Research on News Value Feature Item Based on User Attention

With the update of network technologies and the large increase in the number of news releases, it is more important to choose news topics that are more in line with the audience entry point and higher news value. Based on the random forest algorithm, this experiment utlizes the news value knowledge system to construct the news value feature items, and formulates the algorithm to quantitatively calculate the feature importance of the new value feature items to provide a more accurate and usable method for the research of news value prediction, which can help journalism works better and more effectively.