Financial Marketing Management Service Method and System Based on Multi-Source Data Fusion

Abstract

This invention relates to the field of financial marketing management technology, specifically to a financial marketing management service method and system based on multi-source data fusion. The method includes the following steps: acquiring transaction records and calculating intervals to generate transaction rhythm tags; filtering customers with increasing frequency and volatility to determine high-conversion time periods; analyzing touchpoint activity to generate priority order and matching time points; and unifying multi-terminal interactive data to generate a marketing management plan. In this invention, by serializing and analyzing the mean fluctuation of transaction behavior time intervals, changes in transaction activity are dynamically captured. Combined with latency curves, concentrated response intervals are determined and high-conversion time periods are extracted. These intervals are then analyzed for overlap with multi-touchpoint activity peaks to generate a priority touchpoint sequence and match precise time points. Through multi-source interactive data normalization and sequence rearrangement, the touchpoint and time delivery order is optimized on a unified timeline, improving reach relevance and conversion efficiency.

Description

Technical Field

[0001] This invention relates to the field of financial marketing management technology, and in particular to a financial marketing management service method and system based on multi-source data fusion. Background Technology

[0002] The field of financial marketing management technology refers to a comprehensive methodology that utilizes information technology and data analysis to organize and implement planning, resource allocation, channel management, and customer information management in the marketing, customer development, relationship maintenance, and sales activities of financial products and services. This field encompasses core aspects such as market segmentation and positioning analysis, customer behavior and value assessment, marketing campaign design and execution, channel operation and optimization management, and marketing performance monitoring and adjustment. Its overall technological development emphasizes guiding financial institutions through the collection, integration, and analysis of various types of business data to achieve targeted marketing strategy development and implementation. Traditional financial marketing management service methods based on multi-source data fusion refer to the unified processing of various types of data from different business systems, channel terminals, and external markets in financial business marketing management activities to support marketing strategy development. Common data fusion methods include correlation analysis based on customer characteristic data and transaction record data, joint modeling based on market information data and customer interaction data, pattern matching analysis based on behavioral trajectory data and product attribute data, and trend analysis based on time series characteristics. By constructing and comprehensively utilizing the correlations between these data, a data foundation is provided for the marketing management process.

[0003] Existing technologies in multi-source data fusion applications lack serialized modeling of the time distribution of transaction behavior, making it impossible to dynamically identify changes in transaction activity based on rhythmic characteristics. In response interval extraction, there is a lack of correlation analysis between delay features and concentrated high-conversion time periods. The correspondence between touchpoint activity and ad placement timing is not finely characterized, making it difficult to match the placement order with the actual response peak. The time differences in multi-touchpoint data are not processed uniformly, which can easily lead to decreased touchpoint utilization efficiency and resource allocation deviation, affecting the overall marketing accuracy and conversion performance. Summary of the Invention

[0004] To address the technical problems existing in the prior art, embodiments of the present invention provide a financial marketing management service method based on multi-source data fusion, comprising the following steps:

[0005] To achieve the above objectives, the present invention adopts the following technical solution: a financial marketing management service method based on multi-source data fusion, comprising the following steps:

[0006] S1: Obtain the account transaction records of the target customer group within a fixed time window, sort them in ascending order by transaction time, calculate the time interval between adjacent transactions to form an interval sequence, calculate the mean and volatility values ​​respectively, compare the mean difference between adjacent time windows with the transaction rhythm change threshold, and compare the volatility value change with the transaction volatility threshold to generate a customer group transaction rhythm label.

[0007] S2: Based on the customer group transaction rhythm tags, filter customer groups with increased transaction frequency and rising fluctuations, collect the time of marketing information sent before marketing campaign and the time of the first transaction, calculate the time difference to generate a response delay curve, compare the position difference of the curve to obtain the concentrated response period, count the number of transactions in the period, and generate a high conversion transaction period.

[0008] S3: Based on the high-conversion transaction period, collect the daily activity curves of similar customers at financial service touchpoints, calculate the overlap ratio between the activity peak and the high-conversion period, and sort them to generate the priority touchpoint delivery order;

[0009] S4: Select time points according to the priority touchpoint delivery order, calculate the time difference with the peak of the delay curve of the target customer, filter the matching time points, and generate a delivery time list.

[0010] As a further aspect of the present invention, the customer group transaction rhythm label includes the average transaction change range, the transaction fluctuation value change range, and the transaction rhythm type; the high conversion transaction time period includes the concentrated response period, the number of concentrated response transactions, and the peak transaction conversion period; the priority touchpoint placement order includes the touchpoint overlap ratio, touchpoint priority, and touchpoint identifier; and the placement time list includes the matching time point, the time difference range, and the priority placement time point.

[0011] As a further aspect of the present invention, the specific steps of S1 are as follows:

[0012] S101: Obtain the account transaction records of the target customer group within a fixed time window, sort them in ascending order by transaction time, extract the time intervals between adjacent transactions and integrate them into a sequence to generate a transaction time interval sequence;

[0013] S102: Based on the transaction time interval sequence, extract the center level and stability features, and integrate the two into a feature set to obtain the transaction interval statistical features;

[0014] S103: Call the transaction interval statistical features, compare the changes in the center level and stability of adjacent time windows, and combine the transaction rhythm and fluctuation threshold to establish a customer group transaction rhythm label.

[0015] As a further aspect of the present invention, the specific steps of S2 are as follows:

[0016] S201: Obtain the transaction rhythm tags of the customer groups, filter customer groups with increased transaction frequency and increased volatility, integrate the customer group identifiers that meet the conditions into a set, and generate a target customer group set.

[0017] S202: Based on the target customer group set, collect the marketing information sending time and the first transaction completion time of the customer group before the multi-channel marketing campaign, perform time difference calculation on the two types of time data and arrange them in the order of customer groups to form a response delay curve;

[0018] S203: Call the response delay curve, compare the position difference of the curve on the time axis to obtain the concentrated response period, and count the number of transactions within the period. Combine the concentrated response period with the number of transactions to establish a high conversion transaction period.

