Quality evaluation method and device of target merchant, electronic equipment, storage medium and program product

Abstract

The invention provides a quality evaluation method and device of a target merchant, electronic equipment, a storage medium and a program product, can be applied to the technical field of artificial intelligence and big data, and relates to application of a large model in a financial science and technology scene. The method comprises the steps that multiple candidate merchants in a preset area range of the position where a target merchant is located are determined, merchant data of the multiple candidate merchants on multiple consumption platforms are acquired, and the merchant data comprise menu data and score data of the menu data; extracting multiple dimension features of the target service based on the merchant data; according to the multiple dimension features, constructing a regional knowledge graph of the target service; in the regional knowledge graph, identifying a target node associated with the target merchant, and determining a characteristic score of the target merchant according to the target node and the regional knowledge graph; determining a basic score of the target merchant based on the score data of the target merchant; and evaluating the quality of the target service of the target merchant according to the characteristic score and the basic score.

Description

Technical Field

[0001] This application relates to the fields of artificial intelligence and big data technology, and to the application of large models in fintech scenarios, specifically to a method, apparatus, electronic device, storage medium, and program product for quality assessment of target merchants. Background Technology

[0002] For financial institutions, accurate and dynamic quality assessment of target merchants (such as micro and small enterprises, supply chain partners, and platform operators) is a crucial foundation for core business decisions such as credit approval, risk management, and resource allocation. Current methods for assessing merchant quality primarily rely on static financial statements and manual verification, which suffer from problems such as information lag, limited dimensions, and low efficiency, making it difficult to dynamically and comprehensively reflect a merchant's true market performance and potential risks.

[0003] With the development of big data and artificial intelligence technologies, publicly available data has begun to be used as an auxiliary assessment in the process of evaluating merchant quality. However, these methods typically only perform simple data aggregation or superficial analysis, failing to assess the relative uniqueness of merchants within a regional competitive network, and lacking in-depth semantic mining of massive amounts of unstructured data, resulting in biased assessment results. Moreover, publicly available data often contains noise, and existing methods still lack effective anomaly identification and data cleaning mechanisms, casting doubt on the credibility of the data sources on which the assessment relies, and severely compromising the objectivity and anti-interference capabilities of the assessment results. Summary of the Invention

[0004] In view of the above problems, this application provides a method, apparatus, electronic device, storage medium and program product for quality assessment of target merchants, which can solve the problems of single data source, delayed assessment results and inability to objectively quantify the unique value of merchants in existing merchant quality assessment methods.

[0005] According to the first aspect of this application, a method for quality assessment of a target merchant is provided. The method includes: identifying multiple candidate merchants within a preset area of ​​the target merchant's location; acquiring merchant data for the multiple candidate merchants on multiple consumer platforms, the merchant data including menu data and rating data for the menu data; wherein the multiple candidate merchants are merchants providing the same target service, and the multiple candidate merchants include the target merchant; extracting multiple-dimensional features of the target service based on the merchant data; constructing a regional knowledge graph of the target service based on the multiple-dimensional features; wherein the regional knowledge graph includes multiple nodes and multiple edges, each node representing one of the multiple candidate merchants, and each edge representing the association between two candidate merchants determined based on the multiple-dimensional features; identifying target nodes associated with the target merchant in the regional knowledge graph, and determining a characteristic rating for the target merchant based on the target nodes and the regional knowledge graph; determining a basic rating for the target merchant based on the target merchant's rating data; and comprehensively evaluating the quality of the target merchant's target service by combining the characteristic rating and the basic rating.

[0006] According to an embodiment of this application, a regional knowledge graph is constructed in the following manner: multiple merchant nodes are created for multiple candidate merchants respectively; based on multiple dimensional features, the feature similarity between any two candidate merchants is calculated; if the feature similarity is higher than a preset threshold, an association edge is established between any two candidate merchants, and the feature similarity is used as the association weight of the association edge; a regional knowledge graph is formed based on the merchant nodes and the association edges.

[0007] According to an embodiment of this application, there are multiple target nodes associated with the target merchant; determining the characteristic score of the target merchant based on the target nodes and the regional knowledge graph includes: obtaining multiple association weights between the target merchant and the multiple target nodes in the regional knowledge graph; calculating the average association weight between the target merchant and the multiple target nodes based on the multiple association weights; and determining the characteristic score of the target merchant based on the average association weight, wherein the lower the average association weight, the higher the characteristic score.

[0008] According to an embodiment of this application, the target merchant's rating data includes multiple rating data of the target merchant within a preset time period; based on the target merchant's rating data, determining the target merchant's basic rating includes: recording the generation time of each rating data in the multiple rating data, determining the rating value of each rating data; calculating the weight of each rating value based on the difference between the generation time and the current time; wherein, the weight value decreases as the difference increases; obtaining the average service quality of the target merchant within the preset time period based on the multiple rating values ​​and the weight of each rating value; and determining the target merchant's basic rating based on the average service quality.

[0009] According to an embodiment of this application, determining the basic score of a target merchant based on average service quality includes: sorting multiple score values ​​and using the interquartile range rule to detect and remove abnormal score values; calculating the standard deviation or variance of the multiple score values ​​after removing abnormal score values, and normalizing the calculation results to obtain the degree of fluctuation of the target merchant's score data within a preset time period; and determining the basic score of the target merchant based on the average service quality and the degree of fluctuation.

[0010] According to an embodiment of this application, multiple score values ​​are sorted, and abnormal score points are detected and removed using the interquartile range rule, including: determining the first quartile and the third quartile of the sorted multiple score values; calculating the interquartile range based on the first quartile and the third quartile, and setting an abnormal value judgment interval based on the interquartile range; marking score values ​​that exceed the judgment interval as abnormal score values ​​and removing them.

