Fund transaction service quality evaluation method and device, storage medium and computer program product

Abstract

The invention discloses a fund transaction service quality evaluation method and device, a storage medium and a computer program product, and relates to the technical field of financial information processing, and the method comprises the steps: collecting full-link log burying point data; inputting the full-link log burying point data into a pre-constructed systematized scoring model to carry out service quality weighted scoring so as to obtain a multi-dimensional service quality score value; and processing the multi-dimensional service quality score value to obtain multi-dimensional visual evaluation data, and quantifying the fund transaction service quality according to the multi-dimensional visual evaluation data. Real-time and accurate quantitative evaluation of fund transaction service quality is realized through full-link log burying point data acquisition, service quality weighted scoring by a systematic scoring model, and multi-dimensional visual evaluation data processing and display.

Description

Technical Field

[0001] This application relates to the field of financial information processing technology, and in particular to methods, equipment, storage media and computer program products for evaluating the quality of fund trading services. Background Technology

[0002] In fund trading scenarios, with the rapid expansion of business, the integration of fund trading systems with numerous fund managers has become increasingly complex. Currently, fund trading systems primarily rely on log monitoring platforms, system alarm notifications, and feedback from clients or business personnel to detect system or business anomalies. This passive response mechanism exhibits significant lag; furthermore, during the operation of fund trading systems, alarm logs generated by issues such as network services and the quality of fund manager clearing documents are numerous and scattered, operational data is complexly categorized, and there is a lack of mechanisms for analyzing and tracing historical issues. Consequently, it is impossible to monitor and quantitatively evaluate the quality of fund trading services provided by fund managers in real time and with high accuracy.

[0003] Therefore, how to conduct real-time and accurate quantitative evaluation of the quality of fund trading services has become a technical problem that this application urgently needs to solve.

[0004] The above content is only used to help understand the technical solution of this application and does not represent an admission that the above content is prior art. Summary of the Invention

[0005] The main purpose of this application is to provide a method, device, storage medium, and computer program product for evaluating the quality of fund trading services, aiming to solve the technical problem of how to conduct real-time and accurate quantitative evaluation of the quality of fund trading services.

[0006] To achieve the above objectives, this application proposes a method for evaluating the quality of fund trading services, the method comprising:

[0007] Collect end-to-end log data;

[0008] The end-to-end log data is input into a pre-built systematic scoring model to perform a weighted service quality score, resulting in a multi-dimensional service quality score.

[0009] The multidimensional service quality score is processed to obtain multidimensional visual evaluation data, and the quality of fund trading services is quantified based on the multidimensional visual evaluation data.

[0010] In one embodiment, the step of collecting end-to-end log data includes:

[0011] Obtain fund trading business scenario information, and configure basic data tracking rules and scenario extension rules based on the fund trading business scenario information;

[0012] generate a dynamic burying point rule set in combination with the basic burying point rule and the scene extension rule;

[0013] deploy a lightweight collection agent into a fund transaction system, and embed lightweight preprocessing logic and the dynamic burying point rule set in the lightweight collection agent to obtain a multi-source log parallel collection framework;

[0014] collect full-link log burying point data in parallel based on the multi-source log parallel collection framework.

[0015] In an embodiment, the step of inputting the full-link log burying point data into a pre-constructed systematic scoring model for service quality weighted scoring to obtain a multi-dimensional service quality score value includes:

[0016] obtain fund transaction scene core requirements, and define scoring dimensions according to the fund transaction scene core requirements;

[0017] call fund transaction historical fault data, assign scoring weights to the scoring dimensions according to the fund transaction historical fault data, and determine respective data sources corresponding to the scoring dimensions;

[0018] define scoring rules according to the scoring dimensions, the scoring weights, and the data sources, and construct an interface standardization scoring model, a data standardization scoring model, a network stability scoring model, and a clearing time efficiency scoring model based on the scoring rules respectively;

[0019] construct a systematic scoring model based on the interface standardization scoring model, the data standardization scoring model, the network stability scoring model, and the clearing time efficiency scoring model

[0020] In an embodiment, the step of inputting the full-link log burying point data into a pre-constructed systematic scoring model for service quality weighted scoring to obtain a multi-dimensional service quality score value includes:

[0021] remove invalid fields and same-type abnormal data in the full-link log burying point data, and unify timestamps of the full-link log burying point data to obtain standardized full-link log burying point data, the standardized full-link log burying point data including interface standardization configuration data, inter-day clearing abnormal data, interaction message logs, connectivity logs, data clearing control data, and fund transfer control data;

[0022] input the interface standardization configuration data into the interface standardization scoring model for service quality weighted scoring to obtain an interface standardization score value;

[0023] input the inter-day clearing abnormal data into the data standardization scoring model for service quality weighted scoring to obtain a data standardization score value;

[0024] inputting the interaction message log and the connectivity log into the network stability scoring model for quality of service weighted scoring to obtain a network stability scoring value;

[0025] inputting the data clearing control data and the fund transfer control data into the clearing time limit scoring model for service weighted scoring to obtain a clearing time limit scoring value;

[0026] adding the interface standardization scoring value, the data standardization scoring value, the network stability scoring value, and the clearing time limit scoring value to obtain a multi-dimensional quality of service scoring value.

[0027] In an embodiment, the processing of the multi-dimensional quality of service scoring value to obtain multi-dimensional visual evaluation data, and the quantification of the quality of fund transaction service according to the multi-dimensional visual evaluation data comprises:

[0028] matching metadata tags to the multi-dimensional quality of service scoring value, and structurally storing the multi-dimensional quality of service scoring value after the matching of metadata tags to a data warehouse to obtain a structured scoring data set;

[0029] performing aggregation calculation on the structured scoring data set according to a time dimension to obtain service quality change trend data;

[0030] calling a visual engine to perform visual processing on the service quality change trend data and the structured scoring data set to obtain multi-dimensional visual evaluation data;

[0031] generating a quality of service label according to the multi-dimensional visual evaluation data, and quantifying the quality of fund transaction service according to the quality of service label.

[0032] In an embodiment, after the step of calling the visual engine to perform visual processing on the service quality change trend data and the structured scoring data set to obtain multi-dimensional visual evaluation data, the method further comprises:

[0033] extracting a historical scoring time sequence from the structured scoring data set;

[0034] adopting a long short-term memory network to predict the historical scoring time sequence to obtain a quality of service trend prediction value;

[0035] superimposing a comparison curve of the quality of service trend prediction value in the multi-dimensional visual evaluation data.

[0036] In an embodiment, after the step of processing the multi-dimensional quality of service scoring value to obtain multi-dimensional visual evaluation data, and quantifying the quality of fund transaction service according to the multi-dimensional visual evaluation data, the method further comprises:

[0037] extracting key variables from the multi-dimensional visualization evaluation data, and constructing a causal diagram according to the key variables;

[0038] quantifying causal strength in the causal diagram by using double difference method, sorting the causal strength, and determining a scoring governance item;

[0039] performing multi-agent reinforcement learning training according to the scoring governance item, and generating a fund transaction service quality governance scheme.

[0040] In addition, to achieve the above-mentioned purpose, the present application also proposes a fund transaction service quality evaluation device, which comprises a memory, a processor and a computer program stored in the memory and executable on the processor, and the computer program is configured to implement the steps of the fund transaction service quality evaluation method as described above.

[0041] In addition, to achieve the above-mentioned purpose, the present application also proposes a storage medium, which is a computer readable storage medium, and the storage medium stores a computer program, and the computer program is executed by a processor to implement the steps of the fund transaction service quality evaluation method as described above.

[0042] In addition, to achieve the above-mentioned purpose, the present application also provides a computer program product, which comprises a computer program, and the computer program is executed by a processor to implement the steps of the fund transaction service quality evaluation method as described above.

