A method for classifying audience of digital signage advertisements based on a neural network and a Huff model

Abstract

The invention relates to a method for classifying audience of digital signboard advertisements based on a neural network and a Huff model. The method comprises the following steps: element selection and processing, model construction and model verification are carried out; spatialized digital signage location factors are constructed to obtain the standard grid grid layer of digital signage location factors, including its pixel value and corresponding coordinate value; the improved neural network model is used to calculate the normalized digital signage location factors, and the probability ofeach location containing a variety of audiences is obtained; then the influence of digital signage location on the audience without digital signage location is calculated by using the improved Huff model; then, the results of the two models, namely, the improved neural network model and the improved Huff model, are fused to complete the classification of the digital signboard advertisement audience; finally, the validity of the model is verified by using five validation indicators of the multi-label classification algorithm. The method of the invention comprehensively considers multi-source elements and spatial distance, has high accuracy of classification result, high efficiency of digital signboard advertisement placement and good digital signboard influence effect.

Description

technical field [0001] The invention belongs to the technical field of digital signage audience classification, and relates to a digital signage audience classification method, in particular to a digital signage audience classification method based on a neural network and a Huff model. Background technique [0002] Digital signage refers to a multimedia professional audio-visual system that releases commercial, financial and entertainment information through digital signage terminal display equipment in public places where people gather. As a new type of media, it has become an important medium for physical advertising in modern cities. Compared with traditional TV advertisements and newspaper advertisements, digital signage is more flexible, and can carry out personalized and customized advertisements according to different audience groups. The development of digital signage has only a history of more than 20 years, but its application covers all fields of work and life. Th...

AI Technical Summary

Problems solved by technology

However, these algorithms have high algorithm complexity and long running time, and it is difficult to solve practical problems. At the same time, the audience of digital signage advertising is a kind of spatial data, and the spatial distance has a certain impact on its classification. Therefore, the spatial distance is introduced into digital signage advertising. Audience classification can effectively improve classification accuracy

Images