Merchant business capability prediction method based on scene perception self-adaption

Abstract

The invention relates to a merchant business capability prediction method and device based on scene perception self-adaption, computer equipment, a computer readable storage medium and a computer program product, and can be applied to the technical field of computers. The method comprises the following steps: constructing a commercial tenant business capability prediction index system, and determining basic weight information of each prediction index in the commercial tenant business capability prediction index system; acquiring historical weight information of each prediction index and scene information of a current service scene; determining current dynamic weight information of each prediction index according to the basic weight information, the historical weight information and the scene information; and in response to a business capability prediction request for the current merchant, performing business capability prediction on the merchant information of the current merchant through the business capability prediction model according to the merchant business capability prediction index system and the current dynamic weight information of each prediction index to obtain a business capability prediction result of the current merchant. By adopting the method, the accuracy of merchant business capability prediction can be improved.

Description

Technical Field

[0001] The present application relates to the field of computer technology, and particularly to a method, device, computer device, computer-readable storage medium, and computer program product for predicting merchant business capabilities based on scenario-aware adaptation. Background Art

[0002] With the development of big data technology, business intelligence systems are playing an increasingly important role in enterprise decision-making. The merchant evaluation system based on business intelligence systems has gradually become an important tool for judging the value and potential of merchants. Therefore, how to accurately predict merchant business capabilities has become an important research direction.

[0003] Traditional technologies usually predict merchant business capabilities by manually setting fixed weight indicators; however, it is difficult to adapt to the differentiated requirements of different business scenarios by this method, resulting in low accuracy of merchant business capability prediction. Summary of the Invention

[0004] Based on this, in view of the above technical problems, it is necessary to provide a method, device, computer device, computer-readable storage medium, and computer program product for predicting merchant business capabilities based on scenario-aware adaptation, which can improve the accuracy of merchant business capability prediction.

[0005] In a first aspect, the present application provides a method for predicting merchant business capabilities based on scenario-aware adaptation. The method includes:

[0006] Construct a merchant business capability prediction index system, and determine the basic weight information of each prediction index in the merchant business capability prediction index system;

[0007] Obtain the historical weight information of each prediction index and the scenario information of the current business scenario;

[0008] According to the basic weight information, the historical weight information, and the scenario information, determine the current dynamic weight information of each prediction index;

[0009] In response to a business capability prediction request for a current merchant, through a business capability prediction model, according to the merchant business capability prediction index system and the current dynamic weight information of each prediction index, predict the business capabilities of the merchant information of the current merchant to obtain a business capability prediction result of the current merchant.

[0010] In one of the embodiments, the determining the current dynamic weight information of each prediction index according to the basic weight information, the historical weight information, and the scenario information includes:

[0011] Determine a recursion term, a scenario perception term, and a weight regression term corresponding to the current dynamic weight information according to the base weight information, the historical weight information, and the scenario information;

[0012] Determine the current dynamic weight information according to the recursion term, the scenario perception term, and the weight regression term.

[0013] In one embodiment, the determining a recursion term, a scenario perception term, and a weight regression term corresponding to the current dynamic weight information according to the base weight information, the historical weight information, and the scenario information includes:

[0014] Determine the recursion term according to the historical weight information;

[0015] Determine the scenario perception term according to the base weight information, the historical weight information, and the scenario information;

[0016] Determine the weight regression term according to the base weight information and the historical weight information.

[0017] In one embodiment, after, in response to a business capability prediction request for a current merchant, a business capability prediction result of the current merchant is obtained by a business capability prediction model according to the merchant business capability prediction index system and the current dynamic weight information of each prediction index, the method further includes:

[0018] Determine the business level information of the current merchant according to the business capability prediction result;

[0019] Send the business level information to an e-commerce platform; the e-commerce platform is configured to perform business recommendation and risk control on the current merchant according to the business level information.

[0020] In one embodiment, the method further includes:

[0021] Obtain the structured data, unstructured data, and time series data of the current merchant;

[0022] Perform fusion processing on the structured data, the unstructured data, and the time series data to obtain the merchant information of the current merchant.

[0023] In one embodiment, the obtaining the structured data, unstructured data, and time series data of the current merchant includes:

[0024] Obtain the order data, user evaluation data, after-sales service data, and penalty record data of the current merchant as the structured data;

[0025] Obtain the customer service conversation text data of the current merchant as the unstructured data;

[0026] Obtain the logistics data of the current merchant as the time-series data.

[0027] In a second aspect, the present application also provides a merchant business capability prediction device based on scenario-aware adaptation. The device includes:

[0028] A system construction module for constructing a merchant business capability prediction index system and determining the basic weight information of each prediction index in the merchant business capability prediction index system;

[0029] An information acquisition module for acquiring the historical weight information of each prediction index and the scenario information of the current business scenario;

[0030] An information determination module for determining the current dynamic weight information of each prediction index according to the basic weight information, the historical weight information, and the scenario information;

[0031] An information prediction module for, in response to a business capability prediction request for the current merchant, predicting the business capability of the merchant information of the current merchant through a business capability prediction model according to the merchant business capability prediction index system and the current dynamic weight information of each prediction index, and obtaining the business capability prediction result of the current merchant.

[0032] In a third aspect, the present application also provides a computer device. The computer device includes a memory and a processor. The memory stores a computer program, and when the processor executes the computer program, the following steps are implemented:

[0033] Construct a merchant business capability prediction index system and determine the basic weight information of each prediction index in the merchant business capability prediction index system;

[0034] Obtain the historical weight information of each prediction index and the scenario information of the current business scenario;

[0035] Determine the current dynamic weight information of each prediction index according to the basic weight information, the historical weight information, and the scenario information;

[0036] In response to a business capability prediction request for the current merchant, predict the business capability of the merchant information of the current merchant through a business capability prediction model according to the merchant business capability prediction index system and the current dynamic weight information of each prediction index, and obtain the business capability prediction result of the current merchant.

[0037] Fourthly, the present application also provides a computer-readable storage medium. The computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the following steps are implemented:

[0038] Construct a merchant business ability prediction index system and determine the basic weight information of each prediction index in the merchant business ability prediction index system;

[0039] Obtain the historical weight information of each prediction index and the scenario information of the current business scenario;

[0040] According to the basic weight information, the historical weight information and the scenario information, determine the current dynamic weight information of each prediction index;

[0041] In response to a business ability prediction request for the current merchant, through a business ability prediction model, according to the merchant business ability prediction index system and the current dynamic weight information of each prediction index, perform a business ability prediction on the merchant information of the current merchant to obtain a business ability prediction result of the current merchant.

