Payment and settlement service system based on big data technology

Abstract

The invention relates to the technical field of financial data processing, in particular to a payment and settlement service system based on a big data technology, and the system comprises a data collection module, a user payment behavior recognition module, a payment channel evaluation module, and a payment execution and monitoring module. According to the invention, through deep analysis of transaction data and dynamic user portrait construction, consumption modes and preference trends of users can be captured and analyzed more accurately, and through extraction of periodic purchase behaviors and sudden change preferences, payment experience is highly personalized, customer service quality is greatly improved, payment channel assessment is comprehensive, and user experience is improved. The method includes real-time performance monitoring and cost analysis, guarantees the optimization of the payment process, effectively improves the processing speed and safety, reduces the fraud risk, enhances the transaction safety and transparency through the implementation of full-chain encryption and the real-time tracking of the payment progress, and improves the response speed to potential abnormalities.

Description

Technical Field

[0001] The present invention relates to the field of financial data processing technology, and in particular to a payment settlement service system based on big data technology. Background Art

[0002] Financial data processing is a technical field that focuses on managing and analyzing large amounts of data related to financial transactions. It captures, stores and analyzes data from various financial transactions with the aim of improving transaction efficiency, risk management, customer service and compliance. It uses advanced computing technologies, such as big data analysis, cloud computing and artificial intelligence, to process fast, large-volume and complex financial information, helping banks, investment companies, insurance companies and other institutions better understand market dynamics, customer behavior and internal operational efficiency.

[0003] Among them, the payment settlement service system based on big data technology involves the use of big data technology to optimize the payment settlement process. By integrating and analyzing large amounts of transaction data from various payment channels, it aims to improve the speed and security of payment processing while reducing the risk of fraud. Its main uses include automating payment processes, enhancing transaction transparency, providing personalized customer payment experience, and supporting real-time decision-making. It is crucial for e-commerce platforms to process large-scale transaction data.

[0004] While existing financial data processing technologies support the analysis and management of large amounts of data, their capabilities in high-frequency trading environments are often insufficient. This is particularly true when it comes to instantly updating user profiles and payment preferences. This lack of flexibility and real-time performance means payment processing processes fail to promptly reflect market conditions or user needs. Traditional systems often lack continuous performance evaluation of payment channels, resulting in limited processing capacity during periods of high demand and delayed identification and response to security threats. This not only reduces the user experience but also undermines risk management efficiency, placing institutions at a competitive disadvantage. Summary of the Invention

[0005] The present invention provides a payment settlement service system based on big data technology.

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

[0007] A payment and settlement service system based on big data technology includes:

[0008] The data collection module collects user transaction data from the e-commerce platform through the API interface, synchronizes the data with time, aligns the source information of differences, unifies the transaction record format, and generates formatted transaction sequence data;

[0009] The user payment behavior recognition module collects statistics on the user's purchase frequency, transaction amount, and selected payment time based on the formatted transaction sequence data, identifies the user's consumption pattern, analyzes the user's consumption preference trend based on the recognition results, captures periodic purchase behavior and preference mutations, reveals the user's payment habits in different times and situations, and generates a dynamic profile of the user's payment;

[0010] The payment channel evaluation module combines the user payment dynamic profile to perform performance evaluation on all available payment channels, calculates the processing speed, transaction success rate and load capacity of each payment channel, calculates the usage cost of each payment channel, ranks all payment channels based on the calculation results, and simultaneously monitors real-time market dynamics and payment environment changes to dynamically select payment channels, capture payment channels that match current user needs, and generate a list of selected payment channels;

[0011] The payment execution and monitoring module implements full-chain encryption protection for payment data based on the selected payment channel list, tracks the progress of each transaction on the e-commerce platform in real time, verifies transaction approval, monitors fund transfer and confirms the final status of the transaction based on the real-time tracking results, detects potential anomalies in the transaction process and responds quickly, and comprehensively generates a payment settlement execution report.

[0012] As a further solution of the present invention, the steps for obtaining the formatted transaction sequence data are:

[0013] Through the API interface of the e-commerce platform, the original information of user transaction data is extracted, the field data is parsed and the transaction timestamp, transaction amount, transaction category and user ID are extracted to establish a basic transaction information set;

[0014] The basic transaction information set is formatted uniformly, transaction timestamps are parsed into a standardized time format, time zone conversion and correction are performed, and the order and consistency of data records are adjusted based on the association between transaction category and user identification fields to generate a standardized transaction record set;

[0015] The standardized transaction record set is used to analyze the time interval distribution of the transaction records, verify the integrity of the transaction category field and the user identification field, perform normalized identification on the transaction category field, complete the user identification field information, and generate formatted transaction sequence data.

[0016] As a further solution of the present invention, the steps of identifying the user consumption pattern are:

[0017] Based on the formatted transaction sequence data, the transaction timestamp, transaction amount, and payment time fields are extracted, the transaction data of each user is grouped by parsing the user identification field, the grouped data is sorted according to the timestamp field, and the number of transactions, total transaction amount, and payment time distribution of each user within a fixed time period are counted to generate a preliminary transaction feature set for the user;

[0018] The transaction number, transaction amount and payment time distribution characteristics in the user's preliminary transaction feature set are called to comprehensively analyze each user's transaction behavior and adopt the formula:

[0019]

[0020] Calculate each user's consumption pattern score C u , generate a user consumption pattern score set, where M i is the amount of each transaction, T avg is the average time interval between user transactions, P var is the variance of the payment time distribution, F u The user's transaction frequency;

[0021] Based on the user consumption pattern score set, the user consumption pattern scores are classified and processed, and user categories with different consumption patterns are identified in combination with the score results to generate a user consumption pattern identification result.

[0022] As a further solution of the present invention, the steps for obtaining the user payment dynamic portrait are:

[0023] Based on the user consumption pattern recognition results, the consumption pattern category information of each user is extracted, the transaction records are reordered in time series, and the total transaction amount and transaction frequency distribution of each user in different time periods are counted to generate a basic set for user consumption trend analysis;

[0024] Based on the user consumption trend analysis basic set, analyze the user's transaction behavior within a fixed time period, identify the user's periodic purchase behavior and transaction amount changes, and use the formula:

[0025]

[0026] Calculate the periodic fluctuation range V of the user's consumption amount cycle , capture the changing trend of consumption amount and generate the user's periodic purchasing behavior and preference mutation feature set, where M i is the i-th transaction amount, M i-1 is the transaction amount of the i-1th time, and N is the total number of transactions;

[0027] Based on the user's periodic purchasing behavior and preference mutation feature set, the user's payment habits at different times and situations are analyzed, the user's transaction behavior characteristics in the target time period are extracted, and a dynamic payment profile of the user is generated.

