Operation Management System and Methods Based on Financial Marketing Big Data

Abstract

This invention relates to the field of big data analytics, specifically to an operation and management system and method based on financial marketing big data. The system includes a data standardization module, a customer profiling disambiguation module, a strategy matching module, a real-time transmission module, and an operation strategy generation module. In this invention, a unified data table is constructed through joint judgment of field semantics and numerical rules to improve data consistency. Customer information integration is performed based on the time, credibility level, and logical verification of field conflicts to enhance identification accuracy. Transaction amount, product relevance, and interaction frequency are combined with strategy attributes to determine matching relationships, enabling dynamic strategy filtering. Push content is filtered and arranged according to strategy urgency and channel status to improve transmission efficiency. Customer clicks, conversions, and feedback behavior trends are determined, supporting linked strategy updates. The overall processing chain has a closed-loop structure, improving fusion accuracy and strategy response timeliness under conditions of data fragmentation and multi-source heterogeneity.

Description

Technical Field

[0001] This invention relates to the field of big data analytics, and in particular to an operation management system and method based on financial marketing big data. Background Technology

[0002] The field of big data analytics encompasses a technological system that leverages massive amounts of heterogeneous data to extract valuable information and drive decision-making. Its core lies in constructing high-precision analytical models through multi-source data collection, cleaning, and structured processing, and then combining this with business scenarios to achieve data-driven strategy optimization. This field systematically covers the collection, storage, computation, and visualization processes of user behavior data, transaction data, and market dynamic data in financial marketing scenarios. It focuses on solving problems such as data fragmentation, lagging analysis, and static strategies, providing financial institutions with real-time, dynamic marketing decision support.

[0003] The operation and management system based on big data in financial marketing refers to a technical solution for marketing financial products and services. It integrates structured correlations of user transaction records, social media interaction information, and market trend data to analyze and construct customer behavior prediction models and generate dynamic marketing strategies. This system addresses technical aspects such as cross-platform data standardization and fusion, multi-dimensional customer characteristics, and quantitative analysis of potential demand correlations. Specifically, it cleanses unstructured data to build a customer tagging system, establishes classification and prediction rules based on historical behavioral data, and optimizes marketing paths by updating the strategy library with real-time data streams.

[0004] Existing technologies generally suffer from insufficient structured entry points, static data fusion rules, and delayed strategy response during data processing. Faced with multi-source heterogeneous data, the lack of unified parsing standards for field naming and values ​​often leads to inconsistent behavioral data structures, impacting the standard modeling process. In the customer profiling stage, data field conflicts from different sources fail to be assessed for reliability, and field integration often relies on chronological order, lacking logical matching principles and easily causing user characteristic distortion. Regarding strategy matching, tag binding is often based on fixed rules, failing to reflect the current customer behavior state. Strategy pushes generally use static path settings, ignoring network fluctuations and transmission resource load, resulting in strategy response delays or failures. At the result feedback level, click and conversion data are only used for statistical purposes and fail to participate in strategy adjustment and optimization logic, resulting in a one-way marketing process and an inability to achieve real-time strategy updates and customer behavior linkage. Taking customer transaction volume as an example, when it fluctuates drastically within a short period, existing technologies lack the ability to adjust strategy recommendations based on behavioral fluctuation trends, easily causing strategy failures and affecting the continuity of the customer response chain. Summary of the Invention

[0005] The purpose of this invention is to address the shortcomings of existing technologies by proposing an operation management system and method based on big data in financial marketing.

[0006] To achieve the above objectives, the present invention adopts the following technical solution: An operation and management system based on financial marketing big data includes:

[0007] The data standardization module acquires user transaction records, social media interaction records, and market trend data in financial marketing scenarios, analyzes the semantic matching degree of field names, verifies whether the field distribution conforms to business rules, and generates a standardized customer behavior data table.

[0008] The customer profiling disambiguation module, based on the standardized customer behavior data table, identifies conflicting information of the same customer from different data sources, compares data update time, data source credibility level and field verification results, and generates a unified customer profile table.

[0009] The strategy matching module calls the transaction amount, product relevance, and interaction frequency in the unified customer profile table, combines them with the target product, calculates the customer value and strategy fit, and obtains the strategy matching list;

[0010] The real-time transmission module filters the unique identification code of the customer to be transmitted and the policy content based on the policy urgency in the policy matching list, and performs channel priority sorting according to the transmission path delay level, packet loss and node load status to generate a policy transmission queue.

[0011] The operation strategy generation module calls the customer click records, conversion rates and market feedback information in the strategy transmission queue, and combines them with the return on investment and customer retention period to generate a financial marketing strategy optimization plan.

[0012] As a further aspect of the present invention, the standardized customer behavior data table includes field semantic mapping results, field value distribution rationality verification results, and unified field format data; the unified customer profile table includes conflict information disambiguation results, data credibility level assessment results, and logical consistency verification results; the strategy matching list includes customer value assessment results, strategy fit score, and target strategy scheme items; the strategy transmission queue includes a unique customer identifier, strategy content summary, and transmission path scheduling parameters; and the financial marketing strategy optimization scheme includes customer click behavior data, conversion effect feedback, and strategy benefit assessment results.

[0013] As a further aspect of the present invention, the data standardization module includes:

[0014] The semantic recognition submodule acquires user transaction record fields, social media interaction fields, and market trend fields, calculates the degree of character similarity between field names, the number of times field names appear in combination in the total data, and the degree of comparison with data source labels, selects field pairs whose character similarity and number of times they appear in combination both exceed the field fit benchmark value and the combination frequency benchmark value, and generates field semantic consistency matching degree.

[0015] The numerical judgment submodule analyzes the change range, frequency of occurrence and data fluctuation of field values ​​at different time points based on the transaction frequency value, keyword frequency value and trend indicator value corresponding to the field pairs in the semantic consistency matching degree of the field. It compares the overlap between the actual change range of the field value and the logical benchmark range to obtain the logical conformity ratio of the field value.