[0043] The one or more technical solutions proposed in the present application have at least the following technical effects:

[0044] The full-link log point data is collected, the full-link log point data is input into a pre-constructed systematic scoring model to perform service quality weighted scoring, and multi-dimensional service quality score values are obtained. The multi-dimensional service quality score values are processed to obtain multi-dimensional visual evaluation data, and the fund transaction service quality is quantified according to the multi-dimensional visual evaluation data. First, by collecting the full-link log point data, the real-time collection and integration of the full-link data of the fund transaction system are realized, the integrity and consistency of the data are ensured, and high-quality original data is provided for subsequent quantitative evaluation. Further, the pre-constructed systematic scoring model performs multi-dimensional service quality weighted scoring based on the full-link log point data, and the multi-dimensional service quality score values obtained quantify the fund transaction service quality, realizing multi-dimensional and fine evaluation of the fund transaction service quality. Further, the multi-dimensional service quality score values are processed to obtain multi-dimensional visual evaluation data displayed in a visual manner, the fund transaction service quality is intuitively and comprehensively monitored, and the fund transaction service quality is quantified. In summary, by collecting the full-link log point data, the systematic scoring model performs service quality weighted scoring, and the multi-dimensional visual evaluation data is processed and displayed, realizing real-time and accurate quantitative evaluation of the fund transaction service quality. BRIEF DESCRIPTION OF DRAWINGS

[0045] The accompanying drawings, which are incorporated into and form part of the specification, illustrate embodiments consistent with the present application and, together with the description, serve to explain the principles of the application.

[0046] In order to more clearly illustrate the technical solutions in the embodiments of the present application or the prior art, the accompanying drawings required to be used in the embodiments or prior art description will be briefly introduced. Obviously, for those skilled in the art, other drawings can also be obtained based on these drawings without creative labor.

[0047] Figure 1 A flowchart is provided for the first embodiment of the fund transaction service quality evaluation method of the present application;

[0048] Figure 2 A flowchart is provided for the second embodiment of the fund transaction service quality evaluation method of the present application;

[0049] Figure 3 A flowchart is provided for the fourth embodiment of the fund transaction service quality evaluation method of the present application;

[0050] Figure 4 A flowchart is provided for the fifth embodiment of the fund transaction service quality evaluation method of the present application;

[0051] Figure 5A feasible implementation process schematic diagram is provided for the fifth embodiment of the fund transaction service quality evaluation method of the application.

[0052] Figure 6 A brief process schematic diagram of the fund transaction service quality evaluation method provided by the application is provided.

[0053] Figure 7 A module structure schematic diagram of the fund transaction service quality evaluation device of the embodiment of the application is provided.

[0054] Figure 8 A device structure schematic diagram of the hardware running environment involved in the fund transaction service quality evaluation method in the embodiment of the application is provided.

[0055] The purpose implementation, functional features and advantages of the application will be further described with reference to the accompanying drawings in conjunction with the embodiments. DETAILED DESCRIPTION

[0056] It should be understood that the specific embodiments described herein are only used to explain the technical solutions of the application and are not used to limit the application.

[0057] In order to better understand the technical solutions of the application, the following will be described in detail in conjunction with the drawings of the specification and specific embodiments.

[0058] The main solution of the embodiment of the application is: collecting full-link log point data; inputting the full-link log point data into a pre-constructed systematic scoring model to perform service quality weighted scoring, obtaining multi-dimensional service quality score values; processing the multi-dimensional service quality score values to obtain multi-dimensional visual evaluation data, and quantifying the fund transaction service quality according to the multi-dimensional visual evaluation data.

[0059] The embodiment of the application considers that: in the fund transaction scenario, as the business rapidly expands, the docking of the fund transaction system with numerous management institutions becomes increasingly complex. At present, the fund transaction system mainly relies on log monitoring platforms, system alarm notifications and customer or business personnel feedback to discover system or business abnormalities. Such a passive response mechanism has obvious hysteresis; and due to the complex and scattered alarm log information generated by network services, management clearing file data quality and other problems in the running process of the fund transaction system, the operation data classification is complex, and there is a lack of analysis and tracing mechanism for historical problems, so that the quality of the management fund transaction service cannot be monitored and quantitatively evaluated in real time and accurately.

[0060] The application provides a solution. First, by collecting full-link log data, real-time collection and integration of full-link data of a fund transaction system are realized, the integrity and consistency of the data are ensured, and high-quality original data is provided for subsequent quantitative evaluation. Further, a pre-constructed systematic scoring model performs multi-dimensional weighted scoring of service quality based on full-link log data, and a multi-dimensional service quality score is obtained to quantify the fund transaction service quality and realize multi-dimensional and fine evaluation of the fund transaction service quality. Further, the multi-dimensional service quality score is processed to obtain multi-dimensional visual evaluation data displayed in a visual manner, the fund transaction service quality is intuitively and comprehensively monitored, and the fund transaction service quality is quantified. In summary, by collecting full-link log data, performing weighted scoring of service quality by a systematic scoring model, and processing and displaying multi-dimensional visual evaluation data, real-time and accurate quantitative evaluation of the fund transaction service quality is realized.

[0061] It should be noted that the execution subject of the embodiment can be a computing service device with data processing, network communication and program running functions, such as a tablet computer, a personal computer, a mobile phone, etc., or an electronic device, a fund transaction service quality evaluation system, etc. capable of realizing the above functions. The fund transaction service quality evaluation system is taken as an example to describe the embodiment and the following embodiments.

[0062] Based on this, the application embodiment provides a fund transaction service quality evaluation method, which is described with reference to Figure 1 , Figure 1 The flowchart of the first embodiment of the fund transaction service quality evaluation method of the application is shown in FIG. 1.

[0063] In the embodiment, the fund transaction service quality evaluation method includes steps S10-S30.

[0064] Step S10, collect full-link log data.

[0065] Full-link log data refers to log information generated by all key nodes in the entire process from the user initiating a transaction request to the system completing transaction processing in the fund transaction system.

[0066] In a possible implementation, the step of collecting full-link log data can be: setting a log point at a key node of the system, tracking the transaction process by using a distributed tracking technology, and thus realizing efficient collection of full-link log data.

[0067] It should be noted that in order to ensure the real-time of data, the acquisition system needs to have high concurrent processing capability, and can quickly respond and record various events in the transaction process. At the same time, in order to ensure the accuracy of the data, the collected log data needs to be preliminarily cleaned and verified to remove invalid or duplicate data.

[0068] Step S20, inputting the full-link log point data into a pre-constructed systematic scoring model to perform quality of service weighted scoring, and obtaining a multi-dimensional service quality score value;

[0069] The systematic scoring model is a quantitative evaluation model that comprehensively considers multiple scoring dimensions. The model is pre-constructed based on the core needs of the fund transaction scene, and by assigning appropriate weights to different scoring dimensions, the weighted scoring of the service quality is realized, so as to obtain a multi-dimensional service quality score value that can reflect the comprehensive service quality. The multi-dimensional service quality score value directly shows the service quality level of the manager.

[0070] The scoring dimensions of the systematic scoring model can include but are not limited to interface standardization, data standardization, network stability, clearing time efficiency, etc. Each scoring dimension corresponds to specific evaluation indicators and weight distribution, and the evaluation indicators and weight distribution can be dynamically adjusted according to specific business scenarios and needs. It can be understood that the entire set of scoring logic in the scoring model (including data cleaning, dimension scoring, weight weighting, and score summarizing steps) can be written into SQL script language for storage, so that using the scoring model can complete the calculation inside the database, without the need to pull data to external programs for processing, reducing data movement and improving real-time performance.

