Research On The Optimization Of Time And Commodity Granularity In The Supervision Of Commodity Trading Market

In recent years,the commodity trading market has become increasingly prosperous,involving a wide range of commodities,and has the characteristics of large number of transactions,large price fluctuations and high transaction risk,which poses a huge challenge to the transaction supervision.Especially in the actual process of supervision,there are some problems,such as fuzzy setting of supervision time range,random selection,single setting of supervision commodity range,unclear direction and so on,which lead to the waste of supervision resources and deviation of supervision results.In this thesis,the above problems are studied on the granularity optimization of supervision time and supervision commodity,and the corresponding system is developed for algorithm deployment and function module integration.Firstly,this thesis studies the optimization of regulatory time granularity in commodity trading market.This thesis establishes a time series model for commodity prices,uses PAA dimensionality reduction and standardization methods to process the series and analyze the trend,and then calculates the approximate period of commodity price fluctuation,which is regarded as the recommended regulatory time granularity.The experimental results show that the algorithm can effectively reduce the fitting error to retain the trend of the original series and improve the contraction expansion ratio.Then,this thesis studies the optimization of regulatory commodity granularity in commodity trading market.In the past,most of the algorithms only analyze the commodity transaction data,ignoring the analysis of the commodity market data,so the reliability of the associated commodity is low.In this thesis,firstly,through the analysis of different commodity market data and transaction data,we get the price fluctuation related commodities and transaction event related commodities,then construct the commodity correlation table.Then,according to the requirements of regulatory tasks,the recommended regulatory products are selected from the table to optimize the granularity of regulatory products.Experimental results show that the algorithm can effectively improve the number of associated products and improve the coverage success rate and recommendation matching rate of recommendation results.Finally,this thesis develops an optimization system for commodity transaction supervision time and commodity granularity.On the basis of the previous algorithm research,we design and develop the commodity trading supervision time and commodity granularity optimization system to verify the engineering applicability of the algorithm.Firstly,the operation process and main data structure of the system are designed,and the supervision time granularity optimization algorithm and supervision commodity granularity optimization algorithm are deployed to the corresponding modules.Then the specific implementation of each module,including the data table of each module,the main page view and system integration.Finally,the function and performance of the system are tested and analyzed,which proves the effectiveness and stability of the system,and verifies the good applicability of the algorithm.