[0042] Fifthly, the present application also provides a computer program product. The computer program product includes a computer program, and when the computer program is executed by a processor, the following steps are implemented:

[0043] Construct a merchant business ability prediction index system and determine the basic weight information of each prediction index in the merchant business ability prediction index system;

[0044] Obtain the historical weight information of each prediction index and the scenario information of the current business scenario;

[0045] According to the basic weight information, the historical weight information and the scenario information, determine the current dynamic weight information of each prediction index;

[0046] In response to a business ability prediction request for the current merchant, through a business ability prediction model, according to the merchant business ability prediction index system and the current dynamic weight information of each prediction index, perform a business ability prediction on the merchant information of the current merchant to obtain a business ability prediction result of the current merchant.

[0047] The above-mentioned merchant business ability prediction method, device, computer equipment, computer-readable storage medium and computer program product based on scenario-aware adaptation construct a merchant business ability prediction index system and determine the basic weight information of each prediction index in the merchant business ability prediction index system; obtain the historical weight information of each prediction index and the scenario information of the current business scenario; determine the current dynamic weight information of each prediction index according to the basic weight information, the historical weight information and the scenario information; in response to a business ability prediction request for a current merchant, through a business ability prediction model, according to the merchant business ability prediction index system and the current dynamic weight information of each prediction index, perform a business ability prediction on the merchant information of the current merchant to obtain a business ability prediction result of the current merchant. This solution constructs a merchant business ability prediction index system and realizes dynamic adjustment of index weights, which is beneficial to solving the problem that traditional static weight settings cannot adapt to changing business scenarios; this solution uses a three-dimensional fusion of basic weight, historical weight and scenario information for dynamic weight calculation, enabling the index weights to automatically respond to the characteristic changes of different business scenarios; this dynamic adaptive weight configuration method can more accurately reflect the true business ability of merchants in different scenarios and effectively improve the accuracy of merchant business ability prediction. BRIEF DESCRIPTION OF THE DRAWINGS

[0048] In order to more clearly illustrate the technical solutions in the embodiments of the present application or related technologies, the following will briefly introduce the drawings required for use in the description of the embodiments or related technologies. Obviously, the drawings in the following description are only some embodiments of the present application. For those of ordinary skill in the art, other drawings can be obtained based on these drawings without creative efforts.

[0049] Figure 1 It is a schematic flowchart of a merchant business ability prediction method based on scenario-aware adaptation in an embodiment;

[0050] Figure 2 It is a schematic overall architecture diagram of a merchant business ability prediction method based on scenario-aware adaptation in an embodiment;

[0051] Figure 3 It is a schematic flowchart of data processing steps in an embodiment;

[0052] Figure 4 It is a structural block diagram of a merchant business ability prediction device based on scenario-aware adaptation in an embodiment;

[0053] Figure 5 It is an internal structure diagram of a computer device in an embodiment. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0054] To make the objectives, technical solutions and advantages of this application more clear and understandable, the following further elaborates on this application in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely used to explain this application and are not used to limit this application.

[0055] It should be noted that the user information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data for analysis, stored data, displayed data, etc.) involved in this application are all information and data authorized by the user or fully authorized by all parties, and the collection, use, and processing of relevant data need to comply with relevant regulations.

[0056] In an exemplary embodiment, as Figure 1 shown, a merchant business ability prediction method based on scenario-aware adaptation is provided. In this embodiment, an example is given where this method is applied to a terminal; it can be understood that this method can also be applied to a server, and can also be applied to a system including a terminal and a server, and is implemented through the interaction between the terminal and the server. Among them, the terminal can be but is not limited to various personal computers, laptop computers, smart phones, tablet computers, etc.; the server can be an independent physical server, or a server cluster or distributed system composed of multiple physical servers, or a cloud server providing cloud computing services. In this embodiment, the method includes the following steps:

[0057] Step S101, construct a merchant business ability prediction index system, and determine the basic weight information of each prediction index in the merchant business ability prediction index system.

[0058] Step S102, obtain the historical weight information of each prediction index and the scenario information of the current business scenario.

[0059] Step S103, determine the current dynamic weight information of each prediction index according to the basic weight information, historical weight information, and scenario information.

[0060] Step S104, in response to a business ability prediction request for the current merchant, through a business ability prediction model, according to the merchant business ability prediction index system and the current dynamic weight information of each prediction index, predict the business ability of the merchant information of the current merchant, and obtain the business ability prediction result of the current merchant.

[0061] Among them, the merchant business ability prediction index system can be a multi-dimensional index set for evaluating and predicting the merchant business ability. For example, the merchant business ability prediction index system can include transaction scale (transaction amount, number of transaction customers, number of orders), merchant activity (number of times the merchant logs in to the mall), transaction efficiency (order confirmation speed, order delivery speed, order delivery speed, order collection speed), etc.

[0062] Among them, the basic weight information can be the initial set weight values of each indicator in the prediction index system. For example, the basic weight information can be the preset weight values such as the basic weight of the transaction amount indicator being 40, the basic weight of the order number indicator being 5, and the basic weight of the number of times the merchant logs in to the mall being 10, etc.

[0063] Among them, the historical weight information can be the actual application weight records of each prediction indicator at previous time points. For example, the historical weight information can be expressed as the indicator weight value at the previous moment, which is used to ensure the continuity of weight changes.

[0064] Among them, the scenario information can be a data set describing the characteristics of the current business environment. For example, the scenario information can be indicators reflecting the intensity of the current business scenario such as the scenario coefficient S(t)=0.8 during the promotion period and the scenario coefficient S(t)=0 during the daily period.

[0065] Among them, the current dynamic weight information can be the latest indicator weight calculated based on the basic weight, historical weight, and scenario information.

[0066] Among them, the business ability prediction model can be a model used to predict the business ability of merchants based on each indicator data and dynamic weights. For example, the business ability prediction model can be a pre-trained neural network model or algorithm model, which is used to identify the performance data of merchants on each indicator according to merchant information, and then combine the performance data with the current dynamic weights to predict the business ability prediction result of the current merchant.

[0067] Among them, the merchant information can be an information set reflecting the basic situation and business data of the merchant. For example, the merchant information can include structured data (order data, user evaluation data, after-sales service data, penalty record data), unstructured data (customer service conversation text), and time series data (logistics data), etc.

