Self-service terminal payment data analysis method based on big data and self-service terminal

Abstract

The invention discloses a self-service terminal payment data analysis method based on big data and a self-service terminal, and the payment data analysis method comprises the steps: S1, collecting payment data from self-service terminals distributed at different geographic positions, and generating a comprehensive payment behavior data set containing user payment behavior characteristics; s2, based on the comprehensive payment behavior data set, analyzing rule characteristics of user payment behaviors, and generating a user payment behavior rule analysis result; and S3, based on the user payment behavior rule analysis result, in combination with the terminal geographic position characteristics and historical payment data, predicting the payment demand of the self-service terminal at preset time and place, and generating a payment demand prediction result. The accuracy of user behavior analysis is enhanced by mining the user payment habit recessive mode and combining with the social platform behavior. Payment habits and consumption characteristics of the user in different areas and different time periods can be reflected.

Description

Technical Field

[0001] The present invention relates to the technical field of self-service terminals, and in particular to a self-service terminal payment data analysis method based on big data and a self-service terminal. Background Art

[0002] With the widespread use of self-service terminals and mobile payments, the demand for payment systems to improve convenience, security, and data analysis capabilities is growing. The proliferation of self-service terminals (such as vending machines and unmanned supermarkets) and mobile payment technology has enabled users to complete transactions anytime, anywhere. This has not only increased payment convenience but also generated a massive amount of payment data. This data provides valuable resources for optimizing payment systems, enhancing user experience, and preventing fraud risks. However, efficiently processing and analyzing this data has become a key challenge in the development of payment systems.

[0003] Due to the widespread distribution of self-service terminals and the fragmented data sources, traditional methods struggle to collect and integrate raw payment data, including transaction time, amount, and terminal identification, in real time. This results in inefficient data cleaning and preprocessing, impacting the quality of the resulting comprehensive payment behavior dataset. Existing technologies lack systematic analysis of user payment behavior, effectively extracting patterns in the timing and amount of user payments, and struggle to accurately grasp trends in user payment behavior. This is particularly evident in the analysis of payment behavior in different geographic locations. Existing technologies often only consider simple historical data when predicting payment demand at self-service terminals, failing to fully integrate the terminal's geographic location characteristics and user payment behavior patterns. This results in low accuracy and reliability in the prediction results, making it impossible to provide effective decision support for terminal operations. Traditional payment demand forecasting methods lack dynamic optimization mechanisms, making it impossible to flexibly adjust the prediction model based on actual payment scenarios, and lack an effective confidence assessment system for prediction results.

[0004] Therefore, there is an urgent need for a self-service terminal payment data analysis method and a self-service terminal based on big data. Summary of the Invention

[0005] The present invention provides a self-service terminal payment data analysis method based on big data and a self-service terminal to solve the above-mentioned problems existing in the prior art.

[0006] In order to achieve the above object, the present invention provides the following technical solutions:

[0007] A self-service terminal payment data analysis method based on big data, comprising:

[0008] S1: Collect payment data from self-service terminals distributed in different geographical locations to generate a comprehensive payment behavior dataset containing user payment behavior characteristics;

[0009] S2: Based on the comprehensive payment behavior dataset, analyze the regular characteristics of user payment behavior and generate analysis results of user payment behavior patterns;

[0010] S3: Based on the analysis results of user payment behavior patterns, combined with the terminal's geographical location characteristics and historical payment data, predict the payment demand of the self-service terminal at the preset time and location, and generate payment demand prediction results.

[0011] The S1 step includes:

[0012] S11: Collecting raw payment data streams including transaction time, transaction amount, and terminal identification from multiple self-service terminals in real time;

[0013] S12: Cleaning and preprocessing the original payment data stream to generate cleaned payment data;

[0014] S13: Integrate the cleaned payment data to generate a comprehensive payment behavior dataset.

[0015] Among them, step S2 includes:

[0016] S21: Grouping the comprehensive payment behavior dataset by user and geographic location to generate grouped payment data;

[0017] S22: Based on the grouped payment data, extract the time pattern and amount pattern characteristics of the user's payment to generate payment pattern characteristics;

[0018] S23: Based on the payment pattern characteristics, analyze the trend of the user's payment behavior and generate the user's payment behavior pattern analysis results.

[0019] The S3 step includes:

[0020] S31: Based on the analysis results of user payment behavior patterns and combined with the terminal's geographical location characteristics, a payment demand prediction model is constructed;

[0021] S32: Based on the payment demand prediction model and historical payment data, predict the payment demand at the preset time and location, and generate a preliminary payment demand prediction;

[0022] S33: Conduct a confidence assessment on the preliminary payment demand forecast based on historical deviation rules to generate a payment demand forecast result including a confidence interval.

