Electronic paper coupon issuing prediction method and system based on big data

Abstract

The invention relates to the technical field of electronic paper coupon management, and discloses an electronic paper coupon issuing prediction method and system based on big data, and the system comprises a data collection module and a data analysis module. The prediction method comprises the following steps: step 1, a data acquisition module captures electronic paper coupon big data platform information through a KnowLM algorithm model; step 2, the data acquisition module performs data cleaning and then performs classification to form a data set; 3, the data analysis module calculates and generates a year-on-year difference # imgabs0 # to reflect the long-term use trend of the electronic paper coupons; 4, the data analysis module calculates and generates a link ratio difference # imgabs1 #, the short-term use trend of the electronic paper coupons is reflected, and the big data analysis and prediction accuracy is higher; 5, the data analysis module calculates the annual average traffic flow of the toll station; and step 6, the RNN algorithm model calculates and generates a prediction data set # imgabs2, the issuing number of the electronic paper coupons is predicted, and the algorithm fusion generalization ability is high.

Description

Technical Field

[0001] The present invention relates to the technical field of electronic paper voucher management, and in particular to a method and system for predicting the issuance of electronic paper vouchers based on big data. Background Art

[0002] In modern traffic management, highway toll stations play a vital role. They are not only responsible for collecting tolls, but also related to the regulation of traffic flow and the rapid passage of vehicles. In recent years, with the advancement of technology, traditional manual toll collection methods are gradually being replaced by automated systems such as electronic paper coupons. The automated system not only improves the efficiency of toll collection, but also greatly reduces labor costs and the probability of errors. The self-service electronic paper coupon system allows drivers to collect and recycle paper passes without manual intervention, shortening the processing time from the original 15-25 seconds to 3-5 seconds, significantly improving the speed of passage. The successful upgrade test of the smart lane's self-service printing and reading of paper coupons reflects the application of intelligent traffic management systems in improving the efficiency of highway toll stations. Policies and regulations directly affect the predicted number of electronic paper coupons issued. The toll-free policy for small passenger cars with 7 seats or less during major holidays increases the traffic volume at toll stations during specific periods, so more paper coupons need to be issued during the free transition period to cope with the increased traffic volume. According to the relevant regulations of the Ministry of Transport, from 18:00 to 24:00 on the day before holidays and the last day of holidays, small passenger cars that have not applied for ETC need to issue and collect paper passes. This regulation directly affects the number of electronic paper passes issued and their use. Accurate predictions of traffic flow can better plan the number of electronic paper passes issued and avoid insufficient or excessive paper passes due to inaccurate estimates. However, in the actual prediction process, the factors that affect the number of electronic paper passes issued at highway toll stations are diverse and interrelated. From technical support to policy regulations, to economic costs and traffic demand, each factor needs to be carefully considered to ensure the efficiency and smoothness of the transportation system.

[0003] At present, the traffic flow data collected by the traditional big data-based electronic paper coupon issuance prediction method contains a lot of abnormal data and lacks effective processing methods, which directly affects the accuracy of the prediction results. Traditional prediction models usually use a single time series model. When dealing with complex and changeable traffic flow data, they often show the problem of insufficient generalization ability, especially during holidays and peak periods, when traffic volume changes dramatically, and a single model is difficult to accurately capture these dynamic characteristics. Summary of the invention

[0004] 1. Technical issues to be resolved In view of the shortcomings of the prior art, the present invention provides a method and system for predicting the issuance of electronic paper coupons based on big data, which has the advantages of higher accuracy of big data analysis and prediction and strong algorithm fusion and generalization ability. It solves the problems of low accuracy of traditional big data-based electronic paper coupon issuance prediction methods and insufficient generalization ability of a single algorithm model.

