A service scheme matching method and system based on customer portrait

Abstract

The present application relates to the technical field of computer, in particular to a kind of service scheme matching method and system based on customer portrait, method includes: obtaining the scheme description text, rule text and target customer business data of matching service scheme;The rule text is carried out semantic segmentation and candidate rule identification, and the adaptive condition set is obtained;According to adaptive condition set and preset portrait dimension, evidence item model is constructed, and after being converted into constraint check template, it is written into service semantic atlas;According to customer business data, the customer portrait atlas including business potential index, performance cost index and relationship resilience index is generated;Customer portrait atlas and service semantic atlas are semantically aligned, and the mapping of mapping evidence item and to-be-checked object is established;According to three kinds of indexes, portrait partition is generated and condition check is executed, and according to portrait partition and check result, matching result is generated, and recommendation reason and coordination action are output.

Description

Technical Field

[0001] This invention relates to the field of computer technology, and in particular to a service solution matching method and system based on customer profiles. Background Technology

[0002] With the development of internet service platforms, service recommendation has gradually shifted from manual screening to automated processing based on order data, service data, and user data. In existing technologies, patent application number 2022103553294 discloses an order monitoring method and platform, which generates order information and updates order status using information such as order popularity, warehouse data, and user level; patent application number 2023100150987 discloses an online real-time statistical method and system for home delivery order data, which improves the accuracy of service personnel assignment and service statistics by receiving order information, assigning order tasks, obtaining service personnel service status, and statistically analyzing order data; and patent application number 2022115383275 discloses an intelligent push optimization method based on platform big data feedback, which optimizes push content through user data collection, data classification, and decision tree processing.

[0003] The aforementioned existing technologies can improve the platform's processing efficiency in order monitoring, service process statistics, and information push. However, when service solutions have multiple adaptation conditions, customer data sources are scattered, and the basis for recommendations needs to be explained, they still mainly rely on order status, service records, or user feedback for judgment. It is difficult to uniformly link service rules with customer business data, and it is also difficult to differentiate service solutions from the perspectives of customer development status, fulfillment costs, and relationship stability. As a result, there is still room for improvement in the accuracy, interpretability, and subsequent coordination and processing capabilities of the recommendation results. Summary of the Invention

[0004] In view of the above technical problems, the present invention provides a service solution matching method and system based on customer profiles, aiming to solve the problems in the existing service solution recommendation process, such as the difficulty in using service rules in a structured manner, the difficulty in accurately matching customer information with service conditions, the lack of clear basis for recommendation results, and the lack of effective coordination actions when conditions are not met, thereby improving the accuracy, interpretability and processing efficiency of service solution matching.

[0005] Other features and advantages of the invention will become apparent from the following detailed description, or may be learned in part by practice of the invention.

[0006] According to one aspect of the present invention, a service solution matching method based on customer profiles is proposed, the method comprising: Obtain the solution description text, rule text, and target customer business information corresponding to the service solution to be matched; The rule text is semantically segmented and candidate rule is identified to obtain a set of adaptation conditions related to the service scheme to be matched. Based on the set of adaptation conditions and the preset customer profile dimensions, an evidence item model is constructed. The evidence item model includes semantic associations between customer subjects, business activities, service consumption status, relationship persistence status and development potential status. The evidence item model and the set of adaptation conditions are converted into a constraint verification template, and the service scheme to be matched, the set of adaptation conditions, the evidence item model, and the constraint verification template are associated and written into the service semantic graph. A customer profile map is generated based on the customer business data. The customer profile map includes business potential indicators, performance cost indicators and relationship resilience indicators. The business potential indicators are generated based on changes in customer business activities and the development status of the customer's field. The performance cost indicators are generated based on service reach frequency and service delivery distance. The relationship resilience indicators are generated based on the continuity of the relationship between the customer and the service provider. Semantically align the customer profile graph with the service semantic graph to establish a mapping between the mapping evidence items in the customer profile graph and the objects to be verified in the constraint verification template. Based on the business potential index, the performance cost index, and the relationship resilience index, a profile partition of the target customer is generated, and the constraint verification template is executed to perform condition verification on the evidence items in the customer profile map to obtain the verification result. A service plan matching result is generated based on the profile partition and the verification result. The verification result is used to determine whether the service plan to be matched is suitable for the target customer. The profile partition is used to determine the recommendation order or coordination action type. The recommendation reason and coordination action are output based on the matching conditions that are met or not met in the verification result.

[0007] Furthermore, the rule text is semantically segmented and candidate rule is identified, including generating search suggestions based on the name of the service scheme to be matched, the service object and the processing item, dividing the rule text into rule fragments, extracting the matching condition text from the rule fragments, removing the supporting material descriptions that are irrelevant to the matching condition text, and obtaining the matching condition set.

[0008] Furthermore, constructing the evidence item model includes: converting customer attributes, customer relationships, business activities, performance resources, and historical transaction information in the set of adaptation conditions into basic evidence items; hierarchically associating multiple basic evidence items according to subject relationships, time relationships, and business affiliation relationships; and excluding computational conclusions derived from the basic evidence items that are not used as verification criteria from the evidence item model.