[0028] As a further solution of the present invention, the steps for calculating the usage cost are:

[0029] In combination with the user payment dynamic profile, payment channel history records are collected, and the processing time, transaction status, and load information fields in the transaction records are analyzed. The average processing speed, transaction success rate, and load occupancy rate of each payment channel are calculated to generate a basic set of payment channel performance;

[0030] Based on the average processing speed, transaction success rate and load occupancy rate in the basic set of payment channel performance, a comprehensive analysis of the payment channel's processing capacity and success rate stability under different load levels is conducted using the formula:

[0031]

[0032] Calculate the cost C of using the payment channel z , generate a set of payment channel usage costs, where R s is the transaction success rate, T p is the average processing speed, L c is the load occupancy rate;

[0033] Based on the payment channel usage cost set, combined with the transaction success rate and average processing speed data in the payment channel performance basic set, the comprehensive performance of each payment channel is analyzed to generate a payment channel performance evaluation result.

[0034] As a further solution of the present invention, the steps of obtaining the selected payment channel list are:

[0035] Based on the payment channel usage cost set, extract the cost value of each payment channel one by one, and sort all payment channels from low to high according to cost value by comparing the usage costs. At the same time, combined with the payment channel performance evaluation results, supplement the success rate and processing speed information of each channel, and integrate the sorting results with the performance indicators to generate a payment channel sorting set;

[0036] Based on the payment channel ranking set, combined with payment environment changes in real-time market dynamic data, the current availability status and real-time load information of each channel are analyzed to determine whether the payment channel has become ineffective due to environmental factors. Unavailable channels are removed and the ranking order is readjusted to screen payment channels that match user needs, thereby generating a dynamic selection set of payment channels.

[0037] According to the dynamic selection set of payment channels, the current load and success rate are balanced, and the channels with the fastest processing speed are given priority. All qualified payment channels are integrated, and the channel list is updated in real time according to the dynamic changes of the payment environment to generate a selected payment channel list.

[0038] As a further solution of the present invention, the steps of real-time tracking of transaction progress are as follows:

[0039] Based on the selected payment channel list, the transaction data of each channel is parsed one by one, the payment account, receiving account, transaction amount and transaction time fields in each transaction record are extracted, and a transaction data statistical set is generated;

[0040] Generate a unique identifier for each transaction based on the payment account, receiving account, and transaction amount fields in the transaction data statistical set, combined with the transaction time field, and generate an encrypted transaction record by encrypting the transaction amount, account information, and time fields in segments and then merging them.

[0041] Based on the encrypted transaction records, by associating with the real-time status data of the selected payment channel, the encrypted data of each transaction is tracked in real time, the processing progress and time nodes of each encrypted transaction in the payment channel are recorded, and real-time tracking results of the transaction progress are generated.

[0042] As a further solution of the present invention, the steps for obtaining the payment settlement execution report are:

[0043] Based on the real-time tracking results of the transaction progress, the transaction encryption completion time and the payment channel processing completion time are extracted, the time fields are compared, and the processing time of each transaction is checked to see if it is within the preset range. Timed-out records are removed and marked as records requiring verification. Normal records are simultaneously confirmed to obtain verified transaction records.

[0044] Based on the verified transaction records, the outgoing and incoming times are collected, the time difference between the outgoing and incoming times is calculated, and each time difference is analyzed to determine whether the funds have been successfully transferred. The transferred amount is also checked to see if it is consistent with the original transaction amount, and potential abnormal transaction records are output;

[0045] The verified transaction records are integrated with the potential abnormal transaction records, and the transaction approval information and fund transfer status are linked one by one to track the status of the entire transaction process, review and summarize the integrity and security of the transaction process, and generate a payment settlement execution report.

[0046] Compared with the prior art, the advantages and positive effects of the present invention are:

[0047] In the present invention, through in-depth analysis of e-commerce platform transaction data and construction of dynamic user portraits, it is possible to more accurately capture and analyze users' consumption patterns and preference trends. By extracting periodic purchasing behaviors and sudden changes in preferences, the payment experience can be highly personalized, greatly improving the quality of customer service. Comprehensive payment channel evaluation, including real-time performance monitoring and cost analysis, ensures the optimization of payment processes, effectively improves processing speed and security, and reduces fraud risks. At the same time, the implementation of full-chain encryption and real-time tracking of payment progress enhances transaction security and transparency, and improves the response speed to potential anomalies. BRIEF DESCRIPTION OF THE DRAWINGS

[0048] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the following briefly introduces the drawings required for describing the embodiments. The drawings are only used to illustrate the implementation methods and are not to be considered as limiting the present invention.

[0049] Figure 1 is a system flow chart of the present invention;

[0050] Figure 2 A flowchart for obtaining formatted transaction sequence data according to the present invention;

[0051] Figure 3 This is a flow chart for identifying user consumption patterns of the present invention;

[0052] Figure 4 A flowchart for obtaining a dynamic portrait of user payment according to the present invention;

[0053] Figure 5 A flow chart for calculating the cost of use of the present invention;

[0054] Figure 6 A flowchart for obtaining a list of selected payment channels for the present invention;

[0055] Figure 7 A flowchart for real-time tracking of transaction progress of the present invention;

[0056] Figure 8 This is a flow chart for obtaining the payment settlement execution report of the present invention. DETAILED DESCRIPTION

[0057] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings. Obviously, the described embodiments are only some of the embodiments of the present invention, not all of them. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making any creative efforts shall fall within the scope of protection of the present invention.

[0058] Unless otherwise defined, all technical and scientific terms used in the present invention have the same meaning as commonly understood by those skilled in the art to which the present invention belongs; the terms used in the present invention are only for the purpose of describing specific embodiments and are not intended to limit the present invention. The terms "including" and "having" and any variations thereof in the description and claims of the present invention and the above-mentioned drawings are intended to cover non-exclusive inclusions.