[0019] As a further aspect of the present invention, the specific steps of S3 are as follows:

[0020] S301: Based on the high-conversion transaction period, collect the daily activity curves of similar customers' online financial service touchpoints, identify the peak positions in the curves and record the corresponding time points, integrate the peak time information according to the touchpoint order, and generate a set of touchpoint peak times.

[0021] S302: Call the set of peak times of the touch points, calculate the time interval overlap between the peak times and the conversion transaction time period, calculate the overlap ratio percentage of the touch points, and archive the ratio results in the order of the touch points to obtain the touch point overlap ratio sequence.

[0022] S303: Based on the contact overlap ratio sequence, sort the contacts from largest to smallest to generate a priority contact delivery order.

[0023] As a further aspect of the present invention, the specific steps of S4 are as follows:

[0024] S401: Call the priority touchpoint delivery order, select the touchpoints with higher sorting, obtain the corresponding time point data, and organize the touchpoint time points into a set according to time order to generate a priority touchpoint active time set;

[0025] S402: Based on the set of active times of the priority touchpoints, extract the peak position of the delay curve of the target customer, calculate the time difference with the time point, compare the difference result with the matching threshold, filter the time points that meet the threshold conditions, and obtain the set of matching time points.

[0026] S403: Call the set of matching time points, arrange them in chronological order, remove duplicate records, and create a delivery time list.

[0027] As a further aspect of the present invention, the method further includes:

[0028] S5: Based on the aforementioned campaign time list, calculate the start, peak, and end time differences of customer touchpoint interaction times from multiple acquisition terminals, normalize and adjust them to a unified timeline, rearrange the events, and generate a financial marketing management service plan.

[0029] As a further aspect of the present invention, the financial marketing management service solution includes a multi-touchpoint event sequence under a unified timeline, a touchpoint interaction time distribution, and differentiated adjustment results between touchpoints.

[0030] As a further aspect of the present invention, the specific steps of S5 are as follows:

[0031] S501: Based on the delivery time list, collect the customer touchpoint interaction time recorded by multiple collection terminals, obtain the start time, peak time and end time in sequence, calculate the difference between the three types of time, classify and organize the calculation results according to the collection terminal, and generate a touchpoint time difference set.

[0032] S502: Based on the contact time difference set, identify time difference records with differences between the acquisition terminals, normalize and adjust the differentiated time differences according to the unified time axis reference value, integrate the adjusted records into a unified structure, and obtain a unified time axis record set.

[0033] S503: Call the unified timeline record set, rearrange the multi-touchpoint events according to the adjusted time order, and integrate them into executable marketing management content to establish a financial marketing management service solution.

[0034] A financial marketing management service system based on multi-source data fusion, the system comprising:

[0035] The transaction rhythm analysis module is used to execute S1: obtain the account transaction records of the target customer group in a fixed time window, sort them in ascending order by transaction time, calculate the time interval between adjacent transactions to form an interval sequence, calculate the mean and volatility value respectively, compare the mean difference between adjacent time windows with the transaction rhythm change threshold, and compare the change in volatility value with the transaction volatility threshold to generate a transaction rhythm label for the customer group.

[0036] The frequency fluctuation filtering module is used to execute S2: based on the customer group transaction rhythm tag, filter customer groups with increased transaction frequency and rising fluctuations, collect the marketing information sending time before marketing campaign and the first transaction time, calculate the time difference to generate a response delay curve, compare the curve position difference to obtain the concentrated response period, count the number of transactions in the period, and generate a high conversion transaction period.

[0037] The touchpoint priority sorting module is used to execute S3: based on the high conversion transaction time period, collect the daily activity curve of similar customers at financial service touchpoints, calculate the overlap ratio between the activity peak and the high conversion period, and sort them to generate the priority touchpoint delivery order;

[0038] The delivery time matching module is used to execute S4: select time points according to the priority touchpoint delivery order, calculate the time difference with the peak of the delay curve of the target customer, filter the matching time points, and generate a delivery time list;

[0039] The multi-touchpoint time integration module is used to execute S5: based on the delivery time list, it calculates the start, peak, and end time differences of customer touchpoint interaction times from multiple acquisition terminals, normalizes and adjusts them to a unified time axis, rearranges events, and generates a financial marketing management service plan.

[0040] Compared with the prior art, the advantages and positive effects of the present invention are:

[0041] In this invention, by serializing and analyzing the mean fluctuation of transaction behavior time intervals, changes in transaction activity are dynamically captured. The concentrated response interval is determined by combining the delay curve and high conversion time periods are extracted. The overlap ratio of these intervals with the peak activity of multiple touchpoints is analyzed to generate a priority touchpoint sequence and match precise time points. After normalization and sequence rearrangement of multi-source interactive data, the order of touchpoint and time delivery is optimized on a unified time axis to improve reach relevance and conversion efficiency. Attached Figure Description

[0042] To more clearly illustrate the technical solutions in the embodiments of the present invention, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0043] Figure 1 This is a schematic diagram of the steps of the present invention;

[0044] Figure 2 This is a detailed schematic diagram of S1 of the present invention;

[0045] Figure 3 This is a detailed schematic diagram of S2 of the present invention;

[0046] Figure 4 This is a detailed schematic diagram of S3 of the present invention;

[0047] Figure 5 This is a detailed schematic diagram of S4 of the present invention;

[0048] Figure 6 This is a detailed schematic diagram of S5 of the present invention;

[0049] Figure 7 This is a system module diagram of the present invention. Detailed Implementation

[0050] The technical solution of the present invention will now be described with reference to the accompanying drawings.

