Research On Garlic Price Forecasting And Visualization Based On Big Data

China's garlic production accounts for more than 70% of the world's total output.It is China's export-oriented agricultural products,and its frequent price fluctuations have attracted the attention of the whole society.The fluctuation of garlic price directly affects the vital interests of the garlic industry practitioners and affects the healthy development of the garlic industry.Using garlic industry big data analysis technology,taking garlic price as the research object,summarizing and combing the influencing factors of garlic price and price fluctuation,analyzing the main influencing factors affecting garlic price fluctuation,studying the forecasting method of garlic price change trend,and constructing garlic The specific research contents of the price visualization analysis system are as follows:(1)Research on the fluctuation law of garlic price and its influencing factors.Firstly,according to the peak-valley principle,the garlic price fluctuation cycle is divided.Then,the seasonal adjustment method and HP filter are used to empirically analyze the garlic price,and quantitatively analyze the variation of seasonal components,periodic components,trend components and random components after price series decomposition.And compared with the actual industry verification,the garlic price was finally divided into 8 cycles.On this basis,the paper analyzes the supply,demand and macroeconomic factors that affect the price fluctuation of garlic,and then uses the method of empirical analysis to carry out the correlation test and co-integration test on the selected key influencing factors,and establish a vector autoregressive model.Quantitative angle analysis of the factors affecting garlic price,combined with the impulse response function and variance decomposition method to analyze the contribution of various factors to garlic prices,the study found that the changes in garlic price and exchange rate have a greater impact on garlic price fluctuations.(2)Research on garlic price prediction.Firstly,the paper analyzes the relationship between the price fluctuation of garlic and the price fluctuation of garlic,and then uses the method of empirical analysis to test the correlation between garlic price and garlic price,cointegration test and Granger causality test.The price fluctuation of garlic can predict the future fluctuation of garlic price,and predict the best trend of garlic price fluctuation after the next 2 months.Then combined with the advantages of time series prediction model and machine learning,the ARIMA-SVR combination model is constructed for garlic price.The short-term price forecast is compared with the precision of a single ARIMA and SVR forecasting model.The results show that the ARIMA-SVR combined model has higher prediction accuracy,which proves the validity of ARIMA-SVR model for garlic price prediction.(3)Garlic price visualization analysis system.The system hierarchy structure and functional framework were designed by analyzing the demand of garlic price,and five kinds of databases based on price,region,analysis result,URI and system role were constructed.Based on this,the visualization technology and system development technology were applied to realize the price of garlic.Visual analysis system.The garlic price visualization analysis system more intuitively presents the garlic price and price rules,which is convenient for people to understand and communicate the garlic price and promote the healthy development of the garlic industry.