[0068] Among them, the business ability prediction result can be a quantitative evaluation result of the business ability of merchants in various aspects. For example, the business ability prediction result can include the predicted score (0-100 points) of the merchant's real-time credit ability evaluation or the merchant evaluation level in five grades of S, A, B, C, and D.

[0069] Optionally, the terminal conducts a comprehensive ability assessment of merchants by constructing a merchant business ability prediction index system. Specifically, the terminal first accesses the data sources of multiple shopping malls, including structured data (such as order data, user evaluation data, after-sales service data, penalty record data), unstructured data (such as customer service conversation text), and time-series data (such as logistics data), establishes a unified index caliber, and constructs a merchant business ability prediction index system. This index system covers dimensions such as transaction scale (including transaction amount, number of transaction customers, number of orders), merchant activity (including the number of times the merchant logs in to the shopping mall), transaction efficiency (including order confirmation speed, order delivery speed, order successful delivery speed, order collection speed), etc. Subsequently, the terminal sets the basic weight information for each prediction index in the merchant business ability prediction index system. For example, the basic weight of the transaction amount index is 40, the basic weight of the order number index is 5, the basic weight of the merchant login mall times index is 10, the basic weight of the order confirmation speed index is 10, the basic weight of the order delivery speed index is 10, the basic weight of the order successful delivery speed index is 10, and the basic weight of the order collection speed index is 10. The terminal obtains the historical weight information of each prediction index and the scenario information of the current business scenario based on the existing data, where the historical weight information is the weight value of each index at the previous moment, and the scenario information includes the scenario coefficient during the promotion period (such as S(t)=0.8) or the scenario coefficient during the normal period (such as S(t)=0), and then calculates the current dynamic weight information of each prediction index through a recursive algorithm; when receiving a business ability prediction request for the current merchant, in response to the business ability prediction request for the current merchant, the merchant business ability prediction index system, the current dynamic weight information of each prediction index, and the merchant information of the current merchant are input into the business ability prediction model, and through the business ability prediction model, based on the merchant business ability prediction index system and the current dynamic weight information of each prediction index, the merchant information of the current merchant is predicted for business ability, and the business ability prediction result of the current merchant is obtained.

[0070] In the above merchant business ability prediction method based on scenario-aware adaptation, a merchant business ability prediction index system is constructed, and the basic weight information of each prediction index in the merchant business ability prediction index system is determined; the historical weight information of each prediction index and the scenario information of the current business scenario are obtained; according to the basic weight information, historical weight information and scenario information, the current dynamic weight information of each prediction index is determined; in response to a business ability prediction request for the current merchant, through a business ability prediction model, according to the merchant business ability prediction index system and the current dynamic weight information of each prediction index, the merchant information of the current merchant is predicted for business ability, and a business ability prediction result of the current merchant is obtained. This solution is beneficial to solving the problem that the traditional static weight setting cannot adapt to the changing business scenarios by constructing a merchant business ability prediction index system and realizing the dynamic adjustment of index weights; this solution uses a three-dimensional fusion of basic weight, historical weight and scenario information for dynamic weight calculation, enabling the index weights to automatically respond to the characteristic changes of different business scenarios; this dynamic adaptive weight configuration method can more accurately reflect the true business ability of merchants in different scenarios and effectively improve the accuracy of merchant business ability prediction.

[0071] In an exemplary embodiment, according to the basic weight information, historical weight information and scenario information, determining the current dynamic weight information of each prediction index specifically includes the following content: according to the basic weight information, historical weight information and scenario information, determining a recursion term, a scenario perception term and a weight regression term corresponding to the current dynamic weight information; according to the recursion term, the scenario perception term and the weight regression term, determining the current dynamic weight information.

[0072] Among them, the recursion term can be a computational component that ensures the smooth transition of the current dynamic weight information. For example, the recursion term can be W i (t - 1), which represents inheriting the weight value of the previous moment. For example, W i (t - 1) is the historical weight at time t−1, ensuring the continuity of weight changes.

[0073] Among them, the scenario perception term can be a computational component that dynamically adjusts the weight according to the characteristics of the current business scenario. For example, the scenario perception term can be α*S(t)*((W basei -W i (t - 1)) / W basei ), where S(t) reflects the intensity of the current business scenario (such as the intensity of a promotional activity), W basei represents the basic weight of index i (such as the initial value = 40), and α is the scenario sensitivity coefficient used to control the intensity of the scenario impact, and the overall expression shows the characteristic that the stronger the scenario and the farther the historical weight deviates from the basic value, the greater the adjustment amplitude.

[0074] Among them, the weight regression term can be a computational component that prevents the weight from deviating from the set base value for a long time. For example, the weight regression term can be β * (W basei - W i (t - 1)), where β is the weight regression coefficient used to control the regression speed of the weight towards the base value. This term pulls the weight back to the base value to avoid system instability caused by long-term deviation.

[0075] Optionally, when determining the current dynamic weight information of each prediction indicator, the terminal first constructs a weight calculation model based on the stored data. Specifically, the terminal calculates three components of the current dynamic weight information: the recursion term, the scenario awareness term, and the weight regression term. For the recursion term, the terminal directly obtains the historical weight information W i (t - 1) of each prediction indicator as the value of the recursion term to ensure the continuity of weight change; for the scenario awareness term, the terminal calculates the scenario awareness term α * S(t) * ((W basei - W i (t - 1)) / W basei ), according to the scenario information S(t) of the current business scenario (such as S(t)=0.8 during the promotion period and S(t)=0 during the normal period), the scenario sensitivity coefficient α, the base weight information W i of each prediction indicator, and the historical weight information W basei of each prediction indicator. This term can achieve dynamic adjustment of the weight, making the weight change with the business scenario; for the weight regression term, the terminal calculates the weight regression term β * (W basei - W i (t - 1)) according to the weight regression coefficient β, the base weight information W basei of each prediction indicator, and the historical weight information W i of each prediction indicator. This term ensures that the weight does not deviate from the base set value for a long time and guarantees system stability. The terminal adds the recursion term, the scenario awareness term, and the weight regression term to obtain the current dynamic weight information.

