A method and system for analyzing health degree of telecommunication virtual commodity operation based on big data

Abstract

The present application relates to the technical field of operation analysis and management through cloud computing after the commercialization of service commodities in the telecommunication industry, in particular to a method and system for analyzing the health degree of telecommunication virtual commodities based on big data, comprising the following steps: S1, establishing a tariff level basic model, extracting data from each center of the telecommunication system through a big data platform at regular time, wherein the tariff level basic model comprises a service value basic index model, a service compliance analysis index model and a service analysis index model; the method automatically extracts configuration, transaction, log and complaint data from multiple centers such as sales, customers, orders and auditing through the big data platform, breaking the traditional data silos. It builds a complete analysis chain from the basic index to the converged intermediate model, and objectively determines the index weight by using the entropy method, and finally automatically outputs the health degree comprehensive score and analysis report.

Description

Technical Field

[0001] This invention relates to the field of operational analysis and management technology for the commercialization of services in the telecommunications industry through cloud computing, specifically a method and system for analyzing the operational health of telecommunications virtual goods based on big data. Background Technology

[0002] The telecommunications industry has also been developing rapidly in recent years, with its development model gradually evolving towards a more consumer-friendly, product-oriented approach to business development. As business models become increasingly complex and diverse, single-business models are gradually transforming into more varied and flexible bundled sales methods. So, what kind of product-oriented sales approach can best resonate with users and quickly capture the market? Currently, after productizing and configuring services for market promotion, there is a lack of intelligent operational analysis capabilities regarding market quality. Market planners cannot obtain real-time statistical data during business development and combine it with pre-planning marketing goal analysis, resulting in the inability to promptly identify and warn of revenue quality issues (such as settlement inversion, excessive marketing, double billing, and excessive apportionment), leading to a lag in subsequent market judgments and strategy adjustments.

[0003] To address the aforementioned problems, this invention aims to establish a method and system for analyzing the operational health of telecommunications virtual goods based on big data. Summary of the Invention

[0004] To address the problems in existing technologies, this invention provides a method and system for analyzing the operational health of telecommunications virtual goods based on big data. The technical solution adopted by this invention to solve its technical problem is as follows: On the one hand, this invention provides a method for analyzing the operational health of telecommunications virtual goods based on big data, including the following steps: S1. Establish a tariff-level basic model and extract data from various centers of the telecommunications system on a regular basis through a big data platform. The tariff-level basic model includes a business value basic indicator model, a business compliance analysis indicator model, and a service analysis indicator model. S2. Establish a pricing intermediate model. The big data platform classifies and refines the data from the basic model in step S1, aggregates it into a specified set of indicators, injects logical judgment, and generates an aggregated model and a health model. S3. Use the entropy method to determine the weight of each indicator in step S2, and calculate the comprehensive score of the health of telecommunications virtual goods operation based on the weights. S4. Generate an analysis report based on the comprehensive health score to provide data support for market planning and decision-making.

[0005] On the other hand, the present invention provides a system for analyzing the operational health of telecommunications virtual goods based on big data, applied to the aforementioned method for analyzing the operational health of telecommunications virtual goods based on big data. The system includes: The data acquisition module is used to extract data from various centers of the telecommunications system and transmit it to the model building module; The model building module is used to establish a tariff-level basic model and a tariff intermediate model, wherein the tariff intermediate model is generated based on the data extraction of the tariff-level basic model; The weight calculation module is used to determine the weight of each indicator using the entropy method, and to calculate the comprehensive health score in combination with the indicator data. The analysis output module is used to generate and output an analysis report based on the comprehensive health score.

[0006] The beneficial effects of this invention are: This method automatically extracts configuration, transaction, log, and complaint data from multiple centers such as sales, customers, orders, and auditing through a big data platform, breaking down traditional data silos. It constructs a complete analysis chain from basic indicators to converged intermediate models, and uses the entropy method to objectively determine indicator weights, ultimately automatically outputting a comprehensive health score and analysis report.

[0007] This invention changes the past reliance on manual and fragmented analysis, and realizes normalized, automated panoramic monitoring and quantitative evaluation of the operation status of telecommunications virtual goods, which greatly improves the efficiency of operation analysis and can provide early warning of potential risks, providing forward-looking data support for strategy adjustment. Attached Figure Description

[0008] The present invention will be further described below with reference to the accompanying drawings and embodiments.