[0011] According to an embodiment of this application, determining the basic rating of a target merchant based on the rating data of the target merchant includes: inputting multiple rating data into a pre-trained large model and outputting sentiment scores of multiple rating data; the sentiment score represents the quantitative intensity of the sentiment tendency of the rating data towards the target merchant; calculating the weighted average of the sentiment scores of multiple rating data as the reputation score; and determining the basic rating of the target merchant based on the reputation score.

[0012] The second aspect of this application provides a quality assessment device for a target merchant. The device includes: an acquisition module, used to determine multiple candidate merchants within a preset area of ​​the target merchant's location, and acquire merchant data of the multiple candidate merchants on multiple consumer platforms, the merchant data including menu data and rating data for the menu data; wherein the multiple candidate merchants are merchants providing the same target service, and the multiple candidate merchants include the target merchant; a graph construction module, used to extract multiple-dimensional features of the target service based on the merchant data; and construct a regional knowledge graph of the target service based on the multiple-dimensional features; wherein the regional knowledge graph includes multiple nodes and multiple edges, each node representing one of the multiple candidate merchants, and each edge representing the association between two candidate merchants determined according to the multiple-dimensional features; a feature assessment module, used to identify target nodes associated with the target merchant in the regional knowledge graph, and determine the feature score of the target merchant based on the target nodes and the regional knowledge graph; a basic assessment module, used to determine the basic score of the target merchant based on the rating data of the target merchant; and a comprehensive assessment module, used to combine the feature score and the basic score to assess the quality of the target service of the target merchant.

[0013] A third aspect of this application provides an electronic device comprising: one or more processors; and a memory for storing one or more computer programs, wherein the one or more processors execute the one or more computer programs to implement the steps of the method described above.

[0014] A fourth aspect of this application also provides a computer-readable storage medium having a computer program or instructions stored thereon, which, when executed by a processor, implement the steps of the above-described method.

[0015] The fifth aspect of this application also provides a computer program product, including a computer program or instructions that, when executed by a processor, implement the steps of the above-described method. Attached Figure Description

[0016] The above-mentioned contents, other objects, features and advantages of this application will become clearer from the following description of embodiments with reference to the accompanying drawings, in which:

[0017] Figure 1 The illustration schematically depicts an application scenario of a quality assessment method, apparatus, device, storage medium, and program product for target merchants according to embodiments of this application.

[0018] Figure 2 A flowchart illustrating a quality assessment method for target merchants according to an embodiment of this application is shown schematically.

[0019] Figure 3 A flowchart illustrating the construction of a regional knowledge graph according to an embodiment of this application is shown schematically.

[0020] Figure 4 This illustration schematically shows one of the flowcharts for the target merchant basic score in the target merchant quality assessment method according to an embodiment of this application;

[0021] Figure 5 This illustration schematically shows a second flowchart of the target merchant basic score in the target merchant quality assessment method according to an embodiment of this application;

[0022] Figure 6 This schematically illustrates a structural block diagram of a quality assessment device for a target merchant according to an embodiment of this application; and

[0023] Figure 7 A block diagram schematically illustrates an electronic device suitable for implementing a quality assessment method for target merchants according to an embodiment of this application. Detailed Implementation

[0024] The embodiments of this application will now be described with reference to the accompanying drawings. However, it should be understood that these descriptions are exemplary only and are not intended to limit the scope of this application. In the following detailed description, numerous specific details are set forth to provide a thorough understanding of the embodiments of this application for ease of explanation. However, it will be apparent that one or more embodiments may be implemented without these specific details. Furthermore, descriptions of well-known structures and technologies are omitted in the following description to avoid unnecessarily obscuring the concepts of this application.

[0025] The terminology used herein is for the purpose of describing particular embodiments only and is not intended to limit the scope of this application. The terms “comprising,” “including,” etc., as used herein indicate the presence of the stated features, steps, operations, and / or components, but do not exclude the presence or addition of one or more other features, steps, operations, or components.

[0026] All terms used herein (including technical and scientific terms) have the meanings commonly understood by those skilled in the art, unless otherwise defined. It should be noted that the terms used herein are to be interpreted in a manner consistent with the context of this specification, and not in an idealized or overly rigid way.

[0027] When using expressions such as "at least one of A, B and C", they should generally be interpreted in accordance with the meaning that is commonly understood by those skilled in the art (e.g., "a system having at least one of A, B and C" should include, but is not limited to, a system having A alone, a system having B alone, a system having C alone, a system having A and B, a system having A and C, a system having B and C, and / or a system having A, B and C, etc.).

[0028] In the technical solution of this application, the user information (including but not limited to user personal information, user image information, user device information, such as location information) and data (including but not limited to data used for analysis, stored data, and displayed data) involved are all information and data authorized by the user or fully authorized by all parties. Furthermore, the collection, storage, use, processing, transmission, provision, disclosure, and application of related data all comply with relevant laws, regulations, and standards, take necessary confidentiality measures, do not violate public order and good morals, and provide corresponding operation entry points for users to choose to authorize or refuse.

[0029] In scenarios involving automated decision-making using personal information, the methods, devices, and systems provided in this application all offer users corresponding entry points for choosing to agree to or reject the automated decision-making results. If the user chooses to reject, the process proceeds to the expert decision-making stage. Here, "automated decision-making" refers to the activity of automatically analyzing and evaluating an individual's behavioral habits, interests, or economic, health, and credit status through computer programs, and then making a decision. Here, "expert decision-making" refers to the activity of making decisions by personnel who specialize in a particular field, possess specialized experience, knowledge, and skills, and have reached a certain level of professional expertise.

