VARIABLE QUEST: Network Visualization of Variable Labels Unifying Co-occurrence Graphs

摘要

Data Jacket (DJ) is a technique for sharing information about data and for considering the potential value of datasets, with the data itself hidden, by describing the summary of data in natural language. In DJs, variables are described by variable labels (VLs), which are the names/meanings of variables. In the previous study, the matrix-based method for inferring VLs in DJs whose VLs are unknown, using the texts in outlines of DJs was proposed. However, the previous method showed only the list of VLs with the similarity scores to queries. Moreover, the relationships among VLs were not displayed in the lists, which was difficult for users to understand the connections of VLs. In this paper, we propose VARIABLE QUEST which is the network visualization system of VLs using the matrix-based inferring method of VLs by unifying co-occurrence graphs. VARIABLE QUEST represents the co-occurrence and the frequency between VLs in DJs. The co-occurrence of VLs is a feature that there may be a highly frequent pair of VLs appearing at the same time in data, e.g., "latitude" and "longitude," or "weather" and "temperature." The network visualization unifying co-occurrence graphs may not only support those who want to obtain new data in the Market of Data but also function as a tool for aiding the design of experiments of the researchers.