[0016] The behavior integration submodule, based on the logically consistent data items in the field value logical consistency ratio, calls the behavior tag, time tag, and source tag to which the field belongs, and performs same-dimensional merging, cross-category integration, and unified interval format processing on the transaction frequency value, interaction frequency value, and trend indicator value to obtain a standardized customer behavior data table.

[0017] As a further aspect of the present invention, the specific calculation formula for the variation range of the analysis field value at the difference time node is as follows:

[0018] ;

[0019] in, Represents the field value in the field pair The change range index, Represents field pairs At the point of time field values, Represents field pairs At the point of time field values, Represents field pairs The frequency value of the associated keywords, Represents field pairs The transaction frequency value within the analysis period, Represents field pairs Trend indicator values ​​within the analysis period.

[0020] As a further aspect of the present invention, the customer profiling disambiguation module includes:

[0021] The field validation submodule obtains customer identifier, contact information, and transaction information field data from the standardized customer behavior data table. Based on the field structure, internal code consistency, and filling frequency, it calculates the format matching rate, internal code matching rate, and completeness value, judges the difference with the field consistency benchmark value, and generates the field consistency deviation rate.

[0022] The data optimization submodule calls the field consistency deviation rate, combines the field update time interval, the confidence level ranking value and the consistency offset value, and sorts and filters according to the corresponding confidence interval to generate the field confidence level ranking value after filtering.

[0023] The Customer Unification Submodule performs priority matching and conflict filtering on field values ​​of the same customer under different data sources based on the sorted confidence level values ​​of the filtered fields, merges field values, and generates a unified customer profile table.

[0024] As a further aspect of the present invention, the strategy matching module includes:

[0025] The transaction feature recognition submodule obtains the transaction amount, product relevance, and interaction frequency from the unified customer profile table, and performs data fusion processing on the customer dimension by combining the three items. By setting the order, a unified representation value is formed to obtain the customer behavior intensity value.

[0026] The value fit calculation submodule calls the customer behavior intensity value, combines it with the target product list in the marketing strategy library, and performs item-level screening based on the product correlation information in the customer profile, identifies the density of corresponding relationships, and performs mapping transformation based on the customer behavior intensity value to obtain the strategy fit intensity value.

[0027] The strategy matching list generation submodule locates the top-ranked customer identification information based on the strategy fit strength value, summarizes the target product and customer profile number content, and combines them with the strategy fit strength value to generate a strategy matching degree list.

[0028] As a further aspect of the present invention, the specific formula for calculating the density of the identified correspondence is as follows:

[0029] ;

[0030] Where D represents the degree of relationship between the customer and the target product in the item-level screening. This represents the customer's rating of the relevance of the Kth product feature. This represents the weighted score assigned to the target product on the Kth product feature. This represents the customer's average relevance score across all product features. This represents the average set score for the target product across all product features, where n represents the total number of product features. This represents the weighted total match value between the customer and the target product across all feature dimensions. This represents the product of the average customer preference value and the average target value of the product. It represents the square root of the product.

[0031] As a further aspect of the present invention, the real-time transmission module includes:

[0032] The urgency filtering submodule obtains the urgency level of the strategies in the strategy matching list and the corresponding unique customer identification code, compares the urgency level with the strategy urgency threshold, filters strategies and customer identification codes with urgency levels higher than the threshold, and generates a high-urgency strategy tag set.

[0033] The path evaluation submodule calls the unique customer identification code in the high urgency policy tag set to obtain the delay level, packet loss status and node load status of the corresponding transmission path. It compares these with the delay threshold, packet loss rate standard value and node load benchmark value, records the paths that meet the standards, and generates a set of available path identification codes.

[0034] The queue generation submodule sorts the policies and customer identification codes in the available path identification code set according to the policy urgency, arranges the transmission order, binds the path identifier, policy content and customer identification code, and generates a policy transmission queue.

[0035] As a further aspect of the present invention, the operation strategy generation module includes:

[0036] The click behavior extraction submodule obtains customer click records in the strategy transmission queue, identifies the distribution characteristics of click time, page position and click frequency under product category, and generates customer click offset rate by combining the change in the number of clicks and page dwell time in the same time period.

[0037] The conversion rate matching calculation submodule calls the customer click offset rate, identifies the conversion time and status type in the conversion record, extracts the temporal relationship between click behavior and conversion, and performs matching analysis with the difference in the range of customer retention time. Combined with the fluctuation of the number of conversion statuses, the conversion linkage change amount is obtained.

[0038] The strategy priority ranking submodule calls the conversion linkage change amount, refers to the investment return rate of financial products in the current period, selects the product set in the high range according to the ranking of the return rate in the product set, extracts the conversion linkage change amount of related customers, identifies the value performance of customers under different strategy configurations, and arranges the strategy execution order according to the value performance to generate a financial marketing strategy optimization plan.

[0039] The operational management method based on big data in financial marketing includes the following steps:

[0040] S1: Obtain the field content from transaction, interaction and market data in financial marketing scenarios, make consistency judgments based on the semantic matching of field names and the logical distribution of field values, and generate a standardized customer behavior data table;

[0041] S2: Based on the standardized customer behavior data table, identify the same customer entry from the sources of difference, determine field conflict information according to data time, source level and field verification status, and generate a unified customer profile table;

[0042] S3: Call the key fields in the unified customer profile table, combine them with the product tags in the strategy library, determine the matching relationship between customer characteristics and strategy content, and generate a strategy matching list;

[0043] S4: Based on the urgency level in the policy matching list, filter the content to be pushed, and perform channel priority sorting in combination with path delay, node load and packet loss to generate a policy transmission queue;

[0044] S5: Call the push records in the strategy transmission queue, collect click, conversion and feedback data, combine input and retention information to judge changes in customer behavior, and generate a financial marketing strategy optimization plan.

