How to quickly and effectively mine valuable insights from massive data to better guide decision-making is an important goal of big data analytics. Visual analytics takes advantage of the characteristics of human visual perception, utilizes visualization charts to encode complex data intuitively, and supports human-centered interactive data analytics, which is one of the important ways of big data analytics. However, visual analytics still faces several challenges, such as the high cost of data preparation, the high latency of interaction response, the steep learning curve of visual analytics, and the low e?iciency of interaction mode. To address the above challenges, recent studies propose to optimize the human-computer interaction mode and improve the visual analytics system's intelligence by leveraging data management and artificial intelligence techniques. In this paper, we systematically survey recent advances in this field and present the basic concepts and key technical framework of intelligent data visual analytics. Secondly, we summarize the overall architecture of intelligent data visual analytics, which comprises effective data preparation for visual analytics, intelligent data visualization, e?icient visual analytics, and intelligent visual analytics interfaces. Next, we go through the prior art, paying particular attention to problems that may attract interest from data management, visualization, and machine learning communities. Finally, we discuss the research opportunities of intelligent data visual analytics.