[0030] Figure 1 The illustration schematically depicts an application scenario of a quality assessment method, apparatus, device, storage medium, and program product for a target merchant according to embodiments of this application.

[0031] like Figure 1 As shown, application scenario 100 according to this embodiment may include a first terminal device 101, a second terminal device 102, a third terminal device 103, a network 104, and a server 105. The network 104 serves as a medium for providing a communication link between the first terminal device 101, the second terminal device 102, the third terminal device 103, and the server 105. The network 104 may include various connection types, such as wired or wireless communication links, or fiber optic cables, etc.

[0032] Users can use the first terminal device 101, the second terminal device 102, and the third terminal device 103 to interact with the server 105 via the network 104 to perform data queries, analysis requests, or receive results related to the quality assessment of target merchants. Various client applications can be installed on the first terminal device 101, the second terminal device 102, and the third terminal device 103, such as merchant management software, data analysis tools, web browsers, financial service platform applications, etc. (for example only).

[0033] The first terminal device 101, the second terminal device 102, and the third terminal device 103 can be various electronic devices with displays and supporting network communication and data interaction, including but not limited to smartphones, tablets, laptops, and desktop computers.

[0034] Server 105 may be a server that provides quality assessment services for target merchants, such as a server that acquires, stores, analyzes, and processes merchant-related data received from the first terminal device 101, the second terminal device 102, the third terminal device 103, or other data sources (this is just an example). Server 105 may analyze and process the received merchant data and other data, and feed the processing results back to the terminal devices.

[0035] It should be noted that the target merchant quality assessment method provided in this application embodiment can generally be executed by server 105. Correspondingly, the target merchant quality assessment device provided in this application embodiment can generally be installed in server 105. The target merchant quality assessment method provided in this application embodiment can also be executed by a first terminal device 101, a second terminal device 102, or a third terminal device 103 capable of acquiring and processing relevant data. Correspondingly, the target merchant quality assessment device provided in this application embodiment can also be installed in the first terminal device 101, the second terminal device 102, or the third terminal device 103.

[0036] It should be understood that Figure 1 The number of terminal devices, networks, and servers shown is merely illustrative. Depending on implementation needs, any number of terminal devices, networks, and servers can be included.

[0037] It should be noted that this application relates to the fields of artificial intelligence, big data, and information processing technology, specifically the application of knowledge graph construction, natural language processing, and data mining technologies in business intelligence and fintech scenarios. For example, when banks conduct credit risk assessments for micro and small enterprises, screen partners for core enterprises in the supply chain, or conduct tiered management of acquiring merchants, they can apply the target merchant quality assessment method proposed in this application. By analyzing publicly available data of merchants on multiple consumer platforms, a dynamic knowledge graph is constructed and combined with multi-dimensional scoring to generate a comprehensive and objective merchant quality profile. This provides precise data support for credit decisions, risk monitoring, and resource allocation, effectively improving the intelligence level and risk management capabilities of financial services.

[0038] It should be noted that the merchant quality assessment methods and devices described in this application embodiment can be widely applied to fintech and other related technical fields. Especially in fintech scenarios, they can help financial institutions establish a dynamic merchant quality assessment system based on real market feedback, thereby more accurately identifying high-quality merchants, providing early warnings of potential risks, and developing differentiated financial products and services to achieve a better balance between risk and return.

[0039] The following will be based on Figure 1 The described scene, through Figures 2-5 The quality assessment method for the target merchants in the disclosed embodiments is described in detail.

[0040] Figure 2 A flowchart illustrating a quality assessment method for target merchants according to an embodiment of this application is shown schematically.

[0041] like Figure 2 As shown, the quality assessment method for target merchants in this embodiment includes operations S210 to S250.

[0042] In operation S210, multiple candidate merchants within a preset area of ​​the target merchant's location are identified, and merchant data of the multiple candidate merchants on multiple consumption platforms is obtained. The merchant data includes menu data and rating data for the menu data. Among them, the multiple candidate merchants are merchants that provide the same target service, and the multiple candidate merchants include the target merchant.

[0043] In some embodiments, the system receives the geographic location information (such as latitude and longitude coordinates) of the target merchant and determines a preset area range based on the geographic location information. For example, the preset area range may be a circular area formed by setting a fixed radius or a variable radius (such as 1 km to 5 km) with the geographic location information of the target merchant as the center; or the administrative division (such as street, business district, community) to which the target merchant belongs may be used as the preset area range; or the time-accessible range (such as the area covered by a 10-minute driving distance) may be calculated based on traffic network data and used as the preset area range.

[0044] Multiple candidate merchants within a preset area are identified, that is, other merchants within the preset area that are comparable to the target merchant are identified and selected as candidate merchants. In some embodiments, comparability between two merchants is determined by whether their main services or product categories are the same or highly similar. For example, a list of all merchants located within the preset area is obtained from a local database or by calling a third-party map or business information service API (Application Programming Interface); by comparing the merchants' category tags and service description keywords, all merchants providing the same target service as the target merchant are selected from this list. These selected merchants, together with the target merchant itself, constitute a set of multiple candidate merchants.

[0045] For example, if the target merchant is a small or micro enterprise that mainly provides credit loan consulting services, then the set of candidate merchants may include all merchants within the preset area that also provide financial intermediary services such as credit loan consulting and corporate financing planning.