[0045] Compared with the prior art, the advantages and positive effects of the present invention are as follows:

[0046] In this invention, a unified data table is constructed by jointly judging field semantics and numerical rules to improve data consistency. Customer information is integrated based on the time, credibility level, and logical verification of field conflicts to enhance identification accuracy. Transaction amount, product relevance, and interaction frequency are combined with strategy attributes to determine matching relationships, enabling dynamic strategy filtering. Push content is filtered and arranged according to the urgency of the strategy and the channel status to improve transmission efficiency. Customer clicks, conversions, and feedback participation behavior trends are judged to support strategy update linkage. The overall processing chain has a closed-loop structure, improving the fusion accuracy and strategy response time under the conditions of data fragmentation and multi-source heterogeneity. Attached Figure Description

[0047] Figure 1 This is a system flowchart of the present invention;

[0048] Figure 2 This is a flowchart of the sub-modules of the present invention;

[0049] Figure 3 This is a flowchart of the method steps of the present invention. Detailed Implementation

[0050] To make the objectives, technical solutions, and advantages of this invention clearer, the invention will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the invention.

[0051] In the description of this invention, it should be understood that the terms "length," "width," "upper," "lower," "front," "rear," "left," "right," "vertical," "horizontal," "top," "bottom," "inner," and "outer," etc., indicating orientation or positional relationships, are based on the orientation or positional relationships shown in the accompanying drawings and are only for the convenience of describing the invention and simplifying the description, and do not indicate or imply that the device or element referred to must have a specific orientation, or be constructed and operated in a specific orientation, and therefore should not be construed as a limitation of the invention. Furthermore, in the description of this invention, "a plurality of" means two or more, unless otherwise explicitly specified.

[0052] Please see Figure 1 The operation and management system based on financial marketing big data includes:

[0053] The data standardization module detects user transaction records, social media interaction records, and market trend data in financial marketing scenarios, analyzes the semantic matching degree of field names, verifies whether the distribution of field values ​​conforms to business logic rules, and generates a standardized customer behavior data table.

[0054] The customer profiling disambiguation module is based on a standardized customer behavior data table. It identifies conflicting information about the same customer from different data sources, compares data update time, data source credibility level, and field logic validation results, and generates a unified customer profile table.

[0055] The strategy matching module calls the transaction amount, product relevance, and interaction frequency from the unified customer profile table, combines them with the target products in the marketing strategy library, calculates the customer value and strategy fit, and generates a strategy matching list.

[0056] The real-time transmission module filters the unique customer identifier and policy content to be transmitted based on the policy urgency in the policy matching list, evaluates the latency level, packet loss and node load status of the transmission path, and generates a policy transmission queue.

[0057] The operation strategy generation module calls customer click records, conversion rates, and market feedback information in the strategy transmission queue, and combines them with return on investment and customer retention period to generate financial marketing strategy optimization solutions.

[0058] The standardized customer behavior data table includes field semantic mapping results, field value distribution rationality verification results, and unified field format data. The unified customer profile table includes conflict information disambiguation results, data credibility level assessment results, and logical consistency verification results. The strategy matching list includes customer value assessment results, strategy fit score, and target strategy solution items. The strategy transmission queue includes customer unique identifier, strategy content summary, and transmission path scheduling parameters. The financial marketing strategy optimization solution includes customer click behavior data, conversion effect feedback, and strategy benefit assessment results.

[0059] Please see Figure 2 The data standardization module includes:

[0060] The semantic recognition submodule acquires user transaction record fields, social media interaction fields, and market trend fields, calculates the degree of character similarity between field names, the number of times field names appear in combination in the total data, and the degree of comparison with data source labels, selects field pairs whose character similarity and number of times they appear in combination both exceed the field fit benchmark value and the combination frequency benchmark value, and generates field semantic consistency matching degree.

[0061] The semantic recognition submodule extracts the required fields from multiple data sources, including fields such as transaction time and transaction amount in user transaction records, comments and reposts in social media, and popularity changes in market trend data. It then evaluates the character similarity between different field names, comparing the fields "transaction amount" and "transaction volume" and calculating the differences between characters to determine the highest similarity. Next, it statistically analyzes whether these field names frequently appear in combination in historical data. For example, if a combination appears more than 900 times in 10,000 data points, it is considered to have a high frequency. Simultaneously, the field names are compared with existing data source labels, and the comparison results must meet set fitting standards. For example, trend fields must accurately categorize into the "market trend" label. Field combinations that meet the criteria of high character similarity, high combination frequency, and high fitting are selected. For example, "transaction amount" and "popularity changes" meet all three conditions and are included in the subsequent analysis set. Finally, based on the performance of each field pair in character similarity, combination frequency, and label fitting, an overall evaluation value for the semantic consistency of the fields is summarized using an averaging method.

[0062] The numerical judgment submodule analyzes the changes in field values, frequency of occurrence, and degree of data fluctuation at different time points based on the transaction frequency value, keyword frequency value, and trend indicator value corresponding to the field pairs in the semantic consistency matching degree of the field. It compares the overlap between the actual change range of the field value and the logical benchmark range to obtain the logical conformity ratio of the field value.

[0063] The specific formula for calculating the range of changes in the field value at the time points of difference is as follows:

[0064] ;

[0065] in, Represents the field value in the field pair The change range index, Represents field pairs At the point of time field values, Represents field pairs At the point of time field values, Represents field pairs The frequency value of the associated keywords, Represents field pairs The transaction frequency value within the analysis period, Represents field pairs Trend indicator values ​​within the analysis period;

[0066] Detailed explanation of the formula and its calculation derivation: The formula involves the following parameters: , , , , The parameters are obtained and calculated as follows:

[0067] and Data was collected directly through a data monitoring system and set as... =100 and =150 indicates that a certain indicator is based on time. arrive It increased by 50 units.

[0068] The frequency of keyword occurrences was obtained by aggregating keyword occurrences through a data collection system, and set as... =200, based on data monitoring from the previous six months.

[0069] Obtained through transaction system records, assuming the data is... =75.

[0070] The slope is obtained by analyzing historical data using a specific algorithm and is set as... =10.

[0071] Substituting these parameters, the calculation process of the formula is as follows:

[0072] calculate ;

[0073] Calculate absolute value ;

[0074] Calculate the denominator ≈8.064;

[0075] Calculate the final formula result ≈1240.7;

[0076] This result indicates that the field is paired. Considering time nodes arrive During this period, considering the impact of keyword total frequency and trend changes, the field value change index was 1240.7. This value reflects the significant changes in the field pairs within the analysis period, providing a quantitative basis for subsequent data analysis and decision-making.