[0009] Furthermore, the conversion of the constraint verification template includes converting the admission conditions, exclusion conditions, exception conditions and dependency conditions in the adaptation condition set into corresponding verification nodes, configuring the verification nodes with subordinate relationships as parent-child constraint structures, and writing the parent-child constraint structures into the service semantic graph after performing consistency verification.

[0010] Furthermore, generating the customer profile graph includes: querying the data source directory according to the evidence item model, determining the data system that stores each basic evidence item, requesting the minimum evidence data related to the service solution to be matched from the data system, and writing the minimum evidence data into the customer profile graph according to the evidence item model.

[0011] Furthermore, the generation of the business potential index includes forming a potential level based on changes in customer business volume, frequency of changes in customer service content, and development status of the customer's field; the generation of the performance cost index includes forming a cost level based on service reach frequency, service performance distance, and service resource occupancy status; and the generation of the relationship resilience index includes forming a resilience level based on contract continuity status, stable interaction status, and historical service continuity status.

[0012] Furthermore, generating the profile partition includes comparing the business potential index, the performance cost index, and the relationship resilience index with the corresponding adaptive boundary conditions to obtain the level results of each profile dimension, and determining the profile partition in which the customer is located according to the combination of the level results.

[0013] Furthermore, generating the service solution matching result includes: if the verification result shows that the target customer meets the set of adaptation conditions, determining the recommended order of the service solution based on the profile partition, the business potential index, the performance cost index, and the relationship resilience index; if the verification result shows that the target customer does not meet the set of adaptation conditions, generating the corresponding actions among the evidence completion action, manual review action, and solution adjustment action as the coordination action.

[0014] Furthermore, the method also includes recording changes in customer behavior, service performance, and evidence gaps after the service scheme matching result is adopted; updating adaptive boundary conditions and recommendation ranking rules based on the recording results; and, when an update to the rule text is detected, re-segmenting the updated rule text semantically and identifying candidate rules, and updating the constraint verification template and the service semantic graph.

[0015] According to a second aspect of the present invention, a service solution matching system based on customer profiles is provided, the system comprising: The data acquisition module is used to obtain the solution description text, rule text, and customer business information of the target customer corresponding to the service solution to be matched; The semantic parsing module is used to perform semantic segmentation and candidate rule identification on the rule text to obtain a set of adaptation conditions related to the service scheme to be matched; The evidence item modeling module is used to construct an evidence item model based on the set of adaptation conditions and the preset customer profile dimensions. The evidence item model includes semantic associations between customer subjects, business activities, service consumption status, relationship persistence status and development potential status. The service graph construction module is used to convert the evidence item model and the adaptation condition set into a constraint verification template, and associate the service scheme to be matched, the adaptation condition set, the evidence item model and the constraint verification template into the service semantic graph; The customer profile building module is used to generate a customer profile map based on the customer business data. The customer profile map includes business potential indicators, performance cost indicators and relationship resilience indicators. The business potential indicators are generated based on changes in customer business activities and the development status of the customer's field. The performance cost indicators are generated based on service reach frequency and service delivery distance. The relationship resilience indicators are generated based on the continuity of the relationship between the customer and the service provider. The graph mapping module is used to semantically align the customer profile graph with the service semantic graph, so that the mapping evidence items in the customer profile graph are mapped to the objects to be verified in the constraint verification template. The condition verification module is used to generate the profile partition of the target customer based on the business potential index, the performance cost index and the relationship resilience index, and execute the constraint verification template to perform condition verification on the evidence items in the customer profile map to obtain the verification result. The matching decision module is used to generate service solution matching results based on the profile partitioning and the verification results. The verification results are used to determine whether the service solution to be matched is suitable for the target customer. The profile partitioning is used to determine the recommendation order or coordination action type, and outputs recommendation reasons and coordination actions based on the suitability conditions met or not met in the verification results.

[0016] The technical solution of the present invention has the following beneficial effects: This invention links the rules corresponding to service plans with customer business data, enabling customer-side information to form a profile for service plan judgment. Conditional validation determines the compatibility between the customer and the service plan, reducing omissions caused by manual screening and experience-based judgment. Furthermore, this invention can create profile partitions based on the customer's business potential, fulfillment costs, and relationship resilience. This ensures that the matching results not only reflect whether the service plan requirements are met, but also support further coordination processes such as determining recommendation priority, outputting recommendation reasons, and supplementing evidence, manual review, or plan adjustments. Therefore, this invention improves the targeting and traceability of service recommendations and supports continuous updates to matching results when customer status or service rules change. Attached Figure Description

[0017] Figure 1 This is a flowchart of a service solution matching method based on customer profile in an embodiment of this specification; Figure 2 This is a structural block diagram of a service solution matching system based on customer profiles, as described in an embodiment of this specification. Detailed Implementation

[0018] Exemplary embodiments will now be described more fully with reference to the accompanying drawings. However, these exemplary embodiments can be implemented in many forms and should not be construed as limited to the examples set forth herein; rather, these embodiments are provided to make the invention more comprehensive and complete, and to fully convey the concept of the exemplary embodiments to those skilled in the art. The described features, structures, or characteristics can be combined in any suitable manner in one or more embodiments. In the following description, numerous specific details are provided to give a full understanding of embodiments of the invention. However, those skilled in the art will recognize that the technical solutions of the invention may be practiced with one or more of these specific details omitted, or other methods, components, apparatus, steps, etc., may be employed. In other instances, well-known technical solutions are not shown or described in detail to avoid obscuring various aspects of the invention.