[0005] (II) Technical solution To achieve the above object, the present invention provides the following technical solution: a method for predicting the issuance of electronic paper coupons based on big data, comprising the following steps: Step 1: The data collection module uses the KnowLM algorithm model to capture information from the electronic paper coupon big data platform to obtain traffic flow information and holiday policy information; Step 2: The data collection module cleans the acquired information, classifies it into traffic data sets and policy data sets, and then transmits them to the data analysis module; Step 3: The data analysis module is equipped with an RNN algorithm model, and the traffic data set is substituted into the RNN algorithm model for year-on-year analysis to calculate and generate year-on-year differences. , reflecting the long-term trend of the use of electronic paper coupons at toll booths; Step 4: The data analysis module substitutes the traffic data set into the RNN algorithm model for month-on-month analysis and calculates the month-on-month difference , reflecting the short-term trend of the use of electronic paper coupons at toll booths; Step 5: The data analysis module calculates the average annual traffic flow at the toll station based on the traffic data set, and obtains the average annual usage of electronic paper coupons at the toll station; Step 6: RNN algorithm model based on year-on-year differences , month-on-month difference , policy data set and toll station annual average traffic flow calculation to generate a forecast data set , get the number of electronic paper coupons predicted to be issued in the next month of the current year.

[0006] Preferably, in step one, the traffic flow information includes the traffic flow and types of passing vehicles at all time periods at the toll station, and the holiday policy information includes the holiday date and free passage duration.

[0007] Preferably, in the step 2, data cleaning includes deduplication processing, missing data processing and standardization processing. The data acquisition module uses a time series scatter plot to remove duplicate data in traffic flow information and holiday policy information, and then fills in missing values ​​through the mean interpolation method, and uses an SQL database to standardize the data format.

[0008] Preferably, in step 2, the traffic data set includes the traffic volume and types of passing vehicles at all time periods of the toll station, and the expression of the traffic data set is: , to Corresponding to the daily traffic volume and the types of vehicles passing through, Indicates the time period during which vehicles pass through. Indicates the type of vehicles passing through during the time period, with numbers from 1 to Indicates the toll booth from day 1 to day 2 The traffic volume of the day.

[0009] Preferably, in step 2, the policy data set includes holiday dates and free passage duration, and the expression of the policy data set is: , to Corresponding to each holiday date and free passage duration, Indicates holiday dates. Indicates the duration of free passage, with numbers from 1 to Indicates the 1st to Holidays.

[0010] Preferably, in step 3, the year-on-year difference The calculation formula is as follows: According to the traffic data set, the traffic volume and types of vehicles passing through all months of the current year are counted and recorded as ,in, , , numbers 1 to 12 represent the 1st to 12th months of a year; According to the traffic data set, the traffic volume and types of vehicles in all months of the year before the current year are counted and recorded as ,in, , , numbers 1 to 12 represent the 1st to 12th months of a year;

[0011] In the formula, Indicates the year-on-year difference, Indicates the current year Traffic volume and types of vehicles in each month, Indicates the year before the current year. Traffic volume and types of vehicles in each month, It indicates that the long-term trend of the use of electronic paper coupons at toll stations at the same time point in different years was obtained through the RNN algorithm model analysis.

[0012] Preferably, in step 4, the month-on-month difference The calculation formula is as follows:

[0013] In the formula, Indicates the month-on-month difference, Indicates the current year Traffic volume and types of vehicles in each month, It indicates the short-term trend of the use of electronic paper coupons at toll stations at different time points in the same year obtained through the RNN algorithm model analysis.

[0014] Preferably, in step 5, the average annual traffic flow of the toll station includes the current year Average traffic volume per month and the average annual traffic volume for the 12 months before the current year , and its calculation formula is as follows:

[0015]

[0016] In the formula, Indicates the current year The average of the traffic volume and types of vehicles in the month is the average annual usage of electronic paper coupons at toll stations in the current year. It represents the average of the traffic volume and types of vehicles passing through in the 12 months of the year before the current year, that is, the average annual usage of electronic paper coupons at toll stations in the previous year.

