Banknote adding amount prediction method and device, computer device and storage medium

Abstract

The application relates to a cash replenishment amount prediction method and device, computer equipment and a storage medium. The application relates to the technical field of artificial intelligence. The method comprises the following steps: obtaining current deposit information of each cash dispenser of a target network point, environmental factor information of a plurality of influence factor types, and historical operation information of the target network point; determining the current cash shortage state of each cash dispenser based on the environmental factor information of each type and the current deposit information of each cash dispenser through a stock judgment strategy; for each cash dispenser, inputting a cash replenishment amount prediction model based on the environmental factor information of each type, the current cash shortage state of the cash dispenser and historical cash replenishment information of the cash dispenser, and predicting the target cash replenishment amount of the cash dispenser. The method can improve the prediction accuracy of the target cash replenishment amount of each ATM.

Description

Technical Field

[0001] This application relates to the field of artificial intelligence technology, and in particular to a method, apparatus, computer device, and storage medium for predicting the amount of cash added. Background Technology

[0002] As one of the most important offline service channels for commercial banks, ATMs (Automated Teller Machines) primarily aim to divert small-amount customers from bank counters and generate fee-based income. However, if ATMs hold too much cash, it inevitably leads to idle bank funds and affects bank profitability; conversely, if ATMs hold too little cash, it increases maintenance costs, reduces service efficiency, and leads to customer loss. Therefore, determining when and how much cash to keep in ATMs is a key research focus.

[0003] Currently, banks typically monitor the cash levels in each cash box of every ATM through remote platforms, and then rely on staff experience to determine the amount of cash to be added to each box. However, due to limitations in staff experience and the varying operational status of different ATMs, the accuracy of predicting the amount of cash to be added to each ATM is relatively low. Summary of the Invention

[0004] Therefore, it is necessary to provide a method, apparatus, computer equipment, computer-readable storage medium, and computer program product for predicting the amount of cash added, in order to address the aforementioned technical problems.

[0005] Firstly, this application provides a method for predicting the amount of cash to be added. The method includes:

[0006] The system obtains current deposit information for each ATM at the target branch, environmental factor information of multiple influencing factors, and historical operating information of the target branch; the historical operating information of the target branch includes historical cash replenishment information for each ATM at the target branch.

[0007] Based on the environmental factor information of each type and the current deposit information of each ATM, the current cash shortage status of each ATM is determined through a stock judgment strategy.

[0008] For each ATM, based on environmental factor information of each type, the current cash shortage status of the ATM, and the historical cash replenishment information of the ATM, the cash replenishment amount prediction model is input to predict the target cash replenishment amount of the ATM.

[0009] Optionally, obtaining environmental factor information of multiple influencing factor types includes:

[0010] Multiple current environmental data are acquired, and through environmental analysis strategies, the sub-influencing factors of each current environmental data on each ATM of the target network are analyzed. Based on the correlation between the sub-influencing factors of each current environmental data, the current environmental data is clustered to obtain multiple environmental factor groups.

[0011] Identify the sub-influencing factors corresponding to all environmental data information in each environmental factor group, and use the environmental factor group corresponding to each influencing factor type as the environmental factor information for each type.

[0012] Optionally, the determination of the current cash shortage status of each ATM based on environmental factor information of each type and the current deposit information of each ATM, through a stock determination strategy, includes:

[0013] Identify the sub-influencing factor values ​​of all current environmental data information in the environmental factor group corresponding to each influencing factor type;

[0014] Based on the values ​​of all sub-influence factors corresponding to each influencing factor type, determine the influencing factor value of the influencing factor type;

[0015] For each ATM, the influencing factor values ​​of all influencing factor types and the current deposit information of the ATM are input into the cash shortage judgment model to obtain the current cash shortage status of the ATM.

[0016] Optionally, the step of identifying the sub-influencing factor values ​​of all current environmental data information in the environmental factor group corresponding to each influencing factor type includes:

[0017] For each type of influencing factor corresponding to an environmental factor group, the preset influencing factor range corresponding to each current environmental data information of the environmental factor group of the influencing factor type is identified, and the value of the preset influencing factor range corresponding to each current environmental data information is obtained. The value of the preset influencing factor range corresponding to each current environmental data information is used as the sub-influencing factor value corresponding to each current environmental data information.

[0018] Optionally, determining the influencing factor value for each influencing factor type based on all sub-influencing factor values ​​corresponding to each influencing factor type includes:

[0019] Calculate the correlation between the sub-influencing factors of each current environmental data information in the environmental factor group and the influencing factor type;

[0020] The correlation degree corresponding to each sub-influencing factor is normalized to obtain the influence factor weight of each sub-influencing factor. The sub-influencing factor values ​​of each current environmental data information of the environmental factor group are weighted and summed to obtain the influence factor value corresponding to the influence factor type.

[0021] Optionally, before inputting the values ​​of all influencing factor types and the current deposit information of the ATM into the inventory judgment model to obtain the current cash shortage status of the ATM, the method further includes:

[0022] Obtain the sample influencing factor values ​​for multiple sample influencing factor types, sample deposit information for multiple sample ATMs, and sample cash shortage status for each sample ATM;

[0023] For each sample cash shortage state, the sample influencing factor value of each sample influencing factor type and the sample deposit information of the sample ATM corresponding to each sample cash shortage state are input into the initial stock judgment model to obtain the first cash shortage state of each sample ATM.

[0024] Based on the first cash shortage state of each sample ATM and the sample cash shortage state of each sample ATM, the initial inventory judgment model is trained to identify the parameters for each sample cash shortage state, thus obtaining the first inventory judgment model.

[0025] Optionally, for each ATM, based on environmental factor information of each type, the cash shortage status of the ATM, and the historical cash replenishment information of the ATM, a cash replenishment prediction model is used to predict the target cash replenishment amount for the ATM, including:

[0026] Based on the historical cash replenishment information of the ATM, a scatter plot of cash usage of the ATM is established. Based on the scatter plot of cash usage, among multiple initial cash replenishment prediction models, a cash replenishment prediction model that meets the algorithm requirements of the scatter plot of cash usage is selected.

[0027] Determine the weight value of each of the aforementioned influencing factor types, and based on the weight value of each of the aforementioned influencing factor types, perform a weighted summation of the sub-influencing factor values ​​corresponding to all current environmental data information in the environmental factor group of all influencing factor types to obtain the comprehensive influencing factor value corresponding to all influencing factor types;

[0028] The comprehensive influencing factor values ​​are normalized to obtain the influencing factor weights corresponding to the comprehensive influencing factor values. The influencing factor weights, the cash shortage status of the ATM, and the current deposit information of the ATM are then input into the cash replenishment prediction model to obtain the target cash replenishment amount for the deposit machine.

[0029] Optionally, the step of selecting a banknote replenishment prediction model that meets the algorithm requirements of the banknote consumption scatter plot from multiple initial banknote replenishment prediction models includes:

[0030] The scatter plot of cash usage is smoothed, and the curvature of the scatter plot of cash usage is identified;

[0031] If the curvature is greater than the curvature threshold, return to the smoothing process for the scatter plot of cash usage until the curvature is less than the curvature threshold, and obtain the number of times the scatter plot of cash usage has been smoothed.