[0023] Among them, step S11 includes:

[0024] S111: When a payment transaction occurs at the self-service terminal, the original data stream of the payment transaction is collected to generate a payment record set including the transaction time, amount, and terminal identification;

[0025] S112: Based on the payment record set, extract the real-time transaction information of each self-service terminal to generate an original payment data stream;

[0026] S113: Associating the original payment data stream with the geographic location information to form a payment data stream with a location identifier.

[0027] Wherein, step S22 includes:

[0028] S221: Based on the grouped payment data, calculate the transaction frequency and amount distribution of each user in different time periods to generate a time regularity feature;

[0029] S222: Combine temporal regularity features with geographic location data to analyze the temporal and spatial correlation of payment behaviors and generate comprehensive regularity features;

[0030] S223: Standardize the comprehensive regular features to generate payment regular features.

[0031] Wherein, step S23 includes:

[0032] S231: Based on payment pattern characteristics, traverse payment behavior data of different self-service terminals in sequence;

[0033] S232: Each time a traversal is made, the deviation between the payment behavior of the traversed self-service terminal and the historical payment pattern is calculated and used as an abnormality indicator of the terminal;

[0034] S233: After traversing all self-service terminals, the payment pattern characteristics and abnormal indicators of different self-service terminals are integrated to generate user payment behavior pattern analysis results.

[0035] Among them, step S32 includes:

[0036] S321: Generate payment demand impact parameters based on the geographical location characteristics of the preset time and place and in combination with historical payment data;

[0037] S322: Generate multiple fuzzy payment demands and their demand degrees based on the pre-matched prediction optimization template of the payment demand influencing parameters;

[0038] S323: Based on the fuzzy payment demand and demand degree, combined with the weight of the prediction optimization template, the payment demand prediction model is adjusted to generate a preliminary payment demand prediction.

[0039] Among them, self-service terminals include:

[0040] Data collection module, used to collect original payment data from self-service terminals and generate payment data stream;

[0041] The payment data analysis module is used to receive payment data streams, analyze and predict payment demand, and generate payment demand prediction results;

[0042] The execution module is used to receive the payment demand prediction result, generate control instructions and transmit them to the self-service terminal, and the self-service terminal performs corresponding operations based on the control instructions.

[0043] Among them, also include:

[0044] The device identification module is used to verify the permissions of the connected self-service terminal device and generate terminal identification information in a preset coding format.

[0045] Compared with the prior art, the present invention has the following advantages:

[0046] The big data-based self-service terminal payment data analysis method includes the following steps: S1: collecting payment data from self-service terminals distributed across different locations to generate a comprehensive payment behavior dataset containing user payment behavior characteristics; S2: analyzing the regular characteristics of user payment behavior based on the comprehensive payment behavior dataset to generate a payment behavior pattern analysis result; and S3: based on the user payment behavior pattern analysis result, combining the terminal's geographic location characteristics and historical payment data, predicting payment demand at the self-service terminal at a preset time and location to generate a payment demand prediction result. By mining implicit patterns in user payment habits and integrating them with social platform behavior, the accuracy of user behavior analysis is enhanced. This method can reflect user payment habits and consumption characteristics in different regions and time periods.

[0047] Other features and advantages of the present invention will be set forth in the description which follows, and in part will be obvious from the description, or may be learned by practice of the present invention.

[0048] The technical solution of the present invention is further described in detail below through the accompanying drawings and embodiments. BRIEF DESCRIPTION OF THE DRAWINGS

[0049] The accompanying drawings are used to provide a further understanding of the present invention and constitute a part of the specification. Together with the embodiments of the present invention, they are used to explain the present invention and do not constitute a limitation of the present invention. In the accompanying drawings:

[0050] Figure 1 This is a flow chart of a method for analyzing self-service terminal payment data based on big data in an embodiment of the present invention;

[0051] Figure 2 A flowchart of generating a comprehensive payment behavior dataset in an embodiment of the present invention;

[0052] Figure 3 This is a structural diagram of a self-service terminal based on big data in an embodiment of the present invention. DETAILED DESCRIPTION

[0053] The preferred embodiments of the present invention are described below with reference to the accompanying drawings. It should be understood that the preferred embodiments described herein are only used to illustrate and explain the present invention, and are not used to limit the present invention.