[0017] Preferably, in step six, the prediction data set The calculation formula is as follows:

[0018] In the formula, represents the prediction data set, Indicates the current year holidays, Indicates the weight of the forecast data group calculation formula for year-on-year differences. Indicates the weight of the forecast data group calculation formula for the month-on-month difference. It indicates the weight of the free pass duration on holidays in the calculation formula of the prediction data group. , and are three constant terms with unequal values, and , if the current year is If there are no holidays in a month, , Indicates that the RNN algorithm model follows Weight and The weights are used to obtain the weighted average of the usage trend of electronic paper coupons at toll booths. The weighted average value representing the trend of electronic paper coupon usage plus the mean of traffic volume and vehicle type in the previous year is obtained to obtain the current year’s The number of electronic paper coupons to be issued in the current month is predicted. If there are holidays in a month, , Indicates that the RNN algorithm model follows Weight, Weight and The weights are used to obtain the weighted average of the usage trend of electronic paper coupons at toll booths. The weighted average value representing the usage trend of electronic paper coupons plus the mean of the traffic volume and vehicle types in the current year is obtained. The number of e-paper coupons predicted to be issued in each month.

[0019] An electronic paper voucher issuance prediction system based on big data, comprising a data acquisition module and a data analysis module; The data acquisition module is connected to the electronic paper voucher big data platform through the network. The data acquisition module is provided with a KnowLM algorithm model, and the KnowLM algorithm model is used to obtain traffic flow information and holiday policy information. The data acquisition module performs data cleaning on the acquired information, and classifies and forms a traffic data set and a policy data set, and then transmits them to the data analysis module; The data analysis module is provided with an RNN algorithm model, which performs year-on-year analysis and month-on-month analysis based on the traffic data set, and calculates and generates year-on-year differences. and month-on-month difference The data analysis module calculates the average annual usage of electronic paper coupons at toll booths in the current year based on the traffic data set. The average annual usage of electronic paper coupons at toll booths in the previous year , the RNN algorithm model is based on the year-on-year difference , month-on-month difference , the average annual number of electronic paper coupons used at toll booths in the current year , the average annual number of electronic paper coupons used at toll booths in the previous year and policy data sets, and calculates the prediction data set , get the number of electronic paper coupons predicted to be issued in the next month of the current year.

[0020] Compared with the prior art, the present invention provides a method and system for predicting the issuance of electronic paper coupons based on big data, which has the following beneficial effects: 1. The present invention is connected to the electronic paper voucher big data platform through a data acquisition module network. The data acquisition module is provided with a KnowLM algorithm model. The KnowLM algorithm model is used to obtain traffic flow information and holiday policy information. The data acquisition module performs data cleaning on the acquired information, and classifies and forms a traffic data set and a policy data set, and then transmits them to the data analysis module. The data analysis module is provided with an RNN algorithm model. The RNN algorithm model performs year-on-year analysis and month-on-month analysis according to the traffic data set, and calculates and generates year-on-year differences. and month-on-month difference , using a unified 12-month annual timeline for comparative analysis, reveals the usage trends and periodicity of e-paper coupons, and big data analysis has higher prediction accuracy.

[0021] 2. The present invention calculates the average annual usage of electronic paper coupons at toll stations in the current year based on the traffic data set through the data analysis module The average annual usage of electronic paper coupons at toll booths in the previous year , the RNN algorithm model is based on the year-on-year difference , month-on-month difference , the average annual number of electronic paper coupons used at toll booths in the current year , the average annual number of electronic paper coupons used at toll booths in the previous year and policy data sets, and calculates the prediction data set , get the predicted number of electronic paper coupons to be issued in the next month of the current year, flexibly predict the number of issuances according to holidays, avoid insufficient or excessive issuance of paper coupons due to inaccurate estimates, and the algorithm has strong fusion and generalization capabilities. BRIEF DESCRIPTION OF THE DRAWINGS

[0022] Figure 1 It is a step diagram of the method of the present invention. DETAILED DESCRIPTION