[0009] Figure 1 This invention provides an end-to-end product lifecycle diagram; Figure 2 The flowchart for the construction and evaluation of the health model provided by this invention. Detailed Implementation

[0010] To make the technical means, creative features, objectives and effects of this invention easier to understand, the invention will be further described below in conjunction with specific embodiments.

[0011] like Figure 1 - Figure 2 As shown, the method for analyzing the operational health of telecommunications virtual goods based on big data according to the present invention includes the following steps: S1. Establish a tariff-level basic model and extract data from various centers of the telecommunications system on a regular basis through a big data platform. The tariff-level basic model includes a business value basic indicator model, a business compliance analysis indicator model, and a service analysis indicator model. S2. Establish a pricing intermediate model. The big data platform classifies and refines the data from the basic model in step S1, aggregates it into a specified set of indicators, injects logical judgment, and generates an aggregated model and a health model. S3. Use the entropy method to determine the weight of each indicator in step S2, and calculate the comprehensive score of the health of telecommunications virtual goods operation based on the weights. S4. Generate an analysis report based on the comprehensive health score to provide data support for market planning and decision-making.

[0012] First, a tariff-level foundational model covering three dimensions—business value, operational compliance, and user service—is built around the entire lifecycle of virtual goods. This model utilizes a big data platform to selectively capture relevant data from multiple centers within the telecommunications system, laying a solid data foundation for subsequent analysis. Next, the scattered raw data in the foundational model is systematically categorized and core indicators are extracted. Dispersed individual data points are integrated into a set of indicators with analytical value, and then pre-defined logical judgment rules are injected to form a tariff intermediate model that combines data aggregation and diagnostic capabilities. This model includes core aggregated indicators such as order volume and ARPU, as well as specialized diagnostic models for value, compliance, and service, and a comprehensive health index model. Subsequently, the entropy method, an objective weighting technique, is used to determine the weight of each indicator in the health assessment based on the dispersion of the indicator data itself, avoiding biases caused by subjective human judgment. Finally, a weighted summation method is used to calculate the comprehensive operational health score of the virtual goods, achieving a quantitative representation of their health status. Finally, the overall score is transformed into a well-organized and focused analysis report, providing precise data support for market planners to adjust strategies and make decisions, covering the entire process from data collection to practical application.

[0013] This core approach effectively addresses the lack of digital analytics tools in traditional telecom virtual goods operations. For the first time, it enables full lifecycle health tracking from product planning to after-sales service, allowing marketing planners to monitor product operations in real time. Through the construction of a multi-dimensional model, it covers key aspects such as product commercial value, operational compliance, and user experience, avoiding the bias caused by single-indicator evaluations. The application of entropy methods ensures the objectivity and scientific rigor of weight allocation, making the health assessment results more credible. Simultaneously, the analysis reports directly address market decision-making needs, completely changing the previous situation where data lag led to untimely strategy adjustments. This helps operations teams quickly identify revenue quality issues such as settlement discrepancies and excessive marketing, providing clear guidance for subsequent optimization.

[0014] As a preferred technical solution, the telecommunications system centers mentioned in step S1 include at least two of the following: sales center, customer center, user center, channel center, order center, activation center, audit center, log center, integrated asset management platform, and tariff management platform. The extracted data includes at least one of configuration data, transaction data, log data, and complaint data, and the extraction frequency is daily.

[0015] To ensure a comprehensive, relevant, and timely data foundation for health assessment, this claim clarifies the core specifications for data collection. Regarding data sources, it selects key centers directly related to virtual goods operations within the telecommunications system, including sales centers, customer centers, order centers, and complaint centers. By combining at least two types of centers, it achieves coverage of key links in operational data, user data, and service data, avoiding the information bias caused by a single data source. In terms of data types, it focuses on four core data categories: configuration data reflecting product rule settings, transaction data reflecting actual operational status, log data recording operation trajectories, and complaint data reflecting user satisfaction. This comprehensively covers all data dimensions from "static rules to dynamic operations to user feedback," ensuring that the collected data highly aligns with health assessment requirements. Regarding collection frequency, it is set to daily collection, with the big data platform automatically extracting relevant data from the previous day at regular intervals, ensuring real-time synchronization of data with business development and providing a guarantee for subsequent real-time monitoring and timely early warning. The clearly defined data collection scope effectively avoids the redundant accumulation of invalid data, reduces unnecessary workload in the data processing process, and significantly improves overall analysis efficiency. The daily data collection frequency completely solves the pain point of data lag in traditional analysis, allowing market personnel to keep abreast of the latest developments in product operations and ensure the timeliness of strategy adjustments. Meanwhile, the multi-center, multi-type data combination approach provides comprehensive and multi-layered data support for the cost-level basic model, ensuring the reliability and comprehensiveness of subsequent analysis results from the source and avoiding evaluation biases caused by missing data.