[0071] For example, in a specific embodiment, the pre-constructed systematic scoring model can assign 30% weight to the interface standardization dimension, 30% weight to the data standardization dimension, 30% weight to the network stability dimension, and 10% weight to the clearing time efficiency dimension according to the needs of the fund transaction scene. When the full-link log point data is input into the model, the model will analyze and calculate the data according to the evaluation indicators of each dimension. For example, in terms of interface standardization, the model will check whether the manager supports standard interface business types such as account opening application, subscription application, etc., and give appropriate scores according to the support situation. Finally, the multi-dimensional service quality score value obtained by weighted calculation will comprehensively reflect the service quality of the manager, providing strong support for subsequent decision-making and optimization.

[0072] Step S30, processing the multi-dimensional service quality score value to obtain multi-dimensional visual evaluation data, and quantifying the fund transaction service quality according to the multi-dimensional visual evaluation data.

[0073] The multi-dimensional visual evaluation data refers to data obtained by further processing and handling the multi-dimensional service quality score value, which can visually and visually display the service quality status. The processing of the multi-dimensional visual evaluation data involves multiple steps such as data aggregation, normalization processing, trend analysis, etc., to ensure that the data can accurately and intuitively reflect the actual situation of the service quality.

[0074] In a possible implementation, the multi-dimensional visual evaluation data can be displayed through a data large screen, which can display the service quality score, ranking, trend change, etc. of the manager in real time, so that the relevant personnel can timely understand the system operation status and service quality level.

[0075] For example, in a specific implementation, the multi-dimensional visual evaluation data processed can display the monthly service quality score ranking of each manager in the form of a column chart on the front-end large screen, and display the monthly change trend of the service quality score of a manager in the form of a line chart. In addition, the proportion of each score dimension in the total score can be displayed through a pie chart, and the fluctuation of the service quality in different time periods can be displayed through a heat map. Through these intuitive visual displays, key information can be quickly obtained, problems can be found in time and corresponding measures can be taken, so as to effectively improve the fund transaction service quality.

[0076] The embodiment provides a fund transaction service quality evaluation method. First, by collecting full-link log point data, real-time collection and integration of full-link data of a fund transaction system are realized, the integrity and consistency of the data are ensured, and high-quality original data are provided for subsequent quantitative evaluation; further, a pre-constructed systematic scoring model performs multi-dimensional service quality weighted scoring based on the full-link log point data, multi-dimensional service quality score values are obtained, the fund transaction service quality is quantified, and multi-dimensional and fine evaluation of the fund transaction service quality is realized; further, the multi-dimensional service quality score values are processed to obtain multi-dimensional visual evaluation data displayed in a visual manner, the fund transaction service quality is intuitively and comprehensively monitored, and the fund transaction service quality is quantified. In summary, through full-link log point data collection, service quality weighted scoring by a systematic scoring model, and multi-dimensional visual evaluation data processing and display, real-time and accurate quantitative evaluation of the fund transaction service quality is realized.

[0077] In a possible implementation, step S10 can include steps S11-S14:

[0078] In step S11, fund transaction business scenario information is acquired, and basic point rules and scene extension rules are configured according to the fund transaction business scenario information;

[0079] Fund transaction scenario information refers to detailed information on various business scenarios involved in fund transactions, including but not limited to transaction type (such as purchase, redemption, conversion, etc.), transaction channel (such as online platform, offline outlet, etc.), transaction time (such as weekdays, holidays, etc.), and transaction amount range. Fund transaction scenario information reflects the diversity and complexity of fund transaction operations, providing a basis for configuring basic tracking rules and scenario extension rules.

[0080] Basic data tracking rules refer to the general and fundamental data tracking rules used in fund trading systems to collect key data during the trading process. Scenario-extended rules, on the other hand, are data tracking rules designed to meet the specific needs of particular business scenarios. They supplement and refine basic data tracking rules to satisfy data collection requirements in different scenarios.

[0081] For example, in one specific implementation, when the fund transaction business scenario information shows that a fund manager mainly conducts large-scale fund transactions through online platforms, the basic data collection rules can be configured to collect general data such as request time, transaction amount, and interface response time for all online transactions. The scenario extension rules, however, can be tailored to large-scale transaction scenarios by adding specific indicators such as recording transaction approval processes and collecting risk assessment results. This approach ensures that complete and detailed data can be collected in large-scale transaction scenarios, providing strong support for subsequent analysis of issues such as transaction delays and approval efficiency.

[0082] Step S12: Combine the basic tracking rules with the scene extension rules to generate a dynamic tracking rule set;

[0083] A dynamic event tracking rule set refers to a set of event tracking rules that combines basic event tracking rules with scenario-extended rules, dynamically generated based on real-time changes and needs in fund trading business scenarios. The generation and management of dynamic event tracking rule sets are achieved through a rule engine or configuration management center. The rule engine automatically combines basic event tracking rules with scenario-extended rules based on real-time business scenario information and preset rule generation strategies to generate a dynamic event tracking rule set suitable for the current scenario.

[0084] In one possible implementation, the dynamic event tracking rule set can be updated in real time as business scenarios change, ensuring that the event tracking rules always remain consistent with business requirements.

[0085] Step S13: Deploy the lightweight collection agent to the fund trading system, and embed lightweight preprocessing logic and the dynamic data collection rule set into the lightweight collection agent to obtain a multi-source log parallel collection framework.

[0086] Lightweight data acquisition agents are used to collect log data in fund trading systems. Lightweight preprocessing logic refers to the logic that performs preliminary processing on the raw log data during the data acquisition phase, including but not limited to data format conversion, data cleaning, and data compression. Lightweight data acquisition agents can be deployed at various key nodes of the fund trading system to achieve comprehensive collection of log data across the entire data chain.

[0087] The multi-source log parallel acquisition framework deploys lightweight acquisition agents into different components or modules of a fund trading system and combines them with dynamic event tracking rule sets to achieve parallel acquisition of log data from multiple data sources. This framework can be implemented using a distributed architecture, where each lightweight acquisition agent works independently, transmitting the collected data to a data processing center via middleware such as message queues for subsequent analysis and processing.

[0088] For example, in one specific implementation, a lightweight data collection agent is deployed on the application server of the fund trading system to collect transaction-related interface call logs and business processing logs. Simultaneously, a lightweight data collection agent is also deployed on the database server to collect data query and update logs. The lightweight preprocessing logic embedded in the lightweight data collection agent performs format standardization and simple data cleaning on the collected log data, such as removing blank lines and correcting data format errors. Dynamic data tracking rule sets guide the lightweight data collection agent to collect specific log data according to preset rules. In this way, the multi-source log parallel collection framework can efficiently collect log data from different sources and transmit it to the data processing center, providing data support for subsequent service quality assessment.

[0089] Step S14: Collect end-to-end log data in parallel based on the multi-source log parallel collection framework.

[0090] The collection of end-to-end log data covers every stage of the transaction process, from the initiation of user requests and the system's processing to the final transaction result feedback, enabling monitoring and data collection across the entire transaction chain.

[0091] The multi-source log parallel acquisition framework has data synchronization and verification mechanisms to prevent data loss or duplicate acquisition. For example, each lightweight acquisition agent is assigned a unique identifier, which is carried during data transmission to achieve data traceability and verification. Simultaneously, the framework also needs load balancing capabilities; parallel-acquired data is transmitted and temporarily stored through a distributed message queue, and the data processing center consumes data from the message queue for subsequent processing and analysis.

[0092] Based on the first embodiment of the present application, the second embodiment of the present application is proposed. In the second embodiment of the present application, the same or similar contents as the above first embodiment can be referred to the above introduction, and the subsequent will not be described in detail.

