政府開放資料基於時間與空間的跨檔案應用之研究

隨著資通訊技術發展進入大數據時代，開放資料已成為全球資料應用主流，目前政府開放資料平台雖有提供大量資料集提供大眾使用，但目前平台上許多資料格式不一致或無嚴謹架構，資料品質參差不齊，難以直接加以利用，此問題在針對不同時間或不同行政區域的相同資料的項目統計跨檔案應用更為嚴重，這些應用需要對數據進行預處理，但預處理過程相當複雜，因此本研究對基於時空特徵的政府開放資料跨檔案應用進行初步研究，並且開發輔助工具，以用於開放資料的跨檔案處理。本研究首先探討開發跨檔案應用的可能問題，然後提出可能的解決方案，以期降低應用開發的成本與門檻。另外，本研究進行實作開放資料聚合輔助工具，包含時間與空間分析模組及跨檔案資料聚合應用輔助開發模組以促進跨檔案應用的可能性，其中時間與空間分析模組用於萃取政府所有開放資料集名稱上的時間和空間屬性，而跨檔案資料聚合應用輔助開發模組則是讓使用者根據自己的需求自由選擇要合併進行聚合計算資料集，最後並可將處理結果以CSV、XLSX或JSON等檔案格式匯出，以促進資料之再利用及加值應用。

關鍵字

開放資料；跨檔案應用；時間性資料；空間性資料

並列摘要

With the development of information and communication technology, we live in the era of big data, with a great amount of open data. In Taiwan, the government provides an open data platform with many datasets for public use. However, the data formats of the open datasets on the platform are not all consistent and the data structures are not standardized. As a result, it is difficult to make use of these data directly. Especially, this issue will be more crucial for multi-file applications, e.g. statistics of the same data items in different time periods or in different government departments. These applications require the pre-processing of the data, but the pre-processing is quite complicated. This paper conducts a preliminary study on multi-file applications of government open data based on temporal or spatial characteristics. The study considers the issues in the development of a platform for implementing multi-file applications of open data. In this preliminary study, we will first identify the issues in developing multi-file applications, and then propose possible solutions to reduce cost and overheads in the implementation of the applications. This study further implements an aggregation tool for developing multi-file applications. This tool includes a s time-spatial analysis module and a development module for multi-file data aggregation applications. The time-spatial analysis module is used to extract the time and space attributes from the name of a data set. The development module allows users to freely select multiple datasets to be combined for aggregation calculation. The result of aggregation calculation can be exported in file formats such as CSV, XLSX or JSON to promote data reuse and value-added applications of Taiwan’s government open data.