[0032] Based on the smoothing times of the scatter plot of cash usage, the initial cash replenishment prediction model corresponding to the smoothing times is selected from multiple initial cash replenishment prediction models to obtain the cash replenishment prediction model.

[0033] Secondly, this application also provides a device for predicting the amount of cash to be added. The device includes:

[0034] The acquisition module is used to acquire the current deposit information of each ATM at the target branch, environmental factor information of multiple influencing factors, and the historical operation information of the target branch; the historical operation information of the target branch includes the historical cash replenishment information of each ATM at the target branch.

[0035] The determination module is used to determine the current cash shortage status of each ATM based on environmental factor information of each type and the current deposit information of each ATM, through a stock judgment strategy.

[0036] The prediction module is used to predict the target amount of cash to be added to each ATM based on environmental factor information of each type, the current cash shortage status of the ATM, and the historical cash replenishment information of the ATM, by inputting a cash replenishment prediction model.

[0037] Optionally, the acquisition module is specifically used for:

[0038] Multiple current environmental data are acquired, and through environmental analysis strategies, the sub-influencing factors of each current environmental data on each ATM of the target network are analyzed. Based on the correlation between the sub-influencing factors of each current environmental data, the current environmental data is clustered to obtain multiple environmental factor groups.

[0039] Identify the sub-influencing factors corresponding to all environmental data information in each environmental factor group, and use the environmental factor group corresponding to each influencing factor type as the environmental factor information for each type.

[0040] Optionally, the determining module is specifically used for:

[0041] Identify the sub-influencing factor values ​​of all current environmental data information in the environmental factor group corresponding to each influencing factor type;

[0042] Based on the values ​​of all sub-influence factors corresponding to each influencing factor type, determine the influencing factor value of the influencing factor type;

[0043] For each ATM, the influencing factor values ​​of all influencing factor types and the current deposit information of the ATM are input into the cash shortage judgment model to obtain the current cash shortage status of the ATM.

[0044] Optionally, the determining module is specifically used for:

[0045] For each type of influencing factor corresponding to an environmental factor group, the preset influencing factor range corresponding to each current environmental data information of the environmental factor group of the influencing factor type is identified, and the value of the preset influencing factor range corresponding to each current environmental data information is obtained. The value of the preset influencing factor range corresponding to each current environmental data information is used as the sub-influencing factor value corresponding to each current environmental data information.

[0046] Optionally, the determining module is specifically used for:

[0047] Calculate the correlation between the sub-influencing factors of each current environmental data information in the environmental factor group and the influencing factor type;

[0048] The correlation degree corresponding to each sub-influencing factor is normalized to obtain the influence factor weight of each sub-influencing factor. The sub-influencing factor values ​​of each current environmental data information of the environmental factor group are weighted and summed to obtain the influence factor value corresponding to the influence factor type.

[0049] Optionally, the device further includes:

[0050] The sample acquisition module is used to acquire the sample influencing factor values ​​of multiple sample influencing factor types, the sample deposit information of multiple sample ATMs, and the sample cash shortage status of each sample ATM.

[0051] The input module is used to input the sample influencing factor values ​​of each sample influencing factor type and the sample deposit information of the sample ATM corresponding to each sample cash shortage state into the initial stock judgment model to obtain the first cash shortage state of each sample ATM.

[0052] The training module is used to train the initial inventory judgment model for the recognition parameters of each sample cash shortage state based on the first cash shortage state of each sample ATM and the sample cash shortage state of each sample ATM, so as to obtain the first inventory judgment model.

[0053] Optional, the predicted module is specifically used for:

[0054] Based on the historical cash replenishment information of the ATM, a scatter plot of cash usage of the ATM is established. Based on the scatter plot of cash usage, among multiple initial cash replenishment prediction models, a cash replenishment prediction model that meets the algorithm requirements of the scatter plot of cash usage is selected.

[0055] Determine the weight value of each of the aforementioned influencing factor types, and based on the weight value of each of the aforementioned influencing factor types, perform a weighted summation of the sub-influencing factor values ​​corresponding to all current environmental data information in the environmental factor group of all influencing factor types to obtain the comprehensive influencing factor value corresponding to all influencing factor types;

[0056] The comprehensive influencing factor values ​​are normalized to obtain the influencing factor weights corresponding to the comprehensive influencing factor values. The influencing factor weights, the cash shortage status of the ATM, and the current deposit information of the ATM are then input into the cash replenishment prediction model to obtain the target cash replenishment amount for the deposit machine.

[0057] Optional, the predicted module is specifically used for:

[0058] The scatter plot of cash usage is smoothed, and the curvature of the scatter plot of cash usage is identified;

[0059] If the curvature is greater than the curvature threshold, return to the smoothing process for the scatter plot of cash usage until the curvature is less than the curvature threshold, and obtain the number of times the scatter plot of cash usage has been smoothed.

[0060] Based on the smoothing times of the scatter plot of cash usage, the initial cash replenishment prediction model corresponding to the smoothing times is selected from multiple initial cash replenishment prediction models to obtain the cash replenishment prediction model.

[0061] Thirdly, this application also provides a computer device. The computer device includes a memory and a processor, the memory storing a computer program, and the processor executing the computer program to implement the steps of the method described in any one of the first aspects.

[0062] Fourthly, this application also provides a computer-readable storage medium. The computer-readable storage medium stores a computer program thereon, which, when executed by a processor, implements the steps of the method described in any one of the first aspects.

[0063] Fifthly, this application also provides a computer program product. The computer program product includes a computer program that, when executed by a processor, implements the steps of the method described in any one of the first aspects.

[0064] The aforementioned cash replenishment prediction method, apparatus, computer equipment, and storage medium acquire current deposit information of each ATM at a target branch, environmental factor information of multiple influencing factors, and historical operational information of the target branch. Based on the type of each environmental factor, the environmental factor information is categorized. The historical operational information of the target branch includes historical cash replenishment information for each ATM. Based on the environmental factor information of each type and the current deposit information of each ATM, a stock assessment strategy is used to determine the current cash shortage status of each ATM. For each ATM, based on the environmental factor information of each type, the current cash shortage status of the ATM, and the historical cash replenishment information of the ATM, a cash replenishment prediction model is used to predict the target cash replenishment amount for that ATM. By analyzing environmental factor information of multiple influencing factors and the current deposit information of each ATM, the current cash shortage status of the ATM is determined. Then, by combining the environmental factor information of multiple influencing factors and the current cash shortage status, a cash replenishment prediction model is used to predict the target cash replenishment amount for that ATM. This not only eliminates the need for manual judgment of cash replenishment, but also analyzes environmental factors of multiple influencing factors for each ATM when judging the cash shortage status and the target cash replenishment amount. This ensures that the target cash replenishment amount meets the cash withdrawal needs of the current environment of the ATM, thereby improving the prediction accuracy of the target cash replenishment amount for each ATM. Attached Figure Description

[0065] Figure 1 This is a flowchart illustrating a method for predicting the amount of cash to be added in one embodiment;

[0066] Figure 2 This is a flowchart illustrating an example of predicting the amount of cash added in one embodiment;

[0067] Figure 3 This is a structural block diagram of a banknote addition prediction device in one embodiment;

[0068] Figure 4 This is an internal structural diagram of a computer device in one embodiment. Detailed Implementation

[0069] To make the objectives, technical solutions, and advantages of this application clearer, the following detailed description is provided in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the scope of this application.

