Examination system in the cloud computing platform based on data mining

With the establishment of open universities and development of colleges of online education in China, massive user data are stored in web-based learning cloud platforms and network examination systems. There will have the problem of very slow processing rate if such a huge volume of information is discovered only traditional methods, reducing the mining efficiency because of frequent read-in (RI) and read-out (RO) of abundant data. For that reason, this paper presents APRIORI algorithm which is based on Map/Reduce parallel program model, so as to complete the excavation of extensive test data in a more efficient and reliable manner. From experimental analysis, it can be noted that APRIORI algorithm after cloud computing is highly efficient in mining frequent item-sets in the context of cloud computing. The proposed technique has better performance than traditional ones.