[0016] In terms of data source selection, the operator chose three core departments: the sales center, the complaint center, and the order center. The sales center is responsible for providing operational data such as the distribution of data package listing channels and daily subscription volume; the complaint center focuses on user complaint data such as delayed data delivery and unfulfilled benefits; and the order center provides transaction-related data such as cancellation order details and subscriber profiles. The types of data collected cover configuration data (cost price, standard price, and discount rules for different channels of data packages), transaction data (daily subscription volume, cancellation volume, and actual user spending), and complaint data (total daily complaints and statistical classification of complaint reasons). Following established specifications, the big data platform automatically extracts all relevant data from the previous day at 2:00 AM daily. After cleaning and integration, preliminary analysis is completed before 9:00 AM that day. This efficient data collection and processing model allows marketing personnel to monitor the operational status of data packages in real time. On one particular day, a 30% surge in complaints compared to the previous day was detected, immediately triggering an alert mechanism. The operations team quickly investigated and found that a system malfunction caused the delayed data delivery. They promptly took corrective measures and apologized to users, effectively reducing the risk of user churn.

[0017] As a preferred technical solution, in step S1: The indicators of the basic business value indicator model include product code, product name, product category, product cost price, product standard price, commission expenditure rules, settlement rules, marketing strategy discount information, integrated product business allocation ratio, and listing channel information; The indicators in the business compliance analysis indicator model include the number of users who subscribed to atomic products, the number of users who canceled their atomic products, the fair value of the products, business revenue, business commission expenses, settlement expenses, allocated expenses, amount of outstanding fees, and amount of inactive accounts. The service analysis indicator model includes indicators such as user service activation time, user service subscription and usage time, number of service complaints, and service saturation.

[0018] The core of the tariff-level basic model lies in its comprehensive and detailed depiction of the operational status of telecom virtual goods through three types of subdivided indicator models. The business value basic indicator model focuses on the commercial attributes of the goods themselves. Based on fundamental information such as product code, name, and classification, combined with core commercial data such as cost price, standard price, commission payment rules, and settlement rules, it comprehensively quantifies the core commercial value and profit potential of the goods, providing a basis for judging whether the goods possess market competitiveness. The business compliance analysis indicator model focuses on operational compliance and revenue quality. Through dynamic user data such as the number of subscribers and unsubscribers for atomic goods, combined with financial and operational data such as business revenue, commission payment expenses, settlement fees, arrears, and inactive accounts, it accurately identifies compliance risks such as settlement inversion and double-payment, ensuring the standardization of the operational process. The service analysis indicator model focuses on user experience. Through process data such as user service activation time and service subscription usage time, combined with feedback data such as service complaint volume and service saturation, it intuitively reflects the service quality and user satisfaction of the goods, providing direction for optimizing service processes. The indicators of the three models complement each other and progress layer by layer, jointly forming a complete basic data system. This indicator system comprehensively covers three core dimensions: "product itself," "operational standards," and "user experience." It completely changes the problems of single indicators and one-sided perspectives in traditional evaluations, allowing health analysis to reach every key aspect of operations. All indicators use clear and quantifiable definitions, avoiding the difficulties in data collection and analysis caused by vague descriptions, ensuring the feasibility of data collection and the accuracy of analysis results. Furthermore, the indicators of the three models are highly compatible with the needs of building subsequent intermediate pricing models, providing precise support for indicator extraction and model generation, reducing the workload of secondary data processing, and improving the efficiency of the overall analysis process.

[0019] As a preferred technical solution, the metrics of the aggregation model in step S2 include business subscription volume, business growth volume, package upgrade volume, complaint rate, average monthly call duration, average monthly data traffic, and average revenue per user within a time period. The health model includes a business value diagnostic model, a business development compliance diagnostic model, a user service diagnostic model, and a health index model. The health index model is constructed based on the diagnostic results and corresponding weights of the first three.

