Method and system for intelligently adding and clearing bank notes for bank ATMs based on machine learning
An ATM machine and machine learning technology, applied in the field of big data analysis in the financial industry, can solve the problems of waste of funds, increase in the cost of adding banknotes, and more non-interest-bearing assets, so as to reduce waste and reduce bank operating costs.
- Summary
- Abstract
- Description
- Claims
- Application Information
AI Technical Summary
Problems solved by technology
Method used
Image
Examples
Embodiment 1
[0049] Such as figure 1 As shown, the embodiment of the present invention provides a method for intelligently adding banknotes to bank ATM machines based on machine learning, and the method includes the following steps:
[0050] S1: Wide table processing and sample splitting of bank ATM machine historical transaction data and holiday data. The realization principle and process of the whole steps are as follows:
[0051] 1) Screening, calculating, and summarizing the transaction date, card withdrawal amount of the bank, and card withdrawal amount of other banks in the original transaction data.
[0052] 2) Summarize the data of the bank's card withdrawal amount and other banks' withdrawal amount data to obtain the total daily withdrawal amount; process the daily withdrawal amount of each ATM machine and the value of the date-related dependent variable according to the transaction date and the total daily withdrawal amount.
[0053] 3) According to the annual holiday calendar,...
Embodiment 2
[0092] On the basis of the method in embodiment 1, embodiment 2 provides a method based on machine learning to intelligently add banknotes to bank ATM machines, and the method is as follows Figure 4 As shown, on the basis of the steps in Embodiment 1, the subsequent step S4 is added.
[0093] The implementation process of step S4 is: use the verification sample to generate the clearing and adding banknote strategy. The amount and date of adding and clearing banknotes need to be determined based on three factors: the total daily withdrawal amount, the cost of adding banknotes, and the coverage rate.
[0094] The principle of step S4 is as follows:
[0095] 1. ARIMA can predict the total amount of daily withdrawals in the next 7 days (that is, one week). Considering the actual situation of the business, different recharge times are preset, for example, recharge every Monday and Thursday, recharge every Tuesday and Friday, and recharge every Monday, Wednesday, and Five plus ban...
Embodiment 3
[0101] On the basis of the method in embodiment 2, embodiment 3 provides a kind of method based on machine learning to bank ATM machine intelligently clearing money, described method is as follows Figure 7 As shown, on the basis of the steps in Embodiment 2, the subsequent step S5 is added.
[0102] The process realized in step S5 is: run the model prediction every day to obtain the predicted withdrawal amount in the next 7 days, and obtain the predicted withdrawal amount based on the frequency of adding money, the number of days between intervals, and the rising ratio of the added money amount compared to the predicted withdrawal amount in the money-adding and clearing strategy. The date of the next refill and the amount of refill.
[0103] The realization of step S5 is based on the realization of step S4. Using the verification sample, a strategy of adding and clearing money is selected, and the daily withdrawal total confidence interval rises, and the weekly adding money d...
PUM
Login to View More Abstract
Description
Claims
Application Information
Login to View More 


