Cognitive warfare aims to weaken the ability to make correct decisions by targeting the minds and perceptions of commanders. The spread of false and misleading information is among the most critical tools of this warfare used to create confusion and disruption in the military decision-making process. This research aims to answer how machine learning algorithms can be utilized to identify and neutralize misinformation in cognitive warfare. The research methodology is qualitative, based on a systematic literature review and thematic analysis, through which reliable national and international sources have been analyzed.
The findings indicate that machine learning algorithms, across three layers—content analysis, network analysis, and user behavior analysis—are capable of identifying various types of threats, including fake news, propaganda, rumors, and conspiracy theories. Based on the results, deep learning, natural language processing (NLP), and traditional machine learning algorithms can neutralize the destructive effects of cognitive warfare through operational solutions such as automatic removal, content labeling, and providing correct information. Finally, this research emphasizes that despite the capabilities of artificial intelligence, maintaining "human-in-the-loop" interaction is essential to prevent cognitive laziness and ensure compliance with ethical and legal considerations on the battlefield