The Application Of Data Mining Technology In Analysis Of The Government Procurement

In the past decades,along with swift development of information technology,China's government procurement to promote the construction of information,the majority of the national municipal government procurement by the information management replaced by the manual operation,including the purchase of the Declaration,approval,implementation,funds to pay the entire process of flow,standardized management.The Government procurement management system,in particular the procurement Bulletin text,transaction bulletin text,stored a large number of valuable data,but by the Government Procurement management department to use little data.The Procurement Management Department needs effective data as a basis for the formulation of government procurement macro policy and the next annual budget,however,the data is so large and complex that it can be used to reflect the value of the data,thus providing a basis for government procurement departments to make procurement decisions,and discovering the law of government procurement data through the excavation,to judge whether government procurement has played the guiding role of national financial funds.In order to solve the above problems,this paper uses text mining technology to analyze government procurement data.Firstly,this paper introduces the theory foundation of text mining,the definition of text mining,the process and the common methods.Then the data collection and preprocessing,from Shandong provincial government procurement online collected more than 10,000 transactions bulletin data,text segmentation,text classification,and the construction of a professional dictionary of government procurement.In this paper,the Apriori improved algorithm is put forward,which is used in the government procurement data to analyze whether the purchasing unit and the purchasing agent are dependent on the suppliers.Through the text mining analysis,obtained the government procurement product type,the product place as well as the purchase unit distribution,the result indicated that the government procurement in 2017 the purchase expenditure increased obviously,especially in the scientific research equipment,the engineering,the service,the medical equipment four products to invest more.In purchasing units,government agencies purchase the most,in education,the fastest growth in expenditure,followed by health.The time Series algorithm predicts that the purchase volume of government procurement will increase in the next three years.Through association rules analysis,the results show that some purchasing agent may have a dependency relationship with the supplier.