More about HKUST
A Survey on Data Visualization Collection Retrieval and Mining
PhD Qualifying Examination Title: "A Survey on Data Visualization Collection Retrieval and Mining" by Miss Yanna LIN Abstract: Huge amount of visualizations are created and shared on the web by people from different domains. These visualizations, as a new data format, contain a wealth of knowledge and therefore raise popularity in crawling them as visualization collections. Analyzing such kind of visualization collection is beneficial for many downstream applications, such as suggesting examples for visualization creation or providing textual digest for visualization understanding. However, the analysis is not easy due to the variety (i.e., varied representations, sources, and chart types) and multi-modality (including text, data tables, specifications, and visual effects) of visualizations. Moreover, it will be more challenging, since it requires assembling information from multiple visualizations and exploring the relationships between them. Many researchers have attempted to address these challenges by accessing and mining the hidden collective knowledge contained in a visualization collection. This survey begins with an introduction to the concept and background of data visualization and visualization collection. Next, efforts on data visualization collection research are classified into two categories: information access-oriented retrieval and information analysis-oriented mining. We further divided retrieval into keyword-, structure-, and example-based retrieval, and the mining into summarization, question answering, and design demographic analysis. Based on this taxonomy, we review the pioneering works in each area. This review ends with a discussion about future research opportunities. Date: Monday, 15 November 2021 Time: 4:00pm - 6:00pm Zoom meeting: https://hkust.zoom.us/j/6745980845 Committee Members: Prof. Huamin Qu (Supervisor) Prof. Dik-Lun Lee (Chairperson) Prof. Cunsheng Ding Prof. Raymond Wong **** ALL are Welcome ****