[0059] In the description of the embodiments of the present invention, technical terms such as "first" and "second" are used solely to distinguish between different objects and should not be understood to indicate or imply relative importance or to implicitly specify the quantity, specific order, or primary and secondary relationship of the technical features indicated. In the description of the embodiments of the present invention, "plurality" means more than two, unless otherwise specifically defined.

[0060] In the description of the embodiments of the present invention, the term "and / or" is simply a description of the association relationship between associated objects, indicating that three relationships can exist. For example, A and / or B can represent the following three situations: A exists alone, A and B exists simultaneously, and B exists alone. In addition, the character " / " in this document generally indicates that the associated objects are in an "or" relationship.

[0061] In the description of the embodiments of the present invention, the term "multiple" refers to more than two (including two). Similarly, "multiple groups" refers to more than two groups (including two groups), and "multiple pieces" refers to more than two pieces (including two pieces).

[0062] In the description of the embodiments of the present invention, the technical terms "center", "longitudinal", "lateral", "length", "width", "thickness", "up", "down", "front", "back", "left", "right", "vertical", and "horizontal" are used interchangeably.

[0063] The orientations or positional relationships indicated by “top”, “bottom”, “inside”, “outside”, “clockwise”, “counterclockwise”, “axial”, “radial”, “circumferential”, etc. are based on the orientations or positional relationships shown in the accompanying drawings and are only for the convenience of describing the embodiments of the present invention and simplifying the description. They do not indicate or imply that the device or element referred to must have a specific orientation, be constructed and operate in a specific orientation. Therefore, they cannot be understood as limiting the embodiments of the present invention.

[0064] In the description of the embodiments of the present invention, unless otherwise expressly specified or limited, technical terms such as "installed," "connected," "connect," and "fixed" should be understood in a broad sense. For example, they can refer to fixed connections, detachable connections, or integration; mechanical connections or electrical connections; direct connections or indirect connections through an intermediate medium; and can refer to internal communication between two components or interaction between two components. Those skilled in the art will understand the specific meanings of the above terms in the embodiments of the present invention based on specific circumstances.

[0065] Example 1

[0066] The embodiment of the present invention provides a payment settlement service system based on big data technology, such as Figure 1 Shown, including:

[0067] The data collection module collects user transaction data from the e-commerce platform through the API interface, synchronizes the data with time, aligns the source information of differences, unifies the transaction record format, and generates formatted transaction sequence data;

[0068] The user payment behavior recognition module uses formatted transaction sequence data to count user purchase frequency, transaction amount, and selected payment time, identify user consumption patterns, and analyze user consumption preference trends based on the recognition results. This module captures periodic purchase behaviors and sudden changes in preferences, reveals user payment habits at different times and in different situations, and generates a dynamic profile of user payments.

[0069] The payment channel evaluation module combines user payment dynamic profiles to conduct performance evaluations on all available payment channels, counting the processing speed, transaction success rate, and load capacity of each payment channel, calculating the usage cost of each payment channel, and ranking all payment channels based on the calculated results. It also simultaneously monitors real-time market trends and changes in the payment environment, dynamically selecting payment channels, capturing payment channels that match current user needs, and generating a list of selected payment channels.

[0070] The payment execution and monitoring module implements full-chain encryption protection for payment data based on the selected payment channel list, tracks the progress of each transaction on the e-commerce platform in real time, verifies transaction approval, monitors fund transfers and confirms the final status of transactions based on real-time tracking results, detects potential anomalies in the transaction process and responds quickly, and comprehensively generates payment settlement execution reports.

[0071] Formatted transaction sequence data includes timestamp alignment, data source unification results and format standardization processing results; user payment dynamic portrait includes purchase periodicity analysis records, consumption amount fluctuation analysis results, payment preference and behavior trend mutation analysis results; the selected payment channel list includes channel response speed, successful transaction ratio, cost efficiency and user preference matching; the payment settlement execution report includes transaction confirmation time, funds arrival speed, security incident records and operation abnormality records.

[0072] See also Figure 2 , the steps to obtain formatted transaction sequence data are:

[0073] Through the API interface of the e-commerce platform, the original information of user transaction data is extracted, the field data is parsed and the transaction timestamp, transaction amount, transaction category and user ID are extracted to establish a basic transaction information set;

[0074] By calling the API interface of the e-commerce platform, the data packet returned by the interface is first parsed, and fields such as transaction timestamp, transaction amount, transaction category and user ID are extracted one by one. The original values of the fields are recorded one by one, and the integrity of the field values is checked. For missing field values, the missing fields are completed based on the user's historical transaction data according to the time distribution trend of historical transaction records and the common types of transaction categories. By comparing the user ID fields in the historical records and the current records, it is ensured that the completed data matches the correct user ID. In addition, for fields containing abnormal values in the data returned by the interface, such as negative amounts and timestamps outside a reasonable range, the abnormal fields are eliminated or adjusted in combination with the historical distribution range of the data, and adjusted to a reasonable range. After checking and processing one by one, the data source and integrity correction information are marked at the same time, and finally a basic transaction information set containing complete fields such as transaction timestamp, transaction amount, transaction category and user ID is generated.

[0075] The basic transaction information set is formatted uniformly, transaction timestamps are parsed into a standardized time format, time zone conversion and correction are performed, and the order and consistency of data records are adjusted based on the correlation between transaction category and user identification fields to generate a standardized transaction record set.

[0076] The transaction timestamp field in the basic transaction information set is processed. First, the timestamp format is parsed and timestamps in different formats are uniformly converted into a standard time format. For example, different time units (such as seconds and milliseconds) are uniformly converted into a readable year-month-day-hour-minute-second format. At the same time, the timestamp field is checked for erroneous values. For data with missing timestamps, the time pattern of the transaction category and user identification field is referred to, and the missing timestamp is supplemented by calculating the difference between the previous and next timestamps. Next, the time zone information in the timestamp field is corrected and uniformly adjusted to the standard time zone, and duplicate or conflicting data caused by the time zone adjustment is eliminated. By analyzing the distribution pattern of the timestamp field and the logical order of user transaction records, the records are sorted to ensure that the transaction data of each user is arranged in chronological order. Finally, combined with the user identification field, the records corrected or supplemented during the timestamp processing are marked to ensure the unified format and consistency of the timestamp field in the dataset, and finally a transaction record set with a standard timestamp field is generated.