[0046] After identifying multiple candidate merchants within a pre-defined area encompassing the target merchant's location, publicly available data from these candidate merchants is collected in parallel from multiple independent third-party consumer service platforms or data partners. For example, for each candidate merchant, data requests are initiated in parallel through publicly available application programming interfaces (APIs) on each platform or through authorized data cooperation channels.

[0047] In some embodiments, the acquired merchant data primarily includes menu data and rating data for the menu data. Menu data refers to the core products and services offered by the merchant and their details, such as product names, service packages, and prices—structured or semi-structured information. Rating data mainly refers to text reviews submitted by consumers on the consumption platform regarding merchants or specific products / services. This rating data is unstructured text information with timestamps, containing consumers' subjective evaluations and descriptions. This raw data is acquired through authorized application programming interfaces or compliant data pipelines, and after cleaning, deduplication, and format standardization, it is stored in a local database.

[0048] In operation S220, based on merchant data, multiple dimensional features of the target service are extracted; based on the multiple dimensional features, a regional knowledge graph of the target service is constructed; the regional knowledge graph includes multiple nodes and multiple edges, each node represents one of the multiple candidate merchants, and each edge represents the relationship between two candidate merchants determined based on the multiple dimensional features.

[0049] In some embodiments, the merchant data of each candidate merchant is analyzed to extract multiple dimensional features that can characterize the core attributes of its services. These multiple dimensional features can objectively describe the merchant's positioning and characteristics from different perspectives. For example, each candidate merchant can be represented by a multi-dimensional feature vector, where each dimension of the vector corresponds to a quantified value of a specific feature, such as frequency or proportion.

[0050] For example, based on merchant data, a natural language processing model can be used to identify core service items, product types, or solution categories, and their distribution can be statistically analyzed.

[0051] For example, price information can be extracted from merchant data to calculate their average price, price range, or price index compared with industry benchmarks, thus forming the merchant's positioning characteristics in their market.

[0052] For example, from merchant data, keyword extraction or topic models (such as three-layer Bayesian probability models) can be used to identify the main customer types (such as corporate customers, individual customers, startups) or frequently occurring service scenarios (such as contract review, financing consultation, daily advisory) that the services target.

[0053] For example, from merchant data, identify the expertise, specific tools, methods, or certifications they mention.

[0054] In some embodiments, based on the multiple dimensional features extracted for each candidate merchant, the overall matching degree of the feature vectors between any two candidate merchants on these multiple dimensional features is calculated to determine the association between them. The strength of the association is quantified by the overall matching degree. Subsequently, a regional knowledge graph is constructed based on these associations, where each node represents a candidate merchant, and each edge connects two candidate merchants with a predetermined strength of association. The weight of the edge represents the strength of the association.

[0055] In operation S230, target nodes associated with the target merchant are identified in the regional knowledge graph, and the characteristic score of the target merchant is determined based on the target nodes and the regional knowledge graph.

[0056] After constructing the regional knowledge graph, the node representing the target merchant is located in the regional knowledge graph. Then, the nodes directly connected to the node of the target merchant are identified. The identified nodes represent the candidate merchant nodes that have a certain relationship with the target merchant, that is, the target nodes.

[0057] For example, based on the strength of the association between a target node and a target merchant in a regional knowledge graph, a numerical value reflecting the uniqueness or scarcity of the target merchant's target service in the regional knowledge graph can be calculated, i.e., a feature score. It should be understood that the lower the strength of the association between the target merchant and its connected target nodes, the higher the feature score of the target merchant; conversely, the stronger the association, the lower the feature score.

[0058] In operation S240, the basic rating of the target merchant is determined based on the target merchant's rating data.

[0059] In some embodiments, the rating data of the target merchant may include multiple rating data of the target merchant within a preset time period. The rating data of the target merchant on multiple consumer platforms can be quantitatively analyzed to determine the basic rating of the target merchant. This basic rating can reflect the stable performance and market recognition of the target merchant from different aspects.

[0060] In some embodiments, multiple rating data can be converted into multiple corresponding rating scores, and the average score of the multiple rating scores or the degree of fluctuation of the multiple rating scores within a preset time period can be analyzed. For example, if the target merchant is a gym, the gym's rating data on multiple consumer platforms over the past 6 months can be converted into time-series rating values; by calculating the average score of its rating values ​​over the past 6 months, the average service quality of the gym can be reflected; and its rating stability can be calculated after statistically removing occasional extreme negative or positive reviews. Specifically, within the same preset time period, a gym with a stable rating score above 4.5 will have a significantly higher base rating than other gyms whose rating scores fluctuate wildly between 3.0 and 4.8.

[0061] In some embodiments, a basic rating for the target merchant is determined based on the target merchant's rating data, specifically including operations S241 to S243.

[0062] In operation S241, multiple rating data are input into a pre-trained large model, and the model outputs sentiment scores for multiple rating data; the sentiment scores represent the quantitative strength of the sentiment tendency of the rating data towards the target merchant.

[0063] In operation S242, the weighted average of the sentiment scores of multiple rating data is calculated as the reputation score.

[0064] In operation S243, the base score of the target merchant is determined based on the reputation score.

[0065] For example, if the target merchant is a chain restaurant, banks or financial institutions can input its publicly available ratings data from multiple consumer platforms into a large-scale model. The model analyzes the rating data to determine its overall sentiment and quantifies specific service dimensions implicit in the data (such as food taste, service response speed, hygiene, and price). The model outputs a structured analysis result for each review, containing sentiment scores and keywords for multiple specific service dimensions. The analysis results for each review are then aggregated to determine the chain restaurant's basic rating, ultimately supporting the bank's credit risk assessment or supply chain partner selection decisions.

