A novel social network measurement and perception pattern based on a multi-agent and convolutional neural network

With the rapid advancement of the social network, the total interpersonal relationships among people constitute a social network in real life and the human is the node in this network. Against this background, this paper proposes a novel social network search and perception pattern based on a multi-agent and convolutional neural network. Our research can be regarded as a parallel integration of the multi-agent and CNN. In the CNN part, we adopt prior knowledge that differs from the ordinary convolution neural network and the convolution neural network unique neuron receptive field structure. In the multi-agent part, we combine the characteristics of individual and general-community agents; the establishment and revision of its faith intention is the result of internal thought conditions and interaction with external factors. We apply the proposed model to a social network search, and perception and connection awareness analysis, respectively. The experimental result proves that the proposed method achieves a satisfactory performance.