Relation Mining Between Government Open Data State and the Subject Behavior State in the Data Driven Paradigm postprint

Abstract: [Purpose/significance] In the context of data-driven paradigm, this paper reveals the internal relationship between the open data state of government portal website and the behavior state of its main body, and promotes the effect and process of government data opening.[Method/process] This paper used the crawler method to grab the open data sets in the Shanghai government data portal, and then did correlation analysis and Stepwise regression analysis on the index variables of each data set in turn, and screened out the variables with high correlation degree. At the same time, we further carried out PLS regression test on the variables with significant relationships, and ultimately drew the internal relationship between the state of government open data and the state of its main body's behavior.[Result/conclusion] In the process of government data opening, the subject behavior of government departments has a greater impact on public subject behavior than the object characteristics of data itself. Among the factors affecting the public rating, the sequence from high to low is:government openness and secrecy, machine readability of Data Format, the first opening time of government, the timeliness of government openness. The level of government openness and secrecy has a significant negative impact on the score of government open data; the first opening time of government, the timeliness of government openness, and the machine readability of data format have a significant positive impact on the public score.