[0019] Furthermore, the accompanying drawings are merely illustrative of the invention. The same reference numerals in the drawings denote the same or similar parts, and therefore repeated descriptions of them will be omitted. Some block diagrams shown in the drawings are functional entities and do not necessarily correspond to physically or logically independent entities. These functional entities can be implemented in software, in one or more hardware modules or integrated circuits, or in different network and / or processor devices and / or microcontroller devices.

[0020] This invention provides a service solution matching method based on customer profiles. (Refer to...) Figure 1The diagram shown is a flowchart illustrating a service solution matching method based on customer profiles according to an embodiment of the present invention. This method can be applied to devices such as personal computers and servers. The method can be executed by a device, which can be implemented by software and / or hardware. Specifically, the method may include the following steps S101-S108: In step S101, the solution description text, rule text, and customer business information of the target customer corresponding to the service solution to be matched are obtained.

[0021] The service solutions to be matched can be service items that the platform intends to recommend, configure, or adjust to target customers. The solution description text characterizes the basic content of the service item, and may include information such as service name, service recipients, processing items, service content, scope of application, service version, and effective date. The rule text characterizes the source of the business rules corresponding to the service item, and may include natural language text such as service access requirements, restriction requirements, exception requirements, document requirements, and processing instructions. The target customer's business information characterizes the target customer's current and historical business status, and may include customer entity information, business activity records, contract or cooperation records, service contact records, service performance records, historical transaction records, customer's industry information, and customer feedback information. By simultaneously obtaining the solution description text, rule text, and customer business information in this step, a unified data input foundation can be provided for subsequent rule identification, evidence item modeling, profile generation, and service solution matching, avoiding service recommendations based solely on a single order record, a single customer tag, or human experience. Service rule texts and customer information can respectively correspond to the input basis for service condition modeling, evidence sources, and user profile construction. Three-dimensional customer partition data can correspond to customer status information such as customer development, performance costs, and relationship value.

[0022] In step S102, the rule text is semantically segmented and candidate rules are identified to obtain a set of adaptation conditions related to the service scheme to be matched.

[0023] The process of semantic segmentation and candidate rule identification of the rule text includes generating search suggestions based on the name, service object, and processing item of the service solution to be matched, dividing the rule text into rule fragments, extracting matching condition text from the rule fragments, removing supporting material descriptions that are irrelevant to the matching condition text, and obtaining the matching condition set.

[0024] The rule text can be divided into multiple rule fragments according to clause level, semantic pauses, conditional connectors, and matter boundaries, so that each rule fragment corresponds to a relatively independent service requirement. Search prompts are used to limit the direction of rule identification and can be formed by combining the name of the service solution to be matched, the service recipient, and the processing item. This allows the semantic segmentation process to focus on the text content related to service compatibility judgment, reducing interference from irrelevant explanations, processing procedures, and material listings on rule extraction.

[0025] During candidate rule identification, statements with conditional meanings can be identified from rule fragments. Adaptation condition text can include customer eligibility, business activity status, relationship continuity requirements, service consumption requirements, time requirements, exclusionary situations, exceptions, and dependencies. Material names, certificate names, attachment descriptions, or submission methods used solely to prove a condition are not directly retained as adaptation conditions but are instead processed as data source hints or evidence source information during subsequent evidence item modeling.

[0026] In practical processing, candidate annotations can be applied to rule fragments. Text containing semantics such as "must meet," "must not exist," "except," "depends on," "reaches," "below," "above," "continuous," and "completed" can be identified as candidate adaptation condition texts. Text containing descriptions of application materials, supporting documents, submission paths, processing windows, and format requirements that do not directly limit service adaptation relationships can be eliminated or downgraded. After candidate identification and irrelevant description removal, each adaptation condition in the adaptation condition set can correspond to a type of collectable or verifiable business status of the target customer, thereby avoiding mistaking descriptions of supporting materials for service adaptation rules.

[0027] In step S103, an evidence item model is constructed based on the set of adaptation conditions and the preset customer profile dimensions. The evidence item model includes semantic associations between customer subjects, business activities, service consumption status, relationship persistence status and development potential status.

[0028] Constructing the evidence item model includes: converting customer attributes, customer relationships, business activities, performance resources, and historical transaction information in the set of adaptation conditions into basic evidence items; hierarchically associating multiple basic evidence items according to subject relationships, time relationships, and business affiliation relationships; and excluding computational conclusions derived from the basic evidence items that are not used as verification criteria from the evidence item model.