[0051] In embodiments of the present invention, words such as "exemplarily," "for example," etc., are used to indicate that something is an example, illustration, or description. Any embodiment or design described as "exemplary" in the present invention should not be construed as being more preferred or advantageous than other embodiments or designs. Specifically, the use of the word "exemplary" is intended to present the concept in a concrete manner. Furthermore, in embodiments of the present invention, the meaning expressed by "and / or" can be both, or either one.

[0052] In the embodiments of this invention, the terms "image" and "picture" may sometimes be used interchangeably. It should be noted that, without emphasizing the distinction between them, they convey the same meaning. Similarly, the terms "of," "corresponding (relevant)," and "corresponding" may sometimes be used interchangeably. It should be noted that, without emphasizing the distinction between them, they convey the same meaning.

[0053] In this embodiment of the invention, sometimes a subscript such as W1 may be written in a non-subscript form such as W1. When the difference is not emphasized, the meaning they express is the same.

[0054] To make the technical problems, technical solutions and advantages of the present invention clearer, a detailed description will be given below in conjunction with the accompanying drawings and specific embodiments.

[0055] Please see Figure 1 This invention provides a financial marketing management service method based on multi-source data fusion, comprising the following steps:

[0056] S1: Obtain the account transaction behavior records of the target customer group within a fixed time window, sort the records in ascending order of transaction time, calculate the time interval between adjacent transactions and summarize them into an interval sequence, calculate the mean and volatility value according to the interval sequence, compare the mean difference between adjacent time windows with the transaction rhythm change threshold, compare the change in volatility value with the transaction volatility threshold, and generate a transaction rhythm label for the customer group.

[0057] S2: Based on the customer group transaction rhythm tag, filter the customer group with increased transaction frequency and increased fluctuation. Collect the time of marketing information sent before the multi-channel marketing campaign and the time of the first transaction completed by the customer group. Calculate the time difference between the two to draw a response delay curve. Compare the position difference of the delay curve to obtain the concentrated response period. Count the number of transactions within the concentrated response period to generate high conversion transaction period.

[0058] S3: Based on the high-conversion transaction period, collect the daily activity curve of similar customers' online financial service touchpoints, calculate the overlap ratio between the peak activity position and the high-conversion transaction period, sort the touchpoints according to the ratio, and generate the priority touchpoint placement order;

[0059] S4: Call the priority touchpoint delivery order to select the time point corresponding to the first touchpoint in the sorting, calculate the time difference between the time point and the peak position of the target customer's delay curve, filter the matching time points based on the difference, and generate a delivery time list;

[0060] S5: Calculate the difference between the start, peak and end times of customer touchpoint interaction times recorded by multiple acquisition terminals based on the delivery time list, normalize and adjust the time difference of the different acquisition terminals to a unified time axis, rearrange the multi-touchpoint events according to the adjusted time, and generate a financial marketing management service plan.

[0061] Customer group transaction rhythm tags include the average transaction value change range, transaction volatility value change range, and transaction rhythm type. High conversion transaction time periods include concentrated response periods, concentrated response transaction frequency, and peak transaction conversion periods. Priority touchpoint placement order includes touchpoint overlap ratio, touchpoint priority, and touchpoint identifier. Placement time list includes matching time points, time difference range, and priority placement time points. Financial marketing management service solution includes multi-touchpoint event sequences under a unified timeline, touchpoint interaction time distribution, and results of differentiated adjustments between touchpoints.

[0062] Please see Figure 2 The specific steps of S1 are as follows:

[0063] S101: Obtain the account transaction records of the target customer group within a fixed time window, sort them in ascending order by transaction time, extract the time intervals between adjacent transactions and integrate them into a sequence to generate a transaction time interval sequence;

[0064] When retrieving account transaction records of a target customer group within a fixed time window, first define the scope of the time window, for example, setting it to be divided into 24-hour segments from 0:00 to 24:00 daily. Then, retrieve the transaction details of the target customer group within these hourly segments from the transaction database. The details should include fields such as customer ID, transaction time, transaction amount, and transaction type. Convert the transaction time into a uniform second-count format for easy comparison. Next, sort the transaction records within each time window in ascending order of the number of seconds. After sorting, read the time difference between adjacent transactions sequentially. For example, the interval between 09:10:15 and 09:25:45 is 930 seconds, and the interval between 09:25:45 and 09:45:30 is 1185 seconds. The intervals between all adjacent transactions of a customer within the current window are calculated one by one, and these intervals are arranged sequentially to form a sequence. If a customer has only one transaction within the time window, a null or zero value is recorded directly to ensure the integrity of the data structure. The interval sequences of all customers are summarized into a two-dimensional list by customer ID. Each row represents all intervals of a customer within the time window, and each column represents the interval value at different positions. During the generation process, duplicate transaction records and records that exceed the time window range need to be removed, such as records with timestamps before or after the start or end of the window. Abnormal timestamps (such as records less than 0 or exceeding 86400 seconds) are also cleaned up. Finally, the adjacent transaction time interval sequence is obtained by sorting the transaction times in ascending order.

[0065] S102: Based on the transaction time interval sequence, extract the centrality level and stability features, and integrate the two into a feature set to obtain the transaction interval statistical features;

[0066] When extracting center level and stability features based on transaction time interval sequences, the center level is first calculated for each customer's interval sequence within the current window. The center level can be directly represented by the average of all intervals. For example, if a customer's intervals in a certain window are 930 seconds, 1185 seconds, and 870 seconds, then the center level is 995 seconds. Next, stability is calculated. Stability is represented by the degree of interval fluctuation, which can be obtained by taking the absolute value of the difference between each interval and the center level and then averaging it. For example, if the differences between the aforementioned intervals and the center level are 65 seconds, 190 seconds, and 125 seconds, the average value is 126.67 seconds, yielding the customer's stability value within that window. The above requires dividing the center level and stability into intervals. The interval division is set based on the statistical results of historical transaction behavior. For example, customers with a center level between 0 and 600 seconds are classified as high-frequency transactions, those between 600 and 1800 seconds as medium-frequency transactions, and those greater than 1800 seconds as low-frequency transactions. Stability between 0 and 100 seconds is judged as low volatility, those between 100 and 500 seconds as medium volatility, and those greater than 500 seconds as high volatility. These intervals can be set by analyzing the distribution of historical data. For example, the percentile of all customer interval data in the past year can be selected as the interval boundary. The extracted center level and stability values ​​are combined into a set of features and stored in the feature set for subsequent analysis.

