Method for predicting data based on support vector machine and equipment thereof

Abstract

The invention provides a method for predicting data based on a support vector machine and an equipment thereof. The method comprises the following steps of: reading a plurality of first data (historical data) related to a first period (historical periods); and predicting a second data which corresponds to a second sub-period contained in a second period (predicting period). At least one subset of the read plurality of first data is taken as an input variable of the support vector machine. The time of the second period is later than that of the first period. The embodiment of the invention has effective traffic prediction, thus bringing considerable economic benefits for operators directly or indirectly.

Description

technical field [0001] The present invention relates to a data prediction method and device, more particularly, to a data prediction method and device based on a support vector machine. Background technique [0002] In recent years, wireless communication services have developed rapidly all over the world, the scale of users has continued to expand, and many new services have emerged one after another. However, the increasing traffic volume (showing an obvious upward trend) and the continuous improvement of users' requirements for wireless communication quality have put forward new requirements for wireless communication networks. and so on increase the difficulty. Especially during special holidays (such as Spring Festival, Mid-Autumn Festival, May Day, National Day, New Year's Day, Easter, Thanksgiving, Holy Day, New Year, etc.) or special events (such as Wenchuan Earthquake, Shanghai license plate auction, etc.), wireless communication The sudden increase in the traffic...

AI Technical Summary

Problems solved by technology

[0019] (1) The seasonal model does not take into account the influence of special holidays (such as Chinese New Year, Mid-Autumn Festival, Western Easter, Thanksgiving, and Islamic Holy Days and Eid Days, etc.) on dialogue traffic;

[0020] (2) Although the seasonal model considers the trend, seasonality and randomness of wireless communication traffic data, it does not consider the obvious periodicity of wireless communication traffic data for small granularity (such as days) ;

[0021] (3) The time series of wireless communication traffic has complex and strong nonlinear characteristics, and is affected by many factors. It is difficult to reveal its internal laws by using traditional forecasting techniques, and the generalization ability is poor

Traditional forecasting methods such as linear regression and exponential regression only use the time variable t as the input / independent variable, without considering the influence of other factors, so they can only describe the overall trend of the evolution of the traffic time series, but cannot describe the local details The change

On the basis of linear regression, the seasonal model prediction method can achieve better prediction results than linear regression and exponential regression by considering the changes of seasonal factors at each point. It has also become a commonly used traffic forecasting algorithm, but it is still difficult to obtain Ideal forecasting effect

Images