[0029] Among them, the preset customer profile dimension is used to limit the organizational direction of the evidence item model, so that the rule content in the adaptation condition set can be transformed into collectable, traceable, and verifiable customer-side facts. The customer subject is used to represent the identity, type, domain, and service object attributes of the target customer. Business activities are used to represent the business changes, service content changes, and transaction behaviors of the target customer within a certain time range. Service consumption status is used to represent the status related to service performance costs, such as service reach frequency, service performance distance, and performance resource consumption. Relationship continuity status is used to represent the contract continuity, interaction stability, and historical service continuity. Development potential status is used to represent the target customer's business growth, service demand expansion, and domain development.

[0030] When constructing an evidence item model, customer attributes, customer relationships, business activities, performance resources, and historical transaction information from the set of applicable conditions can be converted into basic evidence items. Basic evidence items are factual data that can be directly verified by customer business information, business system records, or external trusted data sources, such as customer type, contract status, business volume change records, service outreach records, performance resource usage records, and historical transaction records. Multiple basic evidence items are hierarchically linked according to subject relationships, temporal relationships, and business attribution relationships, enabling different business facts under the same target customer to form a traceable chain of evidence. Subject relationships are used to determine that a basic evidence item belongs to the target customer or its related entities; temporal relationships are used to determine the occurrence time, duration, or effective period of the basic evidence item; and business attribution relationships are used to determine the service item, business activity, or performance process corresponding to the basic evidence item.

[0031] Computational conclusions derived from basic evidence items but not directly used as the basis for condition verification are not included in the evidence item model as basic verification objects. These computational conclusions can be used in subsequent customer profile indicator generation, but in this step, only the basic evidence items that can support condition verification and their semantic relationships are retained.

[0032] In step S104, the evidence item model and the set of adaptation conditions are converted into a constraint verification template, and the service scheme to be matched, the set of adaptation conditions, the evidence item model, and the constraint verification template are associated and written into the service semantic graph.

[0033] The process of converting the constraint verification template includes converting the admission conditions, exclusion conditions, exception conditions and dependency conditions in the adaptation condition set into corresponding verification nodes, configuring the verification nodes with subordinate relationships as parent-child constraint structures, and writing the parent-child constraint structures into the service semantic graph after performing consistency verification.

[0034] The constraint verification template is used to transform service requirements in natural language form from the adaptation condition set into an executable verification structure. The evidence item model provides the objects to be verified and their semantic relationships, while the adaptation condition set provides the rules that each object to be verified must meet or exclude. By converting both into a constraint verification template, the adaptation judgment of the service solution can be transformed from manual interpretation to automatic verification based on verification nodes.

[0035] Admission criteria can be converted into verification nodes to determine whether a target customer is eligible for the service; exclusion criteria can be converted into verification nodes to determine whether a target customer is unsuitable for the service; exception criteria can be converted into verification nodes to correct general judgment results; and dependency criteria can be converted into verification nodes to limit the conditions under which other verification nodes take effect. Each verification node can include fields such as node identifier, object to be verified, comparison relationship, threshold or enumeration value, evidence source, verification result type, and explanatory text, enabling the verification node to correspond to the mapped evidence items in the customer profile map.

[0036] Validation nodes with hierarchical relationships can be configured as a parent-child constraint structure. The parent node represents the primary adaptation condition, and the child node represents supplementary, exceptional, exclusionary, or dependent conditions under that primary adaptation condition. During consistency verification of the parent-child constraint structure, it is possible to check whether the reference relationship between parent and child nodes is complete, whether there are conflicting condition directions, whether there are duplicate constraints on the same object to be verified, whether exceptional conditions have their corresponding scope of application, and whether dependent conditions can be confirmed beforehand during the verification process. The constraint verification template, after consistency verification, is associated with the service scheme to be matched, the set of adaptation conditions, and the evidence item model and written into the service semantic graph, enabling the service semantic graph to simultaneously express the correspondence between service schemes, adaptation rules, evidence objects, and verification logic.

[0037] In step S105, a customer profile map is generated based on the customer business data. The customer profile map includes business potential indicators, performance cost indicators, and relationship resilience indicators. The business potential indicators are generated based on changes in customer business activities and the development status of the customer's field. The performance cost indicators are generated based on service reach frequency and service delivery distance. The relationship resilience indicators are generated based on the continuity of the relationship between the customer and the service provider.

[0038] Generating the customer profile map includes: querying the data source directory according to the evidence item model, determining the data system that stores each basic evidence item, requesting the minimum evidence data related to the service plan to be matched from the data system, and writing the minimum evidence data into the customer profile map according to the evidence item model.

[0039] The business potential indicators are generated by forming potential levels based on changes in customer business volume, frequency of changes in customer service content, and development status of the customer's industry. The performance cost indicators are generated by forming cost levels based on service reach frequency, service performance distance, and service resource occupancy status. The relationship resilience indicators are generated by forming resilience levels based on contract continuity status, stable interaction status, and historical service continuity status.

