Knowledge Brewery: Visualizing And Analyzing The Annotations On Map-based Dashboard

With the development of computer science and information technology,a wide variety of visualization tools has been created to help people interact with huge amounts of data in a short time.In this context,dashboards play a significant role in diverse decisionmaking scenarios for the capability of integrating various data sources with analytic functions.Therefore,they have been applied in different research fields including urban planning(e.g.city dashboards)and education(e.g.learning dashboards),etc.However,in most cases,researchers only take the dashboard as a tool to present data thus less attention is paid to user's insights from such interfaces.This inspires our intention to study the diverse comments from different users and further encouraging the communication by sharing the analysis results of their insights to more users.By visualizing the annotations,we aim to build a bridge between users under different scenarios and realize an accumulation process of user knowledge.The innovation lies in the emphasis on the user perspective of the design process through insights.Beyond human-computer interaction,we explore the interaction between different users.This thesis studied the annotation visualizing mainly by a case study of a map-based dashboard In Dash,resulting in a new map-based dashboard “Anno Vis Dashboard”.The new dashboard integrated annotation visualizations through focused topics and presented related geographic and statistic information with interactive functions.A user study made up of a think-aloud experiment(to collect user insights)and a postinterview(to explore topic interest and user backgrounds)was conducted to collect annotation data of 30 users when using the In Dash.Also,segmentation and attention analysis were performed to process and simplify the annotations.Moreover,visualizations of thematic maps,bar charts,and arc diagrams,with interactions of popup tags,selections,and zoom in(out)were combined to visualize the annotations.Finally,a prototype was designed and implemented to integrate the visualizations.The evaluation results of interviews with two visualization experts indicated the high usability of the “Anno Vis Dashboard”.Feedback of designs and functions also organized guidelines and recommendations for future studies on map-based dashboards,including color relation,layout,consistency,and hints for varying panels,etc.