[0066] When operating the S250, the quality of the target merchant's target service is evaluated by combining the feature score and the basic score.

[0067] In some embodiments, after determining the target merchant's distinctive rating and basic rating, weighting coefficients can be configured for the distinctive rating and basic rating respectively (e.g., the distinctive rating has a weight of 'a', the basic rating has a weight of 'b', and a+b=1). These two weighting coefficients can be predefined or dynamically adjusted according to the evaluation focus. For example, in scenarios that encourage innovation and differentiation, the distinctive rating weight 'a' can be increased; in scenarios that pursue stability and customer satisfaction, the basic rating weight 'b' can be increased.

[0068] This application's embodiments identify candidate merchants providing a unified target service within the target merchant's region and aggregate their merchant data across multiple consumer platforms. Based on multiple dimensions of features related to the target service in the merchant data, a regional knowledge graph is constructed, with candidate merchants as nodes and relationships between them as edges. The target merchant's unique score is then calculated by analyzing its associations within this regional knowledge graph. After determining its basic score based on merchant data, the quality of the target merchant's target service is evaluated in conjunction with its unique score. This approach achieves a shift from static, one-dimensional evaluation to dynamic, multi-dimensional evaluation. The regional knowledge graph quantifies the uniqueness of the target merchant in the competitive environment and assesses the stability and market reputation of its target service using merchant data across multiple consumer platforms. This provides a more comprehensive, objective, and practical evaluation basis for the target merchant's quality assessment, effectively improving the accuracy of the assessment and enabling the discovery and mining of high-potential, unique merchants.

[0069] Figure 3 A flowchart illustrating the construction of a regional knowledge graph according to an embodiment of this application is shown.

[0070] like Figure 3 As shown, in the target merchant quality assessment method of this embodiment, the above-mentioned operation S220 specifically includes operations S310 to S330.

[0071] When operating S310, create multiple merchant nodes for each candidate merchant.

[0072] For example, each merchant node stores the basic identification information of the candidate merchant (such as the merchant name) as well as multiple quantified features.

[0073] In operation S321, based on multiple dimensions of features, the feature similarity between any two candidate merchants among multiple candidate merchants is calculated.

[0074] In some embodiments, the multi-dimensional features of candidate merchants are quantized and stored in the corresponding merchant nodes to form multi-dimensional feature vectors. A cosine similarity algorithm is used to quantify the overall similarity between two candidate merchants by measuring the cosine of the angle between their feature vectors in multi-dimensional space. The result is a value between 0 and 1, where a result of 0 indicates that the two candidate merchants are completely dissimilar, and a result of 1 indicates that the two candidate merchants are completely similar.

[0075] In operation S322, if the feature similarity is higher than a preset threshold, an association edge is established between any two candidate merchants, and the feature similarity is used as the association weight of the association edge.

[0076] In some embodiments, the calculated feature similarity is compared with a preset threshold to remove weak and accidental similarities. If the feature similarity between two candidate merchants is higher than the threshold, an association edge is established between the two candidate merchants; at the same time, the specific value of the feature similarity is assigned to this association edge as its association weight.

[0077] When operating S330, a regional knowledge graph is formed based on merchant nodes and associated edges.

[0078] In some embodiments, after operation S330, a distinctive score for the target merchant is determined based on the target nodes associated with the target merchant and the constructed regional knowledge graph; wherein there are multiple target nodes. Specifically, this includes: obtaining multiple association weights between the target merchant and multiple target nodes in the regional knowledge graph; calculating the average association weight between the target merchant and multiple target nodes based on the multiple association weights; and determining the distinctive score of the target merchant based on the average association weight; wherein, the lower the average association weight, the higher the distinctive score. It should be understood that the less similar the average characteristics of the target merchant and the candidate merchants represented by its connected multiple target nodes (i.e., the lower the average association weight), the more unique the target service provided by the target merchant in the regional market, and the stronger its differentiated competitiveness, thus deserving a higher distinctive score.

[0079] Figure 4 One of the flowcharts illustrating the determination of a target merchant's basic score in a quality assessment method for a target merchant according to an embodiment of this application is shown.

[0080] like Figure 4 As shown, in the target merchant quality assessment method of this embodiment, the target merchant's rating data includes multiple rating data of the target merchant within a preset time period, and determining the target merchant's basic rating includes operations S410 to S440.

[0081] In operation S410, the generation time of each rating data in multiple rating data is recorded, and the rating score of each rating data is determined.

[0082] For example, record multiple rating data points generated by the target merchant on multiple consumer platforms within a preset time period; parse the multiple rating data points to obtain multiple corresponding rating scores, and accurately record their corresponding generation times to form a time-series rating sequence, such as {(S1, T1), (S2, T2)…(S n , T n )}, where S i (i=1, 2,…,n) represent the scoring values, T i (i=1, 2,…,n) represents the time when the score was generated, and n is an integer greater than 1.

[0083] In operation S420, the weight of each score is calculated based on the difference between the generated time and the current time; the weight decreases as the difference increases.

[0084] In some embodiments, each rating score (S) i Assign a weight (W) that decays over time. i ), where the time when the rating data was generated (T) i ) Distance from the current time (T) now The longer the time difference (ΔT), the greater the time difference. i = T now - T i The larger the weight (W) is, the higher its weight (W) is. i The smaller the value.

[0085] For example, an exponential decay function can be used to control the rate at which the weights decrease over time. Specifically, Where λ is the attenuation rate parameter, W i As weight.

[0086] For example, fixed weights can be assigned to segments based on time windows, such as a weight of 1.0 for scores within the past month, 0.7 for scores within 1 to 3 months, and 0.4 for scores within 3 to 6 months.