[0040] In this embodiment, the customer profile graph is used to receive factual data and profile metrics from the target customer side. When querying the data source directory based on the evidence item model, the data source directory can record the correspondence between basic evidence items and data systems, enabling customer attributes, customer relationships, business activities, fulfillment resources, and historical transaction information to be located in the customer management system, order system, contract system, fulfillment system, or external trusted data source, respectively. Minimal evidence data refers to data items directly related to the condition verification and profile metric generation of the service solution to be matched. Data not referenced by the constraint verification template and not involved in the generation of profile metrics are not considered necessary request objects. By writing minimal evidence data into the customer profile graph, the introduction of irrelevant data can be reduced.

[0041] The customer profile map can organize minimal evidence data according to the subject relationships, temporal relationships, and business affiliation relationships in the evidence item model. The target customer serves as the subject node in the customer profile map, while business activity records, service contact records, performance resource records, contract records, interaction records, and historical service records serve as evidence nodes associated with the subject node. Each evidence node retains its source system, collection time, validity period, and associated business matter, enabling the customer profile map to express both the target customer's current state and changes in state within a certain time frame.

[0042] Business potential indicators characterize the likelihood of increased demand for or expansion of services from target customers. Potential levels can be established based on changes in customer business volume, frequency of changes in customer service content, and the development status of the customer's industry. Changes in customer business volume characterize the trend of increase or decrease in the target customer's business scale within a preset statistical period. Frequency of changes in customer service content characterizes whether the target customer's customer service needs continue to expand or adjust. The development status of the customer's industry characterizes the growth potential of the target customer's industry or business sector. Potential levels can be represented using discrete levels; the higher the level, the higher the target customer's development potential.

[0043] The performance cost index is used to characterize the intensity of resources required to perform services for target customers. It can be used to form a cost level based on service reach frequency, service performance distance and service resource occupancy status.

[0044] Relationship resilience is used to characterize the stability of the relationship between a target customer and a service provider. It can be categorized into resilience levels based on contract continuity, interaction stability, and historical service continuity. Contract continuity indicates whether the partnership is valid, renewed, or interrupted; interaction stability indicates communication frequency and feedback stability; and historical service continuity indicates whether the target customer has continuously used or received the service throughout the historical period. A higher resilience level indicates a more stable relationship between the target customer and the service provider.

[0045] In step S106, the customer profile graph and the service semantic graph are semantically aligned so that the mapping evidence items in the customer profile graph are mapped to the objects to be verified in the constraint verification template.

[0046] Semantic alignment establishes a correspondence between customer-side facts in the customer profile graph and rule-side objects in the service semantic graph, enabling subsequent condition validation to directly access the target customer's evidence data. Mapped evidence items in the customer profile graph can be data nodes such as customer type, business activity records, service reach records, performance resource records, contract status, interaction records, historical service records, and business volume change records. The objects to be validated in the constraint validation template can be object fields corresponding to admission conditions, exclusion conditions, exception conditions, or dependency conditions.

[0047] When performing semantic alignment, matching can be performed based on evidence item name, evidence item type, customer entity identifier, business affiliation, time range, data unit, and conditional semantics. Evidence item name is used to identify the meaning of data in the customer profile; evidence item type is used to distinguish categories such as entity attributes, business activities, service consumption, relationship continuity, and development potential; customer entity identifier is used to ensure that the mapped evidence item belongs to the target customer; business affiliation is used to determine whether the mapped evidence item corresponds to a service item related to the service plan to be matched; and time range is used to determine whether the mapped evidence item is within the valid period required by the constraint verification template.

[0048] When the mapped evidence items in the customer profile graph differ in name but are semantically consistent with the object to be verified, mappings can be established through synonyms, hierarchical relationships, or business affiliations in the service semantic graph. For example, the contract validity status in the customer profile graph can be mapped to the relationship continuity requirement in the constraint verification template, the number of service outreaches can be mapped to the service outreach frequency requirement, and the business volume growth record can be mapped to the development potential status requirement. After semantic alignment is completed, each object to be verified can be associated with a corresponding mapped evidence item or evidence gap marker, enabling subsequent condition verification to distinguish between conditions met, conditions not met, and insufficient evidence.

[0049] In step S107, a profile partition of the target customer is generated based on the business potential index, the performance cost index and the relationship resilience index, and the constraint verification template is executed to perform condition verification on the evidence items in the customer profile map to obtain the verification result.

[0050] Generating the profile partition includes comparing the business potential index, the performance cost index, and the relationship resilience index with the corresponding adaptive boundary conditions to obtain the level results of each profile dimension, and determining the profile partition in which the customer is located according to the combination of the level results.

[0051] Among them, business potential indicators, fulfillment cost indicators, and relationship resilience indicators serve as the three profile dimensions for the profiling partition. Business potential indicators represent the target customer's business growth and service demand expansion status; fulfillment cost indicators represent the consumption status of contact frequency, fulfillment distance, and resource usage during service fulfillment; and relationship resilience indicators represent the stable state of the relationship between the target customer and the service provider. Each profile dimension can be divided into different levels according to adaptive boundary conditions, and the profile partition in which the target customer belongs is determined by the combination of these levels.