[0020] First, the scattered single-indicator data in the tariff-level basic model are systematically categorized, refined, and integrated. Redundant information is eliminated, and core derived indicators closely related to product health are selected to form a converged model. These converged indicators include business growth data such as service subscription volume, service growth, and package upgrade volume within a time period; service quality data such as complaint rate; and core telecom industry operation indicators such as average monthly call duration (MOU), average monthly data usage (DOU), and average revenue per user (ARPU), transforming scattered data into a set of indicators with direct analytical significance. Based on this, pre-set logical judgment rules are injected into the converged model to construct three specialized diagnostic models: a business value diagnostic model focusing on product profitability potential and market competitiveness; a business development compliance diagnostic model emphasizing the identification of compliance risks in the operational process; and a user service diagnostic model focusing on evaluating service quality and user experience. Finally, by assigning corresponding weights based on the importance of the three specialized diagnostic models, a comprehensive health index model is constructed to achieve a dual analysis of the virtual product's operational health status: "specific problem identification + overall status assessment."

[0021] The convergence model achieves "redundancy removal and core extraction" of basic data, transforming scattered, single data points into highly condensed core indicators. This significantly reduces ineffective calculations in subsequent analysis processes and substantially improves overall analysis efficiency. The health model employs a two-tiered design of "specific diagnosis + comprehensive assessment." It can accurately pinpoint specific problems in the operational process, such as certain compliance risks or service shortcomings, while also providing an overall assessment of the product's operational health status. This meets the dual needs of marketing planners: "precisely optimizing specific problems + making a holistic judgment of the operational situation." Furthermore, the indicators and diagnostic logic in the model are aligned with the operational characteristics of the telecommunications industry, ensuring the industry adaptability and practical application value of the analysis results and avoiding the problem of general models being out of touch with industry realities.

[0022] As a preferred technical solution, the specific steps for determining the weights using the entropy method in step S3 include: The indicator data is standardized using the following formula: in, for The first evaluation object Original values ​​of each indicator; Calculate the weight of the j-th indicator. ; Calculate the entropy value of the j-th index. ,in and >0; Calculate the coefficient of difference for the j-th indicator. ; Calculate the weight of the j-th index ,in This represents the total number of indicators.

[0023] The core of entropy-based weight determination is to quantify the contribution of each indicator to the health assessment through objective data calculation, avoiding subjective bias from manual weighting. First, because different indicators have different units of measurement (e.g., business revenue is measured in "yuan," complaint rate in "%," and order growth rate in "%"), direct calculation would affect the accuracy of the results. Therefore, standardization of the indicator data is necessary. For positive indicators such as business revenue and ARPU (Average Revenue Per User), where "the larger the indicator value, the better," a forward standardization formula is used to eliminate the units of measurement. For negative indicators such as complaint rate, where "the smaller the indicator value, the better," a reverse standardization formula is used to unify the value range of all indicators to [0,1], ensuring that different types of indicators can directly participate in subsequent calculations. Next, the standardized value of each evaluation object on a certain indicator is calculated as a proportion of the sum of the standardized values ​​of all evaluation objects on that indicator; this is the indicator weight, reflecting the relative position of the evaluation object on that indicator. Then, the entropy value of each indicator is calculated based on its weight. The magnitude of the entropy value is related to the dispersion of the indicator data; the smaller the entropy value, the greater the difference between different evaluation objects, the stronger the distinguishing ability of the indicator in health assessment, and the higher the information utility value. Next, the difference coefficient of the indicator is calculated using "1-entropy value" to further quantify the indicator's distinguishing ability. Finally, the difference coefficients of all indicators are normalized to obtain the weight of each indicator. The larger the difference coefficient, the higher the weight, meaning its impact on health assessment is more significant. The application of the entropy method completely abandons the subjective assumptions of traditional manual weighting, determining weights entirely based on the characteristics of the indicator data itself, ensuring the scientific, fair, and objective nature of weight allocation, and making the health assessment results more convincing. Indicator standardization effectively solves the technical problem of not being able to directly compare and calculate indicators of different dimensions, ensuring the rationality and rigor of the entire weight calculation logic. The entire weight calculation process has clear steps and reproducible logic; each step is traceable and verifiable, significantly improving the credibility of the health assessment results and providing a solid foundation for subsequent comprehensive score calculation and decision-making recommendations.