[0054] The embodiment of the present invention provides Figure 1 As shown, the self-service terminal payment data analysis method based on big data includes:

[0055] S1: Collect payment data from self-service terminals distributed in different geographical locations to generate a comprehensive payment behavior dataset containing user payment behavior characteristics;

[0056] S2: Based on the comprehensive payment behavior dataset, analyze the regular characteristics of user payment behavior and generate analysis results of user payment behavior patterns;

[0057] S3: Based on the analysis results of user payment behavior patterns, combined with the terminal's geographical location characteristics and historical payment data, predict the payment demand of the self-service terminal at the preset time and location, and generate payment demand prediction results.

[0058] The working principle of the above technical solution is as follows: S1: Payment data (including fields such as transaction time, amount, and terminal number) is collected from self-service terminals (self-service terminals refer to payment devices deployed in public places, such as vending machines and ATMs) distributed in different geographical locations. This data set is then generated (a structured data set after the integration and cleaning of multiple terminals) containing user payment behavior characteristics. User payment behavior characteristics include transaction time, transaction amount, and terminal identification. Through geographically distributed collection, the data set reflects regional consumption differences.

[0059] S2: Based on a comprehensive payment behavior dataset, analyze the regular characteristics of user payment behavior (regular characteristics refer to periodic transaction patterns extracted from the dataset, including high-frequency transactions during weekday lunchtime hours) and generate analysis results of user payment behavior patterns. The analysis process includes grouping the dataset by user, calculating transaction time distribution density and transaction amount statistics, and then combining geolocation tags to identify spatiotemporal correlations, such as high transaction frequencies during specific time periods and locations. The analysis results include payment pattern models and anomaly indicators for each terminal.

[0060] S3: Based on the analysis results of user payment behavior patterns, combined with the terminal's geographical location characteristics (including the terminal's surrounding environment attributes, such as commercial area / residential area) and historical payment data, the payment demand of the self-service terminal at the preset time and location is predicted, and the payment demand prediction results are generated. Among them, the prediction process inputs the pattern analysis results into the time series model, superimposes the seasonal fluctuation factors in the historical payment data, and outputs the demand in a specific period in the future (such as predicting the number of transactions that the mall terminal needs to process in a certain time period).

[0061] The beneficial effect of this technical solution is that the dataset can reflect users' payment habits and consumption characteristics in different regions and time periods. Through data mining techniques, it can identify payment patterns with significant characteristics. The prediction results can be used to make operational decisions such as terminal equipment maintenance and replenishment, cash reserve management, etc.

[0062] In another embodiment, if Figure 2 As shown, step S1 includes:

[0063] S11: Collecting raw payment data streams including transaction time, transaction amount, and terminal identification from multiple self-service terminals in real time;

[0064] S12: Cleaning and preprocessing the original payment data stream to generate cleaned payment data;

[0065] S13: Integrate the cleaned payment data to generate a comprehensive payment behavior dataset.

[0066] The working principle of the above technical solution is as follows: S11: Real-time collection of raw payment data streams from multiple self-service terminals; wherein self-service terminals refer to payment devices deployed in different geographical locations, including self-service vending machines, ATMs, and other terminal devices with payment functions; each terminal device is equipped with a data collection module for real-time recording of payment data generated during the transaction process; the payment data mainly includes the specific time of the transaction, the actual transaction amount, and a unique identification code for identifying the terminal device; the data collection module transmits the collected payment data to the data processing center in real time via the terminal's built-in communication unit;

[0067] S12: Clean and pre-process the original payment data stream to generate cleaned payment data. The data cleaning process includes: standardizing the format of payment data and converting data transmitted by different terminals into a standard format; deleting duplicate records and outliers, such as data that clearly exceeds the normal transaction amount range; supplementing missing fields and structuring the data. The pre-processing process includes: standardizing timestamps to ensure consistency in time format; normalizing amount data to facilitate subsequent analysis; and encoding terminal identifiers to establish a corresponding relationship between terminal location and identifier.

[0068] S13: Integrate the cleaned payment data to generate a comprehensive payment behavior dataset; the data integration process includes: sorting the cleaned payment data in time series; grouping the data according to the terminal identifier to form payment behavior characteristics in different geographical locations; establishing a payment behavior characteristic index, including a time dimension index, an amount dimension index, and a location dimension index; the generation of the comprehensive payment behavior dataset includes: storing the integrated data as a structured data table, with each record containing fields such as transaction time, transaction amount, and terminal identifier; establishing a query interface for the dataset to support data retrieval and analysis by dimensions such as time, amount, and location.

[0069] The beneficial effect of the above technical solution is that the data set can be used to analyze the payment behavior characteristics in different geographical locations and different time periods, providing data support for business decision-making.

[0070] In another embodiment, step S2 includes:

[0071] S21: Grouping the comprehensive payment behavior dataset by user and geographic location to generate grouped payment data;

[0072] S22: Based on the grouped payment data, extract the time pattern and amount pattern characteristics of the user's payment to generate payment pattern characteristics;

[0073] S23: Based on the payment pattern characteristics, analyze the trend of the user's payment behavior and generate the user's payment behavior pattern analysis results.

