Mobile terminal data traffic prediction method based on arge big data analysis
Mobile data traffic is influenced by various factors,such as user behavior,network conditions,holiday effects,etc.,presenting a high degree of non-stationarity,making direct prediction difficult.Therefore,a mobile terminal data traffic prediction method based on big data analysis has been proposed.Firstly,the Pearson correlation coefficient is used to accurately measure the correlation between adjacent time series,and an intuitive time series graph is drawn to effectively reveal the dynamic changes and potential trends of data flow.Secondly,for non-stationary time series,the concept of moving average is introduced to extract key temporal features,and differential method is used to effectively process non-stationary sequences to ensure data stationarity.Fi-nally,in the stage of feature extraction and traffic prediction,big data technology was applied to deeply explore the complex pat-terns behind data traffic,achieving high-precision prediction.The experimental results show that the error value of this method is only 15 MB,and the prediction results are relatively accurate.
big data analysisdata traffic predictionprediction methodtime seriestime series graph
万方数据
