AI-based network mall recharge method

Abstract

The application discloses an AI-based network mall recharging method, which performs inertia analysis on the consumption records in the purchase inertia data, first divides according to the purchase time in the consumption records to obtain consumption months, then obtains the single-month fee value Fi and the single-month consumption value Xi corresponding to each consumption month Yi according to the fee price and the consumption target, analyzes to obtain the inertia mall and the non-inertia mall, and obtains the target fee value of the inertia mall; then, the recharging first-determination analysis is performed on the inertia mall, the smart user is reminded to recharge the target fee value or the amount of half of the target fee value according to the frequency deviation value of the user balance and the consumption time; the application can provide a mall recharging auxiliary system, help the user to determine whether it is appropriate to recharge currently, and can also timely monitor whether there is an automatic renewal target, so as to avoid being charged without attention; the application is simple, effective and easy to use.

Description

Technical Field

[0001] This invention relates to the field of online shopping mall recharge technology, specifically to an AI-based online shopping mall recharge method. Background Technology

[0002] Patent CN107895304A discloses an online shopping mall management system, including a product management subsystem, a management and delivery subsystem, an online-offline subsystem, a membership management subsystem, an order management subsystem, a marketing management subsystem, a financial management subsystem, a data management subsystem, a store decoration subsystem, a mall distribution subsystem, a task center subsystem, a multi-merchant management subsystem, and a POS management subsystem. The beneficial effect of this invention is that by integrating online and offline channels, it achieves the interconnection of data such as product information, membership information, and transaction information, thereby meeting the diverse shopping experience needs of consumers.

[0003] This patent provides a management system for an online shopping mall that can provide convenience to users both online and offline. However, currently, for smart mobile devices such as mobile phones or tablets, online shopping malls are not only for shopping but also for some virtual items, such as game malls or music app malls. For some users, there is a lack of a recharge assistance system to determine whether it is worthwhile to recharge. This is also the case for recharge activities on online shopping platforms. Based on this, this application provides a solution. Summary of the Invention

[0004] The purpose of this invention is to provide an AI-based online shopping mall recharge method;

[0005] The objective of this invention can be achieved through the following technical solutions:

[0006] An AI-based online shopping mall top-up method, which specifically includes the following steps:

[0007] Step 1: Perform inertial analysis on the consumption records within the purchase inertia data. Based on the purchase time in the consumption records, first divide the data by month to obtain the consumption month; then, based on the cost price and the consumption item, obtain the consumption month Y. i The corresponding monthly fee value F i And monthly consumption X i , i=1...15; and based on the monthly fee value F i And monthly consumption X i The analysis yields the inertial shopping mall and the non-inertial shopping mall, and the target fee value of the inertial shopping mall is obtained.

[0008] Step 2: Conduct a recharge decision analysis for the inertial shopping mall, and determine the target recharge amount or half of the target amount based on the frequency deviation of the user's balance and consumption time;

[0009] Step 3: Perform a recharge count analysis for non-inertial shopping malls, and determine the recharge reminder method based on the results of the recharge count analysis.

[0010] Furthermore, before proceeding to step one, the following steps are also required:

[0011] All user purchase inertia data are obtained, including consumption records and stable records. Consumption records include the purchase item, the corresponding consumption time, and the price. Stable records include stable items and their costs.

[0012] Furthermore, the object of consumption is the corresponding consumer goods, the consumption time is the purchase time of the corresponding object of consumption, and the cost price is the price of the corresponding object of consumption.

[0013] Stable assets are long-term purchases that are automatically charged monthly.

[0014] Furthermore, the specific method for inertial analysis in step one is as follows:

[0015] S1: Obtain all consumption records, accumulate them according to the purchase time, divide all consumption records by month, and include several consumption records in each month;

[0016] S2: Automatically retrieves the consumption records for the past 15 months, omitting or deleting other data; then marks each single month as the consumption month Y. i Given i=1...15, obtain the total number of all consumption items for each month and mark it as the monthly consumption value X. i , i=1...15;

[0017] S3: Obtain the cost price of all consumption items for each month, add them together, and mark the resulting value as the monthly cost value F. i , i=1...15;

[0018] S4: Then obtain the monthly consumption value X. i Partial mean analysis was performed on it to obtain the partial mean;

[0019] S5: Then obtain all monthly fee values ​​F i Perform the same partial mean analysis on it in step S4, and the resulting partial mean is labeled as partial mean two.