[0074] The working principle of the above technical solution is: S21: Group the comprehensive payment behavior data set by user and geographic location to generate grouped payment data; wherein, the comprehensive payment behavior data set includes information such as user ID, transaction time, transaction amount, transaction location, etc.; the grouping process first preliminarily classifies the data according to user ID, and then performs secondary grouping based on the transaction location information to form grouped payment data with user-location as the dimension; the grouped payment data includes all transaction records of each user at different locations.

[0075] S22: Based on the grouped payment data, extract the time pattern and amount pattern characteristics of the user's payment to generate payment pattern characteristics; among them, the time pattern characteristics include: the transaction time distribution of users on weekdays and non-working days, daily transaction peak period, transaction interval, etc.; the amount pattern characteristics include: the transaction amount range of users in different time periods, the statistical characteristics of the single transaction amount, etc.; the payment pattern feature extraction process is to perform statistical analysis on the grouped data, calculate the distribution density of the transaction time, identify the changing trend of the transaction amount, and thus summarize the user's payment habit characteristics.

[0076] S23: Based on the payment pattern characteristics, analyze the trend of user payment behavior and generate the analysis results of the user payment behavior pattern; wherein, the payment behavior trend analysis includes: identifying the user's transaction preferences in specific time and space scenarios, judging the periodic characteristics of transaction behavior, discovering abnormal transaction patterns, etc.; the analysis process is to input the extracted payment pattern characteristics into a pre-trained behavior analysis model, which is established by learning the regular characteristics in historical transaction data and can accurately identify the user's payment behavior pattern; the analysis results include the user's typical payment scenarios, time patterns, amount characteristics and potential abnormal transaction indicators.

[0077] Additionally, it includes:

[0078] Regularly update and optimize the payment behavior analysis model based on historical transaction data;

[0079] Adjust model parameters based on the latest transaction data characteristics to improve analysis accuracy.

[0080] Regular model updates ensure that analysis results promptly reflect changing trends in user payment behavior, improving the efficiency of identifying abnormal transactions. The update process involves collecting new transaction data, extracting features, and then retraining the model to adapt to evolving payment scenarios and user habits. The model update frequency can be dynamically adjusted based on data growth and analysis needs, ensuring timely analysis while avoiding resource waste caused by excessive updates.

[0081] The beneficial effects of the above technical solution are: through multi-dimensional analysis and pattern mining of self-service terminal payment data, the terminal layout can be optimized and more accurate services can be provided, thereby improving the operational efficiency and user experience of the self-service terminals.

[0082] In another embodiment, step S3 includes:

[0083] S31: Based on the analysis results of user payment behavior patterns and combined with the terminal's geographical location characteristics, a payment demand prediction model is constructed;

[0084] S32: Based on the payment demand prediction model and historical payment data, predict the payment demand at the preset time and location, and generate a preliminary payment demand prediction;

[0085] S33: Conduct a confidence assessment on the preliminary payment demand forecast based on historical deviation rules to generate a payment demand forecast result including a confidence interval.

[0086] The working principle of the above technical solution is as follows: S31: constructing a payment demand prediction model based on analysis of user payment behavior patterns; wherein, the user payment behavior patterns are obtained by collecting historical transaction data of self-service terminals, including information such as transaction time, transaction amount, and transaction frequency; the terminal geographical location characteristics include the location environment of the terminal, such as commercial area, residential area, office area, etc., as well as the surrounding pedestrian density, commercial activity, etc.; the construction of the payment demand prediction model is to use the user payment behavior characteristics and the terminal geographical location characteristics as input parameters, and establish a prediction relationship through time series analysis methods;

[0087] S32: Utilizing the constructed payment demand prediction model and combining it with historical payment data, a prediction analysis is performed. The prediction analysis process includes: inputting geographic location characteristics of the terminal to be predicted into the prediction model, overlaying the cyclical variation characteristics of the historical payment data at that location, such as the difference between weekdays and holidays, and fluctuation patterns at different times of the day, to generate a preliminary payment demand prediction result. The prediction result includes information such as the expected number of transactions and transaction amounts within a preset time period.

[0088] S33: Conduct confidence assessment on the prediction results and output the final prediction results. Confidence assessment is achieved by analyzing the deviation patterns between historical prediction results and actual conditions. The assessment process takes into account factors such as the prediction time span, the adequacy of historical data, and the impact of emergencies. The final output prediction results include the predicted value and the corresponding confidence interval, providing a decision-making basis for terminal operation management.