[0067] S103: Call the statistical features of the transaction interval, compare the changes in the central level and stability of adjacent time windows, and combine the transaction rhythm and volatility threshold to establish a transaction rhythm label for the customer group;

[0068] When comparing the changes in centrality and stability between adjacent time windows using transaction interval statistical features, it is necessary to first read the centrality and stability values ​​for each customer in two consecutive time windows. Then, the change is obtained by subtracting the value of the previous window from the value of the later window. For example, if customer A's centrality is 995 seconds in the previous window and 1200 seconds in the later window, the change in centrality is 205 seconds. Their stability is 126.67 seconds in the previous window and 200 seconds in the later window, resulting in a change in stability of 73.33 seconds. The change is then compared with a preset rhythm change threshold. The threshold can be set with reference to the historical data distribution; for example, the 75th percentile of the centrality change over the past year can be used as the threshold. The system takes a value, assuming a center level threshold of 300 seconds and a stability threshold of 150 seconds. Changes exceeding these thresholds are considered significant, while changes not exceeding these thresholds are considered insignificant. Labels are then assigned based on the combination of changes. For example, changes in both center level and stability exceeding the thresholds are labeled "rapid fluctuations," changes in center level not exceeding the threshold but changes in stability exceeding the threshold are labeled "stable fluctuations," changes in center level exceeding the threshold but changes in stability not exceeding the threshold are labeled "rapid and stable," and changes in neither exceeding the threshold are labeled "stable and steady." Finally, the labels corresponding to each customer in each time window are recorded in the transaction rhythm label set, forming complete rhythm change classification data.

[0069] Please see Figure 3 The specific steps of S2 are as follows:

[0070] S201: Obtain customer group transaction rhythm tags, filter customer groups with increased transaction frequency and increased volatility, integrate the customer group identifiers that meet the conditions into a set, and generate a target customer group set.

[0071] When obtaining transaction rhythm tags for a customer group, it is necessary to first read the tag values ​​of each customer from the generated transaction rhythm tag set, one by one, according to their identifier. Then, the tag content of each customer is evaluated. The filtering criteria are set as an increase in transaction frequency and an increase in volatility. The method for determining an increase in transaction frequency is that the change in the center level value between two adjacent time windows is less than zero. For example, if the center level of the previous window is 1200 seconds and the center level of the next window is 900 seconds, the change is 300 seconds, which is less than zero, indicating an increase in frequency. The method for determining an increase in volatility is that the change in stability is greater than zero. For example, if the previous window is stable... The interval is 100 seconds, the next window is 150 seconds, and the change amount is 50 seconds. If it is greater than zero, it is judged as an increase in fluctuation. Both conditions must be met simultaneously to be selected into the target customer group. In addition, a validity threshold for the change amount needs to be set to prevent interference from slight fluctuations. This threshold can be set according to the change distribution of transaction data over the past year. For example, the absolute value of the change amount greater than 60 seconds is taken as a valid change, and less than 60 seconds is judged as no change. This threshold value can be determined by rounding down the standard deviation of the change amount of historical samples. Customers that meet the conditions need to be deduplicated and stored in a new set, which is the target customer group set.

[0072] S202: Based on the target customer group set, collect the marketing information sending time and the first transaction completion time of the customer group before multi-channel marketing campaigns, perform time difference calculation on the two types of time data and arrange them in the order of customer groups to form a response delay curve;

[0073] When collecting data on the marketing message sending time and first transaction completion time of a target customer group before multi-channel marketing campaigns, it is necessary to first extract the marketing message sending time for each target customer from the multi-channel sending records. The multi-channel range includes SMS, in-site messages, push notifications, etc. For each customer, the earliest sending time within the analysis period is selected as the customer's marketing message sending time. Then, the first transaction time of the customer within the analysis period is selected from the transaction records, ensuring that the time is accurate to the second. Subsequently, the marketing message sending time is subtracted from the first transaction time to obtain the response delay value. For example, if the marketing message sending time is 09:00:00 and the first transaction time is 09:45:00, the delay time is 2700 seconds. The delay times of all customers are arranged in ascending order according to the customer identifier to form a delay time sequence. This sequence is then mapped to a curve coordinate system, with the horizontal axis representing the customer sequence number and the vertical axis representing the delay time in seconds. Connecting the points forms a response delay curve, where each point on the curve represents the response delay data of a customer.

[0074] S203: Call the response latency curve, compare the differences in the position of the curve on the time axis to obtain the concentrated response period, and count the number of transactions within the period. Combine the concentrated response period with the number of transactions to establish a high-conversion transaction period.

[0075] When comparing the position differences of the response latency curves on the time axis, the curve data is first scanned to identify the intervals where latency values ​​are concentrated. The threshold for determining concentrated response is set when the number of customers in this interval reaches or exceeds 20% of the total number of customers. This threshold can be determined based on the proportion of concentrated periods in historical activity data. For example, if the average proportion of concentrated periods is found to be around 18% after analyzing multiple activities, the threshold can be set to 20%. If, during the scanning process, it is found that the proportion of customers with latency between 1800 seconds and 3600 seconds reaches 25%, then this interval is determined to be a concentrated response period. Next, the number of transactions within this period is counted. For example, if 50 customers complete transactions in this period, and some customers make multiple transactions, the total number of transactions is counted as 65. The start latency, end latency, and number of transactions of this concentrated response period are combined into a set of data records, and multiple such records are generated sequentially to ultimately form a set of high-conversion transaction time periods.

