Information Theory Assisted Massive Media System for Ideological and Political Online Guiding Affinity Estimation

In this paper, we analyze the information theory assisted massive media system for the estimation of ideological and political online guiding affinity. To construct the efficient system, we integrate the following technologies. (1) Based on the access method of TCP/IP, feedback of the authentication results can be obtained within the acceptable delay range in order to achieve a real-time authentication. (2) We construct an information theory model that remains suitable for calculating the similarity, so as to use the information in the data distribution. (3) Unlike existing point-by-point learning strategies based on convolutional neural networks, the proposed framework can learn core text results and four picture results from the search return. Our online guiding system is tested on the student groups in order to obtain satisfactory results.