Data prediction method based on interval extraction method

Abstract

The invention discloses a data prediction method based on an interval extraction method, and mainly relates to the technical field of data processing and data prediction. According to the data prediction method, a plurality of data sets are extracted from a single group of actual data with different interval numbers on the basis of an interval extraction method, and long-term data prediction is achieved. The method is realized based on seven steps: extracting a first-level data set and a third-level data set from a single group of actual data with different interval numbers based on the interval extraction method; extracting a second-level data set from the first-level data set; training the data prediction model group by using the second-stage data set; and inputting the third-level data set into the trained data prediction model group to obtain a fourth-level data set composed of a plurality of independent model prediction result data sets. According to the technical scheme disclosed by the invention, the mining degree of hidden information in actual data is increased; the problem that long-term prediction of data and long-term prediction accuracy of the data are difficult to guarantee is solved.

Description

technical field [0001] The invention relates to the technical field of data processing and data prediction, in particular to a data prediction method based on an interval extraction method. Background technique [0002] At present, it is a commonly used data forecasting technology to obtain forecasting data through experience judgment, formula calculation or training various data forecasting models through actual data; therefore, in order to obtain better forecasting results, many scholars and technicians are more Focus on formula derivation, inventing or improving various data prediction models, etc.; however, the results of increasing the length of data prediction and improving the accuracy of prediction based on actual data processing are rarely reported. [0003] In the technical scheme of training various data prediction models through actual data and finally obtaining the predicted data, the actual data is the basic data of the data prediction technology, and the proce...

AI Technical Summary

Problems solved by technology

[0005] Aiming at the deficiencies of the existing data prediction technology in terms of long-term data prediction and data long-term prediction accuracy, the technical solution disclosed in the present invention increases the degree of mining hidden information in actual data; solves the problem that the existing data prediction technology is difficult to realize data Long-term forecasting and the difficulty of guaranteeing the accuracy of long-term data forecasting

Images