[0087] In operation S430, the average service quality of the target merchant within a preset time period is obtained based on multiple rating scores and the weight of each rating score.

[0088] For example, the weighted average service quality score of the target merchant over a preset time period can be calculated using existing rating scores and their corresponding weights, and this score can be used as the average service quality score.

[0089] When operating S440, determine the basic score of the target merchant based on the average service quality.

[0090] For example, the weighted average service quality score can be normalized and used as the base score for the target merchant.

[0091] Figure 5 The second flowchart illustrating the target merchant basic score in the target merchant quality assessment method according to an embodiment of this application is shown schematically.

[0092] like Figure 5 As shown, in the target merchant quality assessment method of this embodiment, in order to prevent occasional extreme reviews (such as malicious negative reviews) from interfering with the judgment of the target service quality of the target merchant, the above-mentioned operation S440 may also specifically include operations S510 to S530.

[0093] In operation S510, multiple rating scores are sorted, and abnormal rating scores are detected and removed using the interquartile range rule.

[0094] Specifically, the first and third quartiles of the sorted scores are determined; the interquartile range is calculated based on the first and third quartiles, and an outlier determination interval is set based on the interquartile range; scores that exceed the determination interval are marked as outlier scores and removed.

[0095] For example, all rating scores (S1, S2, ..., S) within a preset time period. n Sort the values ​​in ascending order; determine the first quartile (Q1) and third quartile (Q3) of the sorted sequence, where Q1 represents the 25th percentile and Q3 represents the 75th percentile; calculate the interquartile range (IQR) using the formula IQR = Q3 - Q1; set the outlier judgment interval based on the interquartile range (IQR), such as a lower limit of Q1 - k × IQR and an upper limit of Q3 + k × IQR, where k is a preset coefficient; mark the scores that are lower than the lower limit or higher than the upper limit as outlier scores and remove them.

[0096] In operation S520, the standard deviation or variance of multiple rating scores after removing abnormal rating scores is calculated, and the calculation results are normalized to obtain the degree of fluctuation of the target merchant's rating data within a preset time period.

[0097] When operating S530, a base score for the target merchant is determined based on the average service quality and the degree of volatility.

[0098] Figure 6 A schematic block diagram of a quality assessment device for a target merchant according to an embodiment of this application is shown.

[0099] like Figure 6As shown, the target merchant quality assessment device 600 of this embodiment includes an acquisition module 610, a map construction module 620, a feature assessment module 630, a basic assessment module 640, and a comprehensive assessment module 650.

[0100] The acquisition module 610 is used to determine multiple candidate merchants within a preset area range of the target merchant's location, and to acquire merchant data of the multiple candidate merchants on multiple consumer platforms. The merchant data includes menu data and rating data for the menu data. The multiple candidate merchants are merchants providing the same target service, and the multiple candidate merchants include the target merchant. In one embodiment, the acquisition module 610 can be used to perform the operation S210 described above, which will not be repeated here.

[0101] The graph construction module 620 is used to extract multiple-dimensional features of the target service based on merchant data; and to construct a regional knowledge graph of the target service based on the multiple-dimensional features. The regional knowledge graph includes multiple nodes and multiple edges, where each node represents one of multiple candidate merchants, and each edge represents the association between two candidate merchants determined based on the multiple-dimensional features. In one embodiment, the graph construction module 620 can be used to perform the operation S220 described above, which will not be repeated here.

[0102] The feature evaluation module 630 is used to identify target nodes associated with the target merchant in the regional knowledge graph, and determine the feature score of the target merchant based on the target nodes and the regional knowledge graph. In one embodiment, the feature evaluation module 630 can be used to perform the operation S230 described above, which will not be repeated here.

[0103] The basic assessment module 640 is used to determine the basic rating of the target merchant based on the target merchant's rating data. In one embodiment, the basic assessment module 640 can be used to perform the operation S240 described above, which will not be repeated here.

[0104] The comprehensive evaluation module 650 is used to combine the feature score and the basic score to evaluate the quality of the target merchant's target service. In one embodiment, the comprehensive evaluation module 650 can be used to perform the operation S250 described above, which will not be repeated here.

[0105] According to embodiments of this application, any multiple modules among the acquisition module 610, map construction module 620, feature evaluation module 630, basic evaluation module 640, and comprehensive evaluation module 650 can be merged into one module, or any one of these modules can be split into multiple modules. Alternatively, at least some of the functions of one or more of these modules can be combined with at least some of the functions of other modules and implemented in one module. According to embodiments of this application, at least one of the acquisition module 610, map construction module 620, feature evaluation module 630, basic evaluation module 640, and comprehensive evaluation module 650 can be at least partially implemented as hardware circuitry, such as a field-programmable gate array (FPGA), a programmable logic array (PLA), a system-on-a-chip, a system-on-a-substrate, a system-on-package, an application-specific integrated circuit (ASIC), or implemented in hardware or firmware by any other reasonable means of integrating or packaging the circuitry, or implemented in software, hardware, or firmware, or in any suitable combination of any of these three implementation methods. Alternatively, at least one of the acquisition module 610, the map construction module 620, the feature evaluation module 630, the basic evaluation module 640, and the comprehensive evaluation module 650 can be at least partially implemented as a computer program module, which can perform corresponding functions when the computer program module is run.

[0106] Figure 7 A block diagram schematically illustrates an electronic device suitable for implementing the method according to an embodiment of this application.