[0077] Using a standardized set of transaction records, analyze the time interval distribution of transaction records, verify the integrity of the transaction category field and user identification field, normalize the transaction category field, complete the user identification field information, and generate formatted transaction sequence data;

[0078] Based on the timestamp field in the standard transaction record set, the distribution of timestamps is first analyzed, the time intervals between adjacent transaction records are counted, and abnormal transaction records with excessively large or small time intervals are identified. Based on the transaction category and user identification fields in the records, the possible causes of the abnormal time intervals are analyzed. For example, records with excessively large time intervals may be due to reduced user activity during a specific period, while records with excessively small time intervals may be due to duplicate records or system interference. Next, the distribution of the transaction category field is statistically analyzed, the frequency of occurrence of each category of transactions is calculated, and the balance of the category distribution is analyzed, for example, whether there is an excessive concentration of transactions in a certain category. At the same time, the user identification field is checked for duplication or missing information. If the user identification field is found to be missing, a reasonable user identification value is supplemented based on the correlation characteristics between the transaction timestamp and transaction category. After completing the above processing, the value of the transaction category field is uniformly encoded and normalized to ensure that the category field can support subsequent analysis operations. Finally, the timestamp, transaction category, and user identification fields are integrated, and logical associations are established between the fields to generate formatted transaction sequence data containing all fields.

[0079] See also Figure 3 , the steps for identifying user consumption patterns are:

[0080] Based on the formatted transaction sequence data, the transaction timestamp, transaction amount, and payment time fields are extracted. Each user's transaction data is grouped by parsing the user identification field, and the grouped data is sorted by the timestamp field. At the same time, the number of transactions, total transaction amount, and payment time distribution of each user within a fixed time period are counted to generate a preliminary transaction feature set for the user.

[0081] The transaction timestamp, transaction amount, and payment time fields are extracted from the formatted transaction sequence data. First, the user identification field is parsed one by one to group transaction records by user, ensuring that each user's transaction data can be stored separately for subsequent operations. After grouping, each data set is sorted using the transaction timestamp field, arranging the transaction records in chronological order to ensure the consistency of the transaction events and avoid data errors caused by sequencing issues during subsequent calculations. Next, the number of transactions per user within a specified time period is counted. By comparing the timestamp field, the number of transaction events within different time periods is counted, such as calculating transaction frequency by hour, day, or month. Subsequently, the transaction amount field is used to add the amount values of each group of user transaction records to obtain the total transaction amount for each user within the specified time period. Finally, by analyzing the payment time field, the payment time format is standardized and distribution statistics are calculated by time period, such as morning, noon, and evening. The distribution characteristics of user payment times are obtained by the proportion of transactions in each time period. Finally, all calculation results are integrated to generate a preliminary user transaction feature set containing the transaction frequency, transaction amount, and payment time distribution.

[0082] The transaction count, transaction amount, and payment time distribution characteristics in the user's preliminary transaction feature set are called to comprehensively analyze each user's transaction behavior using the formula:

[0083]

[0084] Calculate each user's consumption pattern score C u , generate a user consumption pattern score set, where M i is the amount of each transaction, T avg is the average time interval between user transactions, P var is the variance of the payment time distribution, F u The user's transaction frequency;

[0085] M i : The amount of each transaction is obtained by extracting the transaction amount field in the transaction record and adding them up one by one.

[0086] T avg : Average transaction time interval, obtained by calculating the time difference of all adjacent timestamps in the user's transaction records and taking the average.

[0087] P var : Variance of payment time distribution. By statistically analyzing the distribution of user payment time fields and dividing them into time periods, the variance of payment time distribution is calculated.

[0088] F u : Transaction frequency, calculated by counting the number of transactions per unit time.

[0089] Transaction amount: M = {200, 150, 180} (unit: yuan).

[0090] Transaction timestamp (unit: hour): T = {1, 3, 6}, from which the time interval is calculated to be {2, 3}.

[0091] Payment time distribution: P = {9:00, 12:00, 15:00}, divided into three time periods with frequencies of {1, 1, 1}.

[0092] Transaction frequency per unit time: If a user trades 3 times within 6 hours, the transaction frequency is:

[0093] Calculate the total transaction amount ∑M i :

[0094] ∑M i =200+150+180=530

[0095] Calculate the average transaction time interval T avg :

[0096]

[0097] Calculate the variance P of the payment time distribution var :

[0098] Calculate the average over the payment period:

[0099]

[0100] Compute the sum of squared deviations from the mean for each time period:

[0101] (9-12) 2 +(12-12) 2 +(15-12) 2 =9+0+9=18

[0102] Calculate the variance:

[0103]

[0104] Calculate consumption pattern score C u :

[0105]

[0106] The results show that the user's consumption pattern score is 58.89. This score integrates multiple dimensions, including transaction amount, transaction frequency, and payment time distribution. A higher score indicates that the user's consumption activities are more frequent, the amount is larger, and the payment time distribution is more concentrated within a certain period of time.

[0107] Based on the user consumption pattern score set, the user consumption pattern scores are classified and processed, and the user categories with different consumption patterns are identified by combining the score results to generate the user consumption pattern recognition results;

[0108] Based on the user consumption pattern score set, the user's score value is classified. First, the score interval is delineated according to the numerical distribution in the score set, for example, the score value is divided into three categories: low, medium, and high, for subsequent classification; after the score interval division is completed, the consumption pattern score value of each user is called, and the correspondence between the score value and the delineated interval is compared, and the user's consumption pattern score is classified into the corresponding category, for example, high-scoring users are classified into the high-frequency and high-amount category, and low-scoring users are classified into the low-frequency and low-amount category; then, the transaction frequency and transaction amount fields in the preliminary transaction feature set are called, and the average transaction frequency and transaction amount are calculated for each category of users to obtain the average consumption behavior characteristics of users in different categories, such as the average transaction amount and average transaction frequency of the high-frequency and high-amount category; finally, the classified user categories are labeled and integrated, and the category information of each user is associated with its original feature data to form a user consumption pattern recognition result that includes user consumption category labeling, transaction frequency and transaction amount features.

