期刊名称：International Journal of Advanced Networking and Applications
电子版ISSN：0975-0290
出版年度：2021
卷号：12
期号：4
页码：4639-4643
DOI：10.35444/IJANA.2021.12405
语种：English
出版社：Eswar Publications
摘要：Questionnaire is one of the most important means to collect data and make information decision. In this paper, the micro expression recognition technology of artificial intelligence is applied in the field of print advertising satisfaction collection and information decision-making. Starting wi th the analysis of micro expression, a new concept of satisfaction factor of print advertising is proposed and then various optimization functions in neural network are analyzed and discussed. In the following part, the Adabound optimization function is applied to the network model in the example of clothing print advertisement and the convergence s peed of the loss function is accelerated in order to improve the detection accuracy of the model. Experimental result shows the satisfaction of the viewers to the clothing print advertisements, allowing researchers to judge their purchase desire at the same time.
