A Study on the Prediction of Book Borrowing Based on ARIMA-SVR Model
Yan Pan, Xiuling Jin, Yuliang Li, Dalian Chen, Juan Zhou
Abstract
Accurate prediction of library borrowing can help libraries improve service quality and save costs. Book borrowing is characterized by time series and nonlinearity, with time-varying, nonlinear and non-stationary characteristics. This paper presents a hybrid prediction model based on Autoregressive Integrated Moving Average (ARIMA) and Support Vector-machine Regression (SVR). The model is verified by monthly record data from Xiamen University Library from 2007 to 2017. The experimental results show that the mixed model based on ARIMA and SVR has high prediction accuracy and can accurately describe the complex change trend of the time series of the number of borrowers.
Topics & Concepts
Autoregressive integrated moving averageComputer scienceSupport vector machineAutoregressive modelTime seriesSeries (stratigraphy)Data miningQuality (philosophy)Service (business)Nonlinear systemArtificial intelligenceMachine learningEconometricsMathematicsPhysicsPhilosophyEconomyEpistemologyQuantum mechanicsBiologyEconomicsPaleontologyHydrological Forecasting Using AIStock Market Forecasting MethodsData Stream Mining Techniques