開放政府以透明、參與、協作三大價值影響近年世界各國政府施政的各項措施，成為深化民主的重要途徑。開放政府理念的落實，除政府本身的努力外，更重要的是透過公私協作模式來實踐。惟對於如何應用公私協作提高政府開放資料利用率的問題，尚缺乏較深入的探討。本研究旨在探討以資料驅動的公私協作機制與類型，希望找尋能擴大公民參與、促進政府與社會對話，進而建立彼此相互信任的協作關係，強化互動與合作之關鍵因素。作者藉由Ruijer等人提出的「開放資料民主行動模型」為研究框架，透過專家訪談解析我國推動資料協作過程的相關議題。本文主要的研究發現有四：第一，近年公私部門對開放資料的可應用範圍及想像逐漸成熟，且政府部門逐漸脫離業務委外的思考模式，開始主動尋求廣大民間社群的協助。第二，協作的障礙在於公私部門彼此間的不了解，特別是欠缺對政府運作模式的認識以及法規定義的解釋與釐清，與對資料使用方式的想像。第三，法規面合規（compliance）使用的要求，機關單位可能採取保守態度。第四，從資料協作成果的擴散面來說，單位主管的關注，示範型專案的標竿學習以及資料競賽活動等都是促進資料協作成果擴散的重點。

Open government has become a major issue for the field of government and public policy research featuring three fundamental values in the relationship between the government and the governed. These values are transparency, participation, and collaboration. In addition to the efforts of the government itself, an important practice is to rethink an open government strategy on how the government can deal with all kinds of open government data and the publication of government information through the mode of public-private collaboration. However, a lack of in-depth discussions on how to use public-private collaboration to increase the utilization rate of government open data exists. This research aims to explore data-driven public-private collaboration mechanisms and types. The author uses the “Open Data Democracy Action Model” proposed by Ruijer et al. as a research framework to analyze the relevant issues related to the process of data collaboration in Taiwan through narrative analysis. The main findings of this study include the following: (1) the government has gradually deviated from its outsourcing strategy and has begun to seek the assistance of the broad community. (2) The obstacles to the process of data collaboration stem from a gap in understanding between the public and private sectors, especially from a lack of understanding of the government’s operation mode and interpretation and clarification of legal definitions. (3) Data collaboration in public sectors is conservative in nature. (4) From the perspective of the diffusion of the results of data collaboration, the attention of unit managers, benchmarking projects, and data competition activities are all key points to promote the diffusion of the results of data collaboration.