[0076] Please see Figure 4 The specific steps of S3 are as follows:

[0077] S301: Collect the daily activity curves of similar customers' online financial service touchpoints based on high conversion transaction periods, identify the peak positions in the curves and record the corresponding time points, integrate the peak time information according to the touchpoint order, and generate a set of touchpoint peak times.

[0078] When collecting daily activity curves for similar customers' online financial service touchpoints based on high-conversion transaction periods, it is necessary to first count the number of times the target customer group is active during different time periods of the day for each touchpoint. The count of activity counts should be accurate to the minute or second, and the time and activity counts should be recorded to generate curve data. Then, the activity values ​​of the touchpoint within the day should be arranged in chronological order to form an activity curve array. For example, if the activity of touchpoint A is 5, 12, 18, 10, and 7 in the period from 0 seconds to 3600 seconds, then the peak activity value is 18, corresponding to the time point. For the 1800th second, the activity curve of the touch point is traversed to find all peak points, that is, the current activity value is greater than the activity value of the previous moment and the next moment. If there are multiple peaks, all peak positions and corresponding time points are recorded. Then, the peak time point information of all touch points is integrated in the order of touch points. Each touch point corresponds to one or more peak time points. For example, the peak of touch point A is 1800 seconds, the peak of touch point B is 2400 seconds and 4200 seconds, and the peak of touch point C is 3000 seconds. These time points are combined and stored to form a set of touch point peak times.

[0079] S302: Call the peak time set of touch points, calculate the time interval overlap between the peak time and the conversion transaction time period, calculate the overlap ratio percentage of touch points, and archive the ratio results in touch point order to obtain the touch point overlap ratio sequence.

[0080] The formula for calculating the percentage overlap of contacts is as follows:

[0081] ;

[0082] in, This represents the percentage overlap of the j-th contact point. This represents the number of peak time points for the j-th contact. This indicates the duration of the conversion transaction time period during which the j-th touchpoint and the i-th peak time occur. This represents the time value at the i-th peak time point of the j-th contact. This represents the time value of the center point of the conversion transaction time period that matches the i-th peak time point of the j-th touchpoint. This represents the discriminant factor indicating whether the peak time of the j-th contact and its matching time period effectively overlap. Select 1 if the value is 1, otherwise select 0. This represents the confidence weight coefficient for the i-th peak time point at the j-th contact. This represents the sum of the durations of all matched conversion transactions at the j-th touchpoint.

[0083] The calculation logic of this formula is as follows: First, by using the absolute value... This value represents the time distance of the peak from the center position, and is then used from the duration. The deduction reflects the effective overlap duration, which is then multiplied by the confidence level. With discriminant factor After filtering out invalid or significantly offset peak contributions, the effective overlap of all peaks is summed and then multiplied by the total duration of all segments. The overlap ratio is calculated and then converted into a percentage.

[0084] Data collection and calculation examples:

[0085] Taking touchpoint B as an example, the peak activity times collected on that day are as follows:

[0086] First peak: Second;

[0087] Second peak: Second;

[0088] The third peak: Second;

[0089] These peak values ​​correspond to the high-conversion transaction periods:

[0090] Time Period 1: Start and End , , Second;

[0091] Time Period 2: Start and End , , Second;

[0092] Time Period 3: Start and End , , Second;

[0093] Based on the number of user interactions within a peak ±300-second window, the interaction volume in each peak neighborhood is calculated as follows:

[0094] Peak 1: 110 times (maximum value);

[0095] Peak 2: 95 times;

[0096] Peak value 3: 80 times;

[0097] Therefore, we can conclude that:

[0098] , , ;

[0099] Then calculate the peak offset:

[0100] ;

[0101] ;

[0102] ;

[0103] Calculate by substituting the effective overlap value for each item:

[0104] Item 1: ;

[0105] Item 2: ;

[0106] Item 3: ;

[0107] Summing up the total numerators, we get:

[0108] ;

[0109] Total duration :

[0110] ;

[0111] Substituting into the formula, we get:

[0112] ;

[0113] Table 1. Contact Point B Parameter Acquisition and Formula Calculation Table

[0114]

[0115] As shown in Table 1, the three peaks effectively overlap, and the total contribution after confidence weighting is 1403.20 seconds. With a normalized denominator of 2400 seconds, the final calculated overlap ratio is 58.47%.

[0116] Comparison and Conclusion:

[0117] Assuming a baseline value This value is set based on the median overlap ratio of 12 historical marketing campaigns. When touchpoint B calculates... If this occurs, it indicates that the contact point is slightly lower than the baseline, and its priority may be ranked below the middle in subsequent contact point sorting.

[0118] Explanation of the formula's innovations:

[0119] The advantage of the formula lies in the introduction of confidence weights. It achieves a quantitative expression of peak behavior intensity, and combines offset absolute value calculation and discriminant factor. By controlling the timing and effectiveness of peak periods, a unified comparison benchmark is obtained under the normalization of the total duration, enabling a quantifiable assessment of the active conversion overlap capability of multiple touchpoints; this design provides a clear numerical basis for the subsequent ranking of deployment strategies.

[0120] S303: Based on the contact overlap ratio sequence, sort the contacts from largest to smallest to generate a priority contact delivery order;

[0121] When determining the contact overlap ratio sequence, the contacts in the sequence must first be compared according to their overlap ratio values. The values ​​are then rearranged from largest to smallest. The comparison process involves taking the overlap ratio of two contacts one by one. If the ratio of the first contact is greater than that of the second contact, the first contact is placed first. If the ratios are equal, the order can be determined by the lexicographical order of the contact identifiers or by a random method. For example, if the overlap ratio of contact A in the sequence is 80%, contact B is 66.67%, and contact C is 50%, the sorting result is A, B, C. After sorting, the new order is recorded as the priority contact deployment order. Each element in this order list is a contact identifier, and the position represents the priority. Finally, the priority contact deployment order is generated.