[0077] The behavior integration submodule, based on logically consistent data items in the field value logical consistency ratio, calls the behavior tag, time tag, and source tag to which the field belongs, and performs same-dimensional merging, cross-category integration, and unified interval format processing on transaction frequency value, interaction frequency value, and trend indicator value to obtain a standardized customer behavior data table.

[0078] The behavior integration submodule filters data items with high logical compliance rates, retains data pairs that meet threshold conditions, and adds category labels to these data items. For example, it labels them as financial platforms or social platforms based on the data source, as specific quarters or months based on the time record, and as consumer behavior or interactive behavior based on the behavior category. On this basis, data items under the same label are merged. For example, multiple transaction frequency fields from different sources but all belonging to the "consumer behavior" category can be merged by averaging. If the data from a certain source is more authoritative, a higher weight can be set to calculate the merged value. Further, the data is formatted uniformly. For example, all transaction frequency and trend values ​​are uniformly converted to percentages, or the value range is compressed to between zero and one, and converted by linear conversion to ensure that all kinds of behavior indicators have the same scale, which is convenient for subsequent horizontal comparison and comprehensive utilization. Finally, a standardized behavior data table with a clear structure and unified content is constructed.

[0079] Please see Figure 2 The customer profiling disambiguation module includes:

[0080] The field validation submodule obtains customer identifier, contact information, and transaction information field data from the standardized customer behavior data table. Based on the field structure, internal code consistency, and filling frequency, it calculates the format matching rate, internal code matching rate, and completeness value, judges the difference with the field consistency benchmark value, and generates the field consistency deviation rate.

[0081] First, customer record data from the past quarter is selected, such as transaction records within the last 90 days. Key fields such as customer identifier, contact information, and transaction information are extracted field by field. In the structure analysis stage, data processing tools are used to determine field types and check whether data values ​​conform to the basic format specifications set for the fields. For example, mobile phone numbers should be 11 digits starting with 1, and transaction times should follow the format "year-month-day". If abnormal characters or structural inconsistencies are found, they are marked as format mismatch data. In the internal encoding consistency stage, internal naming rules are used for verification. For example, customer identifiers should uniformly begin with "CUST" followed by a fixed number of digits. The system compares the field value structure of each record to see if it matches the standard encoding template. If fields that do not conform to the standard format are found, they are included in the deviation data set. Field completeness is assessed statistically. The number of non-empty records for each field in the data table is compared with the total number of records to obtain the fill ratio of each field in the overall data. If a field has a large number of missing records, it is considered to have low completeness. The format matching rate and internal code matching rate are also calculated by comparing the number of compliant records with the total number of records. Consistency benchmark values ​​are set for each field, such as format matching not less than 95%, completeness not less than 90%, and coding consistency rate not less than 98%. The current value of each field is compared with the corresponding benchmark value. If the current format matching value of a field is lower than the benchmark value standard, it can be determined that there is a deviation in field consistency. By calculating the difference of each deviation, the consistency deviation rate of that field can be obtained. After summing all field deviation rates, they are used for subsequent modules to perform data optimization and customer normalization operations.

[0082] The data optimization submodule calls the field consistency deviation rate, combines the field update time interval, the confidence level ranking value and the consistency offset value, and sorts and filters according to the corresponding confidence interval to generate the field confidence level ranking value after filtering.

[0083] The data optimization submodule uses field consistency deviation rate as a reference, and combines information such as field update time interval, consistency offset degree, and preliminary sorting weight value for comprehensive evaluation and analysis. During processing, it first counts the update time points of each field in historical data versions, calculating the update frequency by comparing the update dates of each update. For example, if a contact information field is modified multiple times within two months, the update frequency of that field is high, thus affecting the field's credibility assessment. Regarding offset values, it checks the degree of difference in field values ​​between different periods. For example, if there are many inconsistent records between the current version's customer name field and the previous version, the offset value is high, affecting the overall credibility judgment. Weight parameters are set for each field; for example, the field consistency comparison result is considered a core indicator and given a large weight. Then, combined with update time frequency and value offset degree, each field is scored according to the weight factors, generating a credibility level ranking result. During the ranking process, fields with excellent consistency, reasonable update frequency, and small offset values ​​are given higher priority so that subsequent modules can filter and call the fields. Finally, the credibility level ranking sequence of each field is output for field merging and unified field optimization processing.

[0084] The Customer Unification Submodule sorts the field values ​​by the confidence level after filtering, performs priority matching and conflict filtering on the field values ​​of the same customer under different data sources, merges the field values, and generates a unified customer profile table.

[0085] For a customer's records across multiple business systems, the system first compares the reliability ranking scores of these field values ​​and prioritizes retaining data with higher scores. For example, if a customer's phone number has a high score in the main system but a low score in another system, the data from the main system is selected. If significant conflicts are found between multiple field values ​​from different sources, such as inconsistencies in the address field, the system will retain the field values ​​with higher reliability based on the scoring results and delete or weaken the records with lower scores. Simultaneously, field values ​​present in the high-reliability system but missing in other systems (such as email addresses) can be directly added to the unified data structure from the high-reliability system. After field merging, the system performs standardization processing on the merged field values, such as standardizing field format specifications and removing redundant punctuation. Finally, the integrated field values ​​are written into a unified customer profile table, ensuring that customer records have a unified field structure and optimized content.

[0086] Please see Figure 2 The strategy matching module includes:

[0087] The transaction feature recognition submodule obtains the transaction amount, product relevance, and interaction frequency from the unified customer profile table. It combines these three items to perform data fusion processing at the customer dimension, and forms a unified representation value by setting the order to obtain the customer behavior intensity value.

