Abstract

Studying the current situation of media literacy of college students clearly is an important grasp to deepen media literacy education in colleges and universities. In this paper, by analyzing the relationship between information entropy and the maximum entropy principle, the problem of finding the maximum information entropy is transformed into a conditional extreme value problem and solved by using Lagrange multipliers, so as to find out the maximum information entropy distribution function and its parameter estimation under the one-dimensional condition. The maximum information entropy model is then used to study the media literacy of college students, including the use of media, media criticism ability and media utilization ability. The problem of low validity of the collected data is filtered by MaxEnt distribution function. The time spent by college students using online media is 33.42% higher than that of TV media and 48.16% higher than that of newspapers. When acquiring information through media, those who actively pay attention to information sources are 43.86% less than those who never pay attention, and those who use media for entertainment are 32.68% higher than those who participate in social discussions. The study based on the maximum information entropy can systematically analyze the media literacy of college students, and can target the deficiencies and defects in it, which has implications for the optimization of media literacy education in colleges and universities.