[0023] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0024] Since the traffic flow data collected by the traditional electronic paper voucher issuance prediction method based on big data contains a lot of abnormal data and lacks effective processing methods, it directly affects the accuracy of the prediction results. Traditional prediction models usually use a single time series model. When dealing with complex and changeable traffic flow data, they often show the problem of insufficient generalization ability, especially during holidays and peak periods, when the traffic volume changes dramatically. It is difficult for a single model to accurately capture these dynamic characteristics. Therefore, a method and system for predicting the issuance of electronic paper vouchers based on big data are proposed. Please refer to Figure 1 The electronic paper voucher issuance prediction method based on big data includes the following steps: Step 1: The data acquisition module uses the KnowLM algorithm model to capture information from the electronic paper voucher big data platform to obtain traffic flow information and holiday policy information. The core functions of KnowLM include named entity recognition (NER), relationship extraction (RE), and event extraction (EE). Specifically, NER is used to identify vehicle entities in text, RE is used to identify the relationship between traffic flow and time series, and EE is used to identify holiday policies and related factors that affect traffic flow. Traffic flow information includes the traffic flow and types of vehicles passing through the toll station at all times. Traffic flow during peak hours directly affects the number of electronic paper vouchers issued. Different types of vehicles have different demands for paper vouchers. Small passenger cars usually enjoy free passage policies, while large vehicles that have not applied for ETC may need paper vouchers to pass. Holiday policy information includes holiday dates and free passage duration. The specific date and duration of holidays will affect the prediction of the number of paper vouchers issued. The start and end dates of holidays, especially the time limit for toll-free, are key factors in predicting the demand for paper vouchers. During holidays, traffic flow usually increases, resulting in more paper vouchers that need to be issued and recycled. Step 2: The data collection module performs data cleaning on the acquired information. Data cleaning includes deduplication, missing data and standardization. The data collection module uses a time series scatter plot to remove duplicate data from traffic flow information and holiday policy information, checks the continuity and consistency of time, and then fills in missing values ​​through the mean interpolation method. The SQL database is used to standardize the data format. Traffic flow data is uniformly rounded and measured, and holiday policy data is uniformly measured by days. The data collection module classifies the cleaned traffic flow information and holiday policy information into traffic data sets and policy data sets, and then transmits them to the data analysis module. The traffic data set includes the traffic volume and types of vehicles passing through the toll station at all times. The expression of the traffic data set is: , to Corresponding to the daily traffic volume and the types of vehicles passing through, Indicates the time period when the vehicle passes, specifically a certain year, month, and day. Indicates the types of vehicles passing through during a period of time, specifically the total number of passenger cars passing through on a single day, the total number of trucks passing through on a single day, and the total number of special operation vehicles passing through on a single day, with numbers from 1 to Indicates the toll booth from day 1 to day 2 The daily traffic volume; The policy data set includes holiday dates and free passage duration. The expression of the policy data set is: , to Corresponding to each holiday date and free passage duration, Indicates the holiday date, specifically from a certain month and year to a certain month and year. Indicates the duration of free passage, specifically the number of free passage days, numbered from 1 to Indicates the 1st to holidays; Step 3: The data analysis module is equipped with an RNN algorithm model, and the traffic data set is substituted into the RNN algorithm model for year-on-year analysis to calculate and generate year-on-year differences. , reflecting the long-term trend of the use of electronic paper coupons at toll stations, and its calculation formula is as follows: According to the traffic data set, the traffic volume and types of vehicles passing through all months of the current year are counted and recorded as ,in, , , the numbers 1 to 12 represent the first to twelfth months of a year. The subsequent comparative analysis will be conducted using the 12-month year as the time axis, which reveals the usage trend and periodicity of e-vouchers and effectively improves the accuracy of e-voucher issuance forecasts; According to the traffic data set, the traffic volume and types of vehicles in all months of the year before the current year are counted and recorded as ,in, , , numbers 1 to 12 represent the 1st to 12th months of a year;