[0076] The technical solution provided in this embodiment is beneficial to realizing multi-dimensional control of weight calculation by decomposing the current dynamic weight information into three components: the recursion term, the scenario awareness term, and the weight regression term; the recursion term inherits the historical weight to ensure the continuity and smooth transition of weight change; the scenario awareness term dynamically adjusts the weight according to the business scenario intensity and the historical weight deviation to respond to different scenario requirements; the weight regression term prevents the weight from deviating from the base value for a long time and maintains system stability; the weight adaptive algorithm formed by the combination of these three terms can not only quickly respond to changes in the business scenario (such as the promotion period and the normal period), but also avoid weight out-of-control caused by short-term fluctuations, thus being beneficial to improving the accuracy of merchant business ability prediction.

[0077] In an exemplary embodiment, according to the basic weight information, historical weight information, and scenario information, the recursive term, scenario perception term, and weight regression term corresponding to the current dynamic weight information are determined. Specifically, it includes the following: Determine the recursive term according to the historical weight information; determine the scenario perception term according to the basic weight information, historical weight information, and scenario information; determine the weight regression term according to the basic weight information and historical weight information.

[0078] Optionally, when determining the current dynamic weight information, the terminal calculates three different weight adjustment components respectively. First, the terminal determines the recursive term according to the historical weight information. Specifically, directly obtain the historical weight W i at time t - 1 of each prediction index as the recursive term. This method ensures the continuity of weight change and avoids sudden changes in weight. The terminal determines the scenario perception term according to the basic weight information, historical weight information, and scenario information. Specifically, multiply the scenario sensitivity coefficient α by the scenario coefficient S(t) at time t, and then multiply by the weight deviation degree (W basei -W i at time t - 1) / W basei to form the complete scenario perception term α·S(t)·((W basei -W i at time t - 1) / W basei ). In this calculation process, the scenario coefficient S(t) reflects the current business scenario intensity. For example, during the promotion period, S(t)=0.8, and during the daily period, S(t)=0. The weight deviation degree represents the relative deviation between the historical weight and the basic weight. The product of the two makes the adjustment amplitude larger when the scenario is stronger and the historical weight deviates further from the basic value. The terminal determines the weight regression term according to the basic weight information and historical weight information. Specifically, calculate the product of the weight regression coefficient β and the difference between the basic weight and the historical weight (W basei -W i at time t - 1) to form the weight regression term β·(W basei -W i at time t - 1). The role of this term is to pull the weight back to the basic value and avoid system instability caused by long-term deviation.

[0079] The technical solution provided in this embodiment decomposes the calculation of the current dynamic weight information into three independent but collaborative components, which is conducive to achieving accurate control and multi-dimensional balance of weight adjustment; the recursive term inherits the historical weight to ensure the continuity of weight change; the scenario perception term enables the weight to automatically respond to business scenario changes through the comprehensive calculation of the basic weight, historical weight, and scenario information; the weight regression term maintains the long-term stability of the system according to the deviation between the basic weight and the historical weight; it avoids the limitations of a single weight calculation logic, thereby facilitating the improvement of the accuracy and adaptability of merchant business ability prediction.

[0080] In an exemplary embodiment, after, in response to a business capability prediction request for the current merchant, performing a business capability prediction on the merchant information of the current merchant through a business capability prediction model according to the merchant business capability prediction index system and the current dynamic weight information of each prediction index to obtain a business capability prediction result of the current merchant, the following further includes: determining the business level information of the current merchant according to the business capability prediction result; sending the business level information to an e-commerce platform; and the e-commerce platform is used to perform business recommendation and risk control on the current merchant according to the business level information.

[0081] Among them, the business level information may be a merchant capability level identifier divided according to the business capability prediction result. For example, the business level information may be a real-time evaluation level given to the merchant according to a 100-point score interval corresponding to five levels of S, A, B, C, and D.

[0082] Among them, the e-commerce platform may be a platform system that receives the business level information and conducts merchant management activities based on this. For example, the e-commerce platform may be multiple mall platforms such as E-commerce Platform 1, E-commerce Platform 2, and E-commerce Platform 3.

[0083] Among them, the business recommendation may be business development suggestions and opportunities provided by the e-commerce platform to the merchant based on the merchant business level information. For example, the business recommendation may be operation strategies such as providing more traffic tilt and promotion activity participation qualifications to high-level merchants.

[0084] Among them, the risk control may be risk prevention measures taken by the e-commerce platform on the merchant based on the merchant business level information. For example, the risk control may be restricting specific business permissions for low-level merchants, etc., to achieve the maintenance of a healthy and prosperous platform ecosystem.

[0085] Optionally, after the terminal completes the business capability prediction of the current merchant, a series of subsequent business processes will be carried out. First, the terminal determines the business level information of the current merchant according to the business capability prediction result of the current merchant. Specifically, the terminal divides the business capability prediction result (a quantitative score with a value range of 0-100 points) into different level intervals and divides the current merchant according to the five-level standard of SABCD. For example, the terminal may divide 90-100 points into level S, 80-89 points into level A, 70-79 points into level B, 60-69 points into level C, and below 60 points into level D. Then, the terminal will send the determined business level information to the e-commerce platform through the system interface, including multiple mall platforms such as Sub-e-commerce Platform 1, Sub-e-commerce Platform 2, and Sub-e-commerce Platform 3. This unified merchant business level evaluation result solves the problem of data islands between different e-commerce platforms and realizes the unified evaluation of merchant capabilities across platforms. The e-commerce platform can perform business recommendation and risk control on the current merchant according to the business level information.

[0086] The technical solution provided in this embodiment is beneficial to the standardized application and cross-platform sharing of merchant evaluation results by converting the business capability prediction results into merchant business level information and sending it to the e-commerce platform. This method breaks the problem of data islands between different e-commerce platforms, unifies the merchant capability evaluation criteria, enables the e-commerce platform to implement differentiated business recommendations and risk control measures based on the unified business level information; thus, it is conducive to promoting the effective operation of the merchant survival-of-the-fittest mechanism and maintaining the healthy and prosperous development of the platform ecosystem.

[0087] In an exemplary embodiment, it further includes the following: obtaining the structured data, unstructured data, and time-series data of the current merchant; performing fusion processing on the structured data, unstructured data, and time-series data to obtain the merchant information of the current merchant.

[0088] Among them, the structured data can be merchant-related data with a fixed format and fields. For example, the structured data can be merchant information stored in a predefined pattern such as order data, user evaluation data, after-sales service data, penalty record data, etc.

[0089] Among them, the unstructured data can be merchant-related data without a fixed format. For example, the unstructured data can be customer service conversation texts, and the text content of indicators such as "service attitude" is extracted through NLP (Natural Language Processing) technology.