[0093] On this basis, please refer to Figure 2 , before step S20, the fund transaction service quality evaluation method further comprises steps S01-S04:

[0094] Step S01, acquiring fund transaction scene core requirements, defining scoring dimensions according to the fund transaction scene core requirements;

[0095] The fund transaction scene core requirements refer to the business requirements and user expectations that have a key impact on service quality in fund transaction business, usually including accuracy, timeliness, stability, security and other aspects of transaction. Scoring dimensions are specific aspects or indicators for evaluating service quality determined according to these core requirements.

[0096] The acquisition of fund transaction scene core requirements can be achieved by analyzing historical failure data to identify key factors affecting service quality, and then determining scoring dimensions. For example, through statistical analysis of fund transaction historical failure data, it is found that the standardization degree of transaction interface, data accuracy, network stability and clearing timeliness are the main factors affecting service quality, so these factors can be used as scoring dimensions. In a possible implementation, the definition of scoring dimensions can also be combined with expert opinions and industry best practices to ensure the comprehensiveness and reasonableness of scoring dimensions.

[0097] Step S02, calling fund transaction historical failure data, assigning scoring weights to the scoring dimensions according to the fund transaction historical failure data, and determining the respective data sources corresponding to the scoring dimensions;

[0098] Fund transaction historical failure data refers to various failure information recorded in the past operation of the fund transaction system, including failure type, failure occurrence time, failure duration, failure impact range, failure handling measures, etc. Data source refers to the specific data used to evaluate each scoring dimension.

[0099] The analysis of fund transaction historical failure data can be achieved through data mining and statistical analysis techniques. For example, by clustering analysis and correlation analysis of historical failure data, the relationship between different scoring dimensions and service quality is identified, so as to reasonably allocate scoring weights. For example, if the historical failure data shows that the proportion of transaction failures caused by network stability problems is high and has a greater impact on user experience, a higher weight can be allocated to the network stability dimension. The data source and data collection method corresponding to each scoring dimension can also be explicitly defined by establishing a data dictionary or data mapping table.

[0100] In step S03, a scoring rule is defined according to the scoring dimensions, scoring weights, and data sources, and interface standardized scoring models, data standardized scoring models, network stability scoring models, and clearing time efficiency scoring models are constructed based on the scoring rule.

[0101] The scoring rule refers to the specific evaluation criteria and calculation methods formulated according to the scoring dimensions, scoring weights, and data sources, which explicitly define how to calculate the score of each scoring dimension. Interface standardized scoring models, data standardized scoring models, network stability scoring models, and clearing time efficiency scoring models correspond to different scoring dimensions and are used to quantitatively evaluate the service quality of specific dimensions.

[0102] The definition of the scoring rule can be formulated by setting specific deduction items, bonus items, and weight coefficients, etc. For example, for the interface standardized scoring model, the score can be calculated according to the number of standard interface business types supported by the manager, and a certain number of points will be deducted for each missing standard interface business type. For the data standardized scoring model, points can be deducted according to the number of abnormal data in the clearing log, and the same type of abnormal data is only deducted once. The scoring rule can also be stored and managed in the form of configuration files or database tables, which facilitates adjustment and optimization according to business needs.

[0103] In step S04, a systematic scoring model is constructed based on the interface standardized scoring model, the data standardized scoring model, the network stability scoring model, and the clearing time efficiency scoring model.

[0104] The construction of the systematic scoring model needs to ensure that the weight distribution between each scoring model is reasonable, and the score of each scoring model can accurately reflect the status of the corresponding scoring dimension. The weight distribution and scoring rule can be continuously optimized through backtesting and verification of historical data to improve the accuracy and reliability of the systematic scoring model.

[0105] In this embodiment, by clearly defining the core requirements of the fund trading scenario and establishing corresponding scoring dimensions, reasonable weights are assigned to each scoring dimension based on historical fund trading failure data. This weight allocation method based on actual historical data ensures accurate consideration of the impact of different service quality aspects during the evaluation process, making the evaluation results more objective and realistic. Detailed scoring rules provide clear quantitative standards and calculation methods for each scoring dimension. By integrating the various independent scoring models into a systematic scoring model, a comprehensive quantitative evaluation of fund trading service quality is achieved.

[0106] Based on the first and / or second embodiments of this application, a third embodiment of this application is proposed. In this third embodiment, content that is the same as or similar to the above embodiments can be referred to the above description, and will not be repeated hereafter.

[0107] In this embodiment, step S20, which involves inputting the end-to-end log data into a pre-built systematic scoring model for service quality weighting and scoring to obtain a multi-dimensional service quality score, may include steps S21 to S26:

[0108] Step S21: Remove invalid fields and similar abnormal data from the full-link log data, and unify the timestamps of the full-link log data to obtain standardized full-link log data. The standardized full-link log data includes interface standardized configuration data, intraday settlement abnormal data, interaction message logs, connectivity logs, data settlement control data, and fund transfer control data.

[0109] Invalid fields refer to fields that have no practical meaning in the data processing process or do not contribute to service quality assessment, such as null fields and duplicate fields. Identical anomaly data refers to data in the logs that repeatedly record the same type of problem, such as multiple records of failed calls to the same API.

[0110] Standardized end-to-end log data refers to log data that has been cleaned and formatted uniformly, including standardized interface configuration data, intraday settlement anomaly data, interactive message logs, connectivity logs, data settlement control data, and fund transfer control data.

[0111] By removing invalid fields and similar outliers, the complexity of data processing can be reduced and processing speed improved. Simultaneously, standardized timestamps facilitate accurate comparison and aggregation of time-series data in subsequent analysis. In one possible implementation, data cleaning tools or scripts can be used to automatically identify and remove invalid fields and similar outliers, and to convert timestamps.

[0112] Step S22: Input the interface standardization configuration data into the interface standardization scoring model to perform service quality weighted scoring and obtain the interface standardization score value;

[0113] Standardized interface configuration data refers to the configuration information related to interface standardization in the fund trading system, including supported interface business types, interface calling specifications, and interface response time requirements. The standardized interface configuration data is input into the interface standardization scoring model for service quality weighting. The specific processing logic is as follows:

[0114] First, the corresponding configuration data of the management organization with an effective status are selected for processing. The base score is 100 points. For each sub-item of the standard supported interface business type, confirmation document business type, and reconciliation document receiving business type, 3 points are deducted if no sub-item is supported on the same working day, with a minimum score of 0 points. The total score accounts for 30%. For example, if the management organization does not support N items, the standardized score value of the management organization interface is A = (100 - 3 × N) × 30%. When A is less than 0, it is assigned the value 0; otherwise, the calculated result is used.

[0115] Step S23: Input the daytime clearing anomaly data into the data standardization scoring model to perform service quality weighted scoring and obtain the data standardization score value;

[0116] Intraday clearing anomaly data refers to abnormal records generated during the intraday clearing process of fund trading, including issues such as missing data, incorrect data format, and data inconsistencies. Intraday clearing anomaly data is input into a standardized data scoring model for service quality weighting. The specific processing logic is as follows:

[0117] First, the daily settlement logs are cleaned, filtering for abnormal data containing keywords such as "failure," "error," "missing," "redundant," "not present," and "not configured," and without the word "download." This rule is gradually added as the system is optimized. The purpose is to filter abnormal data, and duplicates of the same type of abnormality are removed. The base score is 100 points. 5 points are deducted for each specific item in the settlement abnormality list, and only one deduction is made for the same type of settlement abnormality on the same workday. For example, if a manager has N settlement abnormalities (of different types), the manager's standardized score B = (100 - 5 × N) × 30%. When B is less than 0, it is assigned the value 0; otherwise, the calculated result is used.

[0118] Step S24: Input the interaction message log and the connectivity log into the network stability scoring model to perform a service quality weighted score and obtain a network stability score value.

