Dynamic Prediction Research of Silicon Content in Hot Metal Driven by Big Data in Blast Furnace Smelting Process under Hadoop Cloud Platform

Table 4

Summarization of statistical results of [Si]% lifting dynamic prediction performance of three different algorithms.

Error range					Time (ms)
Interpolation of SVM algorithm for serial Pegasos	77.50%	72.31%	77.90%	74.92%	9,004,328
Parallel Pegasos solving SVM algorithm	82.55%	63.13%	84.07%	72.62%	1,776,882
Parametric parallel Pegasos solving SVM algorithm	92.2%	85.47%	86.33%	85.90%	1,710,788