[0020] S6: Then, the partial mean and the second partial mean are arbitrarily compared with T1. When either of the two values ​​does not exceed T1, a steady-state signal is generated.

[0021] S7: Mark the corresponding online store as an inertial store; otherwise, mark it as a non-inertial store.

[0022] S8: When it is an inertial shopping mall, the monthly fee value F is automatically obtained. i The system automatically calculates the mean and labels it as U; then it calculates the monthly fee F according to the formula. i The deviation value W is calculated using the following formula:

[0023] ;

[0024] In the formula, |*| represents taking the absolute value of the value within the parentheses;

[0025] When the value of W does not exceed T2, F i The mean value is marked as the target cost value;

[0026] When the W value exceeds T2, it will automatically follow the... Select the corresponding F in descending order. i Value; each selected F i When the value is found to be less than T2, it is automatically deleted, and the value of W is recalculated until the value of W does not exceed T2. Then the remaining F is recalculated. i The average value is marked as the target cost value; T2 is the value preset by the management.

[0027] Furthermore, the specific method of partial mean analysis in step S4 is as follows:

[0028] Automatically obtain X i The mean of is denoted as P;

[0029] Then use the obtained X i The number of values ​​greater than P is recorded as the mean value; then X is obtained. i The specific number of values ​​smaller than P is marked as the mean value;

[0030] Then, the absolute value of the difference between the upper and lower mean values ​​is obtained, and this absolute value is marked as the partial mean.

[0031] Furthermore, the specific method for the initial recharge determination analysis in step two is as follows:

[0032] S01: Obtain the inertial shopping mall and its corresponding target fee value;

[0033] S02: Obtain all of the user's consumption records for the most recent month, and obtain the consumption time corresponding to the consumption item in each consumption record;

[0034] S03: Arrange all consumption times in chronological order to obtain the consumption time sequence L. j j=1...n;

[0035] S04: Then use formula Q i =L i+1 -L i i=1...n-1; obtain n-1 time intervals; then automatically obtain Q. i The mean value is calculated in the same way as in step S8 to obtain the deviation value W. The deviation value at this time is remarked as the frequency deviation value. When the frequency deviation value is lower than T3 and n is greater than 10, a high charge signal is automatically generated; otherwise, a low charge signal is generated. T3 is a preset value.

[0036] S05: When a high recharge signal is generated, if the user's balance in the online store is lower than T4, the target fee value will be automatically obtained, and the user will be given the following option: "Current balance is insufficient. Based on your past spending habits, we suggest recharging + target fee value + relevant amount" to help the user recharge quickly; T4 here is a preset value.

[0037] When a low charge signal is generated, if the user's balance in the online store is lower than T4, the target fee value will be automatically obtained, and the user will be given the following option: "Current balance is insufficient. Based on your past spending habits, we suggest you recharge + half of the target fee value + relevant amount" to help the user recharge quickly; T4 here is a preset value.

[0038] Furthermore, the specific method for determining the number of recharges in step three is as follows:

[0039] SS1: Obtain the non-inertial mall and then monitor all recharge activities of the non-inertial mall;

[0040] SS2: When any recharge event occurs, the recharge record will be automatically retrieved, and the recharge amount and the amount received will be obtained. The recharge amount and the amount received specifically refer to the amount that can be received when a certain amount is recharged in this mall.

[0041] SS3: Then subtract the recharge amount from the amount received, divide the difference by the recharge amount to get the discount rate, and mark it as the current discount rate;

[0042] SS4: After obtaining the current recharge event, recharge events that occurred in the previous three months will be marked as the aforementioned event, and the discount ratio of each of the aforementioned events will be obtained accordingly. All discount ratios will be marked as the previous discount ratio group.

[0043] SS5: When the discount percentage is greater than the values ​​in all previous discount percentage groups, an initial signal is generated;

[0044] SS6: Then obtain the discount rates given by all the previous discount rate groups corresponding to the activities within the same time period of the previous year, and mark them as reference rates. Here, the same time period refers to the activities within one week before and after the time when the corresponding discount rate group was generated. If there is no such activity, delete the corresponding discount rate and do not record it.

[0045] Divide the reference ratio by the corresponding previous discount ratio to obtain several improvement ratios; find the corresponding improvement ratio with the smallest improvement ratio and mark it as the potential improvement ratio.

