Research On Applied Technology Achievement Transformation System Based On Self-Organizing Data Mining

Applied technology achievement is an important part of the science and technology achievement,which is the main engine of technology progress and economic development. Successfultransformation of applied technology achievement into actual productivity can promote thedevelopment of the corresponding industries and the contribution of science and technology to theeconomic development. Applied technology achievement transformation is a complex systemengineering. The objective of this thesis is to study the progress of applied technology achievementtransformation and to present methods to improve the transformation efficiency.The thesis focuses on system modeling and prediction methods, where the advantages ofself-organizing data mining and regression analysis methods, qualitative modeling methods andquantitative modeling methods are combined organically for the development of an appliedtechnology achievement transformation system. Firstly, the research background of appliedtechnology achievement transformation and self-organizing data mining is introduced, and thestructure and mechanism of the applied technology achievement transformation system is analyzed.Secondly, a first-principle system model is developed and improved for constructing an appliedtechnology achievement transformation system. Thirdly, based on the statistical data of the appliedtechnology achievement transformation in Shanghai from2000to2007, a system-level data-basedmodel is developed using regression analysis and the GMDH algorithm, respectively. The simulationresults show that, the performance of the proposed GMDH-based prediction model is superior to theregression model for short-term forecasting. Finally, the thesis presents several policy suggestions forthe government decision-making on how to promote the applied technology achievementtransformation and benefit economic development.