[0025] In the formula, Indicates the year-on-year difference, Indicates the current year Traffic volume and types of vehicles in each month, Indicates the year before the current year. Traffic volume and types of vehicles in each month, It indicates that the long-term trend of the use of electronic paper coupons at toll stations at the same time point in different years is obtained through the RNN algorithm model analysis; Step 4: The data analysis module substitutes the traffic data set into the RNN algorithm model for month-on-month analysis and calculates the month-on-month difference , reflecting the short-term trend of the use of electronic paper coupons at toll stations, and its calculation formula is as follows:

[0026] In the formula, Indicates the month-on-month difference, Indicates the current year Traffic volume and types of vehicles in each month, It means that the short-term trend of the use of electronic paper coupons at toll stations at different time points in the same year can be obtained through RNN algorithm model analysis, and the prediction accuracy of big data analysis is higher; Step 5: The data analysis module calculates the average annual traffic flow at the toll station based on the traffic data set, and obtains the average annual usage of electronic paper coupons at the toll station. The average annual traffic flow at the toll station includes the current year Average traffic volume per month and the average annual traffic volume for the 12 months before the current year , and its calculation formula is as follows:

[0027]

[0028] In the formula, Indicates the current year The average of the traffic volume and types of vehicles in the month is the average annual usage of electronic paper coupons at toll stations in the current year. It represents the average of the traffic volume and types of vehicles in the 12 months before the current year, that is, the average annual usage of electronic paper coupons at toll stations in the previous year, so as to facilitate the subsequent prediction of more accurate issuance quantity based on the average annual usage of electronic paper coupons at toll stations; Step 6: RNN algorithm model based on year-on-year differences , month-on-month difference , policy data set and toll station annual average traffic flow calculation to generate a forecast data set , get the number of electronic paper coupons predicted to be issued in the next month of the current year, the calculation formula is as follows:

[0029] In the formula, represents the prediction data set, Indicates the current year holidays, Indicates the weight of the forecast data group calculation formula for year-on-year differences. Indicates the weight of the forecast data group calculation formula for the month-on-month difference. It indicates the weight of the free pass duration on holidays in the calculation formula of the prediction data group. , and are three constant terms with unequal values, and , if the current year is If there are no holidays in a month, , Indicates that the RNN algorithm model follows Weight and The weights are used to obtain the weighted average of the usage trend of electronic paper coupons at toll booths. The weighted average value representing the trend of electronic paper coupon usage plus the mean of traffic volume and vehicle type in the previous year is obtained to obtain the current year’s The number of electronic paper coupons to be issued in the current month is predicted. If there are holidays in a month, , Indicates that the RNN algorithm model follows Weight, Weight and The weights are used to obtain the weighted average of the usage trend of electronic paper coupons at toll booths. The weighted average value representing the usage trend of electronic paper coupons plus the mean of the traffic volume and vehicle types in the current year is obtained. The number of electronic paper coupons to be issued in each month is predicted, and the number of issuance is flexibly predicted based on holidays to avoid insufficient or excessive issuance of paper coupons due to inaccurate estimates. The algorithm has strong fusion and generalization capabilities.