[0046] SS7: Then, obtain the corresponding time of the previous year in the current month, and automatically obtain the discount ratio of all subsequent activities. Mark them as potential discount ratios, multiply all potential discount ratios by potential increase ratios, and mark the resulting value as pre-discount ratios. All pre-discount ratios constitute a pre-discount ratio group.

[0047] SS8: Compares the current discount rate with the pre-discount rate. If all values ​​in the pre-discount rate group are smaller than the current discount rate, a suggested recharge signal is generated. At the same time, if no more than two values ​​in the pre-discount rate group are larger than the current discount rate, a reminder recharge signal is generated.

[0048] SS9: When a suggested recharge signal is generated, it automatically reminds users that "the current discount rate is at its best, and it is recommended to recharge", and recommends the minimum recharge amount based on the promotion.

[0049] When a recharge reminder signal is generated, the system will automatically remind the user that "the current discount rate may be at its best, and the possibility of exceeding the discount rate in the future is low. We suggest you consider recharging."

[0050] Furthermore, after completing step three, the following steps are also required:

[0051] Once all stable records are obtained, the corresponding stable assets and their fees are retrieved. Before each renewal is required, the system automatically reminds the user whether they need to block the renewal.

[0052] The beneficial effects of this invention are:

[0053] This invention performs inertial analysis on consumption records within purchase inertia data. Based on the purchase time within the consumption records, it first divides the data by month to obtain consumption months; then, based on the cost price and the consumption item, it obtains the consumption month Y. i The corresponding monthly fee value F i And monthly consumption X i , i=1...15; and based on the monthly fee value F i And monthly consumption X i The analysis yields the inertial shopping mall and the non-inertial shopping mall, and the target fee value of the inertial shopping mall is obtained.

[0054] Subsequently, a recharge first-time determination analysis is performed for inertial shopping malls. Based on the frequency deviation of the user's balance and consumption time, an intelligent reminder is given to the user to recharge the target amount or half of the target amount. At the same time, a recharge second-time determination analysis can be performed for non-inertial shopping malls. Based on the results of the second-time determination analysis, it is determined whether the recharge reminder method is for non-inertial shopping malls. This invention can provide a shopping mall recharge assistance system to help users determine whether it is appropriate to recharge at this time. At the same time, it can also monitor whether there are any automatic renewal items in a timely manner to avoid being charged without noticing. This invention is simple, effective, and easy to use. Detailed Implementation

[0055] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0056] This invention relates to an AI-based online shopping mall recharge method, which specifically includes the following steps:

[0057] Step 1: Obtain all user purchase habit data, which includes consumption records and stable records. Consumption records include the purchase item, the corresponding consumption time, and the price. The purchase item is the corresponding product, the consumption time is the purchase time, and the price is the price. Stable records include stable items and their costs. Stable items are the corresponding long-term purchases and items that are automatically charged monthly.

[0058] Step Two: Next, perform inertial analysis on the consumption records within the purchase inertia data. The specific method of inertial analysis is as follows:

[0059] S1: Obtain all consumption records, accumulate them according to the purchase time, divide all consumption records by month, and include several consumption records in each month;

[0060] S2: Automatically retrieves the consumption records for the past 15 months, omitting or deleting other data; then marks each single month as the consumption month Y. i Given i=1...15, obtain the total number of all consumption items for each month and mark it as the monthly consumption value X. i , i=1...15;

[0061] S3: Obtain the cost price of all consumption items for each month, add them together, and mark the resulting value as the monthly cost value F. i, i=1...15;

[0062] S4: Then obtain the monthly consumption value X. i Partial mean analysis was performed on it, specifically:

[0063] Automatically obtain X i The mean of is denoted as P;

[0064] Then use the obtained X i The number of values ​​greater than P is recorded as the mean value; then X is obtained. i The specific number of values ​​smaller than P is marked as the mean value;

[0065] Then, obtain the absolute value of the difference between the upper and lower mean values, and mark this absolute value as the partial mean.

[0066] S5: Then obtain all monthly fee values ​​F i Perform the same partial mean analysis on it in step S4, and the resulting partial mean is labeled as partial mean two.

[0067] S6: Then, the partial mean and the second partial mean are arbitrarily compared with T1. When either of the two values ​​does not exceed T1, a steady-state signal is generated. Here, the value of T1 is preset by the management personnel, usually taking the value of 2 or 3. The specific value is adjusted by the management personnel to ensure the accuracy of the final determination of the data by the partial mean or the second partial mean.

