Research on Prediction Model of Audience Viewing Behavior of New Media Driven by Big Data

The purpose of this paper is to deeply explore and build a new media audience viewing behavior prediction model based on big data technology. With the rapid development of information technology, the audience's viewing behavior in the new media environment is increasingly complex and changeable, which poses new challenges to the traditional audience rating analysis and prediction methods. Based on big data, this paper collects, integrates and mining massive user behavior data on new media platforms of radio and television, including multi-dimensional information such as viewing time, content selection and interactive comments, to explore the audience's viewing behavior pattern and its influencing factors. First of all, the paper introduces the application background of big data in the media field and the characteristics of the audience's viewing behavior in the new media of radio and television in detail, and systematically combs and evaluates the existing viewing behavior prediction models. Then, combining machine learning and deep learning theories, a big data-driven prediction model combining time series analysis and neural network algorithm is designed and implemented, in order to more accurately capture and predict the audience's personalized viewing needs and dynamic trends. The experimental results show that the model can not only effectively improve the accuracy of the prediction of viewing behavior, but also provide in-depth insight into the influence of different program types, time periods, social hot spots and other factors on the audience's viewing choice, so as to provide scientific basis for the planning, arrangement and promotion of radio and television programs, and further promote the transformation of radio and television media from traditional content production to precision and intelligent services. Finally, this paper discusses the practical application effect and optimization direction of the model, and looks forward to the broad application prospects of big data in the future development of new media of radio and television.