[0030] The above-mentioned electronic paper voucher issuance prediction method based on big data is implemented by relying on an electronic paper voucher issuance prediction system based on big data, which includes a data acquisition module and a data analysis module; The data collection module is connected to the electronic paper coupon big data platform through the network. The data collection module is provided with a KnowLM algorithm model, which is used to obtain traffic flow information and holiday policy information. The data collection module performs data cleaning on the acquired information, and classifies and forms traffic data sets and policy data sets, and then transmits them to the data analysis module; The data analysis module is equipped with an RNN algorithm model, which performs year-on-year and month-on-month analysis based on the traffic data set and calculates the year-on-year difference. and month-on-month difference , using the 12-month year as the time axis for comparative analysis, revealing the usage trend and periodicity of electronic paper coupons. Big data analysis has higher prediction accuracy. The data analysis module calculates the average annual usage of electronic paper coupons at toll stations in the current year based on the traffic data set. The average annual usage of electronic paper coupons at toll booths in the previous year , the RNN algorithm model is based on the year-on-year difference , month-on-month difference , the average annual number of electronic paper coupons used at toll booths in the current year , the average annual number of electronic paper coupons used at toll booths in the previous year and policy data sets, and calculates the prediction data set , get the predicted number of electronic paper coupons to be issued in the next month of the current year, flexibly predict the number of issuances according to holidays, avoid insufficient or excessive issuance of paper coupons due to inaccurate estimates, and the algorithm has strong fusion and generalization capabilities.

[0031] Although embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that various changes, modifications, substitutions and variations may be made to the embodiments without departing from the principles and spirit of the present invention, and that the scope of the present invention is defined by the appended claims and their equivalents.

Claims

1. A method for predicting the issuance of electronic paper coupons based on big data, characterized in that: The following steps are involved: Step 1: The data collection module uses the KnowLM algorithm model to capture information from the electronic paper coupon big data platform to obtain traffic flow information and holiday policy information; Step 2: The data collection module cleans the acquired information, classifies it into traffic data sets and policy data sets, and then transmits them to the data analysis module; Step 3: The data analysis module is equipped with an RNN algorithm model, and the traffic data set is substituted into the RNN algorithm model for year-on-year analysis to calculate and generate year-on-year differences. , reflecting the long-term trend of the use of electronic paper coupons at toll booths; Step 4: The data analysis module substitutes the traffic data set into the RNN algorithm model for month-on-month analysis and calculates the month-on-month difference , reflecting the short-term trend of the use of electronic paper coupons at toll booths; Step 5: The data analysis module calculates the average annual traffic flow at the toll station based on the traffic data set, and obtains the average annual usage of electronic paper coupons at the toll station; Step 6: RNN algorithm model based on year-on-year differences , month-on-month difference , policy data set and toll station annual average traffic flow calculation to generate a forecast data set , get the number of electronic paper coupons predicted to be issued in the next month of the current year.

2. The electronic paper coupon issuance prediction method based on big data according to claim 1, characterized in that: In the step 1, the traffic flow information includes the traffic flow and types of passing vehicles at all time periods of the toll station, and the holiday policy information includes the holiday date and free passage duration.

3. The electronic paper coupon issuance prediction method based on big data according to claim 2 is characterized by: In the step 2, data cleaning includes deduplication, missing data and standardization. The data collection module uses a time series scatter plot to remove duplicate data in traffic flow information and holiday policy information, then fills in missing values ​​through the mean interpolation method, and uses an SQL database to standardize the data format.

4. The electronic paper coupon issuance prediction method based on big data according to claim 3 is characterized by: In step 2, the traffic data set includes the traffic volume and types of vehicles passing through the toll station at all times. The expression of the traffic data set is: , to Corresponding to the daily traffic volume and the types of vehicles passing through, Indicates the time period during which vehicles pass through. Indicates the type of vehicles passing through during the time period, with numbers from 1 to Indicates the toll booth from day 1 to day 2 The traffic volume of the day.

5. The method for predicting the issuance of electronic paper coupons based on big data according to claim 4, characterized in that: In step 2, the policy data set includes holiday dates and free passage duration. The expression of the policy data set is: , to Corresponding to each holiday date and free passage duration, Indicates holiday dates. Indicates the duration of free passage, with numbers from 1 to Indicates the 1st to Holidays.