[0089] Based on the cluster analysis of terminal locations, the terminal groups with similar environmental characteristics are modeled as a whole;

[0090] By analyzing the entire terminal group, the generalization ability and prediction accuracy of the prediction model for a single terminal can be improved.

[0091] This group analysis method can make full use of data features in similar scenarios to improve the reliability of the prediction model, and can also provide support for the rapid construction of prediction models for newly added terminals.

[0092] The beneficial effects of the above technical solution are: adjusting the prediction results according to the peak transaction volume, further optimizing the terminal operation efficiency.

[0093] In another embodiment, step S11 includes:

[0094] S111: When a payment transaction occurs at the self-service terminal, the original data stream of the payment transaction is collected to generate a payment record set including the transaction time, amount, and terminal identification;

[0095] S112: Based on the payment record set, extract the real-time transaction information of each self-service terminal to generate an original payment data stream;

[0096] S113: Associating the original payment data stream with the geographic location information to form a payment data stream with a location identifier.

[0097] The working principle of the above technical solution is as follows: S111: When a payment transaction occurs at a self-service terminal, the original data stream of the payment transaction is collected to generate a payment record set containing the transaction time, amount and terminal identification; wherein, the self-service terminal refers to a terminal device with payment function deployed in different geographical locations, including self-service vending machines, ATM machines, and self-service payment machines; each self-service terminal is equipped with a data collection module, which records and transmits transaction data in real time through a built-in communication unit; the data collected by the data collection module includes: the specific timestamp of the transaction, the actual transaction amount paid, an identification code for uniquely identifying the terminal device, and payment method information; the collected data is sent to the data processing center in real time through the built-in communication module of the terminal device using encrypted transmission;

[0098] S112: Based on the payment record set, extract the real-time transaction information of each self-service terminal and generate the original payment data stream. After receiving the payment records, the data processing center first cleans and pre-processes the data, including removing duplicate data, supplementing missing values, and unifying the data format. Then, the data is classified and organized according to the terminal identifier, and the transaction time sequence information of each terminal is extracted, converting the discrete transaction records into a continuous data stream. The data stream contains information such as transaction frequency, transaction amount distribution, and transaction time patterns.

[0099] S113: Associating the original payment data stream with geographic location information to form a payment data stream with location identification; wherein the geographic location information includes the installation address of the self-service terminal, the region to which it belongs, the surrounding environment characteristics, etc.; by associating the location coordinates of each terminal with the payment data stream, a transaction data stream with spatial attributes is formed; this association is achieved by mapping the unique identification code of the terminal device with a pre-established terminal location database; the payment data stream with location identification can reflect the transaction characteristics and patterns of different regions.

[0100] The beneficial effects of the above technical solution are: by real-time collection and association of payment data streams with geographical locations, a complete and accurate data basis is provided for subsequent big data analysis, which helps to fully understand the usage and spatial distribution characteristics of self-service terminals.

[0101] In another embodiment, step S22 includes:

[0102] S221: Based on the grouped payment data, calculate the transaction frequency and amount distribution of each user in different time periods to generate a time regularity feature;

[0103] S222: Combine temporal regularity features with geographic location data to analyze the temporal and spatial correlation of payment behaviors and generate comprehensive regularity features;

[0104] S223: Standardize the comprehensive regular features to generate payment regular features.

[0105] The working principle of the above technical solution is: S221: Based on the grouped payment data, the transaction frequency and amount distribution of each user in different time periods are calculated to generate time regularity features; wherein, the grouped payment data refers to the self-service terminal transaction record data sorted and classified according to dimensions such as user ID, transaction time, and transaction amount; the transaction frequency refers to the number of times a user completes payment within a specific time period (such as different time periods of a day, working days / non-working days); the amount distribution refers to the transaction amount range of the user in different time periods and its frequency of occurrence; the time regularity features include the user's transaction time distribution pattern, transaction interval pattern and other time dimension features.

[0106] S222: Combine temporal regularity features with geographic location data to analyze the spatiotemporal correlation of payment behaviors and generate comprehensive regularity features; wherein, geographic location data refers to the geographic coordinate information of the self-service terminal and the type of scene it is located in (such as shopping malls, communities, transportation hubs, etc.); spatiotemporal correlation analysis refers to the correlation analysis of users' payment behaviors at different times and locations to identify users' payment preferences under specific spatiotemporal conditions; comprehensive regularity features include the combined features of temporal regularity and spatial regularity.

[0107] S223: Standardize the comprehensive regularity features to generate payment regularity features. Standardization refers to unifying feature data from different dimensions into the same numerical range to eliminate dimensionality effects. Payment regularity features are standardized user payment behavior features, including user payment habits in time and space.