[0024] A system for analyzing the health of telecommunications virtual goods operations based on big data includes a data acquisition module, a model building module, a weight calculation module, and an analysis output module, with each module communicating with the others in sequence. The data acquisition module is used to extract data from various centers of the telecommunications system and transmit it to the model building module; The model building module is used to establish a tariff-level basic model and a tariff intermediate model, wherein the tariff intermediate model is generated based on the data extraction of the tariff-level basic model; The weight calculation module is used to determine the weight of each indicator using the entropy method, and to calculate the comprehensive health score in combination with the indicator data. The analysis output module is used to generate and output an analysis report based on the comprehensive health score.

[0025] The data acquisition module, serving as the data entry point for the entire system, undertakes the core task of extracting raw data from various centers within the telecommunications system. Following pre-defined acquisition standards, it ensures the comprehensiveness, timeliness, and relevance of the data and transmits it to the model building module in real time. Upon receiving the data, the model building module first constructs a tariff-level basic model, performing structured storage and classification of the raw data. Then, based on the basic model data, it extracts, aggregates, and injects logic to build an intermediate tariff model, completing the transformation from raw data to an analytical model. The weight calculation module utilizes the built-in entropy calculation engine to standardize, calculate entropy, calculate difference coefficients, and solve for weights in the intermediate model's indicator data. Combining the weights and standardized indicator values, it calculates the comprehensive health score for each virtual product. The analysis output module transforms the comprehensive score into an analysis report containing performance across various dimensions, problem warnings, and targeted strategy suggestions. This report is presented through web clients, company dashboards, and other terminals, allowing marketing personnel to download and view it, forming a complete closed loop of "data input - data processing - result output."

[0026] The operator has officially deployed the "Virtual Goods Health Analysis System" and applied it to the daily operational monitoring of its five core virtual goods. After the system was launched, the data collection module automatically extracted data daily from multiple departments, including the sales center, user center, and complaint center, including product order volume, cancellation volume, consumption amount, and complaint details. The integrated raw data was then transmitted to the model building module in real time. Upon receiving the data, the model building module automatically completed the construction of a tariff-level basic model, covering three categories of indicators: business value, compliance analysis, and service analysis. Next, it refined the basic data to generate core aggregated indicators such as ARPU value, complaint rate, and order growth rate, and injected logical judgment rules to construct a specialized diagnostic model and a comprehensive health index model. The weight calculation module called the entropy method calculation engine to calculate the weights of 12 core indicators for the five goods, ultimately determining a weight of 0.22 for "business revenue," 0.35 for "compliance risk," and 0.28 for "service complaints." The weighted summation then yielded the comprehensive health score for each product. Based on the scores, the analysis output module automatically generates the "May 2024 Operational Health Report for 5 Virtual Products." The report not only lists the scores and rankings of each product but also details the prominent issues for each product, such as the excessively high cancellation rate of Product B, and provides targeted suggestions such as adjusting product benefits and optimizing marketing and promotion strategies. This report is available for download by marketing planners via a web client and is also displayed in real-time on the company's operational monitoring dashboard, allowing relevant personnel to intuitively grasp the dynamics of product operations.

[0027] As a preferred technical solution, the data acquisition module includes a log acquisition unit and a transaction data acquisition unit. The log acquisition unit is used to collect tariff configuration log data and tariff production log data, and the transaction data acquisition unit is used to collect business transaction data, billing data, accounting data, and remuneration settlement data from the production system.

[0028] The data acquisition module, through the collaborative efforts of the log acquisition unit and the transaction data acquisition unit, achieves accurate and comprehensive data collection across the entire virtual goods operation process. The log acquisition unit primarily focuses on "configuration-level" data, specifically collecting tariff configuration logs and tariff production logs. It meticulously records static data such as virtual goods rule settings, configuration changes, and listing channel adjustments. This data forms the foundation for characterizing the core attributes and operational rules of the goods. The transaction data acquisition unit, on the other hand, focuses on dynamic data at the "operational level," primarily collecting business transaction data, billing data, accounting data, and commission settlement data from the production system. This comprehensively reflects the actual operation of virtual goods in the market, including key information such as user order and cancellation behavior, actual charging amounts, total revenue, and channel commission expenditures. The two units perform their respective functions while working closely together. The log acquisition unit provides the "basic rule data" for the goods, while the transaction data acquisition unit provides the "actual operational data," jointly covering the entire data chain from "static rules to dynamic operations," ensuring that the collected data fully supports the subsequent construction of the basic model.

