Research On The Design And Development Of Data Mining Cloud System In Web-based Instruction Platform

In the era of big data, a variety of data are being stored in all kinds of network teaching platforms, which grow in massive measure. Full and integrated use of these data as a basis for educational decision-making can improve the objectivity of decision-making and make it more scientific. Educational data mining cloud system is the software system which helps people achieve these goals, and thus is of important significance. Encompassing fields such as education, data, computer software, etc., educational data mining cloud system is still in its infancy. This paper explores the design and development of data mining cloud system.Through interactive information systems analysis, this thesis first discussed data mining application modes in education from the perspective of interactive information system, proposed an educational theory and data mining engineering framework as the theoretical basis for the design and development of the present education data mining and cloud system. Starting from this framework, combined with the software needs hierarchical model, the software needs of current system took shape.Algorithm is the core component of educational data mining system. On the basis of cloud computing/parallel computing binary SVM and multi-single SVM, this thesis proposed a multi-classification binary SVM parallel algorithm, and after an experiment using UCI standard sample data set, it was found more effective than single algorithm with a bit lower accuracy but still within acceptable limit.In the development of large-scale and complex systems, system architecture is a very important part. Based on the analysis of the infrastructure of and SOA service architecture strategy of cloud computing, this paper proposed a cloud-based SOA service architecture solution in the field of educational data mining, which can be divided into the performance layer, interface service layer, cloud computing business logic layer, and cloud data processing layer.At system development stage, rational development and technology selection strategies, as well as cloud computing platform, data synchronization, backstage management, mining algorithm, service interface and other key technologies were discussed.Finally, under the guidance of the application framework derived from the viewpoint of information interactive system, this system was applied to the net-based English grammar learning platform, net-based writing community and learning performance analysis platform to verify the effectiveness of the system in the network teaching and learning setting, and found easy to incorporate with different teaching systems, to support Vector Classification Mining with educational data in cloud environment, and to provide effective platform and integration modules for educational data mining.