<p>The low entry barriers and sharing mechanisms of social media facilitate the rampant spread of fake news, which poses a serious threat to cyberspace security. Manipulator accounts, as one primary driver of the spread of fake news, exacerbate this issue. Numerous methods exist for detecting fake news and manipulator accounts that manipulate public opinion. However, it is challenging to compare these methods due to variations across datasets, metrics, and methodologies. This research, combining a comprehensive survey with case study validation, systematically analyzes the current state of research on the detection of fake news and manipulator accounts, providing new insights and frontiers for emerging researchers. Specifically, this study first summarizes the laws, regulations, and multi-stakeholder governance models to curb the spread of fake news. Secondly, it focuses on fake news detection frameworks, analyzing multimodal datasets, feature extraction techniques, and detection methods. Next, it explores a categorized approach to detecting manipulator accounts based on content semantics, behavior, social structure, and hybrid features. Furthermore, three real-world datasets are used to validate and analyze the performance of various features and methods for fake news detection. Meanwhile, we collect data on fake news manipulation events through web crawling and empirically validate the distribution of manipulator accounts. Finally, we outline future research directions for fake news and manipulator account detection, providing a reference for cyberspace governance.</p>