Enabling quality-aware consideration for interoperable GIS design

• Main Authors: Yu-TingSu, 蘇郁婷
• Other Authors: Jung-Hong Hong
• Format: Others
• Language: en_US
• Published: 2017
• Online Access: http://ndltd.ncl.edu.tw/handle/kbb2zb

碩士 === 國立成功大學 === 測量及空間資訊學系 === 105 === The fast growth of internet-based sharing mechanism and OpenGIS technology enable users to easily access geospatial data in open formats and develop various applications on platforms, ranging from personal computers to internet-based mechanism. With its superior capabilities to integrate data from different domains, GIS has been widely and extensively used by millions of users from a variety of domains. Furthermore, the majority of increasing users in the coming years are expected to be naive users without prior knowledge about geospatial data and GIS operations. While users’ interaction to the GIS becomes easier with friendly interface, the lack of knowledge about the heterogeneous nature of distributed data and the process on geospatial data handling becomes a major source of risks for GIS. We argue the next generation of GIS-based environment, regardless internet-based or not, must have built-in knowledge to automatically and correctly assess the fitness of data use and present the analyzed results to users in an intuitive and meaning way. “Quality-Aware” is not a new concept in GIS research. The idea of Quality-Aware GIS was proposed to take data quality consideration into GIS design by automatically providing auxiliary information for users’ decision reference. By arguing it is impractical to expect users to build a correct understanding towards the selected data purely based on current mechanism(e.g., manually inspect the metadata), we propose a quality-aware GIS mechanism in this thesis for “automatically” analyzing and prompting visual guidance about the quality of the operation outcomes to aid users’ decision making. To develop a quality-aware GIS, we identify two major issues, GIS operations and visualization of outcomes. From the GIS operation perspective, the workflows of GIS operations should be examined and modified by taking the impacts of data quality into account, such that a correct evaluation about the accuracy and validity of the operation outcomes can be presented. From the perspective of outcome visualization, we proposed to use various visual aids, e.g., virtual layer, informative window, symbol transformation and augmented TOC, to address the challenges of visually explaining data quality of operation outcomes. These visual aids are fully assimilated into the GIS interface design, such that users can easily manipulate the tools or components available in the interface to “visualize” the differences of selected datasets. In a nutshell, the quality-aware GIS can thus not only present the operation outcomes of geospatial data, but also convey quality information for understanding acquired geospatial data more comprehensively. In the research, we choose ten common GIS operations, namely, map overlay, geometric operation, nearest neighbor, query (spatial/attribute), measurement (distance/area), voronoi diagram, join (spatial/attribute), statistics, thematic map and routing, to examine their quality-aware strategies. We then summarize useful guidance about developing quality-aware GIS from GIS operation and visualization perspectives. Six of these operations, which are map overlay, query (i.e. select by region), nearest neighbor, measurement (i.e. line measurement), join (i.e. spatial join) and statistics (i.e. mode), are implemented in the prototype system for testing the feasibility of the proposed approach. As a proof-of-concept, the test results demonstrate that the proposed approach can successfully convey reliable and valid operation outcomes and additional information about acquired data via user-friendly interface to help users gain understanding about acquired data comprehensively and make more reliable decisions. GIS has been long seen as a powerful integration tool, but its achievements would be highly restricted if it fails to provide a friendly and correct working platform. This research contributes to improve the quality of decision making, which is the most demanded features for the naïve GIS users.