[0029] This dual-unit subdivision design makes data collection more targeted, avoiding confusion between different types of data and significantly improving the accuracy and efficiency of data collection. The simultaneous collection of log and transaction data completely changes the shortcomings of traditional data collection, which emphasizes operational data while neglecting configuration data. This ensures data integrity, allowing subsequent analysis to simultaneously consider both product rule design and actual operational effects, avoiding analytical bias caused by missing data. Furthermore, the collected log and transaction data can directly match the indicator requirements of the pricing-level basic model, reducing the workload of data cleaning and format conversion, saving time for subsequent model building and analysis, and further improving the overall system efficiency.

[0030] As a preferred technical solution, the model building module includes a basic model building unit and an intermediate model building unit: the basic model building unit is used to build a business value basic indicator model, a business compliance analysis indicator model, and a service analysis indicator model.

[0031] The intermediate model building unit is used to classify, refine, aggregate, and logically judge the basic model data to generate aggregated models and health models.

[0032] The model building module adopts a layered design with basic model building units and intermediate model building units, realizing the systematic and efficient construction of models. The basic model building unit, as the first layer, is mainly responsible for receiving the raw data transmitted by the data acquisition module. According to the three dimensions of business value, compliance analysis, and service analysis, it builds basic business value indicator models, business compliance analysis indicator models, and service analysis indicator models, respectively. It structures and classifies the scattered raw data, transforming the disordered data into ordered and directly callable basic indicator data, providing a solid data source support for subsequent model building. The intermediate model building unit, as the second layer, performs in-depth processing based on the basic model. First, it categorizes and refines the structured data in the basic model. For example, it combines "number of subscribers" and "number of unsubscribers" to extract "net increase in users," and combines "number of complaints" and "number of subscribers" to calculate "complaint rate." Then, it aggregates the refined indicators to form an aggregated model containing core indicators such as ARPU, business growth, and package upgrades. Finally, it injects preset logical judgment rules, such as "complaint rate > 1% is considered service abnormality" and "unsubscription rate > 8% is considered high compliance risk," to generate three types of specialized models: business value diagnosis model, business development compliance diagnosis model, and user service diagnosis model. It also combines weights to build a comprehensive health index model, completing the in-depth transformation from basic data to analytical models.

[0033] The layered design makes the model building logic clearer, with basic and intermediate units each performing their specific functions, facilitating subsequent system maintenance and feature iteration. For example, when adding new basic indicators, only the basic model building unit needs adjustment, without altering the core logic of the intermediate units, thus reducing system maintenance costs. The intermediate model building unit automates the transformation from "data → indicator → model," significantly reducing manual intervention. This not only improves the efficiency of model building but also avoids errors that may arise from manual operations. Furthermore, a clear correspondence is established between the basic and intermediate models; each intermediate model indicator can be traced back to the original data in the basic model, ensuring the traceability of the analytical logic and facilitating rapid root cause identification when problems are discovered.

[0034] As a preferred technical solution, the weight calculation module has a built-in entropy calculation engine, which supports index data standardization processing, entropy calculation, difference coefficient calculation and weight solution, and also supports a manual adjustment interface for weights.

[0035] The weight calculation module employs a dual design—a built-in entropy calculation engine and a manually adjustable interface—to achieve a weight allocation mechanism that prioritizes objectivity while allowing for flexibility. The entropy calculation engine, the module's core, automates the entire process, from standardizing indicator data and calculating entropy values ​​to calculating difference coefficients and weights. This process is entirely based on the inherent characteristics of the indicator data, requiring no manual intervention and ensuring the objectivity and scientific rigor of the weight allocation. Simultaneously, the module includes a manually adjustable interface. This addresses potential special needs in real-world business scenarios, such as a focus on service quality during specific marketing periods or enhanced compliance assessments following industry policy changes. It allows marketing planners to make appropriate fine-tuning adjustments based on actual business requirements, building upon the objective weights calculated using the entropy method. This design adheres to the objective laws of data while accommodating the flexibility of business scenarios, making the weight allocation more aligned with practical application needs.