[0070] The cash replenishment prediction method provided in this application can be applied to terminals, servers, and systems including terminals and servers, and is implemented through interaction between the terminal and the server. The terminal can be, but is not limited to, various personal computers, laptops, smartphones, tablets, etc. The server can be a standalone server or a server cluster composed of multiple servers. The terminal analyzes environmental factor information of multiple influencing factors and the current deposit information of each ATM to determine the current cash shortage status of the ATM. Then, by combining the environmental factor information of multiple influencing factors and the current cash shortage status, a cash replenishment prediction model is used to predict the target cash replenishment amount for the ATM. This not only avoids the manual process of judging cash replenishment, but also analyzes environmental factor information of multiple influencing factors for each ATM simultaneously when judging the cash shortage status and the target cash replenishment amount, ensuring that the target cash replenishment amount meets the cash withdrawal needs of the current environment of the ATM, thereby improving the accuracy of judging the target cash replenishment amount for each ATM.

[0071] In one embodiment, such as Figure 1 As shown, a method for predicting the amount of cash to be added is provided. Taking the application of this method to a terminal as an example, the method includes the following steps:

[0072] Step S101: Obtain the current deposit information of each ATM at the target branch, environmental factor information of multiple influencing factor types, and historical operation information of the target branch.

[0073] The historical operational information of the target outlet includes the historical cash replenishment information for each ATM at the target outlet.

[0074] In this embodiment, the terminal selects the branch where cash replenishment prediction is needed as the target branch. By receiving data from each ATM at the target branch, it obtains the current deposit information and historical cash replenishment information for each ATM. The historical cash replenishment information includes the amount of cash replenished each time and the time of each replenishment. The terminal collects environmental data from the target branch via the internet and analyzes the influencing factors corresponding to each environmental data point to obtain environmental factor information of different influencing factor types. The environmental data refers to environmental information affecting the operation of the target branch, including but not limited to: the number of main roads near the target branch, the status of basic infrastructure construction near the target branch, the number of residential areas near the target branch, the number of residents in each residential area, the age distribution of residents in each residential area, the number of ATMs at other bank branches near the target branch, the number of ATMs at the target branch, cash replenishment personnel allocation information, banknote sorting information, cash replenishment methods, monitoring system malfunction information, assessment method text information, date information, and weather information. The aforementioned environmental information can be categorized into factors influencing surrounding facilities, community residents' conditions, bank ATM management, and the natural environment. For example, factors influencing surrounding facilities include the number of main roads near the target branch and the status of basic infrastructure development in the vicinity; factors influencing community residents' conditions include the number of residential areas near the target branch, the number of residents in each area, and the age distribution of residents in each area; factors influencing bank ATM management include the number of ATMs at the target branch, the staffing of cash replenishment personnel, cash sorting, cash replenishment methods, monitoring system malfunctions, and performance evaluation methods; and natural factors include date and weather. The specific classification process will be explained in detail later.

[0075] Step S102: Based on various types of environmental factor information and the current deposit information of each ATM, determine the current cash shortage status of each ATM through a stock judgment strategy.

[0076] In this embodiment, the terminal determines the current cash shortage status of each ATM based on various types of environmental factor information and the current deposit information of each ATM, using a stock judgment model based on a stock judgment strategy. The stock judgment model is a support vector machine (SVM) model, specifically a multi-level structured classification SVM model. Cash shortage status is categorized into three types: general cash shortage, severe cash shortage, and no cash shortage. Based on the multi-level structured classification SVM model, the cash shortage status can be divided into two main categories: cash shortage and no cash shortage. The cash shortage category includes general cash shortage and severe cash shortage. The specific processing procedure of the stock judgment model will be explained in detail later.

[0077] Step S103: For each ATM, based on various types of environmental factor information, the current cash shortage status of the ATM, and the historical cash replenishment information of the ATM, input the cash replenishment amount prediction model to predict the target cash replenishment amount for the ATM.

[0078] Then, for each ATM, the terminal uses a cash replenishment prediction model to predict the target cash replenishment amount based on various environmental factors, the current cash shortage status of the ATM, and the ATM's historical cash replenishment information. This cash replenishment prediction model is a time series prediction model, which includes, but is not limited to, Autoregression (AR), Autoregressive Moving Average (ARMA), Autoregressive Integrated Moving Average (ARIMA), Seasonal Autoregressive Integrated Moving-Average (SARIMA), Seasonal Autoregressive Integrated Moving-Average with Exogenous Regressors (SARIMAX), Vector Autoregression (VAR), Vector Autoregression Moving-Average (VARMA), Vector Autoregression Moving-Average with Exogenous Regressors (VARMAX), and Holt Winter's Exponential Smoothing (HWES) prediction model. The specific processing steps of the cash replenishment prediction model will be explained in detail later.

[0079] Based on the above scheme, by analyzing environmental factor information of multiple influencing factors and the current deposit information of each ATM, the current cash shortage status of the ATM is determined. Then, by combining the environmental factor information of multiple influencing factors and the current cash shortage status, a cash replenishment prediction model is used to predict the target cash replenishment amount for each ATM. This not only avoids the process of manually judging the cash replenishment amount, but also analyzes the environmental factor information of multiple influencing factors for each ATM when judging the cash shortage status and the target cash replenishment amount. This ensures that the target cash replenishment amount meets the cash withdrawal needs of the current environment of the ATM, thereby improving the accuracy of judging the target cash replenishment amount for each ATM.

[0080] Optionally, environmental factor information of multiple influencing factor types can be obtained, including: obtaining multiple current environmental data information, and through environmental analysis strategies, analyzing the sub-influencing factors of each current environmental data information on each ATM of the target network, and clustering the current environmental data information according to the correlation between the sub-influencing factors of each current environmental data information to obtain multiple environmental factor groups; identifying the influencing factor types corresponding to the sub-influencing factors of all environmental data information in each environmental factor group, and using the environmental factor group corresponding to each influencing factor type as the environmental factor information of each type.