[0068] S7: Mark the corresponding online store as an inertial store; otherwise, mark it as a non-inertial store.

[0069] S8: When it is an inertial shopping mall, the monthly fee value F is automatically obtained. i The system automatically calculates the mean and labels it as U; then it calculates the monthly fee F according to the formula. i The deviation value W is calculated using the following formula:

[0070] ;

[0071] In the formula, |*| represents taking the absolute value of the value within the parentheses;

[0072] When the value of W does not exceed T2, F i The mean value is marked as the target cost value;

[0073] When the W value exceeds T2, it will automatically follow the... Select the corresponding F in descending order. i Value; each selected F i When the value is found to be less than T2, it is automatically deleted, and the value of W is recalculated until the value of W does not exceed T2. Then the remaining F is recalculated. iThe mean of the values ​​is marked as the target cost value; T2 is a value preset by the management personnel. The specific value can be determined by calculating the W value after a set of data that meets the management personnel's conditions. The specific details are not disclosed here.

[0074] Step 3: Perform initial recharge determination analysis for the inertial shopping mall. The specific initial recharge determination analysis method is as follows:

[0075] S01: Obtain the inertial shopping mall and its corresponding target fee value;

[0076] S02: Obtain all of the user's consumption records for the most recent month, and obtain the consumption time corresponding to the consumption item in each consumption record;

[0077] S03: Arrange all consumption times in chronological order to obtain the consumption time sequence L. j j=1...n;

[0078] S04: Then use formula Q i =L i+1 -L i i=1...n-1; obtain n-1 time intervals; then automatically obtain Q. i The mean value is calculated in the same way as in step S8 to obtain the deviation value W. The deviation value at this time is remarked as the frequency deviation value. When the frequency deviation value is lower than T3 and n is greater than 10, a high charge signal is automatically generated; otherwise, a low charge signal is generated. T3 is a preset value.

[0079] S05: When a high recharge signal is generated, if the user's balance in the online store is lower than T4, the target fee value will be automatically obtained, and the user will be given the following option: "Current balance is insufficient. Based on your past spending habits, we suggest recharging + target fee value + relevant amount" to help the user recharge quickly; T4 here is a preset value.

[0080] When a low recharge signal is generated, if the user's balance in the online store is lower than T4, the target fee value will be automatically obtained, and the user will be given the following option: "Current balance is insufficient. Based on your past spending habits, we suggest you recharge + half of the target fee value + relevant amount" to help the user recharge quickly; T4 here is a preset value.

[0081] Step 4: Conduct a recharge count analysis for non-inertial shopping malls. The specific method for the recharge count analysis is as follows:

[0082] SS1: Obtain the non-inertial mall and then monitor all recharge activities of the non-inertial mall;

[0083] SS2: When any recharge event occurs, the recharge record will be automatically retrieved, and the recharge amount and the amount received will be obtained. The recharge amount and the amount received specifically refer to the amount that can be received when a certain amount is recharged in this mall.

[0084] SS3: Then subtract the recharge amount from the amount received, divide the difference by the recharge amount to get the discount rate, and mark it as the current discount rate;

[0085] SS4: After obtaining the current recharge event, recharge events that occurred in the previous three months will be marked as the aforementioned event, and the discount ratio of each of the aforementioned events will be obtained accordingly. All discount ratios will be marked as the previous discount ratio group.

[0086] SS5: When the discount percentage is greater than the values ​​in all previous discount percentage groups, an initial signal is generated;

[0087] SS6: Then obtain the discount rates given by all the previous discount rate groups corresponding to the activities within the same time period of the previous year, and mark them as reference rates. Here, the same time period refers to the activities within one week before and after the time when the corresponding discount rate group was generated. If there is no such activity, delete the corresponding discount rate and do not record it.

[0088] Divide the reference ratio by the corresponding previous discount ratio to obtain several improvement ratios; find the corresponding improvement ratio with the smallest improvement ratio and mark it as the potential improvement ratio.

[0089] SS7: Then, obtain the corresponding time of the previous year in the current month, and automatically obtain the discount ratio of all subsequent activities. Mark them as potential discount ratios, multiply all potential discount ratios by potential increase ratios, and mark the resulting value as pre-discount ratios. All pre-discount ratios constitute a pre-discount ratio group.