[0119] Interaction message logs refer to the message records generated during interactions between the fund trading system and the manager, including request messages, response messages, message transmission times, and other information. Connectivity logs record the network connectivity between the system and the manager, including information such as network connection establishment, disconnection, and connection delays. The interaction message logs and connectivity logs are input into the network stability scoring model for service quality weighted scoring. The specific processing logic is as follows:

[0120] In the administrator message log data, filter for data where the administrator's response message is empty, and deduct 5 points for each instance of an anomaly. In the SFTP / FTP connectivity log, filter for connectivity test anomaly records from 9:00 AM to 6:00 PM, and deduct 5 points for each anomaly on the same workday. The base score is 100 points, with 5 points deducted for each anomaly, and a minimum score of 0 points. The total score accounts for 30%. For example, if the administrator's response message is empty and occurs N times on the same workday, the administrator's network stability score C = (100 - 5 × N) × 30%. When C is less than 0, assign a value of 0; otherwise, use the calculated result.

[0121] Step S25: Input the data clearing control data and the fund transfer control data into the clearing timeliness scoring model to perform service weighted scoring and obtain the clearing timeliness score value;

[0122] Data clearing control data refers to various control information recorded during the fund transaction data clearing process, including clearing start time, clearing end time, clearing status, and clearing exception handling records. Fund transfer control data refers to control information recorded during the fund transfer process, including transfer request time, transfer confirmation time, transfer status, and transfer amount. The data clearing control data and fund transfer control data are input into the clearing timeliness scoring model for service weighted scoring. The specific processing logic is as follows:

[0123] The system currently ensures that each manager completes settlement by the end of the day. Settlement completion time can be viewed in the data settlement control record. The day is divided into 24 hourly intervals. For each hourly interval where settlement is completed after 12:00 noon, 5 points will be deducted. In the fund purchase and redemption transfer control records, the last update time of records marked as having no confirmed transfer data or successful transfers for the day is filtered. The day is also divided into 24 hourly intervals. For each hourly interval where settlement is completed after 12:00 noon, 3 points will be deducted. The base score is 100 points, with deductions calculated based on the cumulative hourly deductions for each completion time interval. The minimum score is 0 points, and each hourly interval accounts for 10% of the total score. For example, if the data clearing and fund purchase transfer for the administrator are completed at 11:00 AM (no points deducted), and the fund redemption transfer is completed at 2:30 PM, calculated using a 24-hour clock with 12:00 PM as the base point, the delay is calculated as (H+1)-12, with a deduction coefficient of 3 points. The administrator's clearing timeliness score is D = (100-3×((H+1)-12))×10%. If the data clearing timeliness is later than 12:00 PM, additional points will be deducted, with a deduction coefficient of 5 points. When D is less than 0, the value is assigned to 0; otherwise, the calculated result is used.

[0124] Step S26: Add the interface standardization score, the data standardization score, the network stability score, and the settlement timeliness score to obtain a multi-dimensional service quality score.

[0125] The scores from each dimension are aggregated to obtain a multi-dimensional service quality score, which is used to comprehensively evaluate the overall service quality of the administrator. The multi-dimensional service quality score = Interface Standardization Score A + Data Standardization Score B + Network Stability Score C + Settlement Timeliness Score D.

[0126] In this embodiment, data preprocessing and standardization ensure the accuracy and consistency of the evaluation data; multiple scoring models are used to quantitatively score service quality from different dimensions, which can comprehensively reflect the quality of fund trading services; finally, the scores of each dimension are summarized to obtain a multi-dimensional service quality score, realizing a comprehensive, accurate and multi-dimensional quantitative evaluation of the quality of fund trading services.

[0127] Based on the above embodiments of this application, a fourth embodiment of this application is proposed. In this fourth embodiment, content that is the same as or similar to that in the above embodiments can be referred to the above description, and will not be repeated hereafter.

[0128] Based on this, please refer to Figure 3 , Figure 3 This is a schematic flowchart of the fourth embodiment of this application. Figure 3As shown, step S30, which processes the multi-dimensional service quality score to obtain multi-dimensional visual evaluation data, and quantifies the fund transaction service quality based on the multi-dimensional visual evaluation data, includes steps S31 to S34:

[0129] Step S31: Match metadata tags to the multidimensional service quality score value, and store the multidimensional service quality score value after matching metadata tags in a structured manner in the data warehouse to obtain a structured score dataset;

[0130] Metadata tags are additional information tags used to describe multidimensional service quality scores. These tags may include the scoring date, the scoring object (such as the administrator's identifier), the scoring dimensions (such as interface standardization, data standardization, etc.), and the scoring model version. A structured scoring dataset is a collection of multidimensional service quality scores and related metadata that has undergone structured storage processing.

[0131] For example, in one specific implementation, the system generates a set of metadata tags for each multi-dimensional service quality score, such as a score date of "2024-11-20", a score object of "Manager A", and score dimensions including "interface standardization, data standardization, network stability, and settlement timeliness". Then, the system stores these score values ​​and their metadata tags in a data warehouse according to a predefined data table structure. The data table may contain fields such as "score object ID", "score date", and "interface standardization score".

[0132] Step S32: Aggregate and calculate the structured scoring dataset according to the time dimension to obtain service quality change trend data;

[0133] Aggregation calculations by time dimension refer to summarizing and calculating data in a structured scoring dataset based on time fields (such as day, week, month, quarter, year, etc.). For example, this involves calculating statistical indicators such as average score, total score, maximum value, and minimum value for each time period. Service quality trend data refers to statistical data obtained through aggregation calculations that reflects changes in service quality over time. This data can be used to analyze fluctuations in service quality, identify periods of problem, and evaluate the effectiveness of improvement measures.

[0134] For example, in one specific implementation, the system aggregates and calculates the structured scoring dataset monthly, determining the average, maximum, and minimum multidimensional service quality scores for each month. Assume that in November 2024, Manager A's average multidimensional service quality score was 85, the maximum was 92, and the minimum was 78. This data constitutes service quality trend data. By comparing the data from each month, the fluctuation trend of service quality can be observed.

[0135] Step S33: Call the visualization engine to perform visualization processing on the service quality change trend data and the structure score dataset to obtain multi-dimensional visualization evaluation data;

[0136] The visualization engine transforms service quality trend data and structured scoring datasets into intuitive visual charts, resulting in multi-dimensional visual evaluation data that enables users to quickly understand the service quality status and make effective analyses and decisions.

[0137] For example, in one specific implementation, the visualization engine extracts service quality trend data and structured scoring datasets from the data warehouse, and then generates various charts according to preset visualization configurations. For instance, it generates line charts to show the monthly trends of multi-dimensional service quality scores, bar charts to compare the scores of different managers on various scoring dimensions, and pie charts to show the proportion of each scoring dimension in the total score. These charts collectively constitute multi-dimensional visualized evaluation data, allowing users to quickly understand the overall status, trends, and performance of each dimension of service quality. This intuitive presentation enables users to promptly identify problems, recognize strengths and weaknesses, and provide strong support for service quality optimization.

[0138] Step S34: Generate service quality labels based on the multidimensional visualization evaluation data, and quantify the quality of fund trading services based on the service quality labels.

[0139] Service quality labels refer to tags or categories generated based on multi-dimensional visual evaluation data to describe the status of service quality, such as "excellent," "good," "average," and "poor." In quantitative fund trading service quality, these labels translate service quality into specific levels or numerical values.

[0140] In one possible implementation, service quality tags can be generated based on preset scoring thresholds or business rules. For example, a multi-dimensional service quality score of 90 or above is tagged as "Excellent"; 80-89 is "Good"; 70-79 is "Average"; and below 70 is "Poor". Assuming a manager's multi-dimensional service quality score is 85, the system will generate a "Good" service quality tag for them. Furthermore, the system can generate more detailed tags based on the performance of each scoring dimension, such as "Excellent interface standardization," "Good data standardization," "Average network stability," and "Good settlement timeliness." In this way, service quality tags can intuitively reflect the manager's service quality status, facilitating quick understanding of the situation and enabling corresponding optimization and improvement measures.