[0052] When each portrait dimension is represented using discrete ranks, the rank results can be normalized to ensure that different portrait dimensions are within a unified comparison range. The normalized rank value can be determined using the following formula: ; in, Indicates the first The normalized rating value of a target customer in the l-th profile dimension This represents the original rating value of the i-th target customer in the l-th profile dimension. This represents the maximum number of levels corresponding to the l-th portrait dimension. Through this processing, even if different portrait dimensions have different numbers of levels, they can be converted into a unified range of partitioning criteria.

[0053] Adaptive boundary conditions can include potential boundary conditions corresponding to business potential indicators, cost boundary conditions corresponding to performance cost indicators, and resilience boundary conditions corresponding to relationship resilience indicators. Comparing business potential indicators with potential boundary conditions yields potential level results; comparing performance cost indicators with cost boundary conditions yields cost level results; and comparing relationship resilience indicators with resilience boundary conditions yields resilience level results. Profile partitions are determined by combinations of potential level results, cost level results, and resilience level results. For example, when each profile dimension uses a two-tiered division (high and low), the three profile dimensions can be combined to form eight profile partitions, each corresponding to the target customer's combined status in terms of development potential, performance cost, and relationship stability.

[0054] When it's necessary to determine the proximity between a target customer and a profile partition center, a profile vector can be constructed using business potential indicators, fulfillment cost indicators, and relationship resilience indicators. A distance function can then be used to determine the distance between the target customer and the candidate profile partition center. The distance function can be determined using the following formula: ; in, This represents the distance between the i-th image vector and the j-th image vector, where d represents the number of image dimensions. This represents the value of the i-th image vector in the l-th image dimension. This represents the value of the j-th profile vector in the l-th profile dimension, where p represents the distance order. When p is 1, it corresponds to the distance accumulated by the absolute difference of dimensions; when p is 2, it corresponds to the distance calculated by the square root of the difference of squares. This distance function can be used to assign target customers to profile partitions that are closer when they are in the critical boundary area.

[0055] Once the customer profile partitions are determined, the constraint verification template is executed to perform condition verification on the evidence items in the customer profile graph. Condition verification can be performed according to the verification nodes corresponding to admission conditions, exclusion conditions, exception conditions, and dependency conditions, or according to the dependency order determined by the parent-child constraint structure. For admission conditions, it is determined whether the evidence items in the customer profile graph meet the corresponding requirements; for exclusion conditions, it is determined whether there are situations that would cause the service solution to be unsuitable; for exception conditions, it is determined whether the general verification results need to be corrected; for dependency conditions, it is determined whether the associated verification nodes have a basis for effectiveness. Verification results can include states such as satisfied, unsatisfied, missing evidence, and requiring review, and the corresponding adaptation conditions, mapped evidence items, and verification descriptions are recorded for subsequent generation of service solution matching results, recommendation reasons, and coordination actions.

[0056] In step S108, a service plan matching result is generated based on the profile partition and the verification result. The verification result is used to determine whether the service plan to be matched is suitable for the target customer. The profile partition is used to determine the recommendation order or coordination action type, and the recommendation reason and coordination action are output based on the matching conditions that are met or not met in the verification result.

[0057] Generating the service solution matching result includes: if the verification result shows that the target customer meets the set of matching conditions, determining the recommended order of the service solution based on the profile partition, the business potential index, the performance cost index, and the relationship resilience index; if the verification result shows that the target customer does not meet the set of matching conditions, generating the corresponding actions among the evidence completion action, manual review action, and solution adjustment action as the coordination action.

[0058] In this embodiment, the service solution matching result may include an adaptation status, recommendation ranking, recommendation reasons, and coordination actions. The adaptation status is determined by the verification result and indicates whether the service solution to be matched is suitable for the target customer; the recommendation ranking is jointly determined by the profile partitioning and business potential indicators, performance cost indicators, and relationship resilience indicators, and indicates the priority of the service solution to be matched among candidate services; the recommendation reasons are generated by the met or unmet adaptation conditions and corresponding evidence items, and explain the basis for the formation of the matching result; the coordination actions are used to provide processing direction when there is insufficient evidence, conflicting conditions, or incompatible solutions.

[0059] When the verification results indicate that the target customer meets the set of suitability criteria, the recommendation type of the service solution to be matched can be determined based on the customer profile partitioning. Then, the recommendation ranking is determined by combining business potential indicators, fulfillment cost indicators, and relationship resilience indicators. High business potential indicators and high relationship resilience indicators indicate that the target customer has strong service extension value and a relatively stable relationship foundation, thus increasing the recommendation ranking of the service solution to be matched. High fulfillment cost indicators indicate that the service fulfillment requires more resources, thus decreasing the recommendation ranking or marking the service solution to be matched as requiring resource coordination before recommendation. Low business potential indicators and low relationship resilience indicators suggest that the service solution to be matched can be recommended as a low-priority option or an observation recommendation. The reasons for the recommendation can include the entry conditions met by the target customer, untriggered exclusion conditions, met dependency conditions, and applicable exception conditions, making the recommendation results traceable.

