Visualization techniques for collaborative data analysis

点击翻译按钮获取中文摘要

We are continuously generating large amounts of data in different forms, and facing more complicated problems related to the data. With recent technological advances in big data analytics, many well--defined questions can be automatically resolved by machines. However, there are still many data problems ill-defined, vague, and exploratory, and they require close human engagement and supervision. Information Visualization (InfoVis) and Human-Computer Interaction (HCI) are powerful techniques to help people understand abstract data and algorithms with visual representations and user interactions. However, due to the scale and complexity of the data problems, multiple analysts with diverse backgrounds are requested to collaborate together. Further, of the three key players------data, machines, and humans---"people as the center" has become a shared view for the approach to solve real world problems. But human bandwidth for processing information is limited. Additionally, the communication cost is usually high in work collaboration. While many methods have been proposed in InfoVis and HCI, there still lacks a holistic understanding of the relationships between data, machines, and humans, as well as effective techniques to leverage and support them. This proposed research aims to investigate visualization techniques for collaborative data analysis, by promoting the interplay between data, machines, and humans, as an overarching goal. We will pursue three interrelated (but not strictly dependent) objectives. The first is to develop visual analysis tools to provide insights into analysts' collaborative behaviors and patterns. The second is to develop interactive visualization to support communication of abstract data and models in collaborative data analysis. The third is to design algorithms and visual metaphors to improve quality and efficiency of knowledge discovery in collaboration. This research will produce new visualization techniques to leverage and support the interplay of data, machines, and humans in collaborative data analysis activities. The resulting methods and systems will provide understandings to human behaviors during collaboration and to improve analystss' efficiency in collaborative data analysis, thus saving time and money. The outcomes will form an infrastructure or repository to be reused for broader interests in both immediate domains (e.g., data science) and other fields (e.g., social science). This research program will also train students in a diverse and interdisciplinary environment, allowing them to gain theoretical and practical skills in designing new visual representations, interactions, and algorithms.