Innovative Strategies for Cultivating Journalism and Communication Talents in the All-Media Era Based on the Light GBM Model

Abstract

Under the development background of digitalization of big data and information, artificial intelligence and the Internet of Things are rising like mushrooms. It is difficult for the traditional journalism communication teaching mode in colleges and universities to keep pace with the development of the times, and there are several problems in the traditional teaching methods, such as teaching being too rigid, lack of personalized teaching, and teaching means being too backward. This paper focuses on the innovative strategy of journalism communication talents training in the era of full media based on Light GBM model. Firstly, we understand the principle and construction idea of Light GBM model by viewing the information, comparing the difference between XG Boost algorithm and Light GBM two algorithms, analyzing the advantages and limitations of Light GBM, optimizing the algorithm by establishing the hyperparameters of Light GBM model under grid search algorithm, understand the teaching status of college journalism broadcasting majors through survey and analysis, based on Light model under the analysis of feedback to propose targeted innovation strategies, 30 times ten-fold hierarchical cross-validation study of the three algorithms. The mean and standard deviation results of 11 dichotomous classification models accuracy and F1 values under the three algorithms were compared. The results show that in 11 sets of dichotomous data sets, the optimized Light GBM algorithm has 9 sets of experimental accuracy wins and 7 sets of experimental F1 values wins, with 81.8% accuracy and 63.6% of F1 values wins. The optimized Light GBM algorithm had seven experiments that beat the other two groups of algorithms in comparing the accuracy and F1 value. This study can accurately predict students’ learning behaviors to improve the professional performance of journalism and communication students in a targeted manner and has a reference value for the innovation of journalism and communication teaching mode in colleges and universities in the context of big data and all-media era, and this study has great historical significance for the development of Chinese journalism industry.