[0090] Among them, the time-series data can be merchant-related data that changes over time and is recorded in chronological order. For example, the time-series data can be logistics data, which reflects the time-series information in the merchant's order processing and delivery process.

[0091] Among them, the fusion processing can be a process of integrating and unifying different types of data.

[0092] Optionally, before starting the merchant business capability prediction, the terminal first needs to obtain the comprehensive data resources of the current merchant. The terminal accesses multiple mall data sources through the middleware API (Application Programming Interface) gateway and obtains three types of core data of the current merchant: structured data, unstructured data, and time-series data. Specifically, the structured data obtained by the terminal includes the order data, user evaluation data, after-sales service data, penalty record data, etc. of the current merchant, and these data have fixed formats and fields; the unstructured data obtained by the terminal includes the customer service conversation texts of the current merchant, and the text content of indicators such as service attitude is extracted from them through NLP (Natural Language Processing) technology; the time-series data obtained by the terminal includes the logistics data of the current merchant, which reflects the time-series information in the merchant's order processing and delivery process. Subsequently, the terminal performs fusion processing on the obtained different types of data to obtain the merchant information of the current merchant.

[0093] The technical solution provided in this embodiment is conducive to building a comprehensive and multi-dimensional merchant information system by simultaneously acquiring three types of merchant data of different natures, namely structured data, unstructured data and time series data; it breaks the problem of data silos, realizes unified processing of multi-platform and multi-data types, and maintains consistency in indicator caliber; structured data provides standardized merchant operating indicators, unstructured data such as customer service texts capture difficult-to-quantify service quality characteristics through NLP technology, and time series data reflects changes in the dynamic operational capabilities of merchants; this is conducive to forming a more comprehensive, more accurate and more timely merchant capability evaluation basis, avoiding one-sided evaluation caused by a single data source, and providing multi-dimensional and accurate data support for subsequent merchant business capability predictions.

[0094] In an exemplary embodiment, the structured data, unstructured data and time series data of the current merchant are obtained, specifically including the following contents: obtaining the order data, user evaluation data, after-sales service data and penalty record data of the current merchant as structured data; obtaining the customer service conversation text data of the current merchant as unstructured data; obtaining the logistics data of the current merchant as time series data.

[0095] Among them, order data can be data that records merchant transaction behavior and order processing status. For example, order data can be transaction information such as the total amount of merchant transaction orders, order quantity, and single order amount in the past 30 days.

[0096] Among them, user evaluation data can be data records of consumers' evaluation of merchant services and product quality. For example, user evaluation data can be customer rating data, evaluation content, praise rate and other information reflecting satisfaction with the merchant.

[0097] Among them, after-sales service data can be data information that records the merchant's after-sales processing capabilities. For example, after-sales service data can be the merchant's after-sales response time, return and exchange processing time, after-sales problem resolution rate, and other data that reflects the quality of after-sales service.

[0098] Among them, the penalty record data may be the platform's penalty record for merchants' violations, for example, the penalty record data may be information such as the number of complaints and penalties against the merchant, the reasons for the penalties, penalty points deducted, etc.

[0099] Among them, the customer service conversation text data can be the text of the conversation content between the merchant and the customer. For example, the customer service conversation text data can be the text content containing indicators such as service attitude extracted by NLP technology, which is used to analyze the merchant's service quality.

[0100] Among them, the logistics data can be time - series data recording the entire process of commodity distribution. For example, the logistics data can be time - series information such as order confirmation speed, delivery speed, successful delivery speed, and collection speed, which reflect the order processing and distribution efficiency of merchants.

[0101] Optionally, the terminal accesses multiple mall data sources through the middleware API gateway respectively to obtain three types of data with different natures: structured data, unstructured data, and time - series data. For structured data, the terminal obtains the order data of the current merchant (such as the total amount of transaction orders, the number of orders, the amount of a single order), user evaluation data (such as customer score, evaluation content, favorable comment rate), after - sales service data (such as after - sales response time, return and exchange processing duration, after - sales problem solution rate), and penalty record data (such as the number of complaints, reasons for penalties, penalty points). These data have predefined formats and fields and can be directly used for index calculation. For unstructured data, the terminal obtains the customer service conversation text data of the current merchant. This type of data does not have a fixed structure format and requires NLP (Natural Language Processing) technology to extract index information such as service attitude from the text content. For time - series data, the terminal obtains the logistics data of the current merchant, including time - series information such as order confirmation speed, order delivery speed, order successful delivery speed, and order collection speed, which is used to evaluate the transaction efficiency of merchants.

[0102] The technical solution provided in this embodiment clearly divides merchant data into three major categories: structured data (order data, user evaluation data, after - sales service data, and penalty record data), unstructured data (customer service conversation text data), and time - series data (logistics data), making data collection more targeted and systematic; breaking the data island problem existing in the prior art and realizing the unified acquisition and management of multi - dimensional data; thus facilitating the formation of a more comprehensive, accurate, and timely merchant ability evaluation basis, avoiding one - sided prediction caused by a single data source, and providing multi - dimensional and accurate data support for subsequent merchant business ability prediction.

[0103] The following uses an application example to illustrate the merchant business ability prediction method based on scenario - aware adaptation provided in this application. This application example takes the application of this method to a terminal as an example.

[0104] In the field of platform - based e - commerce, merchant credit and ability evaluation is a very important link in platform operation. The merchant credit and ability evaluation system generated based on merchant basic data and platform operation process data promotes the effective operation of the merchant survival - of - the - fittest mechanism. It is the key pillar for the platform to continuously improve customer satisfaction and operation level, and is the basis for the healthy operation of the platform ecosystem.

[0105] With the development of the e-commerce market, various major promotions, activities, and live streaming sales are very frequent and increasingly important. This business format poses challenges to the short-term sales supply capabilities of merchants. For the platform to scientifically evaluate merchants, it is also necessary to scientifically evaluate the performance of merchants in corresponding scenarios and design a more reasonable weight calculation method. The larger the platform, the more necessary it is to avoid static weight settings or adopt manual temporary configuration for business scenarios.

[0106] This application example uses a recursive algorithm and realizes the dynamic adaptive adjustment of weights through a combination of "historical weight inheritance + scenario-aware correction + base value regression". It can not only quickly respond to changes in business scenarios but also prevent the system from getting out of control due to short-term fluctuations.

