Optimization and Application of SPRINT Big Data Classification Algorithm in Cloud Computing Framework

When processing massive data, existing classification methods frequently take too long to compute, making their performance challenging to satisfy the actual needs of big data applications. In this paper, an SPRINT optimization classification algorithm based on a cloud computing framework is studied. The idea is to build a three-level cloud computing framework, and the optimization SPRINT classification algorithm is implanted into it so that the SPRINT classification algorithm has better stability and communication efficiency, as well as a higher-performance parallel computing environment in the process of processing big data. SPRINT classification algorithm optimization the process is as follows: First, the segmentation of GiNi value nodes is used to enhance the parallel computing capability of the classification algorithm and the optimization ability of continuous attributes. Next, the subsets of the big data set are reclassified, and the decision tree construction is finished based on the optimal segmentation point to enhance the big data classification processing ability. According to the experimental findings, the cloud computing framework's optimized SPRINT classification algorithm outperforms the more established big data classification algorithm in terms of classification performance and efficiency, as well as execution rate and parallel performance.