Open AccessPredicting and Evaluating the Online News Popularity based on Random Forest
Author(s) -
Yuchao Zhang,
Kun Lin
Publication year - 2021
Publication title -
journal of physics. conference series
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.21
H-Index - 85
eISSN - 1742-6596
pISSN - 1742-6588
DOI - 10.1088/1742-6596/1994/1/012040
Subject(s) - popularity , random forest , computer science , construct (python library) , dimension (graph theory) , the internet , data science , machine learning , artificial intelligence , data mining , world wide web , mathematics , social psychology , pure mathematics , programming language , psychology
With the rapid development of science and technology, the internet has become a large media of information spread. There is a large quantity message on this platform. And online articles are the main form of information propagation. If the press can know what kind of articles will be more popular, they can construct an article that can help them spread the information they want to spread. Therefore, it’s very important to predict the popularity of these articles. Some models in machine learning could be applied to this problem. In this paper, it will introduce an approach based on Random Forest. To avoid too much calculation, the experiment first uses PCA to make dimension reduction. Then the model evaluation uses the ROC area values to assess the accuracy of the model. Its performance is better than CART and C4.5.