[0081] In this embodiment, the terminal collects multiple current environmental data points via the internet and analyzes the sub-influencing factors of each environmental data point on each ATM at the target branch using a pre-set environmental analysis strategy. Specifically, the terminal presets collection parameters for each current environmental data point and collects the corresponding parameter values ​​via the internet to obtain each current environmental data point. The terminal also presets an analysis strategy for each current environmental data point and then analyzes the sub-influencing factors of each environmental data point on each ATM at the target branch based on this strategy. This analysis strategy can be, but is not limited to, factor analysis and principal component analysis. Each current environmental data information has sub-influencing factors, such as: the number of main roads near the target branch (the sub-influencing factor is the impact of the number of main roads on the number of people passing by the target branch); the basic infrastructure construction near the target branch (the sub-influencing factor is the impact of basic infrastructure on the pedestrian flow of the target branch); the number of residential areas near the target branch (the sub-influencing factor is the impact of the population coverage of the target branch); the number of residents in each residential area (the sub-influencing factor is the impact of the population density of the target branch); the age distribution of residents in each residential area (the sub-influencing factor is the impact of the target population of the target branch); the number of ATMs in other bank branches near the target branch (the sub-influencing factor is the impact of the cash demand of the target branch); the number of ATMs at the target branch (the sub-influencing factor is the impact of the total cash reserves of the target branch); the date (the sub-influencing factor is the impact of the business demand of the target branch); and the weather (the sub-influencing factor is the impact of the business volume of the target branch). The impact of different influencing factors varies. In terms of time, they are generally divided into the beginning, middle, and end of the month. Generally, when pension payments are concentrated in the middle or end of the month, the cash demand of ATMs increases significantly, and ATM shortages frequently occur. The presence of ATMs from other banks in the vicinity reduces the amount of cash required from the target branch. The demand for ATMs increases significantly when the nearby community residents are older, as some elderly customers are resistant to electronic payments. Furthermore, internal bank management factors related to ATMs, including but not limited to staffing for cash replenishment, banknote sorting, replenishment methods, monitoring system malfunctions, and performance evaluation methods, also greatly influence ATM throughput. ATM cash replenishment is calculated per branch. For example, if a branch has one ATM, the replenishment amount is for that single ATM. If there are two ATMs, considering that one can be used after the other runs out of cash, the replenishment amount is for both ATMs.

[0082] The terminal clusters each current environmental data point based on its sub-influencing factors, resulting in multiple environmental factor groups. For example, if the sub-influencing factors of the number of ATMs at other bank branches near the target branch (affecting the target branch's cash demand) and the number of ATMs at the target branch (affecting the target branch's total cash reserves) have a high correlation, then the current environmental data points corresponding to these sub-influencing factors constitute one environmental factor group. Similarly, if the sub-influencing factors of the number of main roads near the target branch (affecting the number of main roads on the number of people passing by the target branch), the sub-influencing factors of the infrastructure development near the target branch (affecting the infrastructure on pedestrian traffic), the sub-influencing factors of the number of residential areas near the target branch (affecting the population coverage of the target branch), and the sub-influencing factors of the number of residents in each residential area (affecting the population density of the target branch) have a high correlation, then the current environmental data points corresponding to these sub-influencing factors constitute one environmental factor group. The terminal identifies the sub-influencing factor with the highest correlation to each sub-influencing factor in each environmental factor group as the influencing factor type of that environmental factor group, and uses that environmental factor group as environmental factor information to obtain environmental factor information for multiple influencing factor types.

[0083] Based on the above scheme, by analyzing each current environmental data information, the current environmental data information is classified, avoiding the problem of collinearity due to a large number of influencing factors, and improving the accuracy of subsequent cash shortage status and target cash replenishment amount.

[0084] Optionally, based on information on various types of environmental factors and the current deposit information of each ATM, a stock judgment strategy is used to determine the current cash shortage status of each ATM. This includes: analyzing the sub-influence factor values ​​of all environmental data information in the environmental factor group corresponding to each influencing factor type, and determining the influencing factor value of each influencing factor type based on all sub-influence factor values ​​corresponding to each influencing factor type; for each ATM, inputting the influencing factor values ​​of all influencing factor types and the current deposit information of the ATM into the stock judgment model to determine the current cash shortage status of the ATM.

[0085] In this embodiment, the terminal analyzes the sub-influence factor values ​​of all environmental data information in the environmental factor group corresponding to each influencing factor type, and determines the influencing factor value of each influencing factor type based on all sub-influence factor values. The specific analysis process will be explained in detail later. For each ATM, the terminal inputs the influencing factor values ​​of all influencing factor types and the current deposit information of the ATM into the inventory judgment model to obtain the current cash shortage status of the ATM.

[0086] Based on the above scheme, the current cash shortage status of ATMs can be determined by the types of influencing factors, thus improving the accuracy of determining the current cash shortage status.

[0087] Optionally, identify the sub-influencing factor values ​​of all current environmental data information in the environmental factor group corresponding to each influencing factor type, including:

[0088] For each type of influencing factor corresponding to the environmental factor group, the preset influencing factor range corresponding to each current environmental data information of the environmental factor group of the influencing factor type is identified, and the value of the preset influencing factor range corresponding to each current environmental data information is obtained. The value of the preset influencing factor range corresponding to each current environmental data information is used as the sub-influencing factor value corresponding to each current environmental data information.

[0089] In this embodiment, the terminal presets different levels of influence factor ranges corresponding to each sub-influencing factor. Then, for each type of influencing factor corresponding to an environmental factor group, it identifies the preset influence factor range corresponding to each current environmental data information of that influencing factor group, and uses the degree corresponding to the preset influence factor range of each current environmental data information as the sub-influencing factor value for each current environmental data information. For example, the value corresponding to a small target population is 0, and the influence factor range is 0%-30% of the elderly population; the value corresponding to a medium target population is 1, and the influence factor range is 30%-70% of the elderly population; the value corresponding to a large target population is 2, and the influence factor range is 70%-99% of the elderly population. In the age distribution of residents in each residential area, the elderly population accounts for 54%, so the sub-influencing factor value corresponding to this current environmental data information is 1.

[0090] Based on the above scheme, by pre-setting the range of different degrees of influence factors corresponding to each sub-influencing factor, the weight of each sub-influencing factor is determined, thereby obtaining the influence factor values ​​corresponding to different types of influence factors. This quantifies the degree of influence of current environmental data information on ATMs and improves the accuracy of calculating the current cash shortage status of ATMs.

[0091] Optionally, based on the values ​​of all sub-influencing factors corresponding to each influencing factor type, the influencing factor value of the influencing factor type is determined, including: calculating the correlation degree between the sub-influencing factors of each current environmental data information in the environmental factor group and the influencing factor type; normalizing the correlation degree corresponding to each sub-influencing factor to obtain the influencing factor weight of each sub-influencing factor; and weighting and summing the sub-influencing factor values ​​of each current environmental data information in the environmental factor group to obtain the influencing factor value corresponding to the influencing factor type.

[0092] In this embodiment, the terminal uses a similarity recognition algorithm to calculate the correlation degree between the sub-influencing factors and the influencing factor type for each current environmental data information in the environmental factor group. Then, the terminal normalizes the correlation degree corresponding to each sub-influencing factor to obtain the influencing factor weight for each sub-influencing factor. Finally, the terminal performs a weighted summation of the sub-influencing factor values ​​for each current environmental data information in the environmental factor group to obtain the influencing factor value corresponding to the influencing factor type.

