Research on Trend Analysis and Prediction Algorithm Based on Time Series
Qiang Chen, Qi Xin · 2019 3rd International Conference on Electronic Information Technology and Computer Engineering (EITCE) · 2019
At present, there is an increasing demand for big data analysis in various industries. Big data is characterized by large data scale, long calculation cycle, and high resource performance requirements for distributed computing systems. For the resource performance and state information provided by the distributed computing system of big data, the trend-based time prediction algorithm is used to predict the short-term performance of the resource. We set more weights to track the trend of data change, distinguish the different situations of data increase and decrease, so that the improved hybrid trend prediction algorithm has better performance than the improved one, and effectively improves the accuracy of prediction.