[0107] Existing technical problems:

[0108] 1) Data silos: The data of enterprises' multiple types of shopping malls (such as B2B (business-to-business), B2C (business-to-consumer), cross-border platforms) are scattered, and the indicator calibers are inconsistent, making it impossible to uniformly evaluate.

[0109] 2) Static evaluation: Existing ratings mostly rely on historical transaction data, the weight settings of the indicator system are insensitive, and there is a lack of real-time dynamic tracking (such as service quality and problem-solving timeliness during different business periods).

[0110] 3) Manual maintenance cost: Manual setting by personnel faces problems such as heavy workload and low efficiency of operation personnel, and the accuracy and real-time nature of manual setting cannot be guaranteed.

[0111] 4) Unscientific data: In the face of high-frequency business scenarios that require smooth transitions, static weight settings or manual adjustments result in calculated data that cannot truly reflect the capabilities of merchants.

[0112] This application example uses a recursive algorithm and realizes the dynamic adaptive adjustment of weights through a combination of "historical weight inheritance + scenario-aware correction + base value regression". It can not only quickly respond to changes in business scenarios but also prevent the system from getting out of control due to short-term fluctuations.

[0113] The overall architecture reference of this application example Figure 2 . In Figure 2Among them, there are four main levels from bottom to top: the data access layer, the data processing engineering layer, the model calculation layer, and the result application layer. In the bottom data access layer, the system accesses data sources from multiple sub-e-commerce platforms (including e-commerce platforms such as e-commerce platform 1, e-commerce platform 2, e-commerce platform 3, e-commerce platform 4, etc.) through the middle platform API gateway. These data are divided into three categories: structured data, unstructured data, and time-series data. The second layer is the data processing engineering layer, which consists of the microservice middle platform and includes three key modules: the index standardization module, the weight configuration module, and the real-time calculation engine. This layer is responsible for engineering the authorized data. The third layer is the model calculation layer, which mainly performs index calculation, merchant association analysis, and rule calculation, and conducts comprehensive calculations based on the dynamic index system. The top layer is the result application layer, which includes real-time merchant scoring (in the range of 0-100 points) and scoring result application interfaces, connects to each sub-e-commerce platform, and realizes the actual application of merchant credit ability evaluation.

[0114] The process reference of the data processing steps in this application example Figure 3 . In Figure 3 Among them, it is divided into four main levels as a whole: the application layer, the business layer, the data layer, and the platform data access layer. In the top application layer, the system supports multiple access methods, including H5 (HyperText 5.0), mini programs, and PC (personal computer). The second layer is the business layer, which provides multiple functional modules, including the user center, job posting, job search, recruitment posting, personnel search, and data authorization. The third layer is the data layer, which is responsible for data processing and management, and includes functions such as data cleaning, data caching, matching logic algorithms, data storage, and reading and writing databases. The bottom layer is the platform data access layer, including project resumes, personnel qualifications, personnel violations, training records, whether on the blacklist, enterprise evaluations, salary payments, and personnel training (data one-to-one authorization is required).

[0115] The process of this application example is as follows:

[0116] I. Data access:

[0117] Access multi-mall data sources through the middle platform API (Application Programming Interface) gateway, including:

[0118] Structured data: order data, user evaluation data, after-sales service data, penalty record data;

[0119] Unstructured data: customer service conversation text (NLP extracts service attitude indicators);

[0120] Time-series data: logistics data.

[0121] II. Data processing engineering configuration:

[0122] The middle platform engineers the authorized data. Including:

[0123] Standardization module: Define and configure a unified metric caliber.

[0124] Real-time calculation engine: Calculate dynamic metrics based on FLINK (stream processing engine).

[0125] Weight configuration module: Adaptive weight algorithm based on scenarios and other weight strategy configurations.

[0126] III. Model Calculation:

[0127] Comprehensively calculate the merchant quantitative score based on the dynamic metric system, the merchant penetration risk quantitative calculation, and the plus and minus points triggered by rule configuration, and calculate the real-time credit ability evaluation score of the merchant.

[0128] IV. Application of Evaluation Results:

[0129] Give the merchant a real-time evaluation grade according to the five levels of SABCD, corresponding to a 100-point value range. At the same time, provide an API call interface. Each sub-e-commerce platform executes the operation strategy according to the merchant level result, realizes the survival of the fittest of the merchants, and maintains the healthy and prosperous platform ecosystem.

[0130] Among them, the adaptive weight algorithm based on business scenarios is as follows:

[0131] The application of the adaptive weight algorithm based on business scenarios is the core of the weight configuration module in the data processing engineering layer. On the basis of data cleaning and the establishment of a standard quantitative evaluation dimension system, this algorithm is applied to the weights of some indicators related to business scenarios to realize the dynamic adaptive adjustment of the weights of merchant evaluation indicators, and avoid the operation pressure and subjective deviation brought by manual adjustment.

[0132] I. Business Scenario Indicators and Basic Weights:

[0133] Table 1 (Business Scenario Indicators and Basic Weights)

[0134]

[0135] II. Weight Adaptive Algorithm:

[0136] 1) Adaptive Algorithm:

[0137] W i (t) = W i (t - 1) + α * S(t) * ((W basei - W i (t - 1)) / W basei ) + β * (W basei - W i (t - 1)).

[0138] Assume that it is necessary to dynamically adjust the weight of the i-th indicator (such as transaction volume), then:

[0139] W i (t): The dynamic weight at time t;

[0140] W i (t - 1): The historical weight at time t - 1;

[0141] W basei : The basic weight of indicator i (e.g., initial value = 40);

[0142] S(t): The scenario coefficient at time t (e.g., S(t) = 0.8 during the promotion period, S(t) = 0 during normal days);

[0143] α: The scenario sensitivity coefficient (controlling the intensity of the scenario impact);

[0144] β: The weight regression coefficient (controlling the regression speed of the weight towards the basic value).

[0145] II) Detailed description:

[0146] 1. Recursive term: W i (t - 1):

[0147] Function: Inherit the weight value of the previous moment to ensure the continuity of weight changes.

[0148] Recursiveness manifestation: The calculation of the current weight W i (t) directly depends on the historical weight W i (t - 1), forming autocorrelation in the time series.

[0149] 2. Scenario perception term: α * S(t) * ((W basei - W i (t - 1)) / W basei ):

[0150] Dynamic adjustment logic:

[0151] Scenario coefficient S(t): Reflects the intensity of the current business scenario (such as the intensity of the promotion activity);

[0152] Weight deviation ((W basei - W i (t - 1)) / W basei : Measures the relative deviation between the historical weight and the basic weight;

[0153] Multiplication effect: The stronger the scenario (the larger S(t)) and the farther the historical weight deviates from the basic value, the greater the adjustment amplitude.