[0088] The transaction feature identification submodule extracts three indicators from unified customer profile information: transaction amount, product relevance, and interaction frequency. The transaction amount field aggregates all customer transaction records within a specified period, for example, accumulating all consumption records over three consecutive months to form the total transaction amount at the customer level. Product relevance is determined by establishing a matching model between product categories and customer usage behavior. Based on the product types involved in the customer's historical transactions, each customer is matched and coded against a product tag library, and their preference for different products is assessed through similarity calculations. Interaction frequency is calculated based on customer behaviors on the platform, such as inquiries, clicks, service usage, and proactive contact. The average number of interactions over three months is calculated, and the distribution of behavioral channels and times is recorded. Subsequently, [the system]... Different weighting coefficients are set for the above three indicators. The total transaction amount of each customer is standardized based on its position in the full sample. For example, the highest amount is set as the upper limit and the lowest amount as the lower limit. The values ​​are converted to a unified scoring range through linear mapping. The interaction frequency is set with the maximum value as the standard. The number of interactions of each customer within this range is converted into a score. The product relevance is evaluated numerically based on the similarity calculation results. After standardization, all scoring items are weighted and integrated with the set weighting coefficients to form a unified behavior intensity score. The range of behavior intensity score values ​​is divided into intervals based on the integrated scores. The division results are three levels: high intensity, medium intensity, and low intensity. The behavior intensity values ​​constitute a representation sequence on the customer dimension and are stored in the profile information as a basis for subsequent matching and screening.

[0089] The value fit calculation submodule calls the customer behavior intensity value, combines it with the target product list in the marketing strategy library, and performs item-level screening based on the product correlation information in the customer profile to identify the density of corresponding relationships. It then performs mapping and transformation based on the customer behavior intensity value to obtain the strategy fit intensity value.

[0090] The specific formula for calculating the density of the identification correspondence is as follows:

[0091] ;

[0092] Where D represents the degree of relationship between the customer and the target product in the item-level screening. This represents the customer's relevance score to the k-th product feature. This represents the weighted score assigned to the target product on the k-th product feature. This represents the customer's average relevance score across all product features. This represents the average set score for the target product across all product features, where n represents the total number of product features. This represents the weighted total match value between the customer and the target product across all feature dimensions. This represents the product of the average customer preference value and the average target value of the product. Represents the square root of the product;

[0093] Detailed explanation of the formula and its calculation derivation:

[0094] Parameter definition and acquisition method:

[0095] This is derived through customer behavior data analysis, such as monitoring the frequency of customer interactions and purchase records related to a certain product feature, and quantifying them using a standardized scoring mechanism with a scoring range of 0 to 10 points.

[0096] Determined by the product strategy team based on market research and product positioning, this feature reflects its importance to the overall attractiveness of the product, with a score range of 0 to 10.

[0097] n is the number of specific dimensions categorized according to product characteristics, such as functionality, appearance design, brand influence, etc., totaling 5 items.

[0098] The calculation formula is:

[0099] ;

[0100] The average weighted score of the target product across all product features. The calculation formula is:

[0101] ;

[0102] Parameter values ​​and their basis:

[0103] The following scores were obtained through monitoring and analysis of customers' past behavioral data:

[0104] Feature 1: 8 points (high customer interaction frequency on this feature);

[0105] Feature 2: 6 points (the customer interacts with this feature at a moderate frequency);

[0106] The third characteristic: 7 points (customers interact with this characteristic frequently);

[0107] Feature 4: 5 points (Customers generally interact with this feature on a moderate frequency).

[0108] Item 5: 9 points (Customers interact with this feature very frequently);

[0109] Based on market research and product positioning, the product strategy team sets the following weights:

[0110] Feature 1: 7 points (This feature is important for product attractiveness);

[0111] Feature 2: 8 points (This feature is important for product attractiveness);

[0112] Feature 3: 6 points (This feature has moderate appeal to the product);

[0113] Feature 4: 7 points (This feature is important for product attractiveness);

[0114] Feature 5: 9 points (This feature is very important for product attractiveness);

[0115] n: Total number of product features, 5 in total.

[0116] Parameter calculation process:

[0117] calculate :

[0118] ;

[0119] calculate :

[0120] ;

[0121] calculate :

[0122] ;

[0123] calculate :

[0124] ;

[0125] Substitute into the formula to calculate D:

[0126] ;

[0127] Results analysis:

[0128] The calculated D≈25.73 indicates a high degree of match between the customer and the target product across various feature dimensions. A higher value indicates a stronger customer preference for the target product and a greater strategy fit. This result allows for the development of more targeted marketing strategies to improve customer conversion rates and satisfaction.

[0129] The strategy matching list generation submodule locates the top-ranked customer identification information based on the strategy fit strength value, summarizes the target product and customer profile number content, and combines it with the strategy fit strength value to generate a strategy matching list.

[0130] The strategy matching list generation submodule uses the strategy fit strength value obtained in the previous step as the sorting basis, arranging them in descending order of value. Targets ranked higher are prioritized for matching. During the filtering process, a lower threshold for the fit strength value is set, retaining only customer identification information above this threshold. In the matching list generation process, the profile number of the selected customer is used as the primary key field. Each product in the corresponding strategy target product list is aggregated one by one, and the customer identification information and product list are merged at the field level to generate a structured combined data table. If a customer identification number is C001, and it matches target products P1, P3, and P5 in the matching list with a fit score of 0.72, the combined result structure is: [C001, P1, P3, P5, 0.72]. This process is completed in batches at the system level using data scripts. Field matching, data merging, and sorting operations are completed through standard data processing logic. The final aggregated list can be used as a filtering list before strategy execution, containing customer identification, strategy product combinations, and strategy fit strength values. The data table can be updated over time to maintain the dynamic correlation between the strategy and the customer profile.

[0131] Please see Figure 2 The real-time transmission module includes:

[0132] The urgency filtering submodule obtains the urgency level of the strategies in the strategy matching list and the corresponding unique customer identification code, compares the urgency level with the strategy urgency threshold, filters the strategies and customer identification codes with urgency levels higher than the threshold, and generates a high-urgency strategy tag set.

