부산대학교 도서관

Business- Intelligence (BI) tools have become an integral part in data-driven organizations as it can transform raw data into visual reports to support decision makings. While a picture is worth a thousand words, interpreting visualized information sometimes leads to miscommunication between author and readers since an image chart may have many interesting features and readers may perceive different points of view from the author. Text narration can provide additional detailed descriptive on the topics of interest but requires reading time. To convey points of interest to readers without reducing the effectiveness of chart visualization, synergizing between visualization and its associate text narration is required. In this paper, a design of business-intelligence framework for data visualization and its associate text narration is presented. The key idea is to highlight related chart as response to reading (mouse) position on text area and vice versa. The framework is implemented as a JavaScript wrapper which acts as a relationship control between charts and texts. The wrapper can manipulate various software libraries to work together, thus reducing the complexity in the development. The comparisons between web pages from renowned web sites such as WorldBank.org and ones generated from our proposed framework using similar layout and data demonstrates that the design meet expected outcomes