[0093] Based on the above scheme, the weight of each current environmental data information is determined by calculating the correlation between the sub-influencing factors and the influencing factor type of each current environmental data information in the environmental factor group, thereby improving the accuracy of determining the weight of the current environmental data information.

[0094] Optionally, before determining the current cash shortage status of an ATM by inputting the influencing factor values ​​of all influencing factor types and the current deposit information of the ATM into the inventory judgment model, the following steps are also included: obtaining the sample influencing factor values ​​of multiple sample influencing factor types, the sample deposit information of multiple sample ATMs, and the sample cash shortage status of each sample ATM; for each sample cash shortage status, inputting the sample influencing factor values ​​of each sample influencing factor type and the sample deposit information of the sample ATM corresponding to each sample cash shortage status into the initial inventory judgment model to obtain the first cash shortage status of each sample ATM; based on the first cash shortage status and the sample cash shortage status of each sample ATM, training the initial inventory judgment model for the recognition parameters of each sample cash shortage status to obtain the first inventory judgment model.

[0095] In this embodiment, the terminal acquires sample influencing factor values ​​for multiple sample influencing factor types, sample deposit information for multiple sample ATMs, and the sample cash shortage status for each sample ATM. Then, the terminal categorizes the acquired sample influencing factor values ​​for multiple sample influencing factor types, sample deposit information for multiple sample ATMs, and the first cash shortage status for each sample ATM into two main categories. That is, severe cash shortage and moderate cash shortage are grouped into a single cash shortage category, represented by 1, while no cash shortage is grouped into another category, represented by 0. Then, the model training process begins.

[0096] During the first training process, the terminal inputs the sample influencing factor values ​​of each sample's influencing factor type and the sample deposit information of the sample ATMs corresponding to each sample's cash shortage state into the initial inventory judgment model to obtain the first cash shortage state of each sample ATM. Based on the first cash shortage state and the sample cash shortage state of each sample ATM, the terminal trains the initial inventory judgment model to identify the parameters for each sample cash shortage state. The terminal presets different rate thresholds. If the difference rate between the first cash shortage state and the sample cash shortage state is higher than the first difference rate threshold, the terminal returns to the step of training the initial inventory judgment model to identify the parameters for each sample cash shortage state based on the first cash shortage state and the sample cash shortage state of each sample ATM. This process continues until the difference rate is lower than the difference rate threshold, at which point the terminal stops the iterative operation and enters the second training process.

[0097] During the second training process, for each sample cash shortage state, the terminal randomly selects sample deposit information of a portion of the sample ATMs in that cash shortage state, as well as sample influencing factor values ​​of multiple sample influencing factor types, as the training group. The terminal then returns to the execution terminal and inputs the sample influencing factor values ​​of each sample influencing factor type and the sample deposit information of the sample ATMs corresponding to each sample cash shortage state into the initial inventory judgment model to obtain the first cash shortage state step for each sample ATM. Then, the terminal selects sample deposit information of a portion of the sample ATMs in that cash shortage state, as well as sample influencing factor values ​​of multiple sample influencing factor types, as the detection group, and inputs it into the second-trained initial inventory judgment model. The detection group calculates whether the difference rate between each sample cash shortage state and the first cash shortage state of each ATM reaches a second difference rate. If the difference rate reaches the second difference rate, the terminal uses the initial inventory judgment model as the inventory judgment model.

[0098] Based on the above scheme, the terminal trains the initial stock judgment model through two training sessions and one detection session to obtain the stock judgment model, thereby improving the judgment accuracy of the stock judgment model.

[0099] Optionally, for each ATM, based on various types of environmental factor information, the ATM's cash shortage status, and the ATM's historical cash replenishment information, a cash replenishment prediction model is used to predict the target cash replenishment amount for the ATM. This includes: establishing a scatter plot of cash usage for the ATM based on its historical cash replenishment information; selecting a cash replenishment prediction model that meets the algorithm requirements of the cash usage scatter plot from multiple initial cash replenishment prediction models based on the scatter plot; determining the weight values ​​of each influencing factor type; and performing weighted summation on the sub-influencing factor values ​​corresponding to all current environmental data information in the environmental factor groups of all influencing factor types based on the weight values ​​of each influencing factor type to obtain the comprehensive influencing factor value corresponding to all influencing factor types; normalizing the comprehensive influencing factor value to obtain the influencing factor weight corresponding to the comprehensive influencing factor value; and inputting the influencing factor weight, the ATM's cash shortage status, and the ATM's current deposit information into the cash replenishment prediction model to obtain the target cash replenishment amount for the ATM.

[0100] In this embodiment, the terminal establishes a scatter plot of cash usage for the ATM based on its historical cash replenishment information. Based on this scatter plot, it selects a cash replenishment prediction model from multiple initial cash replenishment prediction models that meets the algorithmic requirements of the scatter plot. The specific selection process will be explained in detail later. Then, the terminal analyzes the weight values ​​of each influencing factor type. Next, the terminal multiplies the sub-influencing factor values ​​corresponding to all current environmental data information in the environmental factor group for each influencing factor type by the weight value of each influencing factor type, and then sums them to obtain the comprehensive influencing factor value corresponding to all influencing factor types. This weight value can be a proportional weight. The terminal normalizes the comprehensive influencing factor value to obtain the influencing factor weight corresponding to the comprehensive influencing factor value. Then, based on the influencing factor weights, the cash shortage status of the ATM, and the current deposit information of the ATM, the terminal determines the target cash replenishment amount for the deposit machine using the cash replenishment prediction model. This cash replenishment prediction model is a trained time series prediction model.

[0101] The formula corresponding to this time series prediction model is:

[0102]

[0103] Where 0≤α≤1, Q t+1 Q is the target amount of cash added on T+1 day. t Q represents the current deposit information value on day T. t , Let α be the predicted amount of cash added on day T, and α be the weight of the influencing factors. is a constant coefficient.

[0104] In the above formula, Based on sample data and multiple preset parameters In the process, the constant coefficient value corresponding to the error value of the minimum cash addition prediction value is selected through trial and error algorithm.

[0105] Based on the above scheme, by adding weights to influencing factors, the target cash replenishment amount for each ATM can be predicted, thereby improving the accuracy of determining the target cash replenishment amount.

[0106] Optionally, based on the scatter plot of cash usage, among multiple initial cash replenishment prediction models, a cash replenishment prediction model that meets the algorithm requirements of the scatter plot of cash usage is selected. This includes: smoothing the scatter plot of cash usage and identifying the curvature of the scatter plot of cash usage; if the curvature is greater than a curvature threshold, returning to the smoothing step of the scatter plot of cash usage until the curvature is less than the curvature threshold, thus obtaining the number of times the scatter plot of cash usage has been smoothed; based on the number of times the scatter plot of cash usage has been smoothed, among multiple initial cash replenishment prediction models, the initial cash replenishment prediction model corresponding to the number of smoothing times is selected to obtain the cash replenishment prediction model.