[0109] See also Figure 4 , the steps to obtain the user payment dynamic portrait are:

[0110] Based on the results of user consumption pattern recognition, we extract each user's consumption pattern category information, reorder the transaction records by time series, and calculate the total transaction amount and transaction frequency distribution of each user in different time periods to generate a basic set for user consumption trend analysis;

[0111] Based on the results of user consumption pattern recognition, the consumption pattern category information of each user is extracted. First, the transaction timestamp and transaction amount fields in the user transaction record are parsed, and the transaction data is rearranged in chronological order to ensure the integrity and logic of the time series data. At the same time, the total transaction amount and number of transactions of the user in different time periods (such as day, week, and month) are counted. By calling the user payment time field, the payment time is divided into fixed time periods (such as morning, noon, and evening). The transaction distribution of the user in different time periods is analyzed, and the proportion of the user's transaction amount in different time periods is calculated one by one to reflect the user's preference characteristics in these time periods; then, according to the number of transactions and amount distribution in the time series, statistical information of transaction behavior in the time period is generated by time period, including total amount, number of transactions and proportion data, and associated with the user's consumption pattern category to form a basic set for user consumption trend analysis.

[0112] Based on the basic set of user consumption trend analysis, analyze the user's transaction behavior within a fixed time period, identify the user's periodic purchase behavior and transaction amount changes, and use the formula:

[0113]

[0114] Calculate the periodic fluctuation range V of the user's consumption amount cycle , capture the changing trend of consumption amount and generate the user's periodic purchasing behavior and preference mutation feature set, where M i is the i-th transaction amount, M i-1 is the transaction amount of the i-1th time, and N is the total number of transactions;

[0115] User transaction amount sequence: M = {200, 300, 250, 400} (unit: yuan).

[0116] Total number of transactions: N=4.

[0117] Calculate the absolute value of the difference between adjacent transaction amounts:

[0118] |M2-M1|=|300-200|=100

[0119] |M3-M2|=|250-300|=50

[0120] |M4-M3|=|400-250|=150

[0121] Calculate the sum of the absolute values of the differences:

[0122]

[0123] Calculate the periodic fluctuation amplitude:

[0124]

[0125] The results show that the periodic fluctuation range of the user's consumption amount is 75 yuan, which can be used to judge the stability and changing trend of the user's consumption amount, providing an important basis for subsequent preference analysis and dynamic portrait generation.

[0126] Based on the user's periodic purchasing behavior and preference mutation feature set, analyze the user's payment habits at different times and situations, extract the user's transaction behavior characteristics in the target time period, and generate a dynamic profile of the user's payment;

[0127] Based on the user's periodic purchasing behavior and preference mutation feature set, the user's payment time field is first called to classify the payment time in the transaction record, and the number of transactions and the amount distribution of the user in different scenarios are calculated by time period (such as working hours and rest time); then, combined with the user's consumption pattern category field, the total transaction amount and payment time concentration of each type of user in different time periods are analyzed one by one. For example, by comparing the transaction frequency during peak and non-peak periods, the user's payment habits in a specific time period are identified; finally, the analysis results are combined with the periodic behavior characteristics, and by integrating the transaction amount, time period transaction characteristics and consumption pattern category labels, a dynamic user payment portrait containing the user's transaction characteristics in a specific time and situation is generated.

[0128] See also Figure 5 , the calculation steps of usage cost are:

[0129] Combined with user payment dynamic profiles, payment channel history records are collected, and the processing time, transaction status, and load information fields in the transaction records are analyzed. The average processing speed, transaction success rate, and load occupancy rate of each payment channel are calculated to generate a basic set of payment channel performance.

[0130] Call the payment channel history records in the user payment dynamic profile, parse the processing time, transaction status and load information fields in the records one by one, calculate the processing time of each transaction one by one by separating the start time and end time fields in the records, count the processing time of all transaction records in each payment channel and calculate the average value to obtain the average processing speed of the channel; then call the transaction status field, count the number of successful transactions and the total number of transactions, divide the number of successful transactions by the total number of transactions to calculate the transaction success rate of each payment channel; then parse the load information field, extract the channel load occupancy value in each time period, summarize the values and calculate the average value of the load occupancy in the time period to reflect the channel load capacity; after completing the above steps, integrate the processing speed, transaction success rate and load occupancy rate data into a unified performance basic set to ensure the consistency of the channel identification corresponding to the data, and form a complete set of all channel performance indicators.

[0131] Based on the average processing speed, transaction success rate, and load occupancy rate in the payment channel performance base set, a comprehensive analysis of the payment channel's processing capacity and success rate stability under different load levels is conducted using the formula:

[0132]

[0133] Calculate the cost C of using the payment channel z , generate a set of payment channel usage costs, where R s is the transaction success rate, T p is the average processing speed, L cis the load occupancy rate;

[0134] Transaction success rate: R s =0.95 (calculated by counting the number of successful transactions and the total number of transactions);

[0135] Average processing speed: T p = 2 seconds (by calculating the processing time item by item and taking the average);

[0136] Load occupancy rate: L c =0.8 (calculated by counting the load occupancy in each time period).

[0137] Calculate the inverse term of the transaction success rate:

[0138]

[0139] Calculate the square root of the load occupancy:

[0140]

[0141] Calculate the overall formula:

[0142]

[0143] The results show that the cost of using the payment channel is 2.841. This value reflects the performance and cost characteristics of the payment channel under the combined effect of success rate, processing speed and load occupancy rate. A lower value indicates that the channel is more efficient and cost-effective, providing an important reference for subsequent performance analysis and selection.

[0144] Based on the payment channel usage cost set, combined with the transaction success rate and average processing speed data in the payment channel performance base set, the comprehensive performance of each payment channel is analyzed to generate a payment channel performance evaluation result;

[0145] Based on the payment channel usage cost set, by extracting the usage cost value of each payment channel, a usage cost index table is established, and the cost values corresponding to all channels are listed one by one; then the transaction success rate field and processing speed field in the performance basic set are called to extract the success rate and processing speed of each payment channel respectively, and the performance indicators are associated with the usage cost index table. Through data matching, a comprehensive data structure containing the cost and performance characteristics of each payment channel is formed; then the usage cost and performance characteristics of each channel are sorted, and by comparing the cost values one by one, the channel with the lowest cost is identified, and the channels are sorted in order from low to high, and the corresponding success rate and processing speed characteristics are marked; finally, the sorted results are output as the payment channel performance evaluation results, and the correlation between the cost and performance indicators of each channel is supplemented.