[0060] When the verification results indicate that the target customer does not meet the set of matching conditions, a coordination action can be generated based on the type of non-compliance. If the non-compliance result stems from a lack of mapping evidence items in the customer profile map, an evidence completion action is generated to prompt for supplementing customer information, business records, or performance data. If the non-compliance result stems from conflicts between evidence items, abnormal time ranges, or unconfirmed parent-child constraint structures, a manual review action is generated to forward the request to business personnel for confirmation. If the non-compliance result indicates a significant deviation between the target customer's current status and the admission conditions of the service plan to be matched, a plan adjustment action is generated to prompt for changing the service plan, adjusting the service content, or postponing the recommendation.

[0061] Recommendation reasons can be generated based on the adaptation conditions corresponding to the verification nodes. For met adaptation conditions, the corresponding mapped evidence items, evidence sources, and satisfaction status are output; for unmet adaptation conditions, the reasons for non-satisfaction, missing or conflicting evidence are output, along with corresponding coordination actions. Through this process, the service solution matching results can not only provide a conclusion on whether to recommend a solution, but also explain why it is recommended, why it is not recommended at this time, and the subsequent processing actions that need to be taken. In one embodiment, the method further includes recording changes in customer behavior, service performance, and evidence gaps after the service scheme matching result is adopted; updating adaptive boundary conditions and recommendation ranking rules based on the recording results; and, when an update to the rule text is detected, re-segmenting the updated rule text semantically and identifying candidate rules, and updating the constraint verification template and the service semantic graph.

[0062] In this embodiment, after the service plan matching result is adopted, subsequent changes in customer behavior, service performance, and evidence gaps of the target customer can be recorded as feedback. Changes in customer behavior may include whether the target customer views recommended content, accepts the service plan, continues the service, generates new business activities, or adjusts the service content. Changes in service performance may include whether the service is completed on schedule, whether performance resources exceed expectations, whether the number of service outreaches increases, whether the service delivery distance changes, and the feedback status after service completion. Changes in evidence gaps may include whether missing evidence is supplemented, whether abnormal evidence is confirmed, whether conflicting evidence is corrected, and whether the results of manual review are written back.

[0063] When updating adaptive boundary conditions based on recorded results, the profile partitions corresponding to the adopted service solutions can be compared with subsequent fulfillment results. If target customers within a certain profile partition consistently demonstrate high service acceptance rates and low fulfillment costs, the boundary conditions for the corresponding profile dimension can be appropriately adjusted to make it easier for similar customers to enter that profile partition. If evidence gaps, excessive fulfillment resource consumption, or poor service feedback frequently occur within a certain profile partition, the corresponding boundary conditions can be tightened or the recommendation priority of that profile partition can be reduced. Updating adaptive boundary conditions does not change the basic evidence items in the customer profile map; it only changes the judgment boundaries used when forming grade results for business potential indicators, fulfillment cost indicators, and relationship resilience indicators.

[0064] When updating recommendation ranking rules based on recorded results, changes in customer behavior and service performance can be used as feedback on recommendation effectiveness, ensuring that the ranking rules reflect the actual impact of service solutions after adoption. For service solutions that demonstrate stable service completion, positive customer feedback, and reduced evidence gaps after adoption, their recommendation ranking within the same profile area can be increased. Conversely, for service solutions that experience a significant increase in performance costs, repeated evidence gaps, or a high proportion of manual review after adoption, their recommendation ranking can be decreased or they can be reclassified as requiring further coordination. Updates to recommendation ranking rules can retain the original rule version and update time, facilitating subsequent tracing of the basis for matching results.

[0065] When an update to the rule text is detected, the updated rule text can be reprocessed as a rule recognition object. The system re-segments the updated rule text semantically and identifies candidate rules to obtain a new set of adaptation conditions, and updates the constraint verification template based on the new set of adaptation conditions. If the newly added adaptation conditions involve new objects to be verified, corresponding verification nodes and evidence item associations can be added to the service semantic graph; if the original adaptation conditions are deleted or changed, the corresponding verification nodes can be marked as invalid or replaced with versions; if the parent-child constraint structure changes, consistency verification can be performed again before writing it into the service semantic graph.

[0066] Based on the same line of thought, such as Figure 2 The diagram shown is a structural block diagram of a service solution matching system based on customer profiles provided in an embodiment of the present invention. The system includes: Data acquisition module 201 is used to acquire the solution description text, rule text and target customer business information corresponding to the service solution to be matched; The semantic parsing module 202 is used to perform semantic segmentation and candidate rule recognition on the rule text to obtain a set of adaptation conditions related to the service scheme to be matched; The evidence item modeling module 203 is used to construct an evidence item model based on the set of adaptation conditions and the preset customer profile dimensions. The evidence item model includes semantic associations between customer subjects, business activities, service consumption status, relationship persistence status and development potential status. The service graph construction module 204 is used to convert the evidence item model and the set of adaptation conditions into a constraint verification template, and associate the service scheme to be matched, the set of adaptation conditions, the evidence item model and the constraint verification template into the service semantic graph; The customer profile building module 205 is used to generate a customer profile map based on the customer business data. The customer profile map includes business potential indicators, performance cost indicators and relationship resilience indicators. The business potential indicators are generated based on changes in customer business activities and the development status of the customer's field. The performance cost indicators are generated based on service reach frequency and service delivery distance. The relationship resilience indicators are generated based on the continuous relationship status between the customer and the service provider. The graph mapping module 206 is used to semantically align the customer profile graph with the service semantic graph, so that the mapping evidence items in the customer profile graph are mapped to the objects to be verified in the constraint verification template. The condition verification module 207 is used to generate the profile partition of the target customer based on the business potential index, the performance cost index and the relationship resilience index, and execute the constraint verification template to perform condition verification on the evidence items in the customer profile map to obtain the verification result. The matching decision module 208 is used to generate service solution matching results based on the profile partition and the verification results. The verification results are used to determine whether the service solution to be matched is suitable for the target customer. The profile partition is used to determine the recommendation order or coordination action type, and outputs recommendation reasons and coordination actions based on the suitability conditions met or not met in the verification results.

