant colony algorithm optmization speed ,and it is easy to fall into local optimum.The second one is attribute reduction algorithm based on adaptive ant colony algorithm, by introducing new mechanism and chance of information exchange , this strategies enable each group ant to choose partner group to communicate and update the pheromone adaptively.Last,the experiment are shown to prove the two new algorithm is more excellence and they are more accurate and efficient to solve the problem of attribute reduction in decision table.