[0108] The beneficial effects of the above technical solution are: by analyzing the time distribution pattern of user payment data, combining it with geographic location information to conduct spatiotemporal correlation analysis, and finally generating standardized payment pattern characteristics, the user's payment behavior pattern is effectively extracted, providing a reliable feature basis for subsequent payment behavior analysis.

[0109] In another embodiment, step S23 includes:

[0110] S231: Based on payment pattern characteristics, traverse payment behavior data of different self-service terminals in sequence;

[0111] S232: Each time a traversal is made, the deviation between the payment behavior of the traversed self-service terminal and the historical payment pattern is calculated and used as an abnormality indicator of the terminal;

[0112] S233: After traversing all self-service terminals, the payment pattern characteristics and abnormal indicators of different self-service terminals are integrated to generate user payment behavior pattern analysis results.

[0113] The working principle of the above technical solution is as follows: payment regularity refers to the fixed behavioral pattern characteristics exhibited by users when making payment transactions on self-service terminals, such as transaction time patterns, transaction amount ranges, and transaction frequency; payment behavior data includes information such as the time, location, amount, and type of transaction; historical payment patterns refer to the set of normal user payment behavior characteristics obtained by analyzing historical transaction data; and the deviation degree indicates the degree of difference between the current payment behavior and the historical payment pattern. The greater the deviation degree, the less consistent the current payment behavior is with the user's historical payment habits.

[0114] Specifically, the system first obtains payment behavior data from each self-service terminal and extracts payment pattern characteristics. It then compares the current payment pattern characteristics with the terminal's historical payment patterns, calculates the degree of deviation between the two, and obtains anomaly indicators reflecting the degree of abnormal payment behavior. Finally, it integrates and analyzes the payment pattern characteristics and corresponding anomaly indicators of all terminals to generate a complete analysis of user payment behavior patterns. This result can be used to determine whether there are any abnormalities in user payment behavior. In this way, abnormal payment behavior can be detected in a timely manner, providing a basis for risk prevention and control.

[0115] The beneficial effects of the above technical solution are: by traversing and analyzing the payment behavior data of different self-service terminals, calculating the deviation between payment behavior and historical payment patterns, and integrating them to generate analysis results of user payment behavior patterns, the analysis accuracy of self-service terminal payment behavior and the ability to identify anomalies are effectively improved.

[0116] In another embodiment, step S32 includes:

[0117] S321: Generate payment demand impact parameters based on the geographical location characteristics of the preset time and place and in combination with historical payment data;

[0118] S322: Generate multiple fuzzy payment demands and their demand degrees based on the pre-matched prediction optimization template of the payment demand influencing parameters;

[0119] S323: Based on the fuzzy payment demand and demand degree, combined with the weight of the prediction optimization template, the payment demand prediction model is adjusted to generate a preliminary payment demand prediction.

[0120] The working principle of the above technical solution is as follows: in S321, the geographical location characteristics of the preset time and place refer to the environmental characteristics of the location of the self-service terminal, such as: commercial area, residential area, school area, etc.; historical payment data refers to the payment transaction information recorded in the area, including transaction time, amount, frequency and other data; payment demand influencing parameters refer to key indicators that can affect the change of payment demand, such as: the change pattern of regional pedestrian flow, consumer payment habits, and the surrounding business activity. Specifically, by analyzing the distribution characteristics of historical payment data of a specific area in different time periods, combined with the geographical location attributes of the area, key parameters that can characterize the change pattern of payment demand are extracted;

[0121] In S322, the prediction optimization template refers to a pre-established payment demand prediction basic model framework, which includes multiple preset demand scenario patterns; fuzzy payment demand refers to possible payment transactions inferred based on existing data, such as morning dining consumption demand, holiday shopping demand, etc.; demand degree refers to the probability of a certain type of payment demand occurring; specifically, the payment demand influencing parameters are input into the prediction optimization template, and the scenario matching rules preset in the template are used to identify various types of payment demands that may occur under the current conditions, and the probability of occurrence of each type of demand is calculated to form a demand degree index;

[0122] In S323, the weight of the prediction optimization template refers to the importance of different prediction factors in the template, which is used to adjust the influence of each factor on the final prediction result; the payment demand prediction model refers to a mathematical model that can predict payment transactions in the future; the preliminary payment demand prediction refers to the expected payment transaction data output by the model, including the expected number of transactions, amount and other information; specifically, based on the identified fuzzy payment demand and its demand degree, combined with the weight parameters of each prediction factor in the prediction optimization template, the payment demand prediction model is optimized and adjusted to generate a payment demand prediction result that is more in line with the actual situation.