[0154] For example: If S(t) = 0.8 during the major promotion period, and the historical weight W i(t - 1) = 0.7 (base weight W basei = 0.5), then this item is a positive correction, pushing the weight to be further adjusted towards the scenario requirements.

[0155] 3. Regression term: β * (W basei - W i (t - 1)):

[0156] Function: Pull the weight back to the base value to prevent the system from becoming unstable due to long-term deviation.

[0157] Example: If W i (t - 1) = 0.8 (base value W basei = 0.5), then this item is a negative correction (β * (-0.3)), prompting the weight to gradually return.

[0158] 4. Coefficient setting:

[0159] The values of the scenario sensitivity coefficient α and the weight regression coefficient β in this scenario are determined through experimental optimization or dynamic adaptive strategies based on the real-time requirements of the business scenario, the historical data distribution, and the system stability requirements. Specifically, α controls the intensity of the scenario impact, and β ensures that the long-term trend of the weight converges to the base value. The two work together to achieve the agility and robustness of weight adjustment.

[0160] The technical solution provided by this application example achieves: 1) Dynamic adaptability: By recursively depending on historical weights, smooth transition is achieved, avoiding weight jumps caused by manual management; the scenario coefficient S(t) can respond to business activities (such as promotions, risk events) in real time; 2) Stability guarantee: The regression term prevents long-term deviation from the base weight and maintains the system robustness; the parameters α and β are adjustable to balance scenario sensitivity and stability; 3) Fit with business scenarios: During the promotion period, increase the weight of price competitiveness (S(t) increases) to promote traffic tilt; during the risk period: increase the weight of penalty points to quickly identify low-credit merchants.

[0161] It should be understood that although the steps in the flowcharts involved in the above-described embodiments are shown in sequence according to the arrows, these steps do not necessarily have to be executed in the order indicated by the arrows. Unless there is a clear indication in this article, the execution of these steps has no strict order limit, and these steps can be executed in other orders. Moreover, at least some of the steps in the flowcharts involved in the above-described embodiments may include multiple steps or multiple stages. These steps or stages do not necessarily have to be executed at the same time, but can be executed at different times. The execution order of these steps or stages does not necessarily have to be sequential, but can be executed alternately or in turn with at least some of the steps or stages in other steps.

[0162] Based on the same inventive concept, an embodiment of the present application further provides a merchant business capability prediction device based on scenario awareness adaptation for implementing the above-mentioned merchant business capability prediction method based on scenario awareness adaptation. The implementation solutions provided by this device to solve problems are similar to those described in the above method. Therefore, the specific limitations in one or more embodiments of the merchant business capability prediction device based on scenario awareness adaptation provided below can refer to the limitations on the merchant business capability prediction method based on scenario awareness adaptation in the above text, and will not be repeated here.

[0163] In an exemplary embodiment, as Figure 4 shown, a merchant business capability prediction device based on scenario awareness adaptation is provided. The merchant business capability prediction device 400 based on scenario awareness adaptation may include:

[0164] A system construction module 401, configured to construct a merchant business capability prediction index system and determine the basic weight information of each prediction index in the merchant business capability prediction index system;

[0165] An information acquisition module 402, configured to acquire the historical weight information of each prediction index and the scenario information of the current business scenario;

[0166] An information determination module 403, configured to determine the current dynamic weight information of each prediction index according to the basic weight information, historical weight information, and scenario information;

[0167] An information prediction module 404, configured to respond to a business capability prediction request for the current merchant, and through a business capability prediction model, predict the business capability of the merchant information of the current merchant according to the merchant business capability prediction index system and the current dynamic weight information of each prediction index, so as to obtain a business capability prediction result of the current merchant.

[0168] In an exemplary embodiment, the information determination module 403 is further configured to determine a recursion term, a scenario awareness term, and a weight regression term corresponding to the current dynamic weight information according to the basic weight information, historical weight information, and scenario information; and determine the current dynamic weight information according to the recursion term, scenario awareness term, and weight regression term.

[0169] In an exemplary embodiment, the information determination module 403 is further configured to determine the recursion term according to the historical weight information; determine the scenario awareness term according to the basic weight information, historical weight information, and scenario information; and determine the weight regression term according to the basic weight information and historical weight information.

[0170] In an exemplary embodiment, the device 400 further includes: an information sending module, configured to determine the business level information of the current merchant according to the business capability prediction result; and send the business level information to the e-commerce platform; the e-commerce platform is configured to perform business recommendation and risk control on the current merchant according to the business level information.

[0171] In an exemplary embodiment, the device 400 further includes: a data acquisition module, configured to acquire the structured data, unstructured data, and time-series data of the current merchant; and perform fusion processing on the structured data, unstructured data, and time-series data to obtain the merchant information of the current merchant.

[0172] In an exemplary embodiment, the data acquisition module is further configured to acquire the order data, user evaluation data, after-sales service data, and penalty record data of the current merchant as structured data; acquire the customer service session text data of the current merchant as unstructured data; and acquire the logistics data of the current merchant as time-series data.

[0173] Each module in the above-mentioned merchant business capability prediction device based on scenario-aware adaptation can be implemented in whole or in part by software, hardware, and their combination. The above-mentioned modules can be embedded in or independent of the processor in the computer device in the form of hardware, or stored in the memory of the computer device in the form of software, so that the processor can call and execute the operations corresponding to the above-mentioned modules.

[0174] In an exemplary embodiment, a computer device is provided. The computer device can be a terminal, and its internal structure diagram can be as Figure 5As shown in the figure. The computer device includes a processor, a memory, an input / output interface, a communication interface, a display unit, and an input device. Among them, the processor, the memory, and the input / output interface are connected through a system bus, and the communication interface, the display unit, and the input device are connected to the system bus through the input / output interface. Among them, the processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system and computer programs. The internal memory provides an environment for the operation of the operating system and computer programs in the non-volatile storage medium. The input / output interface of the computer device is used to exchange information between the processor and external devices. The communication interface of the computer device is used to communicate with external terminals in a wired or wireless manner, and the wireless manner can be implemented through WIFI, a mobile cellular network, NFC (Near Field Communication), or other technologies. When the computer program is executed by the processor, it implements a method for predicting merchant business capabilities based on scenario awareness adaptation. The display unit of the computer device is used to form a visually visible picture, which can be a display screen, a projection device, or a virtual reality imaging device. The display screen can be a liquid crystal display screen or an electronic ink display screen. The input device of the computer device can be a touch layer covering the display screen, or a button, a trackball, or a touchpad provided on the computer device housing, or an external keyboard, touchpad, or mouse, etc.