[0122] Please see Figure 5 The specific steps of S4 are as follows:

[0123] S401: Call the priority touchpoint delivery order, select the touchpoints with the highest order, obtain the corresponding time point data, and organize the touchpoint time points into a set according to the time order to generate the priority touchpoint active time set;

[0124] When invoking the priority touchpoint delivery order, the system first reads the generated priority touchpoint delivery order list, selecting the top-ranked touchpoints sequentially from the first position in the list. For example, selecting the top three touchpoints A, B, and C. Then, it accesses the activity database or log file to retrieve the time point data for each touchpoint. The time points must be recorded in seconds to ensure time precision. For example, the time points for touchpoint A are 1200 seconds, 2400 seconds, and 3600 seconds; for touchpoint B, they are 1800 seconds and 3000 seconds; and for touchpoint C, they are 1500 seconds, 2700 seconds, and 3900 seconds. Subsequently, these time points are rearranged in ascending order of time value. During the rearrangement process, the touchpoint identification information corresponding to the time points must be retained for subsequent traceability. At the same time, duplicate time points are checked and merged. Finally, the processed time point data is stored in a set in chronological order. This set is the priority touchpoint active time set.

[0125] S402: Based on the set of active times of priority touchpoints, extract the peak position of the delay curve of the target customer, calculate the time difference with the time point, compare the difference result with the matching threshold, filter the time points that meet the threshold conditions, and obtain the set of matching time points;

[0126] When extracting the peak position of the delay curve for a target customer based on the set of active times of priority touchpoints, it is necessary to first identify the peak points from the delay curve data. The criterion for determining a peak is that the delay value at that time point is greater than the delay values ​​of the two adjacent time points. After identifying all peak positions, the peak time points are extracted by customer or group dimension. For example, the peak times of the delay curve for the target customer are 1300 seconds, 2500 seconds, and 3700 seconds. Then, the difference between each peak time point and each time point in the set of active times of priority touchpoints is calculated. The calculation method is to take the absolute difference. For example, the difference between the peak of 1300 seconds and the active point of 1200 seconds is 10. The difference between 0 seconds, 2500 seconds (peak time), and 2400 seconds (active time) is 100 seconds. The difference between 3700 seconds (peak time) and 3600 seconds (active time) is also 100 seconds. After calculation, the difference is compared with the matching threshold. The matching threshold can be set based on the synchronicity distribution of historical delays and active times. For example, if the median of the difference between 50% of the peak time and the active time in past activities is 150 seconds, then the threshold is set to 150 seconds. The difference is less than or equal to the threshold and is considered a matching point. The time points that meet the conditions are recorded in the matching time point set. Each element in the set contains the matched peak time and the corresponding active time.

[0127] S403: Call the set of matching time points, arrange them in chronological order, remove duplicate records, and create a list of delivery times;

[0128] When calling the matching time point set, all matching time points in the set must first be re-sorted in ascending order of time value. During the sorting process, the time point values ​​are directly compared. For example, if the time points in the set are 1200 seconds, 1300 seconds, 2500 seconds, 2400 seconds, and 3700 seconds, then after sorting, they will be 1200 seconds, 1300 seconds, 2400 seconds, 2500 seconds, and 3700 seconds. After sorting, it is necessary to check for duplicate records. The condition for determining duplicates is that the time point values ​​are the same and the corresponding peak value and the source of the active point are consistent. If duplicates exist, one record is kept and the extra records are deleted. The deduplicated time points are stored in the delivery time list in order of sorting results. Each time point in the list corresponds to a specific delivery opportunity, and finally, a complete delivery time list is obtained.

[0129] Please see Figure 6 The specific steps of S5 are as follows:

[0130] S501: Based on the delivery time list, collect customer touchpoint interaction times recorded by multiple collection terminals, sequentially obtain the start time, peak time and end time, calculate the difference between the three types of times, classify and organize the calculation results according to the collection terminal, and generate a touchpoint time difference set.

[0131] When collecting customer touchpoint interaction times from multiple collection terminals based on the campaign time list, it is necessary to first read each time point in the campaign time list, and then use that time point as an index to extract the corresponding touchpoint's interaction record from the interaction logs of multiple collection terminals. Each record must include a start time, peak time, and end time. The start time is the time when the touchpoint first triggers an interaction, the peak time is the time when the activity level reaches its maximum during the interaction, and the end time is the time when the touchpoint interaction is completed or interrupted. For example, in collection terminal 1, a customer's start time is 1200 seconds, peak time is 1500 seconds, and end time is 1800 seconds. In collection terminal 2, the same touchpoint's start time... The start time is 1250 seconds, the peak time is 1550 seconds, and the end time is 1820 seconds. Then, the peak delay is obtained by subtracting the start time from the peak time for each record within the same acquisition terminal, the end delay is obtained by subtracting the peak time from the end time, and the total interaction time is obtained by subtracting the start time from the end time. For example, the peak delay of acquisition terminal 1 is 300 seconds, the end delay is 300 seconds, and the total duration is 600 seconds. These difference results are organized separately according to the acquisition terminal, and the difference results of each acquisition terminal are stored separately. Finally, a set of touch point time differences is generated. This set is based on the acquisition terminal as a dimension, and each dimension contains three types of time difference data for all touch points within that terminal.

[0132] S502: Based on the contact time difference set, identify the time difference records that differ between the acquisition terminals, normalize and adjust the differentiated time differences according to the unified time axis reference value, and integrate the adjusted records into a unified structure to obtain a unified time axis record set.