[0133] First, each strategy item needs to be read from the system database, and its urgency level and unique customer identification code are extracted. For example, after reading a strategy, its urgency level is identified as 82, and the corresponding customer identification code is CID-20240401. Then, based on the strategy's category (e.g., real-time trading, data backhaul, video transmission, etc.), the urgency threshold set in the system for that type of strategy is found. For example, the urgency threshold for video transmission strategies is set to 70, and for real-time trading strategies, it is set to 90. By comparing the urgency level of each strategy with the urgency threshold of its category, strategies with urgency levels greater than the threshold are identified, and these strategies and their customer identification codes are recorded. During the comparison process, by traversing the entire strategy list and performing logical judgments on each record, strategy items that meet the conditions and their customer identification codes are extracted to form a data tag set. In actual use, such as when processing financial trading strategies in a bank's data center, if the urgency level of a strategy is set to 95, which is higher than the threshold of 90 for its trading strategy category, its corresponding identification code CID will be tagged along with the strategy ID and included in the high-urgency strategy set. When processing multiple strategies, they can be read and processed in batches to form a structured table record, such as {CID, strategy ID, urgency}, for use in the subsequent path evaluation stage.

[0134] The path evaluation submodule calls the unique customer identification code in the high urgency policy tag set to obtain the delay level, packet loss status and node load status of the corresponding transmission path. It compares these with the delay threshold, packet loss rate standard value and node load benchmark value, records the paths that meet the standards, and generates a set of available path identification codes.

[0135] After receiving the high-urgency policy tag set, the path evaluation submodule needs to call the transmission monitoring system or network status interface for each customer's unique identifier within the set to obtain the current latency, packet loss level, and node load status of all available transmission paths. The system sets the latency upper limit to 80 milliseconds, the packet loss ratio upper limit to 1%, and the node load baseline to 85%. If the obtained latency value is less than or equal to the upper limit, and the packet loss ratio does not exceed the standard value and the node load is within the baseline value, then the path can be considered to meet the usage standards. Taking customer CID-20240401 as an example, it corresponds to two paths. Path A has a latency of 60 milliseconds, a packet loss rate of 0.5%, and a node load of 68%, all within the standard range; Path B has a latency of 92 milliseconds, a packet loss rate of 1.2%, and a node load of 84%, where both the latency and packet loss values ​​exceed the range. Therefore, Path A is recorded as a usable path. Throughout the evaluation process, the system compares each customer's path with its performance parameters and standard limits. Paths that meet the criteria are organized into a set of structured data, which together with the path identifier, customer identification code, and corresponding indicators form a set of available path records, which are then used in the queue scheduling process.

[0136] The queue generation submodule sorts the policies and customer identification codes in the available path identification code set according to the policy urgency, arranges the transmission order, binds the path identifier, policy content and customer identification code, and generates a policy transmission queue.

[0137] The queue generation submodule receives the set of available path identification codes and prioritizes all policies based on the urgency values ​​recorded in the policies. The sorting operation arranges the policies from highest to lowest urgency value, and then binds the sorted results one-to-one with the path identifiers and customer identification codes obtained in the previous stage to construct a policy queue for transmission. Taking three policies as an example, Policy 1 corresponds to customer CID-10001, with an urgency of 90 and path identifier P1; Policy 2 corresponds to customer CID-10002, with an urgency of 85 and path identifier P2; and Policy 3 corresponds to customer CID-10003, with an urgency of 70 and path identifier P3. The sorted order is Policy 1, Policy 2, and Policy 3. Each queue item consists of a customer identification code, a path identifier, and policy content. After binding, they form the transmission queue structure, such as the first item being {CID-10001, P1, Policy 1 content}. This queue storage structure can be in key-value table form or converted into a read-write data table for downstream scheduling module calls. The queue generation process is based on two inputs: path availability judgment and policy urgency ranking. The policy scheduling order is determined through a dual binding mechanism.

[0138] Please see Figure 2 The operation strategy generation module includes:

[0139] The click behavior extraction submodule obtains customer click records from the strategy transmission queue, identifies the distribution characteristics of click time, page position, and click frequency under product category, and generates customer click offset rate by combining the changes in the number of clicks and page dwell time within the same time period.

[0140] When the click behavior extraction submodule retrieves customer click records from the strategy transmission queue, it needs to extract and process each record item by item. Each click data entry includes a customer identifier, the click time, page identifier, click area location code, the visited page address, the page loading completion time, and the total page dwell time. After extraction, all records are sorted chronologically to form a continuous click sequence. Next, based on the product tags attached to each click data entry, such as fund, insurance, and wealth management categories, each type of click is categorized and aggregated, and the number of page clicks under each category is counted. To further refine the page click characteristics, the page can be divided into top, content area, and bottom, and the number of clicks and percentages of each area can be counted. For example, the top clicks of a certain type of page may account for 40% of all clicks. Based on this, the total number of clicks within each time period is counted according to hourly or half-hourly periods, and the increase or decrease in the number of clicks within each period is recorded. Next, the page dwell time is statistically analyzed for different product categories. The concentration and volatility can be assessed using the mean and median. For example, the average dwell time for wealth management pages is 38 seconds, and the median is 35 seconds. After integrating the above information, an indicator is constructed based on the percentage change in click volume and the fluctuation range of page dwell time to measure the stability and shift in customer clicks. This indicator comprehensively considers the fluctuation range of click volume and the fluctuation range of dwell time. For example, if a customer's click volume increases by 20% between 9 and 10 o'clock, and the fluctuation range of page dwell time is relatively high, then their shift indicator will increase accordingly, reflecting the degree of change in the customer's click behavior within the current product category.

[0141] The conversion rate matching calculation submodule calls the customer click offset rate, identifies the conversion time performance and status type in the customer conversion record, extracts the temporal relationship between click behavior and conversion performance, performs matching analysis based on the difference between the temporal relationship and the customer retention time range, and obtains the conversion linkage change amount by referring to the fluctuation of the number of conversion statuses.