[0107] like Figure 7 As shown, an electronic device 700 according to an embodiment of this application includes a processor 701, which can perform various appropriate actions and processes according to a program stored in a read-only memory (ROM) 702 or a program loaded from a storage portion 708 into a random access memory (RAM) 703. The processor 701 may include, for example, a general-purpose microprocessor (e.g., a CPU), an instruction set processor and / or an associated chipset and / or a special-purpose microprocessor (e.g., an application-specific integrated circuit (ASIC)), etc. The processor 701 may also include onboard memory for caching purposes. The processor 701 may include a single processing unit or multiple processing units for performing different actions of the method flow according to an embodiment of this application.

[0108] RAM 703 stores various programs and data required for the operation of electronic device 700. Processor 701, ROM 702, and RAM 703 are interconnected via bus 704. Processor 701 executes various operations of the method flow according to embodiments of this application by executing programs in ROM 702 and / or RAM 703. It should be noted that the program may also be stored in one or more memories other than ROM 702 and RAM 703. Processor 701 may also execute various operations of the method flow according to embodiments of this application by executing programs stored in one or more memories.

[0109] According to embodiments of this application, the electronic device 700 may further include an input / output (I / O) interface 705, which is also connected to a bus 704. The electronic device 700 may also include one or more of the following components connected to the I / O interface 705: an input section 706 including a keyboard, mouse, etc.; an output section 707 including a cathode ray tube (CRT), liquid crystal display (LCD), etc., and a speaker, etc.; a storage section 708 including a hard disk, etc.; and a communication section 709 including a network interface card such as a LAN card, modem, etc. The communication section 709 performs communication processing via a network such as the Internet. A drive 710 is also connected to the I / O interface 705 as needed. A removable medium 711, such as a disk, optical disk, magneto-optical disk, semiconductor memory, etc., is installed on the drive 710 as needed so that computer programs read from it can be installed into the storage section 708 as needed.

[0110] This application also provides a computer-readable storage medium, which may be included in the device / apparatus / system described in the above embodiments; or it may exist independently and not assembled into the device / apparatus / system. The computer-readable storage medium carries one or more programs, which, when executed, implement the method according to the embodiments of this application.

[0111] According to embodiments of this application, the computer-readable storage medium can be a non-volatile computer-readable storage medium, such as including but not limited to: portable computer disks, hard disks, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), portable compact disk read-only memory (CD-ROM), optical storage devices, magnetic storage devices, or any suitable combination thereof. In this application, the computer-readable storage medium can be any tangible medium containing or storing a program that can be used by or in conjunction with an instruction execution system, apparatus, or device. For example, according to embodiments of this application, the computer-readable storage medium may include ROM 702 and / or RAM 703 and / or one or more memories other than ROM 702 and RAM 703 described above.

[0112] Embodiments of this application also include a computer program product comprising a computer program containing program code for performing the methods shown in the flowchart. When the computer program product is run on a computer system, the program code enables the computer system to implement the target merchant quality assessment method provided in the embodiments of this application.

[0113] When the computer program is executed by the processor 701, it performs the functions defined in the system / apparatus of this application embodiment. According to the embodiments of this application, the systems, apparatuses, modules, units, etc., described above can be implemented by computer program modules.

[0114] In one embodiment, the computer program may rely on a tangible storage medium such as an optical storage device or a magnetic storage device. In another embodiment, the computer program may also be transmitted and distributed in the form of signals over a network medium, and may be downloaded and installed via the communication section 709, and / or installed from a removable medium 711. The program code contained in the computer program can be transmitted using any suitable network medium, including but not limited to: wireless, wired, etc., or any suitable combination thereof.

[0115] In such an embodiment, the computer program can be downloaded and installed from a network via the communication section 709, and / or installed from the removable medium 711. When the computer program is executed by the processor 701, it performs the functions defined in the system of this application embodiment. According to the embodiments of this application, the systems, devices, apparatuses, modules, units, etc., described above can be implemented by computer program modules.

[0116] According to embodiments of this application, program code for executing the computer programs provided in the embodiments of this application can be written in any combination of one or more programming languages. Specifically, these computational programs can be implemented using high-level procedural and / or object-oriented programming languages, and / or assembly / machine languages. Programming languages ​​include, but are not limited to, languages ​​such as Java, C++, Python, "C", or similar programming languages. The program code can be executed entirely on the user's computing device, partially on the user's device, partially on a remote computing device, or entirely on a remote computing device or server. In cases involving remote computing devices, the remote computing device can be connected to the user's computing device via any type of network, including a local area network (LAN) or a wide area network (WAN), or it can be connected to an external computing device (e.g., via the Internet using an Internet service provider).

[0117] The flowcharts and block diagrams in the accompanying drawings illustrate the architecture, functionality, and operation of possible implementations of systems, methods, and computer program products according to various embodiments of this application. In this regard, each block in a flowchart or block diagram may represent a module, segment, or portion of code containing one or more executable instructions for implementing a specified logical function. It should also be noted that in some alternative implementations, the functions indicated in the blocks may occur in a different order than those indicated in the drawings. For example, two consecutively indicated blocks may actually be executed substantially in parallel, and they may sometimes be executed in reverse order, depending on the functions involved. It should also be noted that each block in a block diagram or flowchart, and combinations of blocks in a block diagram or flowchart, may be implemented using a dedicated hardware-based system that performs the specified function or operation, or using a combination of dedicated hardware and computer instructions.

[0118] Those skilled in the art will understand that the features described in the various embodiments of this application can be combined and / or combined in various ways, even if such combinations or combinations are not explicitly described in this application. In particular, the features described in the various embodiments of this application can be combined and / or combined in various ways without departing from the spirit and teachings of this application. All such combinations and / or combinations fall within the scope of this application.