[0175] Those skilled in the art can understand that Figure 5 the structure shown in the figure is only a block diagram of some structures related to the solution of this application, and does not constitute a limitation on the computer device to which the solution of this application is applied. The specific computer device may include more or fewer components than those shown in the figure, or combine some components, or have different component arrangements.

[0176] In an exemplary embodiment, a computer device is further provided, including a memory and a processor. A computer program is stored in the memory, and when the processor executes the computer program, the steps in the above method embodiments are implemented.

[0177] In an exemplary embodiment, a computer-readable storage medium is provided, on which a computer program is stored, and when the computer program is executed by the processor, the steps in the above method embodiments are implemented.

[0178] In an exemplary embodiment, a computer program product is provided, including a computer program, and when the computer program is executed by the processor, the steps in the above method embodiments are implemented.

[0179] Those of ordinary skill in the art can understand that all or part of the processes in the methods of the above embodiments can be completed by instructing relevant hardware through a computer program. The computer program can be stored in a non-volatile computer-readable storage medium. When the computer program is executed, it can include the processes of the embodiments of the above methods. Among them, any reference to a memory, database, or other medium used in the embodiments provided in this application can include at least one of non-volatile and volatile memories. Non-volatile memory can include read-only memory (ROM), magnetic tape, floppy disk, flash memory, optical memory, high-density embedded non-volatile memory, resistive random access memory (ReRAM), magnetoresistive random access memory (MRAM), ferroelectric random access memory (FRAM), phase change memory (PCM), graphene memory, etc. Volatile memory can include random access memory (RAM) or external cache memory, etc. By way of illustration and not limitation, RAM can be in various forms, such as static random access memory (SRAM) or dynamic random access memory (DRAM), etc. The databases involved in the embodiments provided in this application can include at least one of relational databases and non-relational databases. Non-relational databases can include distributed databases based on blockchain, etc., without limitation. The processors involved in the embodiments provided in this application can be general-purpose processors, central processing units, graphics processing units, digital signal processors, programmable logic devices, data processing logics based on quantum computing, etc., without limitation.

[0180] The technical features of the above embodiments can be combined arbitrarily. For the sake of concise description, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, it should be considered as the scope described in this specification.

[0181] The above-described embodiments only represent several implementation manners of this application. The description is relatively specific and detailed, but it should not be construed as a limitation on the patent scope of this application. It should be noted that for those of ordinary skill in the art, without departing from the concept of this application, several modifications and improvements can still be made, and these all belong to the protection scope of this application. Therefore, the protection scope of this application should be subject to the appended claims.

Claims

1. A method for predicting the business capabilities of merchants based on scenario-aware adaptation, characterized in that, The method includes: Constructing a merchant business ability prediction index system and determining the basic weight information of each prediction index in the merchant business ability prediction index system; Obtaining the historical weight information of each prediction index and the scenario information of the current business scenario; Determining the current dynamic weight information of each prediction index according to the basic weight information, the historical weight information and the scenario information; In response to a business ability prediction request for a current merchant, through a business ability prediction model, predicting the business ability of the merchant information of the current merchant according to the merchant business ability prediction index system and the current dynamic weight information of each prediction index, and obtaining the business ability prediction result of the current merchant.

2. The method according to claim 1, wherein The determining the current dynamic weight information of each prediction index according to the basic weight information, the historical weight information and the scenario information includes: Determining a recursion term, a scenario perception term and a weight regression term corresponding to the current dynamic weight information according to the basic weight information, the historical weight information and the scenario information; Determining the current dynamic weight information according to the recursion term, the scenario perception term and the weight regression term.

3. The method according to claim 2, wherein The determining the recursion term, the scenario perception term and the weight regression term corresponding to the current dynamic weight information according to the basic weight information, the historical weight information and the scenario information includes: Determining the recursion term according to the historical weight information; Determining the scenario perception term according to the basic weight information, the historical weight information and the scenario information; Determining the weight regression term according to the basic weight information and the historical weight information.

4. The method according to claim 1, wherein After predicting the business ability of the merchant information of the current merchant in response to a business ability prediction request for the current merchant, through a business ability prediction model, according to the merchant business ability prediction index system and the current dynamic weight information of each prediction index, and obtaining the business ability prediction result of the current merchant, it further includes: Determining the business level information of the current merchant according to the business ability prediction result; Sending the business level information to an e-commerce platform; the e-commerce platform is used to perform business recommendation and risk control on the current merchant according to the business level information.

5. The method according to any one of claims 1 to 4, characterized in that, The method further includes: Obtaining the structured data, unstructured data and time series data of the current merchant; Performing fusion processing on the structured data, the unstructured data and the time series data to obtain the merchant information of the current merchant.

6. The method according to claim 5, characterized in that, The obtaining the structured data, unstructured data and time series data of the current merchant includes: Obtaining the order data, user evaluation data, after-sales service data and penalty record data of the current merchant as the structured data; Obtaining the customer service session text data of the current merchant as the unstructured data; Obtaining the logistics data of the current merchant as the time series data.

7. A merchant business capability prediction device based on scenario-aware adaptation, characterized in that, The device includes: A system construction module, configured to construct a merchant business ability prediction index system and determine the basic weight information of each prediction index in the merchant business ability prediction index system; An information acquisition module, configured to acquire historical weight information of each of the prediction metrics and scenario information of the current business scenario; An information determination module, configured to determine current dynamic weight information of each of the prediction metrics according to the basic weight information, the historical weight information, and the scenario information; An information prediction module, configured to, in response to a business capability prediction request for a current merchant, perform a business capability prediction on merchant information of the current merchant through a business capability prediction model according to the merchant business capability prediction index system and the current dynamic weight information of each of the prediction metrics, so as to obtain a business capability prediction result of the current merchant.

8. A computer device, comprising a memory and a processor, the memory storing a computer program, characterized in that, When the processor executes the computer program, the steps of the method according to any one of claims 1 to 6 are implemented.

9. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 6 are implemented.

10. A computer program product, comprising a computer program, characterized in that, When the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 6 are implemented.

Images