[0036] The automated operation of the entropy calculation engine significantly reduces manual workload, avoids subjective biases caused by manual weighting, and ensures the fairness and accuracy of weight calculation results, providing a core guarantee for the reliability of the overall health score. The addition of a manual adjustment interface effectively addresses the pain point that "purely objective calculations cannot adapt to special business scenarios," improving the system's practicality and flexibility. It allows weight allocation to be adjusted promptly according to changes in business priorities, avoiding the problem of analysis results becoming disconnected from actual needs due to fixed weights. Furthermore, the module automatically records the complete logs of engine calculations and manual adjustment operations, ensuring that the entire weight allocation process is traceable and verifiable, improving the transparency and credibility of the analysis results.

[0037] As a preferred technical solution, the data acquisition module is interactively connected to an external platform, which includes at least three of the following: a unified billing platform, a comprehensive accounting platform, a commission / settlement configuration platform, a marketing OP platform, a unified product operation platform, and a complaint management platform, to achieve synchronous collection of configuration content, rule data, and operational data.

[0038] The data acquisition module establishes interactive connections with core external platforms of telecommunications operations, enabling synchronous collection and integration of data from multiple platforms, further expanding the breadth and depth of data sources. The selected external platforms are all core systems of key aspects of telecommunications operations, including converged billing platforms, integrated accounting platforms, commission / settlement configuration platforms, marketing OP platforms, unified product operation platforms, and complaint management platforms. These platforms cover key operational aspects such as billing, accounting, marketing, and service, providing rich and professional data. Through interface integration technology, the data acquisition module achieves real-time data synchronization with these external platforms. The collected data covers multiple dimensions, including configuration content (such as billing rules and package benefits configuration), rule data (such as commission calculation rules and settlement standards), and operational data (such as marketing activity participation and complaint handling progress). Simultaneously, the synchronous collection of data from multiple platforms also enables cross-validation of data. For example, billing data from the converged billing platform and revenue data from the integrated accounting platform can be cross-checked, ensuring the consistency and accuracy of the collected data and providing high-quality data support for subsequent analysis.

[0039] The data source has been expanded from the internal center of the telecommunications system to core external platforms, completely solving the problem of "incomplete data from a single center." This allows the collected data to cover the entire operational process of "billing, accounting, marketing, and service," significantly improving data completeness and depth, and enabling multi-dimensional and comprehensive health analysis. Synchronous collection and cross-validation of data from multiple platforms effectively avoids errors or omissions that may exist with data from a single platform, ensuring data consistency and accuracy, and improving the reliability of health assessment results from the source. Furthermore, the specialized data from external platforms provides more detailed support for analysis. For example, the collection of complaint reason classification data from the complaint management platform allows the operations team to accurately pinpoint service shortcomings, further enhancing the practical value of the analysis reports.

[0040] In the description of this specification, references to terms such as "one embodiment," "some embodiments," "example," "specific example," or "some examples," etc., indicate that a specific feature, structure, material, or characteristic described in connection with that embodiment or example is included in at least one embodiment or example of the invention. In this specification, the illustrative expressions of the above terms do not necessarily refer to the same embodiment or example. Furthermore, the specific features, structures, materials, or characteristics described may be combined in any suitable manner in one or more embodiments or examples.

[0041] The foregoing has shown and described the basic principles, main features, and advantages of the present invention. Those skilled in the art should understand that the present invention is not limited to the above embodiments. The embodiments and descriptions in the specification are merely illustrative of the principles of the invention. Various changes and modifications can be made to the invention without departing from its spirit and scope, and all such changes and modifications fall within the scope of protection claimed by the present invention. The scope of protection of the present invention is defined by the appended claims and their equivalents.

Claims

1. A method for analyzing the operational health of telecommunications virtual goods based on big data, characterized in that, Includes the following steps: S1. Establish a tariff-level basic model and extract data from various centers of the telecommunications system on a regular basis through a big data platform. The tariff-level basic model includes a business value basic indicator model, a business compliance analysis indicator model, and a service analysis indicator model. S2. Establish a pricing intermediate model. The big data platform classifies and refines the data from the basic model in step S1, aggregates it into a specified set of indicators, injects logical judgment, and generates an aggregated model and a health model. S3. Use the entropy method to determine the weight of each indicator in step S2, and calculate the comprehensive score of the health of telecommunications virtual goods operation based on the weights. S4. Generate an analysis report based on the comprehensive health score to provide data support for market planning and decision-making.