[0123] The beneficial effects of the above technical solution are: based on the geographical location characteristics of the preset time and place, combined with historical payment data, payment demand influencing parameters are generated, and based on the parameters, the prediction optimization template is pre-matched to generate multiple fuzzy payment demands and their demand degrees. Finally, based on the fuzzy payment demands and demand degrees, combined with the weights of the prediction optimization template, the payment demand prediction model is adjusted to generate a preliminary payment demand forecast, thereby realizing accurate analysis and prediction of self-service terminal payment data and improving the pertinence and efficiency of payment services.

[0124] In another embodiment, if Figure 3 As shown, the self-service terminal includes:

[0125] Data collection module, used to collect original payment data from self-service terminals and generate payment data stream;

[0126] The payment data analysis module is used to receive payment data streams, analyze and predict payment demand, and generate payment demand prediction results;

[0127] The execution module is used to receive the payment demand prediction result, generate control instructions and transmit them to the self-service terminal, and the self-service terminal performs corresponding operations based on the control instructions.

[0128] The execution module includes:

[0129] The scheduling calculation submodule is used to calculate the task sequence including the device access order and time interval based on the spatiotemporal distribution characteristics in the prediction data and the geographical location of the terminal device;

[0130] The instruction generation submodule is used to convert the task sequence into an instruction data packet containing a preset operation code, an execution timestamp and operation parameters, and transmit the instruction data packet to the corresponding self-service terminal device through a set protocol channel;

[0131] The feedback processing submodule is used to receive the execution result code and device operating parameters returned by the terminal device, convert the operating parameters into a data structure in a set format, and provide it to the payment data analysis module for optimizing the prediction results.

[0132] The working principle of this technical solution is as follows: The data collection module, based on IoT technology, collects payment data from self-service terminals in different geographical locations in real time, including transaction time, transaction amount, terminal number, and other information. During the collection process, the raw data is first cleaned and standardized to remove outliers and redundant information. The processed data is then integrated into a structured payment data stream. This data stream contains comprehensive payment behavior characteristics, reflecting user payment habits in different regions and over different time periods.

[0133] The payment data analysis module receives standardized payment data streams and uses time series analysis to identify the temporal patterns and spatial distribution characteristics of user payment behavior. The analysis process first groups the data by terminal location, calculates the transaction density distribution for each time period, and extracts periodic payment patterns. Then, combining the terminal's geographic location attributes, it builds a payment behavior prediction model. This model, trained using historical data, can predict payment demand in specific areas and time periods.

[0134] Based on payment demand forecasts, the execution module generates targeted terminal control instructions. These instructions include adjusting cash reserves, setting transaction limits, and scheduling maintenance schedules. The execution module transmits these instructions via a secure channel to the corresponding self-service terminal. Upon receiving the instructions, the terminal device executes the corresponding operational adjustments, dynamically optimizing terminal service capabilities.

[0135] The beneficial effects of this technical solution include: through real-time data collection and analysis, accurate prediction and dynamic adjustment of self-service terminal payment demand can be achieved, improving terminal operational efficiency and service quality. The system's various modules work together to form a closed-loop, data-driven management mechanism, which is particularly suitable for the intelligent operation and management of large-scale self-service terminal networks.

[0136] In another embodiment, further comprising:

[0137] The device identification module is used to verify the permissions of the connected self-service terminal device and generate terminal identification information in a preset coding format.

[0138] The device identification module includes:

[0139] The zero-knowledge verification submodule is used to perform an identity verification process based on the zero-knowledge proof protocol on the connected self-service terminal device, ensuring that permission verification is completed without leaking sensitive device information;

[0140] The terminal registration submodule is used to receive the verification result of the zero-knowledge verification submodule, assign unique identification information to the self-service terminal device that passes the verification, and record the geographical location coordinate information of the terminal device;

[0141] The data storage submodule is used to save the unique identification information, geographic location coordinate information and equipment operation records to a data storage space with a preset structure;

[0142] The status monitoring submodule is used to determine the device status by calculating the transaction frequency change rate, capital flow ratio and hardware operating parameters of the terminal device, and send a notification message containing the specific parameter deviation value to the execution module when it detects that the parameters exceed the predetermined range.

[0143] The working principle of this technical solution is as follows: The device identification module performs identity authentication and permission management for self-service terminals connected to the system. The authentication process first verifies the terminal's hardware signature, then checks the validity of the terminal's security certificate, and finally generates unique terminal identification information. This identification information is encrypted to ensure secure data transmission between the terminal and the system.

[0144] The beneficial effects of the above technical solution are: achieving accurate prediction and dynamic adjustment of self-service terminal payment needs, and improving terminal operation efficiency and service quality.