[0133] When identifying time difference records with discrepancies between acquisition terminals based on contact point time difference sets, the time differences of the same contact points in different acquisition terminals are first compared one by one. The comparison method is to directly calculate the absolute difference. For example, if the peak delays of acquisition terminal 1 and acquisition terminal 2 at the same contact point are 300 seconds and 350 seconds respectively, then the difference is 50 seconds. If the difference exceeds the preset difference threshold, it is determined that there is a difference. The difference threshold can be set by analyzing the consistency of historical multi-terminal records. For example, the average and standard deviation of all historical differences are calculated, and the average plus the standard deviation is rounded down to obtain the threshold. Assuming it is 40 seconds, then in this example, 50 seconds has exceeded the threshold. Records with discrepancies are identified as such. For these records, normalization adjustment is required based on a unified timeline benchmark. The benchmark can be set as the average of the earliest start times among all acquisition terminal records. For example, after calculating the earliest start times of multiple touch points on different terminals, the average value is taken as 1000 seconds. Then, the time differences of each terminal are adjusted proportionally to the timeline starting from 1000 seconds. During the adjustment, the proportional relationship between the time differences must be preserved. The adjusted data is archived by touch point and integrated into a data record with a unified structure, ultimately resulting in a unified timeline record set. All touch point time data in this record set are aligned under the same benchmark.

[0134] S503: Call the unified timeline record set, rearrange the multi-touchpoint events according to the adjusted time order, and integrate them into executable marketing management content to establish a financial marketing management service solution;

[0135] When calling a unified timeline record set, the multi-touchpoint events need to be rearranged according to the adjusted time order. Specifically, the start time of all touchpoints in the record set is extracted first and sorted by time value from smallest to largest. When the start times are the same, they are sorted by peak time. If the peak times are also the same, they are sorted by end time. For example, after adjustment, touchpoint A's start time is 1100 seconds, touchpoint B's is 1150 seconds, and touchpoint C's is 1130 seconds, so the sorting result is A, C, B. After sorting, the detailed information of these touchpoints is integrated in sequence, including start time, peak time, end time, and time difference data. The time relationship between adjacent touchpoints is preserved to form a continuously executable event chain. Finally, these sequentially arranged touchpoint event sets are organized into structured marketing management content. Each record contains a touchpoint identifier and its corresponding three types of time information, ultimately establishing a complete financial marketing management service solution.

[0136] Please see Figure 7 A financial marketing management service system based on multi-source data fusion includes:

[0137] The transaction rhythm analysis module is used to execute S1: obtain the account transaction records of the target customer group in a fixed time window, sort them in ascending order by transaction time, calculate the time interval between adjacent transactions to form an interval sequence, calculate the mean and volatility value respectively, compare the mean difference between adjacent time windows with the transaction rhythm change threshold, and compare the change in volatility value with the transaction volatility threshold to generate a transaction rhythm label for the customer group.

[0138] The frequency fluctuation filtering module is used to execute S2: filter customer groups with increasing transaction frequency and rising fluctuations based on customer group transaction rhythm tags, collect the time of marketing information sent before marketing campaign and the time of the first transaction, calculate the time difference to generate a response delay curve, compare the curve position difference to obtain the concentrated response period, count the number of transactions in the period, and generate high conversion transaction period.

[0139] The touchpoint priority sorting module is used to execute S3: based on the high conversion transaction period, collect the daily activity curve of similar customers at financial service touchpoints, calculate the overlap ratio between the activity peak and the high conversion period, and sort them to generate the priority touchpoint delivery order;

[0140] The delivery time matching module is used to execute S4: select time points according to the priority touchpoint delivery order, calculate the time difference with the peak of the delay curve of the target customer, filter the matching time points, and generate a delivery time list;

[0141] The multi-touchpoint time integration module is used to execute S5: based on the delivery time list, it calculates the start, peak, and end time differences of customer touchpoint interaction times from multiple acquisition terminals, normalizes and adjusts them to a unified timeline, rearranges events, and generates a financial marketing management service plan.

[0142] The above description is merely a specific embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the technical scope disclosed in the present invention should be included within the scope of protection of the present invention. Therefore, the scope of protection of the present invention should be determined by the scope of the claims.

Claims

1. A financial marketing management service method based on multi-source data fusion, characterized in that: The following steps are involved: S1: Obtain the account transaction records of the target customer group within a fixed time window, sort them in ascending order by transaction time, calculate the time interval between adjacent transactions to form an interval sequence, calculate the mean and volatility values ​​respectively, compare the mean difference between adjacent time windows with the transaction rhythm change threshold, and compare the volatility value change with the transaction volatility threshold to generate a customer group transaction rhythm label. S2: Based on the customer group transaction rhythm tags, filter customer groups with increased transaction frequency and rising fluctuations, collect the time of marketing information sent before marketing campaign and the time of the first transaction, calculate the time difference to generate a response delay curve, compare the position difference of the curve to obtain the concentrated response period, count the number of transactions in the period, and generate a high conversion transaction period. S3: Based on the high-conversion transaction period, collect the daily activity curves of similar customers at financial service touchpoints, calculate the overlap ratio between the activity peak and the high-conversion period, and sort them to generate the priority touchpoint delivery order; S4: Select time points according to the priority touchpoint delivery order, calculate the time difference with the peak of the delay curve of the target customer, filter the matching time points, and generate a delivery time list.

2. The financial marketing management service method based on multi-source data fusion according to claim 1, characterized in that, The customer group transaction rhythm tags include the average transaction change range, the transaction fluctuation value change range, and the transaction rhythm type. The high conversion transaction time period includes the concentrated response period, the number of concentrated response transactions, and the peak transaction conversion period. The priority touchpoint placement order includes the touchpoint overlap ratio, touchpoint priority, and touchpoint identifier. The placement time list includes the matching time point, the time difference range, and the priority placement time point.