[0090] SS8: Compares the current discount rate with the pre-discount rate. If all values ​​in the pre-discount rate group are smaller than the current discount rate, a suggested recharge signal is generated. At the same time, if no more than two values ​​in the pre-discount rate group are larger than the current discount rate, a reminder recharge signal is generated.

[0091] SS9: When a suggested recharge signal is generated, it automatically reminds users that "the current discount rate is at its best, and it is recommended to recharge", and recommends the minimum recharge amount based on the promotion.

[0092] When a recharge reminder signal is generated, the system will automatically remind the user of the message "The current discount rate may be at its best, and the possibility of exceeding the discount rate in the future is low. We suggest you consider recharging."

[0093] Of course, during steps SS1-SS9, the system can also automatically detect and remind users when there are activities in the inertial shopping mall.

[0094] Step 5: Obtain all stable records, corresponding stable targets and their fees. Before each renewal, automatically remind the user whether they want to block the renewal, allowing the user to make a free choice.

[0095] The foregoing has provided a detailed description of one embodiment of the present invention, but this description is merely a preferred embodiment and should not be construed as limiting the scope of the invention. All equivalent variations and modifications made within the scope of the claims of this invention should still fall within the patent coverage of this invention.

Claims

1. AI-based network mall recharge method, characterized by, The method specifically comprises the following steps: Step one: Inertia analysis on the consumption record in the purchase inertia data, according to the purchase time in the consumption record, first divided by month, get the consumption month; then according to the expense price and the consumption target, get each consumption month Y i The corresponding single month fee value F i And single month consumption value X i , i=1...15; and according to the single month fee value F i And single month consumption value X i Inertia mall and non-inertia mall are analyzed and the target fee value of inertia mall is obtained; Step two: the first judgment analysis of the inertia city, according to the user balance and the frequency offset value of the consumption time to determine the target fee value or half of the target fee value of the smart prompt user recharge target; Step three: the second judgment analysis of the non inertia city, according to the result of the second judgment analysis to determine the recharge reminder mode of the non inertia city; The inertia analysis in step one is specifically as follows: S1: all consumption records are obtained, accumulated according to the purchase time, and all consumption records are divided by month, each month including a plurality of consumption records; S2: automatically obtain the consumption record of the past 15 months, and the rest of the data is omitted or deleted; then mark each single month as consumption month Y i , i = 1... 15, obtain the total number of all consumption targets of each month, and mark it as single-month consumption value X i , i = 1... 15; S3: Obtain the corresponding cost price of all consumption targets in each month, and add them to obtain a value marked as single-month cost value F i i = 1...15; S4: Then the monthly consumption X is obtained i The mean deviation is analyzed to obtain the mean deviation S5: After that, all the monthly fee values F are obtained i The same deviation analysis is performed in the same way as step S4, and the obtained deviation value is marked as deviation value two. S6: then the mean value and the mean value are compared with T1, and when any of the two values does not exceed T1, a steady state signal is generated; S7: the corresponding network mall is marked as an inertia city, otherwise it is marked as a non inertia city; S8: When it is the inertia mall, the single-month fee value F is automatically obtained i , the average value is automatically calculated and marked as U; then the deviation value W of the single-month fee value F is calculated according to the formula i , and the specific calculation formula is: ； In the formula, |*| represents taking the absolute value of the value in the parentheses; When the value of W does not exceed T2, F i the mean value of F is marked as the target fare value; When the W value exceeds T2, it will automatically follow the... Select the corresponding F in descending order. i Value; each selected F i When the value is found to be less than T2, it is automatically deleted, and the value of W is recalculated until the value of W does not exceed T2. Then the remaining F is recalculated. i The average value is used as the target cost value; T2 is a value preset by the management. The first judgment analysis in step two is specifically as follows: S01: the inertia city and its corresponding target fee value are obtained; S02: all consumption records of the user in the last month are obtained, and the consumption time corresponding to each consumption record is obtained; S03: arrange all the consumption times in time sequence to obtain a consumption time sequence L j , j = 1...n; S04: thereafter, the formula Q i = L i+1 - L i , i = 1...n-1; n-1 time intervals are obtained; thereafter, the mean value of Q i is automatically obtained, the deviation W is calculated in the same way as in step S8, and the deviation at this time is re-labeled as the frequency deviation value; when the frequency deviation value is lower than T3, and n is greater than 10, a