EFFECTIVE RECOGNITION OF MISINFORMATION WITH THE HELP OF NEURAL NETWORKS: FOCUS ON THE DETECTION OF EMOTIONAL IMPACT

Abstract

The research focuses on the study of effective methods for automated disinformation recognition using advanced neural network and natural language processing techniques. Particular emphasis is placed on identifying the emotional impact that often accompanies disinformation content. The study applies innovative approaches to text analysis to identify hidden techniques for manipulating the emotional state of the audience used by disseminators of false information. The result of the study is the development of high-precision systems for recognising and categorising the emotional colouring of disinformation-related content. Such systems are able to effectively detect and filter potentially harmful materials, thereby strengthening the resilience of the information environment. The proposed solutions can make an important contribution to the fight against the spread of disinformation and contribute to improving cybersecurity by ensuring the reliability and integrity of the information space. The findings have the potential to improve the means of countering disinformation and strengthening trust in the information environment. The study also includes an analysis of the ethical aspects of using neural networks to recognise disinformation and the development of appropriate standards aimed at protecting user privacy. This comprehensive work offers a robust system for filtering and responding to potential disinformation threats, opening up new perspectives for improving cybersecurity and ensuring the reliability of the information environment.