[0107] In this embodiment, the terminal smooths the scatter plot of cash usage using a time-series exponential smoothing algorithm and identifies the curvature of the scatter plot. Then, the terminal presets a curvature threshold and determines whether the curvature of the scatter plot exceeds the threshold. If the curvature exceeds the threshold, the terminal returns to the smoothing process until the curvature falls below the threshold, at which point the terminal obtains the number of smoothing iterations for the scatter plot. Then, based on the initial cash addition prediction models corresponding to different smoothing iterations stored in the database, the terminal selects the initial cash addition prediction model corresponding to that smoothing iteration. The database contains a preset correspondence between the initial cash addition prediction models for each smoothing iteration.

[0108] Specifically, the terminal uses a pre-set prediction model for the initial cash addition based on different smoothing times in the database. This model includes a single exponential smoothing prediction model for one smoothing process, a double exponential smoothing prediction model for two smoothing processes, and a triple exponential smoothing prediction model for three smoothing processes.

[0109] Based on the above scheme, by using a scatter plot of cash volume to filter the cash replenishment prediction model, the accuracy of predicting the target cash replenishment volume of the same ATM is improved when historical cash replenishment information of the same ATM is available.

[0110] In one embodiment, such as Figure 2 As shown, a road damage detection example is provided, which includes the following steps:

[0111] Step S201: Obtain the current deposit information of each ATM at the target branch and the historical operation information of the target branch.

[0112] Step S202: Obtain multiple current environmental data information, and through environmental analysis strategies, analyze the sub-influencing factors of each current environmental data information on each ATM of the target network, and cluster the current environmental data information according to the correlation between the sub-influencing factors of each current environmental data information to obtain multiple environmental factor groups.

[0113] Step S203: Identify the influencing factor types corresponding to the sub-influencing factors of all environmental data information in each environmental factor group, and take the environmental factor group corresponding to each influencing factor type as the environmental factor information of each type.

[0114] Step S204: For each type of influencing factor corresponding to the environmental factor group, identify the preset influencing factor range corresponding to each current environmental data information of the environmental factor group of the influencing factor type, obtain the value of the preset influencing factor range corresponding to each current environmental data information, and take the value of the preset influencing factor range corresponding to each current environmental data information as the sub-influencing factor value corresponding to each current environmental data information.

[0115] Step S205: Calculate the correlation between sub-influencing factors and influencing factor types for each current environmental data information in the environmental factor group.

[0116] Step S206: Normalize the correlation degree corresponding to each sub-influencing factor to obtain the influence factor weight of each sub-influencing factor, and sum the sub-influencing factor values ​​of each current environmental data information of the environmental factor group to obtain the influence factor value corresponding to the influence factor type.

[0117] Step S207: For each ATM, input the influencing factor values ​​of all influencing factor types and the current deposit information of the ATM into the inventory judgment model to obtain the current cash shortage status of the ATM.

[0118] Step S208: Based on the historical cash replenishment information of the ATM, create a scatter plot of the cash usage of the ATM.

[0119] Step S209: Smooth the scatter plot of cash usage and identify the curvature of the scatter plot of cash usage.

[0120] Step S210: If the curvature is greater than the curvature threshold, return to the step of smoothing the scatter plot of cash usage until the curvature is less than the curvature threshold, and obtain the number of times the scatter plot of cash usage has been smoothed.

[0121] Step S211: Based on the smoothing times of the scatter plot of cash usage, select the initial cash addition prediction model corresponding to the smoothing times from multiple initial cash addition prediction models to obtain the cash addition prediction model.

[0122] Step S212: Determine the weight value of each influencing factor type, and based on the weight value of each influencing factor type, perform weighted summation on the sub-influencing factor values ​​corresponding to all current environmental data information in the environmental factor group of all influencing factor types to obtain the comprehensive influencing factor value corresponding to all influencing factor types.

[0123] Step S213: Normalize the comprehensive influencing factor values ​​to obtain the influencing factor weights corresponding to the comprehensive influencing factor values, and input the influencing factor weights, the cash shortage status of the ATM, and the current deposit information of the ATM into the cash replenishment prediction model to obtain the target cash replenishment amount of the deposit machine.

[0124] It should be understood that although the steps in the flowcharts of the embodiments described above are shown sequentially according to the arrows, these steps are not necessarily executed in the order indicated by the arrows. Unless explicitly stated herein, there is no strict order restriction on the execution of these steps, and they can be executed in other orders. Moreover, at least some steps in the flowcharts of the embodiments described above may include multiple steps or multiple stages. These steps or stages are not necessarily completed at the same time, but can be executed at different times. The execution order of these steps or stages is not necessarily sequential, but can be performed alternately or in turn with other steps or at least some of the steps or stages of other steps.

[0125] Based on the same inventive concept, this application also provides a banknote replenishment prediction device for implementing the aforementioned banknote replenishment prediction method. The solution provided by this device is similar to the implementation described in the above method; therefore, the specific limitations in one or more banknote replenishment prediction device embodiments provided below can be found in the limitations of the banknote replenishment prediction method described above, and will not be repeated here.

[0126] In one embodiment, such as Figure 3 As shown, a device for predicting the amount of cash added is provided, comprising: an acquisition module 310, a determination module 320, and a prediction module 330, wherein:

[0127] The acquisition module 310 is used to acquire the current deposit information of each ATM at the target branch, environmental factor information of multiple influencing factor types, and the historical operation information of the target branch; the historical operation information of the target branch includes the historical cash replenishment information of each ATM at the target branch.

[0128] The determination module 320 is used to determine the current cash shortage status of each ATM based on environmental factor information of each type and the current deposit information of each ATM, through a stock judgment strategy.

[0129] The prediction module 330 is used to predict the target amount of cash to be added to each ATM based on environmental factor information of each type, the current cash shortage status of the ATM, and the historical cash replenishment information of the ATM, by inputting a cash replenishment prediction model.

[0130] Optionally, the acquisition module 310 is specifically used for:

[0131] Multiple current environmental data are acquired, and through environmental analysis strategies, the sub-influencing factors of each current environmental data on each ATM of the target network are analyzed. Based on the correlation between the sub-influencing factors of each current environmental data, the current environmental data is clustered to obtain multiple environmental factor groups.

[0132] Identify the sub-influencing factors corresponding to all environmental data information in each environmental factor group, and use the environmental factor group corresponding to each influencing factor type as the environmental factor information for each type.

[0133] Optionally, the determining module 320 is specifically used for:

[0134] Identify the sub-influencing factor values ​​of all current environmental data information in the environmental factor group corresponding to each influencing factor type;

[0135] Based on the values ​​of all sub-influence factors corresponding to each influencing factor type, determine the influencing factor value of the influencing factor type;

[0136] For each ATM, the influencing factor values ​​of all influencing factor types and the current deposit information of the ATM are input into the cash shortage judgment model to obtain the current cash shortage status of the ATM.

[0137] Optionally, the determining module 320 is specifically used for:

[0138] For each type of influencing factor corresponding to an environmental factor group, the preset influencing factor range corresponding to each current environmental data information of the environmental factor group of the influencing factor type is identified, and the value of the preset influencing factor range corresponding to each current environmental data information is obtained. The value of the preset influencing factor range corresponding to each current environmental data information is used as the sub-influencing factor value corresponding to each current environmental data information.

