关键词: 层次演化, 分层人工蜂群算法, 神经网络, 服装流行色, 色相预测

Abstract:

Aiming at the accuracy defect of current prediction method on fashion color, a hierarchical multi-hive artificial bee colony algorithm (HABC) was proposed based on complex adaptive system theory and biological coordinated evolutionary thought. HABC was applied in calculating the weight values of network and tested by fitting curves. The results show that it can improve the accuracy of neural network simulation of the target problem. Using the data of women international spring and summer clothing fashion color in the years from 2007 to 2016 released by PANTONE, a prediction model of fashionable color hue is established. The influence of nodes number in different hidden layers on hue prediction accuracy is analyzed. By comparing the prediction results with that issued by PANTONE official, it is shown that the proposed method improves the prediction accuracy compared with the other methods.

Key words: hierarchical evolution, hierarchical bee colony algorithm, meural network, clothes fashion color, color prediction