[0145] Obviously, those skilled in the art can make various changes and modifications to the present invention without departing from the spirit and scope of the invention.

Claims

1. A self-service terminal payment data analysis method based on big data, characterized in that: include: S1: Collect payment data from self-service terminals distributed in different geographical locations to generate a comprehensive payment behavior dataset containing user payment behavior characteristics; S2: Based on the comprehensive payment behavior dataset, analyze the regular characteristics of user payment behavior and generate analysis results of user payment behavior patterns; S3: Based on the analysis results of user payment behavior patterns, combined with the terminal's geographical location characteristics and historical payment data, predict the payment demand of the self-service terminal at the preset time and location, and generate payment demand prediction results.

2. The self-service terminal payment data analysis method based on big data according to claim 1, characterized in that: Step S1 includes: S11: Collecting raw payment data streams including transaction time, transaction amount, and terminal identification from multiple self-service terminals in real time; S12: Cleaning and preprocessing the original payment data stream to generate cleaned payment data; S13: Integrate the cleaned payment data to generate a comprehensive payment behavior dataset.

3. The self-service terminal payment data analysis method based on big data according to claim 1, characterized in that: Step S2 includes: S21: Grouping the comprehensive payment behavior dataset by user and geographic location to generate grouped payment data; S22: Based on the grouped payment data, extract the time pattern and amount pattern characteristics of the user's payment to generate payment pattern characteristics; S23: Based on the payment pattern characteristics, analyze the trend of the user's payment behavior and generate the user's payment behavior pattern analysis results.

4. The self-service terminal payment data analysis method based on big data according to claim 1, characterized in that: The S3 steps include: S31: Based on the analysis results of user payment behavior patterns and combined with the terminal's geographical location characteristics, a payment demand prediction model is constructed; S32: Based on the payment demand prediction model and historical payment data, predict the payment demand at the preset time and location, and generate a preliminary payment demand prediction; S33: Conduct a confidence assessment on the preliminary payment demand forecast based on historical deviation rules to generate a payment demand forecast result including a confidence interval.

5. The self-service terminal payment data analysis method based on big data according to claim 2, characterized in that: Step S11 includes: S111: When a payment transaction occurs at the self-service terminal, the original data stream of the payment transaction is collected to generate a payment record set including the transaction time, amount, and terminal identification; S112: Based on the payment record set, extract the real-time transaction information of each self-service terminal to generate an original payment data stream; S113: Associating the original payment data stream with the geographic location information to form a payment data stream with a location identifier.

6. The self-service terminal payment data analysis method based on big data according to claim 3 is characterized in that: Step S22 includes: S221: Based on the grouped payment data, calculate the transaction frequency and amount distribution of each user in different time periods to generate a time regularity feature; S222: Combine temporal regularity features with geographic location data to analyze the temporal and spatial correlation of payment behaviors and generate comprehensive regularity features; S223: Standardize the comprehensive regular features to generate payment regular features.

7. The self-service terminal payment data analysis method based on big data according to claim 3 is characterized in that: Step S23 includes: S231: Based on payment pattern characteristics, traverse payment behavior data of different self-service terminals in sequence; S232: Each time a traversal is made, the deviation between the payment behavior of the traversed self-service terminal and the historical payment pattern is calculated and used as an abnormality indicator of the terminal; S233: After traversing all self-service terminals, the payment pattern characteristics and abnormal indicators of different self-service terminals are integrated to generate user payment behavior pattern analysis results.

8. The self-service terminal payment data analysis method based on big data according to claim 4, characterized in that: Step S32 includes: S321: Generate payment demand impact parameters based on the geographical location characteristics of the preset time and place and in combination with historical payment data; S322: Generate multiple fuzzy payment demands and corresponding demand degrees based on the pre-matched prediction optimization template of the payment demand influencing parameters; S323: Based on the fuzzy payment demand and demand degree, combined with the weight of the prediction optimization template, the payment demand prediction model is adjusted to generate a preliminary payment demand prediction.

9. A self-service terminal using the method for analyzing payment data based on big data self-service terminals according to any one of claims 1 to 8, characterized in that: include: Data collection module, used to collect original payment data from self-service terminals and generate payment data stream; The payment data analysis module is used to receive payment data streams, analyze and predict payment demand, and generate payment demand prediction results; The execution module is used to receive the payment demand prediction result, generate control instructions and transmit them to the self-service terminal, and the self-service terminal performs corresponding operations based on the control instructions.

10. The self-service terminal based on big data according to claim 9, characterized in that: Also includes: The device identification module is used to verify the permissions of the connected self-service terminal device and generate terminal identification information in a preset coding format.

Images