Successfully detecting, analyzing, and reasoning about collective anomalies is important for many real-life application domains (e.g., intrusion detection, fraud analysis, software security). The primary challenges to achieving this goal include the overwhelming number of low-risk events and their multimodal relationships, the diversity of collective anomalies by various data and anomaly types, and the difficulty in incorporating the domain knowledge of experts. In this paper, we propose the novel concept of the faceted High-Order Correlation Graph (HOCG). Compared with previous, low-order correlation graphs, HOCG achieves better user interactivity, computational scalability, and domain generality through synthesizing heterogeneous types of objects, their anomalies, and the multimodal relationships, all in a single graph. We design elaborate visual metaphors, interaction models, and the coordinated multiple view based interface to allow users to fully unleash the visual analytics power of the HOCG. We conduct case studies for three application domains and collect feedback from domain experts who apply our method to these scenarios. The results demonstrate the effectiveness of the HOCG in the overview of point anomalies, the detection of collective anomalies, and the reasoning process of root cause analyses.

Detecting, analyzing and reasoning collective anomalies is important for many real-life application domains such as facility monitoring, software analysis and security. The main challenges include the overwhelming number of low-risk events and their multifaceted relationships which form the collective anomaly, the diversity in various data and anomaly types, and the difficulty to incorporate domain knowledge in the anomaly analysis process. In this paper, we propose a novel concept of high-order correlation graph (HOCG). Compared with the previous correlation graph definition, HOCG achieves better user interactivity, computational scalability, and domain generality through synthesizing heterogeneous types of nodes, attributes, and multifaceted relationships in a single graph. We design elaborate visual metaphors, interaction models, and the coordinated multiple view based interface to allow users to fully unleash the visual analytics power over HOCG. We conduct case studies in two real-life application domains, i.e., facility monitoring and software analysis. The results demonstrate the effectiveness of HOCG in the overview of point anomalies, detection of collective anomalies, and reasoning process of root cause analysis.

Deep learning has achieved impressive success in a variety of tasks and is developing rapidly in recent years. The problem of understanding the deep learning models has become an issue for the development of deep learning, for example, in domains like medicine and finance which require interpretable models. While it is challenging to analyze and interpret complicated deep neural networks, visualization is good at bridging between abstract data and intuitive representations. Visual analytics for deep learning is a rapidly growing research field. To help users better understand this field, we present a mini-survey including a user-based taxonomy that covers state-of-the-art works of the field. Regarding the requirements of different types of users (beginners, practitioners, developers, and experts), we categorize the methods and tools by four visualization goals respectively focusing on teaching deep learning concepts, architecture assessment, tools for debugging and improving models, and visual explanation. Notably, we present a table consisting of the name of the method or tool, the year, the visualization goal, and the types of networks to which the method or tool can be applied, to assist users in finding available tools and methods quickly. To emphasize the importance of visual explanation for deep learning, we introduce the studies in this research field in detail.