A Study on the Method of Press Freedom Value Assessment Based on Artificial Neural Network

Abstract

Freedom of the press embodies an increasingly important role in modern democratic society, and people pay more and more attention to the value of freedom of press. The value of freedom of press has been elaborated from the basic concept, right attributes, and role functions of freedom of press, which are the basis for the existence of the value of freedom of the press and the jurisprudential basis for strengthening its legal construction. Freedom of press is a right extended from the freedom of speech and publication and is the application of freedom of speech and publication in news communication activities. Freedom of the press in the modern sense is considered an institutional right, both a fundamental right and a political right. As the role of press freedom in the process of modern democratic society becomes more and more prominent, the conflict between its value and other social values becomes more and more obvious. In such a context, this study introduces BP neural network technology into the assessment of press freedom value, expecting that a set of assessment method models with press freedom value as the core can be constructed. In order to achieve this goal, the core problem is how to use BP neural network technology to automatically evaluate the value of freedom of the press. A web crawler is used to crawl news from various online sources and the news are scored and evaluated by the system to present the most valuable news first to the readers. A dataset of 1440 data samples was created to be used in the experiments. The proposed GRU with multifeature fusion method has a higher accuracy and lower error rate as compared to LSTM, linear regression, and random forest-based model.