[0142] The conversion rate matching calculation submodule first extracts information from existing customer conversion records, mainly including the time of conversion and the conversion status. Next, this conversion information is correlated with the customer's click behavior sequence, forming a click-conversion chain for each customer. Within this chain, the time difference between the last click and the first conversion is identified. For example, if a customer's last click was at 10:20 AM and the conversion was at 10:35 AM, the interval is 15 minutes. This time difference is recorded and combined with the corresponding customer's retention time in the system to calculate the relative lag between click and conversion. For example, if the retention period is 30 days and the conversion interval is 1 day, the difference is small, indicating relatively concentrated conversion behavior. Subsequently, all customers are categorized by conversion status type, such as completed, processing, and canceled. The aforementioned time difference data for each category is statistically analyzed to identify the differences between the categories. Simultaneously, it's necessary to analyze the quantity changes of each conversion status over a certain time period. For example, if there were 200 "Completed" statuses last week and 250 this week, the quantity fluctuation value is 50. By analyzing the fluctuations over multiple periods, an assessment metric representing the magnitude of status change can be constructed. After combining these three types of information, the performance of customer conversion linkage changes is evaluated. For instance, if a customer's click behavior shows a high degree of deviation, while conversions occur rapidly and the number of statuses is actively increasing, their overall change value will be in a higher range.

[0143] The strategy priority submodule calls the conversion linkage change volume, refers to the investment return rate performance of financial products in the current period, and defines the product set with the return rate in the upper range according to the ranking of the return rate in the product set. It extracts the conversion linkage change volume of related customers in the set, identifies the value performance of customers under the differentiated strategy configuration structure, arranges the strategy implementation order according to the order of value performance, and generates a financial marketing strategy optimization plan.

[0144] The strategy prioritization submodule needs to prioritize products based on the aforementioned conversion-linkage change values ​​and the performance of different financial products within the current period. During execution, first, a set of products is established, and a corresponding yield is assigned to each product, for example, by setting an annualized yield data for a reference period. Then, all products are sorted from highest to lowest yield, and a group of products with higher returns is defined, for example, the top 20% of all products are selected as the upper-level range. Next, the customer sets of these products are analyzed, and the conversion-linkage change value for each customer is extracted. Then, the changes for all customers are summed or averaged as a reference value for the customer response strength of that product. For example, if a financial product has ten customers and their average change value is 1.6, then the linkage weight value for that product is 1.6. All products in the upper-level range are sorted according to this value; the higher the value, the stronger the customer behavior linkage and the higher the priority. Based on this, the fluctuation range of customer linkage values ​​under each high-priority product is extracted. For example, the standard deviation and skewness of these values ​​are statistically analyzed to determine the level of difference under different strategy configuration conditions. Then, the entire strategy deployment order is rearranged according to the priority list, and finally, an optimized strategy order structure suitable for the current product portfolio is generated.

[0145] Please see Figure 3 The operational management method based on big data in financial marketing includes the following steps:

[0146] S1: Obtain the field content from transaction, interaction and market data in financial marketing scenarios, make consistency judgments based on the semantic matching of field names and the logical distribution of field values, and generate a standardized customer behavior data table;

[0147] S2: Based on a standardized customer behavior data table, identify the same customer entry from the sources of difference, determine field conflict information according to data time, source level and field validation status, and generate a unified customer profile table;

[0148] S3: Call the key fields in the unified customer profile table, combine them with the product tags in the strategy library, determine the matching relationship between customer characteristics and strategy content, and generate a strategy matching list;

[0149] S4: Based on the urgency level in the policy matching list, filter the content to be pushed, and combine path delay, node load and packet loss to perform channel priority sorting and generate a policy transmission queue.

[0150] S5: Call the push records in the strategy transmission queue, collect click, conversion and feedback data, combine investment and retention information to judge changes in customer behavior, and generate financial marketing strategy optimization plan.

[0151] The above are merely preferred embodiments of the present invention and are not intended to limit the present invention in any other way. Any person skilled in the art may make changes or modifications to the above-disclosed technical content to create equivalent embodiments that can be applied to other fields. However, any simple modifications, equivalent changes, and modifications made to the above embodiments based on the technical essence of the present invention without departing from the scope of the present invention shall still fall within the protection scope of the present invention.

Claims

1. An operation and management system based on big data in financial marketing, characterized by: The system includes: The data standardization module acquires user transaction records, social media interaction records, and market trend data in financial marketing scenarios, analyzes the semantic matching degree of field names, verifies whether the field distribution conforms to business rules, and generates a standardized customer behavior data table. The customer profiling disambiguation module, based on the standardized customer behavior data table, identifies conflicting information of the same customer from different data sources, compares data update time, data source credibility level and field verification results, and generates a unified customer profile table. The strategy matching module calls the transaction amount, product relevance, and interaction frequency in the unified customer profile table, combines them with the target product, calculates the customer value and strategy fit, and obtains the strategy matching list; The real-time transmission module filters the unique identification code of the customer to be transmitted and the policy content based on the policy urgency in the policy matching list, and performs channel priority sorting according to the transmission path delay level, packet loss and node load status to generate a policy transmission queue. The operation strategy generation module calls the customer click records, conversion rates and market feedback information in the strategy transmission queue, and combines them with the return on investment and customer retention period to generate a financial marketing strategy optimization plan; The data standardization module includes: The semantic recognition submodule acquires user transaction record fields, social media interaction fields, and market trend fields, calculates the degree of character similarity between field names, the number of times field names appear in combination in the total data, and the degree of comparison with data source labels, selects field pairs whose character similarity and number of times they appear in combination both exceed the field fit benchmark value and the combination frequency benchmark value, and generates field semantic consistency matching degree. The numerical judgment submodule analyzes the change range, frequency of occurrence and data fluctuation of field values ​​at different time points based on the transaction frequency value, keyword frequency value and trend indicator value corresponding to the field pairs in the semantic consistency matching degree of the field. It compares the overlap between the actual change range of the field value and the logical benchmark range to obtain the logical conformity ratio of the field value. The behavior integration submodule, based on the logically consistent data items in the field value logical consistency ratio, calls the behavior tag, time tag, and source tag to which the field belongs, and performs same-dimensional merging, cross-category integration, and unified interval format processing on the transaction frequency value, interaction frequency value, and trend indicator value to obtain a standardized customer behavior data table.