6. The electronic paper voucher issuance prediction method based on big data according to claim 5, characterized in that: In step 3, the year-on-year difference The calculation formula is as follows: According to the traffic data set, the traffic volume and types of vehicles passing through all months of the current year are counted and recorded as ,in, , , numbers 1 to 12 represent the 1st to 12th months of a year; According to the traffic data set, the traffic volume and types of vehicles in all months of the year before the current year are counted and recorded as ,in, , , numbers 1 to 12 represent the 1st to 12th months of a year; In the formula, Indicates the year-on-year difference, Indicates the current year Traffic volume and types of vehicles in each month, Indicates the year before the current year. Traffic volume and types of vehicles in each month, It indicates that the long-term trend of the use of electronic paper coupons at toll stations at the same time point in different years was obtained through the RNN algorithm model analysis.

7. The method for predicting the issuance of electronic paper coupons based on big data according to claim 6, characterized in that: In step 4, the month-on-month difference The calculation formula is as follows: In the formula, Indicates the month-on-month difference, Indicates the current year Traffic volume and types of vehicles in each month, It indicates the short-term trend of the use of electronic paper coupons at toll stations at different time points in the same year obtained through the RNN algorithm model analysis.

8. The electronic paper coupon issuance prediction method based on big data according to claim 7, characterized in that: In step 5, the average annual traffic flow of the toll station includes the current year Average traffic volume per month and the average annual traffic volume for the 12 months before the current year , and its calculation formula is as follows: In the formula, Indicates the current year The average of the traffic volume and types of vehicles in the month is the average annual usage of electronic paper coupons at toll stations in the current year. It represents the average of the traffic volume and types of vehicles passing through in the 12 months of the year before the current year, that is, the average annual usage of electronic paper coupons at toll stations in the previous year.

9. The electronic paper coupon issuance prediction method based on big data according to claim 8, characterized in that: In step six, the prediction data set The calculation formula is as follows: In the formula, represents the prediction data set, Indicates the current year holidays, Indicates the weight of the forecast data group calculation formula for year-on-year differences. Indicates the weight of the forecast data group calculation formula for the month-on-month difference. It indicates the weight of the free pass duration on holidays in the calculation formula of the prediction data group. , and are three constant terms with unequal values, and , if the current year is If there are no holidays in a month, , Indicates that the RNN algorithm model follows Weight and The weights are used to obtain the weighted average of the usage trend of electronic paper coupons at toll booths. The weighted average value representing the trend of electronic paper coupon usage plus the mean of traffic volume and vehicle type in the previous year is obtained to obtain the current year’s The number of electronic paper coupons to be issued in the current month is predicted. If there are holidays in a month, , Indicates that the RNN algorithm model follows Weight, Weight and The weights are used to obtain the weighted average of the usage trend of electronic paper coupons at toll booths. The weighted average value representing the usage trend of electronic paper coupons plus the mean of the traffic volume and vehicle types in the current year is obtained. The number of e-paper coupons predicted to be issued in each month.

10. An electronic paper voucher issuance prediction system based on big data, used to implement the electronic paper voucher issuance prediction method based on big data described in claims 1-9, characterized in that: Including data acquisition module and data analysis module; The data acquisition module is connected to the electronic paper voucher big data platform through the network. The data acquisition module is provided with a KnowLM algorithm model, and the KnowLM algorithm model is used to obtain traffic flow information and holiday policy information. The data acquisition module performs data cleaning on the acquired information, and classifies and forms a traffic data set and a policy data set, and then transmits them to the data analysis module; The data analysis module is provided with an RNN algorithm model, which performs year-on-year analysis and month-on-month analysis based on the traffic data set, and calculates and generates year-on-year differences. and month-on-month difference The data analysis module calculates the average annual usage of electronic paper coupons at toll booths in the current year based on the traffic data set. The average annual usage of electronic paper coupons at toll booths in the previous year , the RNN algorithm model is based on the year-on-year difference , month-on-month difference , the average annual number of electronic paper coupons used at toll booths in the current year , the average annual number of electronic paper coupons used at toll booths in the previous year and policy data sets, and calculates the prediction data set , get the number of electronic paper coupons predicted to be issued in the next month of the current year.

Images