[0139] Optionally, the determining module 320 is specifically used for:

[0140] Calculate the correlation between the sub-influencing factors of each current environmental data information in the environmental factor group and the influencing factor type;

[0141] The correlation degree corresponding to each sub-influencing factor is normalized to obtain the influence factor weight of each sub-influencing factor. The sub-influencing factor values ​​of each current environmental data information of the environmental factor group are weighted and summed to obtain the influence factor value corresponding to the influence factor type.

[0142] Optionally, the device further includes:

[0143] The sample acquisition module is used to acquire the sample influencing factor values ​​of multiple sample influencing factor types, the sample deposit information of multiple sample ATMs, and the sample cash shortage status of each sample ATM.

[0144] The input module is used to input the sample influencing factor values ​​of each sample influencing factor type and the sample deposit information of the sample ATM corresponding to each sample cash shortage state into the initial stock judgment model to obtain the first cash shortage state of each sample ATM.

[0145] The training module is used to train the initial inventory judgment model for the recognition parameters of each sample cash shortage state based on the first cash shortage state of each sample ATM and the sample cash shortage state of each sample ATM, so as to obtain the first inventory judgment model.

[0146] Optionally, the prediction module 330 is specifically used for:

[0147] Based on the historical cash replenishment information of the ATM, a scatter plot of cash usage of the ATM is established. Based on the scatter plot of cash usage, among multiple initial cash replenishment prediction models, a cash replenishment prediction model that meets the algorithm requirements of the scatter plot of cash usage is selected.

[0148] Determine the weight value of each of the aforementioned influencing factor types, and based on the weight value of each of the aforementioned influencing factor types, perform a weighted summation of the sub-influencing factor values ​​corresponding to all current environmental data information in the environmental factor group of all influencing factor types to obtain the comprehensive influencing factor value corresponding to all influencing factor types;

[0149] The comprehensive influencing factor values ​​are normalized to obtain the influencing factor weights corresponding to the comprehensive influencing factor values. The influencing factor weights, the cash shortage status of the ATM, and the current deposit information of the ATM are then input into the cash replenishment prediction model to obtain the target cash replenishment amount for the deposit machine.

[0150] Optional, the predicted module is specifically used for:

[0151] The scatter plot of cash usage is smoothed, and the curvature of the scatter plot of cash usage is identified;

[0152] If the curvature is greater than the curvature threshold, return to the smoothing process for the scatter plot of cash usage until the curvature is less than the curvature threshold, and obtain the number of times the scatter plot of cash usage has been smoothed.

[0153] Based on the smoothing times of the scatter plot of cash usage, the initial cash replenishment prediction model corresponding to the smoothing times is selected from multiple initial cash replenishment prediction models to obtain the cash replenishment prediction model.

[0154] Each module in the aforementioned banknote replenishment prediction device can be implemented entirely or partially through software, hardware, or a combination thereof. These modules can be embedded in or independent of the processor in a computer device, or stored in the memory of a computer device as software, so that the processor can call and execute the corresponding operations of each module.

[0155] In one embodiment, a computer device is provided, which may be a terminal, and its internal structure diagram may be as follows: Figure 4 As shown, the computer device includes a processor, memory, communication interface, display screen, and input devices connected via a system bus. The processor provides computing and control capabilities. The memory includes non-volatile storage media and internal memory. The non-volatile storage media stores the operating system and computer programs. The internal memory provides an environment for the operation of the operating system and computer programs stored in the non-volatile storage media. The communication interface is used for wired or wireless communication with external terminals; wireless communication can be achieved through Wi-Fi, mobile cellular networks, NFC (Near Field Communication), or other technologies. When executed by the processor, the computer program implements a method for predicting the amount of cash added. The display screen can be an LCD screen or an e-ink display screen. The input devices can be a touch layer covering the display screen, buttons, a trackball, or a touchpad on the computer device's casing, or an external keyboard, touchpad, or mouse.

[0156] Those skilled in the art will understand that Figure 4 The structure shown is merely a block diagram of a portion of the structure related to the present application and does not constitute a limitation on the computer device to which the present application is applied. Specific computer devices may include more or fewer components than those shown in the figure, or combine certain components, or have different component arrangements.

[0157] In one embodiment, a computer device is provided, including a memory and a processor, wherein the memory stores a computer program, and the processor executes the computer program to implement the steps of the method described in any one of the first aspects.

[0158] In one embodiment, a computer-readable storage medium is provided having a computer program stored thereon, which, when executed by a processor, implements the steps of the method described in any one of the first aspects.

[0159] In one embodiment, a computer program product is provided, including a computer program that, when executed by a processor, implements the steps of the method described in any one of the first aspects.

[0160] It should be noted that the user information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data used for analysis, data stored, data displayed, etc.) involved in this application are all information and data authorized by the user or fully authorized by all parties.

[0161] Those skilled in the art will understand that all or part of the processes in the methods of the above embodiments can be implemented by a computer program instructing related hardware. The computer program can be stored in a non-volatile computer-readable storage medium, and when executed, it can include the processes of the embodiments of the above methods. Any references to memory, databases, or other media used in the embodiments provided in this application can include at least one of non-volatile and volatile memory. Non-volatile memory can include read-only memory (ROM), magnetic tape, floppy disk, flash memory, optical memory, high-density embedded non-volatile memory, resistive random access memory (ReRAM), magnetic random access memory (MRAM), ferroelectric random access memory (FRAM), phase change memory (PCM), graphene memory, etc. Volatile memory can include random access memory (RAM) or external cache memory, etc. By way of illustration and not limitation, RAM can take many forms, such as Static Random Access Memory (SRAM) or Dynamic Random Access Memory (DRAM). The databases involved in the embodiments provided in this application may include at least one type of relational database and non-relational database. Non-relational databases may include, but are not limited to, blockchain-based distributed databases. The processors involved in the embodiments provided in this application may be general-purpose processors, central processing units, graphics processing units, digital signal processors, programmable logic devices, quantum computing-based data processing logic devices, etc., and are not limited to these.

[0162] The technical features of the above embodiments can be combined in any way. For the sake of brevity, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0163] The embodiments described above are merely illustrative of several implementation methods of this application, and while the descriptions are specific and detailed, they should not be construed as limiting the scope of this patent application. It should be noted that those skilled in the art can make various modifications and improvements without departing from the concept of this application, and these all fall within the protection scope of this application. Therefore, the protection scope of this application should be determined by the appended claims.