2. The method for analyzing the operational health of telecommunications virtual goods based on big data according to claim 1, characterized in that, The telecommunications system centers mentioned in step S1 include at least two of the following: sales center, customer center, user center, channel center, order center, activation center, audit center, log center, integrated asset management platform, and tariff management platform. The extracted data includes at least one of the following: configuration data, transaction data, log data, and complaint data, and the extraction frequency is daily.

3. The method for analyzing the operational health of telecommunications virtual goods based on big data according to claim 2, characterized in that, In step S1: The indicators of the basic business value indicator model include product code, product name, product category, product cost price, product standard price, commission expenditure rules, settlement rules, marketing strategy discount information, integrated product business allocation ratio, and listing channel information; The indicators in the business compliance analysis indicator model include the number of users who subscribed to atomic products, the number of users who canceled their atomic products, the fair value of the products, business revenue, business commission expenses, settlement expenses, allocated expenses, amount of outstanding fees, and amount of inactive accounts. The service analysis indicator model includes indicators such as user service activation time, user service subscription and usage time, number of service complaints, and service saturation.

4. The method for analyzing the operational health of telecommunications virtual goods based on big data according to claim 3, characterized in that, The metrics of the convergence model mentioned in step S2 include business subscription volume, business growth volume, package upgrade volume, complaint rate, average monthly call duration, average monthly data traffic, and average revenue per user within a time period. The health model includes a business value diagnostic model, a business development compliance diagnostic model, a user service diagnostic model, and a health index model. The health index model is constructed based on the diagnostic results and corresponding weights of the first three.

5. The method for analyzing the operational health of telecommunications virtual goods based on big data according to claim 4, characterized in that, The specific steps for determining the weights using the entropy method in step S3 include: The indicator data is standardized using the following formula: in, for The first evaluation object Original values ​​of each indicator; Calculate the weight of the j-th indicator. ; Calculate the entropy value of the j-th index. ,in and >0; Calculate the coefficient of difference for the j-th indicator. ; Calculate the weight of the j-th index ,in This represents the total number of indicators.

6. A system for analyzing the operational health of telecommunications virtual goods based on big data, applied to the method for analyzing the operational health of telecommunications virtual goods based on big data as described in any one of claims 1-5, characterized in that, The system includes: The data acquisition module is used to extract data from various centers of the telecommunications system and transmit it to the model building module; The model building module is used to establish a tariff-level basic model and a tariff intermediate model, wherein the tariff intermediate model is generated based on the data extraction of the tariff-level basic model; The weight calculation module is used to determine the weight of each indicator using the entropy method, and to calculate the comprehensive health score in combination with the indicator data. The analysis output module is used to generate and output an analysis report based on the comprehensive health score.

7. The system for analyzing the operational health of telecommunications virtual goods based on big data according to claim 6, characterized in that, The data acquisition module includes a log acquisition unit and a transaction data acquisition unit. The log acquisition unit is used to collect tariff configuration log data and tariff production log data. The transaction data acquisition unit is used to collect business transaction data, billing data, accounting data, and remuneration settlement data from the production system.

8. The system for analyzing the operational health of telecommunications virtual goods based on big data according to claim 6, characterized in that, The model building module includes a basic model building unit and an intermediate model building unit: The basic model building unit is used to build a basic business value indicator model, a business compliance analysis indicator model, and a service analysis indicator model. The intermediate model building unit is used to classify, refine, aggregate, and logically judge the basic model data to generate aggregated models and health models.

9. A system for analyzing the operational health of telecommunications virtual goods based on big data, as described in claim 6, is characterized in that... The weight calculation module has a built-in entropy calculation engine, which supports standardization of indicator data, entropy calculation, difference coefficient calculation and weight solution, and also supports a manual adjustment interface for weights.

10. A system for analyzing the operational health of telecommunications virtual goods based on big data, as described in claim 6, characterized in that... The data acquisition module interacts with external platforms, which include at least three of the following: a unified billing platform, a comprehensive accounting platform, a commission / settlement configuration platform, a marketing OP platform, a unified product operation platform, and a complaint management platform, to achieve synchronous collection of configuration content, rule data, and operational data.

Images