[0141] In this embodiment, metadata tags are added to the multi-dimensional service quality scores and stored in a structured manner, laying the foundation for efficient data querying and analysis, and making data management more standardized and orderly. Aggregated calculations by the time dimension clearly reveal the changing trends of service quality, facilitating the timely detection of fluctuations. A visualization engine transforms the data into intuitive charts, greatly improving data understandability and enabling users to quickly grasp key information about service quality. By generating service quality tags, a quantitative assessment of service quality is achieved, providing management with a concise and clear basis for decision-making.

[0142] In one specific implementation, such as Figure 4 As shown, Figure 4 This is a schematic diagram illustrating a specific implementation of the fourth embodiment of this application. Step S33 is followed by steps S331 to S333:

[0143] Step S331: Extract historical rating time series from the structured rating dataset;

[0144] Historical rating time series refers to a sequence of rating data extracted from a structured rating dataset and arranged chronologically. This data records changes in service quality at different points in time. The extraction of historical rating time series aims to provide foundational data for subsequent time series analysis and forecasting, helping to identify patterns and trends in service quality changes.

[0145] In one possible implementation, historical rating time series can be generated by extracting rating data from the data warehouse in chronological order by writing SQL query statements or using data extraction tools.

[0146] Step S332: Use a long short-term memory network to predict the historical rating time series to obtain a service quality trend prediction value;

[0147] Service quality trend prediction refers to the estimated service quality score for a future period obtained by predicting historical score time series using an LSTM model. Specifically, an LSTM model is built and trained using a deep learning framework (such as TensorFlow or PyTorch), taking historical score time series as input, to predict service quality scores for multiple future time points.

[0148] Step S333: Overlay the comparison curve of the service quality trend prediction value onto the multidimensional visualization evaluation data.

[0149] A comparison curve refers to overlaying the historical service quality score change curve with the predicted service quality trend curve in a visualization chart to visually compare the changing trends of the actual and predicted values.

[0150] For example, when generating a line chart showing service quality trends, the historical score time series and the LSTM-predicted service quality trend values ​​are plotted on the same chart. The historical score curve is represented by a solid line, and the predicted score curve is represented by a dashed line, with the starting point of the prediction marked on the chart.

[0151] Based on the above embodiments of this application, a fifth embodiment of this application is proposed. In this fifth embodiment, content that is the same as or similar to the above embodiments can be referred to the above description, and will not be repeated hereafter.

[0152] Based on this, please refer to Figure 5 , Figure 5 This is a schematic flowchart of the fifth embodiment of this application. Figure 5 As shown, after step S30, which processes the multi-dimensional service quality score to obtain multi-dimensional visual evaluation data, and then quantifies the fund transaction service quality based on the multi-dimensional visual evaluation data, steps S40 to S60 are further included:

[0153] Step S40: Extract key variables from the multidimensional visualization evaluation data and construct a causal graph based on the key variables;

[0154] Key variables refer to factors or indicators that have a significant impact on service quality in multidimensional visualization assessment data, such as interface standardization scores and network stability scores. Cause-and-effect graphs are graphical tools used to represent causal relationships between variables. By constructing cause-and-effect graphs, the interactions between key variables and their impact on service quality can be visually displayed.

[0155] Key variables were extracted from the multidimensional visualization assessment data: interface standardization score, data standardization score, network stability score, and settlement timeliness score. The correlations and interactions between these variables were then analyzed to construct a causal graph. For example, the causal graph might show that a decrease in the network stability score leads to an increase in interface call failures, thus affecting the interface standardization score. Simultaneously, a decrease in the data standardization score may lead to an increase in settlement anomalies, affecting the settlement timeliness score. This visualization of causal relationships clearly shows the mutual influence between the key variables, allowing for targeted measures to address the problems.

[0156] Step S50: The causal intensity in the causal graph is quantified using the difference-in-differences method, the causal intensity is sorted, and the scoring governance items are determined.

[0157] The difference-in-differences (DID) method quantifies the causal strength in the causal graph and ranks them according to their strength to determine scoring governance items, providing clear direction and priorities for service quality optimization. In one possible implementation, statistical analysis software (such as Python's statsmodels library) can be used to perform the DID calculation and rank the causal strength.

[0158] Specifically, the system uses the difference-in-differences method to quantitatively analyze the causal relationships in the causal graph. Assume there are three main causal relationships in the graph: A. Decreased network stability leads to increased API call failures; B. Data standardization issues lead to increased settlement anomalies; C. Settlement time delays lead to increased customer complaints. Using the difference-in-differences method, the causal strengths are calculated as follows: A: 0.65, B: 0.78, C: 0.45. Ranking the causal strengths, B (data standardization issues leading to increased settlement anomalies) is determined as the primary governance item, followed by A (decreased network stability leading to increased API call failures), and finally C (settlement time delays leading to increased customer complaints).

[0159] Step S60: Perform multi-agent reinforcement learning training based on the scoring governance items to generate a fund transaction service quality governance scheme.

[0160] Multi-agent reinforcement learning training involves multiple agents learning through trial and error in an environment to optimize their behavioral strategies in order to maximize cumulative rewards. In service quality governance, agents represent different governance measures or strategies, and through continuous trial and evaluation, the optimal governance solution is found. A fund trading service quality governance solution refers to a series of specific measures and strategies generated through multi-agent reinforcement learning training to improve the quality of fund trading services.

[0161] For example, in one specific implementation, the system constructs a multi-agent reinforcement learning training environment based on scoring governance items (such as data standardization and network stability issues). Each agent represents a governance measure; for example, agent 1 represents "optimizing the data cleaning process," agent 2 represents "increasing network bandwidth," and agent 3 represents "adjusting interface call timeouts," etc. The agents in the environment try different combinations of measures, observe changes in service quality indicators, and adjust their strategies according to a preset reward function (such as an improvement in service quality score). After multiple iterations of training, the agents learn the optimal combination of governance measures; for example, "optimizing the data cleaning process" and "increasing network bandwidth" can significantly improve the data standardization score and network stability score. Based on these training results, the system generates a fund trading service quality governance scheme, including specific measures, implementation steps, and expected results.

[0162] In this embodiment, by extracting key variables from multidimensional visualized evaluation data and constructing a causal graph, the core factors affecting the quality of fund trading services and their interrelationships are accurately identified, providing a clear direction for subsequent governance measures. The difference-in-differences method is used to quantify and rank the causal strength in the causal graph, making the priority of service quality issues clearly visible and ensuring the reasonable allocation and effective utilization of governance resources. Multi-agent reinforcement learning is used for training to generate an intelligent fund trading service quality governance scheme based on the scoring governance items.

[0163] For example, to help understand the implementation process of the fund transaction service quality assessment method obtained by combining this embodiment with the above embodiments, please refer to... Figure 6 , Figure 6 This application provides a simplified flowchart of a method for evaluating the quality of fund trading services. Specifically, it provides a scoring model based on collected daily log data from fund trading systems, after cleaning and processing, to assess the service level and technical capabilities of fund managers. The specific implementation process is as follows:

[0164] First, log data collection is performed: standardized configuration data of interfaces are collected throughout the day, configuration data supporting file business types are confirmed, data settlement logs are collected, data of communication with the administrator is collected, SFTP / FTP connectivity logs are collected, data settlement control records are collected, data of fund purchase direction transfer control records are collected, and data of fund redemption direction transfer control records are collected.

