Full metadata record
DC FieldValueLanguage
dc.contributor.author張紀秋en_US
dc.contributor.authorChang, Chi-Chiuen_US
dc.contributor.author王昱舜en_US
dc.contributor.authorWang, Yu-Shuenen_US
dc.date.accessioned2014-12-12T02:42:58Z-
dc.date.available2014-12-12T02:42:58Z-
dc.date.issued2013en_US
dc.identifier.urihttp://140.113.39.130/cdrfb3/record/nctu/#GT070156604en_US
dc.identifier.urihttp://hdl.handle.net/11536/75300-
dc.description.abstract在現今的資訊時代中,將多變量時間序列資料視覺化是一個必要的技術, 視覺化讓我們了解資料的趨勢,找出多變量的異常變化,預測序列資料的未來走向。因為這些優點,過去也發展了許多視覺化的方法,但過去大部分研究分別著重在時間序列資料或是多變量資料,很少有結合兩者性質的可視化,其原因在於會增加視覺化的複雜度。 為了克服這個問題,我們介紹了一個新的方法,用顏色表示序列資料一個時間點的多變量,相似的顏色則表示相似的多變量;如此可以用一色彩條表示一筆序列資料,從左至右的顏色變化表示隨著時間而改變的多變量。 若不同序列資料的色彩條有相似的顏色與顏色變化,我們可以說這兩筆資料是相似的,反之則可以說是不相似的。 這種節省空間的視覺化方法允許所有序列資料同時顯示在螢幕上,資料分析師可以以全域的觀點找出整體資料的多變量走向;我們系統也提供了其他功能幫助資料分析師了解顏色代表的多變量意義,多虧了互動式的設計,我們系統可以讓使用者很直覺的從全域觀點到局部觀點觀察與分析資料,在最後的案例研究與使用者研究證明了我們方法的可行性。zh_TW
dc.description.abstractTime varying multivariate data visualization is essential because the identified trends, anomalies, or structures usually enable future prediction and are beneficial to developments. Given by such advantages, many techniques have been presented. However, previous methods mainly focus on the depiction of either time varying data or multivariate data because the combination of these two properties would rapidly increase the visualization complexity. To prevent this problem, we introduce a new visualization method that abstracts each data series using a color bar, which extends from left to right as time goes. The color on a bar is determined based on its variates at an instant, in which similar data items have similar colors and the others do not. This space-efficient visualization technique allows all the data series to be displayed at the same time so that the global evolutions between series could be easily obtained by data analysts. Our system also allows analysts to investigate the variates that a color represents for further understanding the details. Thanks to the interactive performance, this from-global-to-local visualization is intuitive. The case studies and the conducted user studies demonstrate the feasibility of our technique.en_US
dc.language.isoen_USen_US
dc.subject視覺化zh_TW
dc.subject多維資料zh_TW
dc.subjectVisualizationen_US
dc.subjectmultivariate dataen_US
dc.title多變量時間序列資料可視化zh_TW
dc.titleTime varying multivariate data visualizationen_US
dc.typeThesisen_US
dc.contributor.department多媒體工程研究所zh_TW
Appears in Collections:Thesis