Abstract:

We exploit the basic concepts of the rough sets theory, genetic algorithm and particle swarm optimization algorithm. Attributes reduction is one of the key issues in knowledge discovery. Traditional reduction algorithms feature serial search, low efficiency and slow convergence speed. By combining computational intelligence and rough sets, we propose a minimum attributes reduction algorithm which bases on the rough sets, genetic algorithm and particle swarm optimization algorithm. In order to solve the minimum attribute reduction, this algorithm regulates the function parameters dynamically and calculates attribute core using attribute dependability, thus restricting the initialized population. Experimental results prove the efficiency of the proposed algorithm in attribute reduction for high dimensionality and big data.

Key words: attribute reduction;rough sets;genetic algorithm;particle swarm optimization;dependability of attributes