[0165] Further, data is entered into the data warehouse, cleaned, and processed: At 1:00 AM daily (T+1 day), data from the previous day (T day) is imported into the data warehouse, and data cleaning and processing are performed using a scoring model. The data is divided into four dimensions: interface standardization, data standardization, network stability, and settlement timeliness. The total score is 100 points, and each manager's final score for each day is generated through detailed evaluation of the four dimensions.

[0166] Furthermore, the scoring data is shipped out daily: the data warehouse cleans and processes the results based on the scoring model, and then ships them to the application database according to the date dimension. This data is then displayed on the front-end dashboard according to different dimensions. Currently, the display dimensions include the monthly total score ranking, the monthly average total score trend, and the monthly average score trend displayed according to four dimensions: interface standardization, data standardization, network stability, and settlement timeliness.

[0167] Finally, the scoring governance items are analyzed: by analyzing the scoring results of each manager's total score and each sub-score, as well as the historical score change trends, we can analyze the current technical service weaknesses of the managers in terms of interface standardization, data standardization, network stability, and settlement timeliness. Combined with detailed log data, we can generate a comprehensive governance plan to promote managers to improve service quality and technical capabilities, and close the loop on the problem.

[0168] It should be noted that the above examples are only for understanding this application and do not constitute a limitation on the fund transaction service quality assessment method of this application. Any simple modifications based on this technical concept are within the protection scope of this application.

[0169] This application also provides a device for evaluating the quality of fund trading services; please refer to [reference needed]. Figure 7 The fund transaction service quality assessment device includes:

[0170] The acquisition module 10 is used to collect end-to-end log data.

[0171] The scoring module 20 is used to input the full-link log data into a pre-built systematic scoring model to perform service quality weighted scoring and obtain a multi-dimensional service quality score value.

[0172] The quantitative module 30 is used to process the multi-dimensional service quality score to obtain multi-dimensional visual evaluation data, and to quantify the quality of fund trading services based on the multi-dimensional visual evaluation data.

[0173] The fund trading service quality assessment device provided in this application, employing the fund trading service quality assessment method described in the above embodiments, can solve the technical problems related to fund trading service quality assessment. Compared with the prior art, the beneficial effects of the fund trading service quality assessment device provided in this application are the same as those of the fund trading service quality assessment method provided in the above embodiments, and other technical features in the fund trading service quality assessment device are the same as those disclosed in the methods of the above embodiments, and will not be repeated here.

[0174] This application provides a fund trading service quality assessment device, which includes: at least one processor; and a memory communicatively connected to the at least one processor; wherein the memory stores instructions executable by the at least one processor, and the instructions are executed by the at least one processor to enable the at least one processor to perform the fund trading service quality assessment method in the above embodiment 1.

[0175] The following is for reference. Figure 8The diagram illustrates a structural schematic of a fund transaction service quality assessment device suitable for implementing embodiments of this application. The fund transaction service quality assessment device in this application may include, but is not limited to, mobile terminals such as mobile phones, laptops, digital radio receivers, PDAs (Personal Digital Assistants), PADs (Portable Application Description), PMPs (Portable Media Players), in-vehicle terminals (e.g., in-vehicle navigation terminals), and fixed terminals such as digital TVs and desktop computers. Figure 8 The fund transaction service quality assessment device shown is merely an example and should not impose any limitations on the functionality and scope of use of the embodiments of this application.

[0176] like Figure 8 As shown, the fund trading service quality assessment device may include a processing unit 1001 (e.g., a central processing unit, a graphics processing unit, etc.), which can perform various appropriate actions and processes according to a program stored in a read-only memory 1002 or a program loaded from a storage device 1003 into a random access memory 1004. The random access memory 1004 also stores various programs and data required for the operation of the fund trading service quality assessment device. The processing unit 1001, the read-only memory 1002, and the random access memory 1004 are interconnected via a bus 1005. An input / output interface 1006 is also connected to the bus. Typically, the following systems can be connected to the input / output interface 1006: input devices 1007 including, for example, a touch screen, touchpad, keyboard, mouse, image sensor, microphone, accelerometer, gyroscope, etc.; output devices 1008 including, for example, a liquid crystal display (LCD), speaker, vibrator, etc.; storage devices 1003 including, for example, magnetic tape, hard disk, etc.; and communication devices 1009. Communication device 1009 allows the fund trading service quality assessment device to communicate wirelessly or wiredly with other devices to exchange data. Although the figure shows fund trading service quality assessment devices with various systems, it should be understood that implementation or possession of all the systems shown is not required. More or fewer systems may be implemented alternatively.

[0177] Specifically, according to the embodiments disclosed in this application, the processes described above with reference to the flowcharts can be implemented as computer software programs. For example, embodiments disclosed in this application include a computer program product comprising a computer program carried on a computer-readable medium, the computer program containing program code for performing the methods shown in the flowcharts. In such embodiments, the computer program can be downloaded and installed from a network via a communication device, or installed from storage device 1003, or installed from read-only memory 1002. When the computer program is executed by processing device 1001, it performs the functions defined in the methods of the embodiments disclosed in this application.

[0178] The fund trading service quality assessment device provided in this application, employing the fund trading service quality assessment method described in the above embodiments, can solve the technical problems of fund trading service quality assessment. Compared with the prior art, the beneficial effects of the fund trading service quality assessment device provided in this application are the same as those of the fund trading service quality assessment method provided in the above embodiments, and other technical features of the fund trading service quality assessment device are the same as those disclosed in the method of the previous embodiment, and will not be repeated here.

[0179] It should be understood that the various parts disclosed in this application can be implemented using hardware, software, firmware, or a combination thereof. In the description of the above embodiments, specific features, structures, materials, or characteristics can be combined in any suitable manner in one or more embodiments or examples.

[0180] The above description is merely a specific embodiment of this application, but the scope of protection of this application is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the scope of the technology disclosed in this application should be included within the scope of protection of this application. Therefore, the scope of protection of this application should be determined by the scope of the claims.

[0181] This application provides a computer-readable storage medium having computer-readable program instructions (i.e., a computer program) stored thereon, the computer-readable program instructions being used to execute the fund transaction service quality assessment method in the above embodiments.

[0182] The computer-readable storage medium provided in this application may be, for example, a USB flash drive, but is not limited to, electrical, magnetic, optical, electromagnetic, infrared, or semiconductor systems or devices, or any combination thereof. More specific examples of computer-readable storage media may include, but are not limited to: electrical connections having one or more wires, portable computer disks, hard disks, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), optical fibers, portable compact disk read-only memory (CD-ROM), optical storage devices, magnetic storage devices, or any suitable combination thereof. In this embodiment, the computer-readable storage medium may be any tangible medium containing or storing a program that can be used by or in conjunction with an instruction execution system or device. The program code contained on the computer-readable storage medium may be transmitted using any suitable medium, including but not limited to: wires, optical cables, RF (Radio Frequency), etc., or any suitable combination thereof.

[0183] The aforementioned computer-readable storage medium may be included in the fund transaction service quality assessment equipment; or it may exist independently and not be assembled into the fund transaction service quality assessment equipment.

[0184] The aforementioned computer-readable storage medium carries one or more programs. When the aforementioned one or more programs are executed by the fund trading service quality assessment device, the fund trading service quality assessment device causes the following: it collects end-to-end log data; it inputs the end-to-end log data into a pre-built systematic scoring model to perform service quality weighted scoring, thereby obtaining a multi-dimensional service quality score value; it processes the multi-dimensional service quality score value to obtain multi-dimensional visual assessment data, and it quantifies the fund trading service quality based on the multi-dimensional visual assessment data.

[0185] Computer program code for performing the operations of this application can be written in one or more programming languages ​​or a combination thereof, including object-oriented programming languages ​​such as Java, Smalltalk, and C++, as well as conventional procedural programming languages ​​such as "C" or similar programming languages. The program code can be executed entirely on the user's computer, partially on the user's computer, as a standalone software package, partially on the user's computer and partially on a remote computer, or entirely on a remote computer or server. In cases involving remote computers, the remote computer can be connected to the user's computer via any type of network—including a Local Area Network (LAN) or a Wide Area Network (WAN)—or can be connected to an external computer (e.g., via the Internet using an Internet service provider).