[0146] See also Figure 6 , the steps to obtain the selected payment channel list are:

[0147] Based on the payment channel usage cost set, the cost value of each payment channel is extracted one by one. By comparing the usage costs, all payment channels are sorted from low to high according to the cost value. At the same time, combined with the payment channel performance evaluation results, the success rate and processing speed information of each channel are supplemented. The sorting results are integrated with the performance indicators to generate the payment channel sorting set;

[0148] According to the usage cost field in the payment channel usage cost set, the usage cost value of each payment channel is extracted one by one, and the usage cost corresponding to each channel is recorded. By establishing an index table, the identifier of each payment channel and its usage cost are stored in correspondence; then the success rate field in the performance evaluation result set is called, and the success rate of each payment channel is calculated by parsing the historical transaction records of the payment channel one by one. The success rate data is obtained by dividing the number of successful transactions by the total number of transactions. At the same time, the processing time information in the processing speed field is extracted, and the average processing speed of the payment channel is obtained by calculating the processing time of each transaction and taking the average; the extracted success rate, processing speed and usage cost are matched to form a complete payment channel comprehensive performance data table; then, the comprehensive performance characteristics of the payment channels are compared one by one. By comparing the relationship between success rate, processing speed and usage cost, low-cost and high-performance channels are marked, and priority information is marked according to performance quality to generate a payment channel sorting set with comprehensive performance and cost characteristics to support subsequent screening and dynamic selection processes.

[0149] Based on the payment channel sorting set and combined with payment environment changes in real-time market dynamic data, the current availability and real-time load information of each channel are analyzed to determine whether the payment channel has become invalid due to environmental factors. Unavailable channels are removed and the sorting order is readjusted to select payment channels that match user needs, thus generating a dynamic selection set of payment channels.

[0150] Based on the sorting results in the payment channel sorting set, the availability of the current channel is verified by reading the comprehensive performance score and usage cost value of each payment channel one by one, combined with the payment environment change field in the market dynamic data; the load information in the payment environment change field is parsed to check whether the load occupancy rate of each payment channel exceeds the normal range. If the load occupancy is too high or the channel fails due to market changes, the channel is eliminated and the sorting order of the remaining channels is readjusted; then the user demand field in the user payment dynamic portrait is called to extract the performance characteristics of user preferences, such as giving priority to high success rate channels or low-cost channels, and comparing the performance indicators of each payment channel with the priority of the user demand field one by one. By comparing the user demand characteristics with the comprehensive performance of the channel, the payment channels that can meet the user's specific needs are screened out; finally, the screened channels are integrated into the payment channel dynamic selection set, and the set content is dynamically updated according to real-time market data to ensure that the channels in the set always adapt to user needs and environmental changes.

[0151] Dynamically select a set of payment channels, balancing the current load and success rate, and giving priority to the channels with the fastest processing speed. All eligible payment channels are integrated, and the channel list is updated in real time according to the dynamic changes in the payment environment to generate a list of selected payment channels.

[0152] The filtering results in the payment channel dynamic selection set are called, and the channel information fields in the set are parsed one by one to extract information such as the matching score, success rate, processing speed and load status of each channel, and further analyze the filtered payment channels. Combined with the real-time load status field in the market dynamic data, the current load of each payment channel is checked one by one, and high-load channels are eliminated, while channels with low success rates are further filtered out. Subsequently, the user demand field in the user payment dynamic portrait is called, and the performance priority in the user demand, such as fast payment priority, low cost priority or high success rate priority, is compared with the performance characteristics of each payment channel one by one, and channels that match the user demand are selected first. The filtered payment channels are then arranged in order of processing speed from high to low, and these channels are integrated to form a list of selected payment channels. On this basis, the channel list content is checked and adjusted according to the real-time update of the market dynamic data to ensure that the payment channels in the list can always meet the requirements of the current payment environment and user needs, and finally the selected payment channel list is output for subsequent calls.

[0153] See also Figure 7 , the steps for real-time tracking of transaction progress are:

[0154] Based on the selected payment channel list, the transaction data of each channel is parsed one by one, and the payment account, receiving account, transaction amount and transaction time fields in each transaction record are extracted to generate a statistical set of transaction data;

[0155] Call the payment channel field in the selected payment channel list, parse the transaction data one by one, extract the payment account, receiving account, transaction amount and transaction time fields, first perform format verification on the payment account and receiving account fields to ensure that they comply with specific format specifications, such as checking whether it is a valid account structure and eliminating records that do not conform to the format; then perform numerical verification on the transaction amount field to detect whether it is a valid numerical type, remove null values or abnormal values, and ensure data integrity; then combine the payment account and receiving account fields to form an independent identifier, associate it with the transaction amount and transaction time fields, and construct a basic transaction data set; perform segmentation operations on the transaction amount field, divide the amount into multiple parts according to fixed units, encrypt each part separately, and generate an encrypted string through the encrypted result; bind each encrypted string with the transaction time and independent identifier to form a statistical set of transaction data that has undergone preliminary encryption processing, to provide support for subsequent encryption integrity verification.

[0156] Generate a unique identifier for each transaction based on the payment account, receiving account, and transaction amount fields in the transaction data statistical set, combined with the transaction time field. Then, generate an encrypted transaction record by encrypting the transaction amount, account information, and time fields in segments and merging them.

[0157] Call the encrypted basic transaction data set, parse each transaction record one by one, read the encrypted string, transaction time field and unique identifier, and check whether the encryption is complete by comparing the encrypted string with the original field content; parse the structure of the encrypted string, split it into multiple independent parts, and verify whether each part complies with the encryption rules to ensure that there are no omissions; then calculate the encryption time of each transaction by recording the encryption start time and encryption completion time, and store the result in the record; count the encryption time of all transaction records, and calculate the average encryption time of the entire transaction batch by accumulating the time values one by one and taking the average; match the encryption time and encryption string results with the unique identifier of the transaction to form real-time encryption tracking information; generate an encryption progress set containing encryption status and time information by marking the encryption progress one by one to ensure that the encryption processing process of each transaction can be fully tracked and verified.