high confidence signal is automatically generated, otherwise a low confidence signal is generated; T3 is a preset value; S05: when the high charge signal is generated, if the balance of the user in the network mall is lower than T4, the target fee value is automatically obtained, and the following selection is sent to the user: "current balance is insufficient, according to your past consumption habit, it is suggested to recharge + target fee value + related amount", to help the user quickly recharge; here T4 is a preset value; when the low charge signal is generated, if the balance of the user in the network mall is lower than T4, the target fee value is automatically obtained, and the following selection is sent to the user: "current balance is insufficient, according to your past consumption habit, it is suggested to recharge + half of the target fee value + related amount", to help the user quickly recharge; here T4 is a preset value; The second judgment analysis in step three is specifically as follows: SS1: the non inertia city is obtained, and then all recharge activities of the non inertia city are monitored; SS2: when any recharge activity occurs, the recharge record is automatically obtained, and the recharge amount and the account amount are obtained, the recharge amount and the account amount specifically refer to the amount of money when the amount of money is charged in the mall, and the amount of money that can be accounted for is the account amount; SS3: then the account amount is subtracted from the recharge amount, and the difference is divided by the recharge amount to obtain the discount ratio, which is marked as the current discount ratio; SS4: then the recharge activities occurring in the previous three months before the current recharge activity are obtained, each activity is marked as the preceding activity, and the discount ratio of each activity in the preceding activity is obtained, and all the discount ratios are marked as the preceding discount ratio group; SS5: when the discount ratio is greater than all the values in the preceding discount ratio group, an initial signal is generated; SS6: then the discount ratio given by all the activities corresponding to the preceding discount ratio group in the same time of the previous year is obtained, which is marked as the reference ratio, and the same time here refers to the activities within one week before and after the time when the corresponding preceding discount ratio group is generated, if not, the corresponding preceding discount ratio is deleted together, and is not recorded. Divide the reference ratio by the corresponding front discount ratio to obtain several promotion ratios; obtain the corresponding promotion ratio with the smallest promotion ratio, and mark it as a potential upgrade ratio; SS7: Then obtain the corresponding time of the previous year in the current month, and then obtain the discount ratio of all subsequent activities, mark it as a potential discount ratio, multiply all potential discount ratios by the potential upgrade ratio, and mark the obtained value as a pre-discount ratio. All pre-discount ratios form a pre-discount ratio group; SS8: Compare the current discount ratio with the pre-discount ratio. When all values in the pre-discount ratio group are smaller than the current discount ratio, a suggestion recharge signal is generated. At the same time, if the number of values in the pre-discount ratio group that are larger than the current discount ratio does not exceed two, a reminder recharge signal is generated; SS9: When the suggestion recharge signal is generated, automatically remind the user that "the current discount ratio is in the best period, and it is suggested to recharge", and recharge the minimum amount according to the activity; When the reminder recharge signal is generated, automatically remind the user that "the current discount ratio may be in the best period, and there is a low possibility of exceeding the ratio in the future. It is suggested to consider recharging" words. 2.The AI-based network mall recharge method of claim 1, wherein, Before step one, the following steps also need to be performed: Obtain all purchase inertia data of the user, including consumption records and stable records; consumption records include consumption targets, corresponding consumption times and fee prices; stable records include stable targets and target fees. 3.The AI-based network mall recharge method of claim 2, characterized in that, The consumption target is the corresponding consumption commodity, the consumption time is the purchase time of the corresponding consumption target, and the fee price is the price of the corresponding consumption target; The stable target is the corresponding long-term purchased item, and the monthly automatic deduction item. 4.The AI-based network mall recharge method of claim 1, wherein, The specific way of the bias analysis in step S4 is: The mean of X is automatically obtained and marked as P i . After obtaining the number of values of X i greater than the P value, mark it as the even-up value; then obtain the specific number of values of X i less than the P value, mark it as the even-down value; Then obtain the absolute value of the difference between the upper and lower average values, and mark the absolute value as the bias value. 5.The AI-based network mall recharge method of claim 2, wherein, After completing the processing steps of step three, the following steps also need to be performed: Obtain all stable records, and obtain the stable target and target fee corresponding to the stable record. Before each time the user needs to renew the fee, automatically remind the user whether to stop this time of fee renewal.

Images