.png){kind=icon}
Return
Model for detecting fake news using a machine learning algorithm
.png){kind=icon}
PRE
AI
DOI：10.1007/s40012-025-00421-z.png)
.png){kind=icon}
Original request for help.png){kind=icon}
Share.png){kind=icon}
Save
Abstract
En 中文
Fake news has recently gained popularity, and people frequently circulate it without verifying its veracity to advance particular ideologies or political goals. Media organizations rely on attracting viewers to their websites to generate online advertising revenue. Therefore, it is essential to be able to recognize fake news to uphold credibility and integrity. Developing these strategies and technologies is crucial to ensure the veracity and integrity of news reports in the digital age. This paper presents a cutting-edge research model and tools for detecting fake news in real-world settings. This model extracts features in various directions, applies information gain to reduce the feature set, and utilizes cross-validation to compare the efficiency of different models. Finally, the model employs sequential forward selection to evaluate its optimal accuracy. The results of the experiment show that the Adaboost classifier achieves optimal accuracy with only 24 features, compared to 40 features.
Keywords:
Fake news
Natural language processing
Techniques and tools
Machine learning algorithm
Information gain
Journal
C
IF:
0
Papers:
31
Citations:
3