[0158] Based on the encrypted transaction records, by associating with the real-time status data of the selected payment channel, the encrypted data of each transaction is tracked in real time, the processing progress and time nodes of each encrypted transaction in the payment channel are recorded, and the real-time tracking results of the transaction progress are generated;

[0159] Call the encryption progress collection, and combine it with the real-time channel status information in the selected payment channel list to parse the encryption completion time, unique identifier and encryption status of each transaction one by one. By comparing the encryption completion time with the payment channel processing start time, calculate the time interval from encryption to transmission of each transaction; for the channel status information of each transaction, parse the channel processing time node to check whether the transmission time interval exceeds the preset range; for transaction records that exceed the range, mark them as records requiring further inspection, and record the reasons for the abnormality, such as transmission delay or channel failure; then, match the normal transaction records with the payment channel processing status one by one, extract the channel processing completion time, and integrate the processing completion time with the records in the encryption progress collection to generate complete transaction progress information; finally, organize this information into a comprehensive data set including transaction encryption status, transmission time and channel processing progress to ensure that the full chain status of each transaction can be fully recorded and verified at any time.

[0160] See also Figure 8 , the steps to obtain the payment settlement execution report are:

[0161] Based on the real-time tracking results of transaction progress, the transaction encryption completion time and payment channel processing completion time are extracted, and the time fields are compared to check whether the processing time of each transaction is within the preset range. Timed-out records are removed and marked as records requiring verification. Normal records are simultaneously confirmed to obtain verified transaction records.

[0162] Based on the encryption progress and channel processing time information in the real-time tracking results of transaction progress, the encryption completion time and payment channel processing completion time of each transaction are extracted, and the records are parsed one by one according to the unique identifier. By calculating the difference between the encryption completion time and the payment channel start time, the time interval from encryption completion to channel processing is obtained, and it is checked whether the time interval is within the preset range. If the time interval exceeds the range, the record is marked as a potential anomaly; then the time interval result of the normal record is bound to the transaction unique identifier and organized into a set containing the time interval and channel completion time. This set is used to further refine the analysis of whether the transmission time of each transaction is valid; by comparing the transmission time with the encryption completion time one by one, a set of transaction records marked with time anomalies and normal times is generated.

[0163] Based on verified transaction records, the outgoing and incoming times are collected, and the time difference between the outgoing and incoming times is calculated. Each time difference is analyzed one by one to determine whether the funds have been successfully transferred. At the same time, the transfer amount is checked to see if it is consistent with the original transaction amount, and potential abnormal transaction records are output;

[0164] Call the verified transaction records, parse the funds transfer fields of each transaction one by one, extract the funds outgoing time, incoming time and transfer amount fields, first check whether the difference between the outgoing time and the incoming time exceeds the preset funds transfer time range, mark the records that exceed the range, and calculate the deviation of the transfer time; then, by comparing the transfer amount and the original transaction amount field, check whether there is any inconsistency in the amount, mark the records with inconsistent amounts as potential abnormal transaction records, and record the reasons for the abnormality; classify and organize the normal records that have passed the verification and the potential abnormal records, and associate the record status with the funds transfer time, amount consistency and transaction unique identifier respectively, generate a tag set containing the funds transfer status, and ensure that all transactions have complete status records.

[0165] Integrate verified transaction records with potentially abnormal transaction records, link transaction approval information and fund transfer status one by one, track the status of the entire transaction process, review and summarize the integrity and security of the transaction process, and generate a payment settlement execution report;

[0166] Based on verified transaction records and potential abnormal transaction records, the approval status, funds transfer status and final transaction status fields of each transaction are extracted. By parsing the approval status fields one by one, it is verified whether the transaction is approved. If the status is not approved, the record is marked as a final invalid transaction. Then, the transactions with abnormal funds transfer status are marked, and an explanation of the abnormal transaction status is generated, including the reason for the delay, the reason for the incorrect amount, and the specific problems of the inconsistent status. The data of all normal transactions and potential abnormal transactions are reintegrated, and the transaction approval status, funds transfer status and final status fields are summarized into a unique identifier. Finally, based on this collection, they are classified one by one by time, channel and funds status to comprehensively generate a payment settlement execution report.

[0167] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, not to limit them. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some or all of the technical features therein. These modifications or replacements do not deviate the essence of the corresponding technical solutions from the scope of the technical solutions of the embodiments of the present invention, and they should all be included in the scope of the claims and description of the present invention. In particular, as long as there is no structural conflict, the various technical features mentioned in the various embodiments can be combined in any way. The present invention is not limited to the specific embodiments disclosed herein, but includes all technical solutions that fall within the scope of the claims.

Claims

1. A payment and settlement service system based on big data technology, characterized in that: The system comprises: The data collection module collects user transaction data from the e-commerce platform through the API interface, synchronizes the data with time, aligns the source information of differences, unifies the transaction record format, and generates formatted transaction sequence data; The user payment behavior recognition module collects statistics on the user's purchase frequency, transaction amount, and selected payment time based on the formatted transaction sequence data, identifies the user's consumption pattern, analyzes the user's consumption preference trend based on the recognition results, captures periodic purchase behavior and preference mutations, reveals the user's payment habits in different times and situations, and generates a dynamic profile of the user's payment; The payment channel evaluation module combines the user payment dynamic profile to perform performance evaluation on all available payment channels, calculates the processing speed, transaction success rate and load capacity of each payment channel, calculates the usage cost of each payment channel, ranks all payment channels based on the calculation results, and simultaneously monitors real-time market dynamics and payment environment changes to dynamically select payment channels, capture payment channels that match current user needs, and generate a list of selected payment channels; The payment execution and monitoring module implements full-chain encryption protection for payment data based on the selected payment channel list, tracks the progress of each transaction on the e-commerce platform in real time, verifies transaction approval, monitors fund transfer and confirms the final status of the transaction based on the real-time tracking results, detects potential anomalies in the transaction process and responds quickly, and comprehensively generates a payment settlement execution report.