Claims

1. A method for quality assessment of target merchants, characterized in that, The method includes: The system identifies multiple candidate merchants within a preset area of ​​the target merchant's location, and acquires merchant data for these multiple candidate merchants on multiple consumer platforms. The merchant data includes menu data and rating data for the menu data. The multiple candidate merchants are merchants providing the same target service, and the multiple candidate merchants include the target merchant. Based on the merchant data, multiple dimensional features of the target service are extracted; based on the multiple dimensional features, a regional knowledge graph of the target service is constructed; wherein, the regional knowledge graph includes multiple nodes and multiple edges, each node represents one of the multiple candidate merchants, and each edge represents the association between two candidate merchants determined according to the multiple dimensional features; In the regional knowledge graph, target nodes associated with the target merchant are identified, and the characteristic rating of the target merchant is determined based on the target nodes and the regional knowledge graph. Based on the rating data of the target merchant, determine the basic rating of the target merchant; The quality of the target service from the target merchant is evaluated by combining the feature rating and the base rating.

2. The quality assessment method for target merchants according to claim 1, characterized in that, The regional knowledge graph is constructed in the following way: Create multiple corresponding merchant nodes for each of the multiple candidate merchants; Based on the multiple dimensions of features, calculate the feature similarity between any two candidate merchants among the multiple candidate merchants; If the feature similarity is higher than a preset threshold, an association edge is established between any two candidate merchants, and the feature similarity is used as the association weight of the association edge. The regional knowledge graph is formed based on the merchant nodes and the associated edges.

3. The quality assessment method for target merchants according to claim 2, characterized in that, There are multiple target nodes associated with the target merchant; The step of determining the characteristic rating of the target merchant based on the target node and the regional knowledge graph includes: In the regional knowledge graph, obtain multiple association weights between the target merchant and multiple target nodes; Based on the multiple association weights, calculate the average association weight between the target merchant and the multiple target nodes; The feature score of the target merchant is determined based on the average association weight; wherein, the lower the average association weight, the higher the feature score.

4. The quality assessment method for target merchants according to claim 1, characterized in that, The rating data of the target merchant includes multiple rating data of the target merchant within a preset time period; The determination of the basic rating of the target merchant based on the rating data of the target merchant includes: Record the generation time of each of the multiple rating data, and determine the rating score of each rating data; The weight of each score is calculated based on the difference between the generation time and the current time; wherein the weight decreases as the difference increases. The average service quality of the target merchant within the preset time period is obtained based on the multiple rating scores and the weight of each rating score. The base score of the target merchant is determined based on the average service quality.

5. The quality assessment method for target merchants according to claim 4, characterized in that, The process of determining the basic score of the target merchant based on the average service quality includes: The multiple rating scores are sorted, and abnormal rating scores are detected and removed using the interquartile range rule; Calculate the standard deviation or variance of multiple rating scores after removing the abnormal rating scores, and normalize the calculation results to obtain the degree of fluctuation of the target merchant's rating data within the preset time period; The base score of the target merchant is determined based on the average service quality and the degree of fluctuation.

6. The quality assessment method for target merchants according to claim 5, characterized in that, The process of sorting the multiple rating scores and using the interquartile range rule to detect and remove abnormal rating points includes: Determine the first and third quartiles of the sorted score values; The interquartile range is calculated based on the first quartile and the third quartile, and an outlier determination interval is set based on the interquartile range. The scores that exceed the judgment range are marked as abnormal scores and removed.

7. The quality assessment method for target merchants according to claim 1, characterized in that, The determination of the basic rating of the target merchant based on the rating data of the target merchant includes: Multiple sets of rating data are input into a pre-trained large model, which outputs sentiment scores for the multiple sets of rating data; the sentiment scores represent the quantitative strength of the sentiment tendency of the rating data towards the target merchant; Calculate the weighted average of the sentiment scores from multiple rating data sets as the reputation score; Based on the reputation score, the basic rating of the target merchant is determined.

8. A quality assessment device for target merchants, characterized in that, The device includes: The acquisition module is used to determine multiple candidate merchants within a preset area of ​​the target merchant's location, and acquire merchant data of the multiple candidate merchants on multiple consumer platforms. The merchant data includes menu data and rating data for the menu data. The multiple candidate merchants are merchants that provide the same target service, and the multiple candidate merchants include the target merchant. The graph construction module is used to extract multiple dimensional features of the target service based on the merchant data; and to construct a regional knowledge graph of the target service based on the multiple dimensional features; wherein the regional knowledge graph includes multiple nodes and multiple edges, each node represents one of the multiple candidate merchants, and each edge represents the association between two candidate merchants determined according to the multiple dimensional features; The feature evaluation module is used to identify target nodes associated with the target merchant in the regional knowledge graph, and determine the feature score of the target merchant based on the target nodes and the regional knowledge graph. The basic assessment module is used to determine the basic score of the target merchant based on the rating data of the target merchant; The comprehensive evaluation module is used to combine the feature score and the basic score to evaluate the quality of the target service of the target merchant.

9. An electronic device, comprising: One or more processors; Memory, used to store one or more computer programs. The characteristic feature is that the one or more processors execute the one or more computer programs to implement the steps of the method according to any one of claims 1 to 7.

10. A computer-readable storage medium having a computer program or instructions stored thereon, characterized in that, When the computer program or instructions are executed by a processor, they implement the steps of the method according to any one of claims 1 to 7.

11. A computer program product, comprising a computer program or instructions, characterized in that, When the computer program or instructions are executed by a processor, they implement the steps of the method according to any one of claims 1 to 7.

Images