Improved Flower Pollination Algorithm and Its Application in User Identification Across Social Networks

Aiming at the shortages of basic flower pollination algorithm (FPA) with slow convergence speed, low search precision, and easy to fall into local optimum, a new adaptive FPA based on opposition-based learning and t-distribution (OTAFPA) was proposed and be applied to the social networks. First, the opposition-based learning strategy is utilized to increase the diversity and quality of the initial population. Then, the adaptive dynamic switching probability is introduced, which can effectively balance the global and local search according to the current number of iterations. Finally, the t-distribution variation is used to increase the population diversity and to help the algorithm jump out of the local optimum. The simulation experiments on eight classical test functions show that OTAFPA has better global optimization ability, which improves the convergence speed and the solution accuracy of the algorithm. The OTAFPA also shows superior performance in practical applications of user identification across social networks.