2. The payment and settlement service system based on big data technology according to claim 1, characterized in that: The steps for obtaining the formatted transaction sequence data are as follows: Through the API interface of the e-commerce platform, the original information of user transaction data is extracted, the field data is parsed and the transaction timestamp, transaction amount, transaction category and user ID are extracted to establish a basic transaction information set; The basic transaction information set is formatted uniformly, transaction timestamps are parsed into a standardized time format, time zone conversion and correction are performed, and the order and consistency of data records are adjusted based on the association between transaction category and user identification fields to generate a standardized transaction record set; The standardized transaction record set is used to analyze the time interval distribution of the transaction records, verify the integrity of the transaction category field and the user identification field, perform normalized identification on the transaction category field, complete the user identification field information, and generate formatted transaction sequence data.

3. The payment and settlement service system based on big data technology according to claim 2, characterized in that: The steps for identifying the user consumption pattern are: Based on the formatted transaction sequence data, the transaction timestamp, transaction amount, and payment time fields are extracted, the transaction data of each user is grouped by parsing the user identification field, the grouped data is sorted according to the timestamp field, and the number of transactions, total transaction amount, and payment time distribution of each user within a fixed time period are counted to generate a preliminary transaction feature set for the user; The transaction number, transaction amount and payment time distribution characteristics in the user's preliminary transaction feature set are called to comprehensively analyze each user's transaction behavior and adopt the formula: Calculate each user's consumption pattern score C u , generate a user consumption pattern score set, where M i is the amount of each transaction, T avg is the average time interval between user transactions, P var is the variance of the payment time distribution, F u The user's transaction frequency; Based on the user consumption pattern score set, the user consumption pattern scores are classified and processed, and user categories with different consumption patterns are identified in combination with the score results to generate a user consumption pattern identification result.

4. The payment and settlement service system based on big data technology according to claim 3 is characterized in that: The steps for obtaining the user payment dynamic portrait are as follows: Based on the user consumption pattern recognition results, the consumption pattern category information of each user is extracted, the transaction records are reordered in time series, and the total transaction amount and transaction frequency distribution of each user in different time periods are counted to generate a basic set for user consumption trend analysis; Based on the user consumption trend analysis basic set, analyze the user's transaction behavior within a fixed time period, identify the user's periodic purchase behavior and transaction amount changes, and use the formula: Calculate the periodic fluctuation range V of the user's consumption amount cycle , capture the changing trend of consumption amount and generate the user's periodic purchasing behavior and preference mutation feature set, where M i is the i-th transaction amount, M i-1 is the transaction amount of the i-1th time, and N is the total number of transactions; Based on the user's periodic purchasing behavior and preference mutation feature set, the user's payment habits at different times and situations are analyzed, the user's transaction behavior characteristics in the target time period are extracted, and a dynamic payment profile of the user is generated.

5. The payment and settlement service system based on big data technology according to claim 4 is characterized in that: The calculation steps of the usage cost are: In combination with the user payment dynamic profile, payment channel history records are collected, and the processing time, transaction status, and load information fields in the transaction records are analyzed. The average processing speed, transaction success rate, and load occupancy rate of each payment channel are calculated to generate a basic set of payment channel performance; Based on the average processing speed, transaction success rate and load occupancy rate in the basic set of payment channel performance, a comprehensive analysis of the payment channel's processing capacity and success rate stability under different load levels is conducted using the formula: Calculate the cost C of using the payment channel z , generate a set of payment channel usage costs, where R s is the transaction success rate, T p is the average processing speed, L c is the load occupancy rate; Based on the payment channel usage cost set, combined with the transaction success rate and average processing speed data in the payment channel performance basic set, the comprehensive performance of each payment channel is analyzed to generate a payment channel performance evaluation result.

6. The payment and settlement service system based on big data technology according to claim 5, characterized in that: The steps for obtaining the selected payment channel list are: Based on the payment channel usage cost set, extract the cost value of each payment channel one by one, and sort all payment channels from low to high according to cost value by comparing the usage costs. At the same time, combined with the payment channel performance evaluation results, supplement the success rate and processing speed information of each channel, and integrate the sorting results with the performance indicators to generate a payment channel sorting set; Based on the payment channel ranking set, combined with payment environment changes in real-time market dynamic data, the current availability status and real-time load information of each channel are analyzed to determine whether the payment channel has become ineffective due to environmental factors. Unavailable channels are removed and the ranking order is readjusted to screen payment channels that match user needs, thereby generating a dynamic selection set of payment channels. According to the dynamic selection set of payment channels, the current load and success rate are balanced, and the channels with the fastest processing speed are given priority. All qualified payment channels are integrated, and the channel list is updated in real time according to the dynamic changes of the payment environment to generate a selected payment channel list.

7. The payment and settlement service system based on big data technology according to claim 6, characterized in that: The steps for real-time tracking of the transaction progress are as follows: Based on the selected payment channel list, the transaction data of each channel is parsed one by one, the payment account, receiving account, transaction amount and transaction time fields in each transaction record are extracted, and a transaction data statistical set is generated; Generate a unique identifier for each transaction based on the payment account, receiving account, and transaction amount fields in the transaction data statistical set, combined with the transaction time field, and generate an encrypted transaction record by encrypting the transaction amount, account information, and time fields in segments and then merging them. Based on the encrypted transaction records, by associating with the real-time status data of the selected payment channel, the encrypted data of each transaction is tracked in real time, the processing progress and time nodes of each encrypted transaction in the payment channel are recorded, and real-time tracking results of the transaction progress are generated.

8. The payment and settlement service system based on big data technology according to claim 7, characterized in that: The steps for obtaining the payment settlement execution report are as follows: Based on the real-time tracking results of the transaction progress, the transaction encryption completion time and the payment channel processing completion time are extracted, the time fields are compared, and the processing time of each transaction is checked to see if it is within the preset range. Timed-out records are removed and marked as records requiring verification. Normal records are simultaneously confirmed to obtain verified transaction records. Based on the verified transaction records, the outgoing and incoming times are collected, the time difference between the outgoing and incoming times is calculated, and each time difference is analyzed to determine whether the funds have been successfully transferred. The transferred amount is also checked to see if it is consistent with the original transaction amount, and potential abnormal transaction records are output; The verified transaction records are integrated with the potential abnormal transaction records, and the transaction approval information and fund transfer status are linked one by one to track the status of the entire transaction process, review and summarize the integrity and security of the transaction process, and generate a payment settlement execution report.

Images