2. The operation and management system based on financial marketing big data according to claim 1, characterized in that: The standardized customer behavior data table includes field semantic mapping results, field value distribution rationality verification results, and unified field format data. The unified customer profile table includes conflict information disambiguation results, data credibility level assessment results, and logical consistency verification results. The strategy matching list includes customer value assessment results, strategy fit score, and target strategy solution items. The strategy transmission queue includes a unique customer identifier, strategy content summary, and transmission path scheduling parameters. The financial marketing strategy optimization solution includes customer click behavior data, conversion effect feedback, and strategy benefit assessment results.

3. The operation and management system based on financial marketing big data according to claim 1, characterized in that: The specific formula for calculating the range of change of the analysis field value at the time point of difference is as follows: ； in, Represents the field value in the field pair The change range index, Represents field pairs At the point of time field values, Represents field pairs At the point of time field values, Represents field pairs The frequency value of the associated keywords, Represents field pairs The transaction frequency value within the analysis period, Represents field pairs Trend indicator values ​​within the analysis period.

4. The operation management system based on financial marketing big data according to claim 3, characterized in that: The customer profiling disambiguation module includes: The field validation submodule obtains customer identifier, contact information, and transaction information field data from the standardized customer behavior data table. Based on the field structure, internal code consistency, and filling frequency, it calculates the format matching rate, internal code matching rate, and completeness value, judges the difference with the field consistency benchmark value, and generates the field consistency deviation rate. The data optimization submodule calls the field consistency deviation rate, combines the field update time interval, the confidence level ranking value and the consistency offset value, and sorts and filters according to the corresponding confidence interval to generate the field confidence level ranking value after filtering. The Customer Unification Submodule performs priority matching and conflict filtering on field values ​​of the same customer under different data sources based on the sorted confidence level values ​​of the filtered fields, merges field values, and generates a unified customer profile table.

5. The operation management system based on financial marketing big data according to claim 4, characterized in that: The strategy matching module includes: The transaction feature recognition submodule obtains the transaction amount, product relevance, and interaction frequency from the unified customer profile table, and performs data fusion processing on the customer dimension by combining the three items. By setting the order, a unified representation value is formed to obtain the customer behavior intensity value. The value fit calculation submodule calls the customer behavior intensity value, combines it with the target product list in the marketing strategy library, and performs item-level screening based on the product correlation information in the customer profile, identifies the density of corresponding relationships, and performs mapping transformation based on the customer behavior intensity value to obtain the strategy fit intensity value. The strategy matching list generation submodule locates the top-ranked customer identification information based on the strategy fit strength value, summarizes the target product and customer profile number content, and combines them with the strategy fit strength value to generate a strategy matching degree list.

6. The operation management system based on financial marketing big data according to claim 5, characterized in that: The specific formula for calculating the density of the identified correspondences is as follows: ； Where D represents the degree of relationship between the customer and the target product in the item-level screening. This represents the customer's relevance score to the k-th product feature. This represents the weighted score assigned to the target product on the k-th product feature. This represents the customer's average relevance score across all product features. This represents the average set score for the target product across all product features, where n represents the total number of product features. This represents the weighted total match value between the customer and the target product across all feature dimensions. This represents the product of the average customer preference value and the average target value of the product. It represents the square root of the product.

7. The operation management system based on financial marketing big data according to claim 5, characterized in that: The real-time transmission module includes: The urgency filtering submodule obtains the urgency level of the strategies in the strategy matching list and the corresponding unique customer identification code, compares the urgency level with the strategy urgency threshold, filters strategies and customer identification codes with urgency levels higher than the threshold, and generates a high-urgency strategy tag set. The path evaluation submodule calls the unique customer identification code in the high urgency policy tag set to obtain the delay level, packet loss status and node load status of the corresponding transmission path. It compares these with the delay threshold, packet loss rate standard value and node load benchmark value, records the paths that meet the standards, and generates a set of available path identification codes. The queue generation submodule sorts the policies and customer identification codes in the available path identification code set according to the policy urgency, arranges the transmission order, binds the path identifier, policy content and customer identification code, and generates a policy transmission queue.

8. The operation management system based on financial marketing big data according to claim 7, characterized in that: The operation strategy generation module includes: The click behavior extraction submodule obtains customer click records in the strategy transmission queue, identifies the distribution characteristics of click time, page position and click frequency under product category, and generates customer click offset rate by combining the change in the number of clicks and page dwell time in the same time period. The conversion rate matching calculation submodule calls the customer click offset rate, identifies the conversion time and status type in the conversion record, extracts the temporal relationship between click behavior and conversion, and performs matching analysis with the difference in the range of customer retention time. Combined with the fluctuation of the number of conversion statuses, the conversion linkage change amount is obtained. The strategy priority ranking submodule calls the conversion linkage change amount, refers to the investment return rate of financial products in the current period, selects the product set in the high range according to the ranking of the return rate in the product set, extracts the conversion linkage change amount of related customers, identifies the value performance of customers under different strategy configurations, and arranges the strategy execution order according to the value performance to generate a financial marketing strategy optimization plan.

9. An operational management method based on big data in financial marketing, characterized in that, The operation and management system based on financial marketing big data according to any one of claims 1-8 includes the following steps: S1: Obtain the field content from transaction, interaction and market data in financial marketing scenarios, make consistency judgments based on the semantic matching of field names and the logical distribution of field values, and generate a standardized customer behavior data table; S2: Based on the standardized customer behavior data table, identify the same customer entry from the sources of difference, determine field conflict information according to data time, source level and field verification status, and generate a unified customer profile table; S3: Call the key fields in the unified customer profile table, combine them with the product tags in the strategy library, determine the matching relationship between customer characteristics and strategy content, and generate a strategy matching list; S4: Based on the urgency level in the policy matching list, filter the content to be pushed, and perform channel priority sorting in combination with path delay, node load and packet loss to generate a policy transmission queue; S5: Call the push records in the strategy transmission queue, collect click, conversion and feedback data, combine input and retention information to judge changes in customer behavior, and generate a financial marketing strategy optimization plan.

Images