Claims

1. A method for predicting the amount of cash added, characterized in that, The method includes: The system obtains current deposit information for each ATM at the target branch, environmental factor information of multiple influencing factors, and historical operating information of the target branch; the historical operating information of the target branch includes historical cash replenishment information for each ATM at the target branch. Based on the environmental factor information of each type and the current deposit information of each ATM, the current cash shortage status of each ATM is determined through a stock judgment strategy. For each ATM, based on the environmental factor information of each type, the current cash shortage status of the ATM, and the historical cash replenishment information of the ATM, the cash replenishment amount prediction model is input to predict the target cash replenishment amount of the ATM; Specifically, for each ATM, based on environmental factor information of each type, the current cash shortage status of the ATM, and the historical cash replenishment information of the ATM, a cash replenishment prediction model is input to predict the target cash replenishment amount for the ATM, including: Based on the historical cash replenishment information of the ATM, a scatter plot of cash usage of the ATM is established. Based on the scatter plot of cash usage, among multiple initial cash replenishment prediction models, a cash replenishment prediction model that meets the algorithm requirements of the scatter plot of cash usage is selected. Determine the weight value of each of the aforementioned influencing factor types, and based on the weight value of each of the aforementioned influencing factor types, perform a weighted summation of the sub-influencing factor values ​​corresponding to all current environmental data information in the environmental factor group of all influencing factor types to obtain the comprehensive influencing factor value corresponding to all influencing factor types; The comprehensive influencing factor values ​​are normalized to obtain the influencing factor weights corresponding to the comprehensive influencing factor values. The influencing factor weights, the cash shortage status of the ATM, and the current deposit information of the ATM are then input into the cash replenishment prediction model to obtain the target cash replenishment amount for the ATM. The method also includes using a similarity recognition algorithm to calculate the correlation between sub-influencing factors and influencing factor types for each current environmental data information in the environmental factor group.

2. The method according to claim 1, characterized in that, Obtain information on environmental factors of multiple influencing factor types, including: Multiple current environmental data are acquired, and through environmental analysis strategies, the sub-influencing factors of each current environmental data on each ATM of the target network are analyzed. Based on the correlation between the sub-influencing factors of each current environmental data, the current environmental data is clustered to obtain multiple environmental factor groups. Identify the sub-influencing factors corresponding to all environmental data information in each environmental factor group, and use the environmental factor group corresponding to each influencing factor type as the environmental factor information for each type.

3. The method according to claim 2, characterized in that, Based on environmental factor information of each type and current deposit information of each ATM, the current cash shortage status of each ATM is determined through a stock assessment strategy, including: Identify the sub-influencing factor values ​​of all current environmental data information in the environmental factor group corresponding to each influencing factor type; Based on the values ​​of all sub-influence factors corresponding to each influencing factor type, determine the influencing factor value of the influencing factor type; For each ATM, the influencing factor values ​​of all influencing factor types and the current deposit information of the ATM are input into the cash shortage judgment model to obtain the current cash shortage status of the ATM.

4. The method according to claim 3, characterized in that, The sub-influence factor values ​​of all current environmental data information in the environmental factor group corresponding to each influencing factor type are identified, including: For each type of influencing factor corresponding to an environmental factor group, the preset influencing factor range corresponding to each current environmental data information of the environmental factor group of the influencing factor type is identified, and the value of the preset influencing factor range corresponding to each current environmental data information is obtained. The value of the preset influencing factor range corresponding to each current environmental data information is used as the sub-influencing factor value corresponding to each current environmental data information.

5. The method according to claim 3, characterized in that, The step of determining the influencing factor value for each influencing factor type based on all sub-influencing factor values ​​corresponding to each influencing factor type includes: Calculate the correlation between the sub-influencing factors of each current environmental data information in the environmental factor group and the influencing factor type; The correlation degree corresponding to each sub-influencing factor is normalized to obtain the influence factor weight of each sub-influencing factor. The sub-influencing factor values ​​of each current environmental data information of the environmental factor group are weighted and summed to obtain the influence factor value corresponding to the influence factor type.

6. The method according to claim 3, characterized in that, Before inputting the values ​​of all influencing factor types and the current deposit information of the ATM into the inventory judgment model to obtain the current cash shortage status of the ATM, the method further includes: Obtain the sample influencing factor values ​​for multiple sample influencing factor types, sample deposit information for multiple sample ATMs, and sample cash shortage status for each sample ATM; For each sample cash shortage state, the sample influencing factor value of each sample influencing factor type and the sample deposit information of the sample ATM corresponding to each sample cash shortage state are input into the initial stock judgment model to obtain the first cash shortage state of each sample ATM. Based on the first cash shortage state of each sample ATM and the sample cash shortage state of each sample ATM, the initial inventory judgment model is trained to identify the parameters for each sample cash shortage state, thus obtaining the first inventory judgment model.

7. The method according to claim 1, characterized in that, The step of selecting a banknote replenishment prediction model that meets the algorithmic requirements of the banknote consumption scatter plot from multiple initial banknote replenishment prediction models includes: The scatter plot of cash usage is smoothed, and the curvature of the scatter plot of cash usage is identified; If the curvature is greater than the curvature threshold, return to the smoothing process for the scatter plot of cash usage until the curvature is less than the curvature threshold, and obtain the number of times the scatter plot of cash usage has been smoothed. Based on the smoothing times of the scatter plot of cash usage, the initial cash replenishment prediction model corresponding to the smoothing times is selected from multiple initial cash replenishment prediction models to obtain the cash replenishment prediction model.

8. A device for predicting the amount of cash added, characterized in that, The device includes: The acquisition module is used to acquire the current deposit information of each ATM at the target branch, environmental factor information of multiple influencing factors, and the historical operation information of the target branch; the historical operation information of the target branch includes the historical cash replenishment information of each ATM at the target branch. The determination module is used to determine the current cash shortage status of each ATM based on environmental factor information of each type and the current deposit information of each ATM, through a stock judgment strategy. The prediction module is used to predict the target amount of cash to be added to each ATM based on environmental factor information of each type, the current cash shortage status of the ATM, and the historical cash replenishment information of the ATM, by inputting a cash replenishment prediction model. The prediction module is further configured to: establish a scatter plot of cash usage for the ATM based on its historical cash replenishment information; and, based on the scatter plot, select a cash replenishment prediction model that meets the algorithm requirements of the scatter plot from multiple initial cash replenishment prediction models; determine the weight values ​​of each type of influencing factor; and, based on the weight values ​​of each type of influencing factor, perform weighted summation on the sub-influencing factor values ​​corresponding to all current environmental data information in the environmental factor groups of all influencing factor types to obtain the comprehensive influencing factor value corresponding to all influencing factor types; normalize the comprehensive influencing factor value to obtain the influencing factor weight corresponding to the comprehensive influencing factor value; and input the influencing factor weight, the cash shortage status of the ATM, and the current deposit information of the ATM into the cash replenishment prediction model to obtain the target cash replenishment amount for the ATM. The device is also used to calculate the correlation between sub-influencing factors and influencing factor types for each current environmental data information in the environmental factor group using a similarity recognition algorithm.

9. A computer device comprising a memory and a processor, wherein the memory stores a computer program, characterized in that, When the processor executes the computer program, it implements the steps of the method according to any one of claims 1 to 7.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by a processor, it implements the steps of the method according to any one of claims 1 to 7.

11. A computer program product, comprising a computer program, characterized in that, When the computer program is executed by a processor, it implements the steps of the method according to any one of claims 1 to 7.

Images