[0186] The flowcharts and block diagrams in the accompanying drawings illustrate the architecture, functionality, and operation of possible implementations of systems, methods, and computer program products according to various embodiments of this application. In this regard, each block in a flowchart or block diagram may represent a module, segment, or portion of code containing one or more executable instructions for implementing a specified logical function. It should also be noted that in some alternative implementations, the functions indicated in the blocks may occur in a different order than those indicated in the drawings. For example, two consecutively indicated blocks may actually be executed substantially in parallel, and they may sometimes be executed in reverse order, depending on the functions involved. It should also be noted that each block in the block diagrams and / or flowcharts, and combinations of blocks in the block diagrams and / or flowcharts, can be implemented using a dedicated hardware-based system that performs the specified function or operation, or using a combination of dedicated hardware and computer instructions.

[0187] The modules described in the embodiments of this application can be implemented in software or hardware. The names of the modules do not necessarily limit the functionality of the unit itself.

[0188] The readable storage medium provided in this application is a computer-readable storage medium that stores computer-readable program instructions (i.e., a computer program) for executing the above-described fund transaction service quality assessment method, thereby solving the technical problem of fund transaction service quality assessment. Compared with the prior art, the beneficial effects of the computer-readable storage medium provided in this application are the same as the beneficial effects of the fund transaction service quality assessment method provided in the above embodiments, and will not be repeated here.

[0189] This application also provides a computer program product, including a computer program that, when executed by a processor, implements the steps of the fund transaction service quality assessment method described above.

[0190] The computer program product provided in this application can solve the technical problem of fund transaction service quality assessment. Compared with the prior art, the beneficial effects of the computer program product provided in this application are the same as the beneficial effects of the fund transaction service quality assessment method provided in the above embodiments, and will not be repeated here.

[0191] The above description is only a part of the embodiments of this application and does not limit the patent scope of this application. All equivalent structural transformations made under the technical concept of this application and using the contents of the specification and drawings of this application, or direct / indirect applications in other related technical fields, are included in the patent protection scope of this application.

Claims

1. A method for evaluating the quality of fund trading services, characterized in that, The methods for evaluating the quality of transaction services include: Collect end-to-end log data; The end-to-end log data is input into a pre-built systematic scoring model to perform a weighted service quality score, resulting in a multi-dimensional service quality score. The multidimensional service quality score is processed to obtain multidimensional visual evaluation data, and the quality of fund trading services is quantified based on the multidimensional visual evaluation data.

2. The fund transaction service quality assessment method as described in claim 1, characterized in that, The steps for collecting end-to-end log data include: Obtain fund trading business scenario information, and configure basic data tracking rules and scenario extension rules based on the fund trading business scenario information; A dynamic set of tracking rules is generated by combining the basic tracking rules with the scenario extension rules. A lightweight data collection agent is deployed to the fund trading system, and lightweight preprocessing logic and the dynamic data collection rule set are embedded in the lightweight data collection agent to obtain a multi-source log parallel collection framework. The multi-source log parallel acquisition framework is used to collect full-link log data in parallel.

3. The fund transaction service quality assessment method as described in claim 1, characterized in that, Before the step of inputting the end-to-end log data into a pre-built systematic scoring model for service quality weighting and scoring to obtain a multi-dimensional service quality score, the following steps are included: Identify the core requirements of fund trading scenarios and define scoring dimensions based on these core requirements. Call up historical fault data of fund transactions, assign scoring weights to the scoring dimensions based on the historical fault data of fund transactions, and determine the data source corresponding to each scoring dimension; Based on the scoring dimensions, scoring weights, and data sources, scoring rules are defined, and based on the scoring rules, interface standardization scoring models, data standardization scoring models, network stability scoring models, and settlement timeliness scoring models are constructed respectively. A systematic scoring model is constructed based on the interface standardization scoring model, the data standardization scoring model, the network stability scoring model, and the settlement timeliness scoring model.

4. The fund transaction service quality assessment method as described in claim 3, characterized in that, The step of inputting the end-to-end log data into a pre-built systematic scoring model for service quality weighting to obtain a multi-dimensional service quality score includes: Invalid fields and similar abnormal data in the full-link log data are removed, and the timestamps of the full-link log data are unified to obtain standardized full-link log data. The standardized full-link log data includes interface standardized configuration data, intraday settlement abnormal data, interaction message logs, connectivity logs, data settlement control data, and fund transfer control data. The standardized interface configuration data is input into the standardized interface scoring model to perform a service quality weighted score, and the standardized interface score value is obtained. The daytime clearing anomaly data is input into the data standardization scoring model to perform a service quality weighted score, and a data standardization score value is obtained. The interaction message log and the connectivity log are input into the network stability scoring model to perform a service quality weighted score, and a network stability score value is obtained. The data clearing control data and the fund transfer control data are input into the clearing timeliness scoring model to perform a service-weighted scoring, and a clearing timeliness score value is obtained. The interface standardization score, the data standardization score, the network stability score, and the settlement timeliness score are added together to obtain a multi-dimensional service quality score.

5. The fund transaction service quality assessment method as described in claim 1, characterized in that, The step of processing the multi-dimensional service quality score to obtain multi-dimensional visual evaluation data, and quantifying the quality of fund trading services based on the multi-dimensional visual evaluation data, includes: Metadata tags are matched to the multidimensional service quality score values, and the multidimensional service quality score values ​​after matching metadata tags are structured and stored in the data warehouse to obtain a structured score dataset. The structured scoring dataset is aggregated and calculated according to the time dimension to obtain service quality change trend data; The visualization engine is invoked to visualize the service quality change trend data and the structure score dataset to obtain multidimensional visualization evaluation data. Service quality labels are generated based on the multidimensional visualization evaluation data, and the quality of fund trading services is quantified based on the service quality labels.

6. The fund transaction service quality assessment method as described in claim 5, characterized in that, Following the step of visualizing the service quality change trend data and the structured scoring dataset using a visualization engine to obtain multidimensional visualization evaluation data, the method further includes: Extract historical rating time series from the structured rating dataset; A long short-term memory network is used to predict the historical rating time series to obtain a service quality trend prediction value; A comparison curve of the service quality trend prediction value is overlaid on the multidimensional visualization evaluation data.

7. The fund transaction service quality assessment method as described in claim 1, characterized in that, After processing the multidimensional service quality score to obtain multidimensional visual evaluation data, and quantifying the fund trading service quality based on the multidimensional visual evaluation data, the process further includes: Key variables are extracted from the multidimensional visualization evaluation data, and a causal graph is constructed based on the key variables; The causal intensity in the causal graph is quantified using the difference-in-differences method, and the causal intensity is ranked to determine the scoring governance items; Based on the aforementioned scoring governance items, multi-agent reinforcement learning training is performed to generate a fund transaction service quality governance scheme.

8. A fund transaction service quality assessment device, characterized in that, The device includes: a memory, a processor, and a computer program stored in the memory and executable on the processor, the computer program being configured to implement the steps of the fund transaction service quality assessment method as described in any one of claims 1 to 7.

9. A storage medium, characterized in that, The storage medium is a computer-readable storage medium, and a computer program is stored on the storage medium. When the computer program is executed by a processor, it implements the steps of the fund transaction service quality assessment method as described in any one of claims 1 to 7.

10. A computer program product, characterized in that, The computer program product includes a computer program that, when executed by a processor, implements the steps of the fund transaction service quality assessment method as described in any one of claims 1 to 7.

Images