3. The financial marketing management service method based on multi-source data fusion according to claim 1, characterized in that, The specific steps of S1 are as follows: S101: Obtain the account transaction records of the target customer group within a fixed time window, sort them in ascending order by transaction time, extract the time intervals between adjacent transactions and integrate them into a sequence to generate a transaction time interval sequence; S102: Based on the transaction time interval sequence, extract the center level and stability features, and integrate the two into a feature set to obtain the transaction interval statistical features; S103: Call the transaction interval statistical features, compare the changes in the center level and stability of adjacent time windows, and combine the transaction rhythm and fluctuation threshold to establish a customer group transaction rhythm label.

4. The financial marketing management service method based on multi-source data fusion according to claim 1, characterized in that, The specific steps of S2 are as follows: S201: Obtain the transaction rhythm tags of the customer groups, filter customer groups with increased transaction frequency and increased volatility, integrate the customer group identifiers that meet the conditions, and generate a target customer group set. S202: Based on the target customer group set, collect the marketing information sending time and the first transaction completion time of the customer group before the multi-channel marketing campaign, perform time difference calculation on the two types of time data and arrange them in the order of customer groups to form a response delay curve; S203: Call the response delay curve, compare the position difference of the curve on the time axis to obtain the concentrated response period, and count the number of transactions within the period. Combine the concentrated response period with the number of transactions to establish a high conversion transaction period.

5. The financial marketing management service method based on multi-source data fusion according to claim 1, characterized in that, The specific steps for S3 are as follows: S301: Based on the high-conversion transaction period, collect the daily activity curves of similar customers' online financial service touchpoints, identify the peak positions in the curves and record the corresponding time points, integrate the peak time information according to the touchpoint order, and generate a set of touchpoint peak times. S302: Call the set of peak times of the touch points, calculate the time interval overlap between the peak times and the conversion transaction time period, calculate the overlap ratio percentage of the touch points, and archive the ratio results in the order of the touch points to obtain the touch point overlap ratio sequence. S303: Based on the contact overlap ratio sequence, sort the contacts from largest to smallest to generate a priority contact delivery order.

6. The financial marketing management service method based on multi-source data fusion according to claim 1, characterized in that, The specific steps of S4 are as follows: S401: Call the priority touchpoint delivery order, select the touchpoints with higher sorting, obtain the corresponding time point data, and organize the touchpoint time points into a set according to time order to generate a priority touchpoint active time set; S402: Based on the set of active times of the priority touchpoints, extract the peak position of the delay curve of the target customer, calculate the time difference with the time point, compare the difference result with the matching threshold, filter the time points that meet the threshold conditions, and obtain the set of matching time points. S403: Call the set of matching time points, arrange them in chronological order, remove duplicate records, and create a list of delivery times.

7. The financial marketing management service method based on multi-source data fusion according to claim 1, characterized in that, The method further includes: S5: Based on the aforementioned campaign time list, calculate the start, peak, and end time differences of customer touchpoint interaction times from multiple acquisition terminals, normalize and adjust them to a unified timeline, rearrange the events, and generate a financial marketing management service plan.

8. The financial marketing management service method based on multi-source data fusion according to claim 7, characterized in that, The financial marketing management service solution includes a multi-touchpoint event sequence under a unified timeline, touchpoint interaction time distribution, and differentiated adjustment results between touchpoints.

9. The financial marketing management service method based on multi-source data fusion according to claim 7, characterized in that, The specific steps of S5 are as follows: S501: Based on the delivery time list, collect the customer touchpoint interaction time recorded by multiple collection terminals, obtain the start time, peak time and end time in sequence, calculate the difference between the three types of time, classify and organize the calculation results according to the collection terminal, and generate a touchpoint time difference set. S502: Based on the contact time difference set, identify time difference records with differences between the acquisition terminals, normalize and adjust the differentiated time differences according to the unified time axis reference value, integrate the adjusted records into a unified structure, and obtain a unified time axis record set. S503: Call the unified timeline record set, rearrange the multi-touchpoint events according to the adjusted time order, and integrate them into executable marketing management content to establish a financial marketing management service solution.

10. A financial marketing management service system based on multi-source data fusion, characterized in that: The system is used to implement the financial marketing management service method based on multi-source data fusion as described in any one of claims 1-9, and the system includes: The transaction rhythm analysis module is used to execute S1: obtain the account transaction records of the target customer group in a fixed time window, sort them in ascending order by transaction time, calculate the time interval between adjacent transactions to form an interval sequence, calculate the mean and volatility value respectively, compare the mean difference between adjacent time windows with the transaction rhythm change threshold, and compare the change in volatility value with the transaction volatility threshold to generate a transaction rhythm label for the customer group. The frequency fluctuation filtering module is used to execute S2: based on the customer group transaction rhythm tag, filter customer groups with increased transaction frequency and rising fluctuations, collect the marketing information sending time before marketing campaign and the first transaction time, calculate the time difference to generate a response delay curve, compare the curve position difference to obtain the concentrated response period, count the number of transactions in the period, and generate a high conversion transaction period. The touchpoint priority sorting module is used to execute S3: based on the high conversion transaction time period, collect the daily activity curve of similar customers at financial service touchpoints, calculate the overlap ratio between the activity peak and the high conversion period, and sort them to generate the priority touchpoint delivery order; The delivery time matching module is used to execute S4: select time points according to the priority touchpoint delivery order, calculate the time difference with the peak of the delay curve of the target customer, filter the matching time points, and generate a delivery time list; The multi-touchpoint time integration module is used to execute S5: based on the delivery time list, it calculates the start, peak, and end time differences of customer touchpoint interaction times from multiple acquisition terminals, normalizes and adjusts them to a unified time axis, rearranges events, and generates a financial marketing management service plan.

Images