[0067] This system links the rules corresponding to service plans with customer business data, enabling customer-side information to form a profile for service plan evaluation. Conditional validation determines the compatibility between customers and service plans, reducing omissions caused by manual screening and experience-based judgment. Simultaneously, the system can create profile partitions based on customer business potential, performance costs, and relationship resilience. This ensures that matching results not only reflect whether service plan requirements are met but also support further coordination processes such as recommendation priority determination, recommendation reason output, evidence supplementation, manual review, or plan adjustment. Therefore, this invention improves the relevance and traceability of service recommendations and supports continuous updates to matching results when customer status or service rules change.

[0068] The specific details of the above system have been described in detail in the method section of the implementation plan. For any undisclosed details, please refer to the implementation plan of the method section, and therefore will not be repeated here.

[0069] The above figures are merely illustrative of the processes included in the method according to exemplary embodiments of the present invention and are not intended to be limiting. It is readily understood that the processes shown in the above figures do not indicate or limit the temporal order of these processes. Furthermore, it is also readily understood that these processes may be executed synchronously or asynchronously, for example, in multiple modules.

[0070] It should be noted that although several modules or units of the system have been mentioned in the detailed description above, this division is not mandatory. In fact, according to exemplary embodiments of the present invention, the features and functions of two or more modules or units described above can be embodied in one module or unit. Conversely, the features and functions of one module or unit described above can be further divided and embodied by multiple modules or units.

[0071] Other embodiments of the invention will readily occur to those skilled in the art upon consideration of the specification and practice of the invention disclosed herein. This application is intended to cover any variations, uses, or adaptations of the invention that follow the general principles of the invention and include common knowledge or customary techniques in the art not disclosed herein. The specification and embodiments are to be considered exemplary only, and the true scope and spirit of the invention are indicated by the claims.

[0072] It should be understood that the present invention is not limited to the precise structure described above and shown in the accompanying drawings, and various modifications and changes can be made without departing from its scope. The scope of the invention is limited only by the appended claims.

Claims

1. A service solution matching method based on customer profiles, characterized in that, The method includes: Obtain the solution description text, rule text, and target customer business information corresponding to the service solution to be matched; The rule text is semantically segmented and candidate rule is identified to obtain a set of adaptation conditions related to the service scheme to be matched. Based on the set of adaptation conditions and the preset customer profile dimensions, an evidence item model is constructed. The evidence item model includes semantic associations between customer subjects, business activities, service consumption status, relationship persistence status and development potential status. The evidence item model and the set of adaptation conditions are converted into a constraint verification template, and the service scheme to be matched, the set of adaptation conditions, the evidence item model, and the constraint verification template are associated and written into the service semantic graph. A customer profile map is generated based on the customer business data. The customer profile map includes business potential indicators, performance cost indicators and relationship resilience indicators. The business potential indicators are generated based on changes in customer business activities and the development status of the customer's field. The performance cost indicators are generated based on service reach frequency and service delivery distance. The relationship resilience indicators are generated based on the continuity of the relationship between the customer and the service provider. Semantically align the customer profile graph with the service semantic graph to establish a mapping between the mapping evidence items in the customer profile graph and the objects to be verified in the constraint verification template. Based on the business potential index, the performance cost index, and the relationship resilience index, a profile partition of the target customer is generated, and the constraint verification template is executed to perform condition verification on the evidence items in the customer profile map to obtain the verification result. A service plan matching result is generated based on the profile partition and the verification result. The verification result is used to determine whether the service plan to be matched is suitable for the target customer. The profile partition is used to determine the recommendation order or coordination action type. The recommendation reason and coordination action are output based on the matching conditions that are met or not met in the verification result.

2. The service solution matching method based on customer profile according to claim 1, characterized in that, The rule text is semantically segmented and candidate rule is identified, including generating search suggestions based on the name, service object and processing item of the service solution to be matched, dividing the rule text into rule fragments, extracting the matching condition text from the rule fragments, removing the supporting material descriptions that are irrelevant to the matching condition text, and obtaining the matching condition set.

3. The service solution matching method based on customer profile according to claim 1, characterized in that, Constructing the evidence item model includes: converting customer attributes, customer relationships, business activities, performance resources, and historical transaction information in the set of adaptation conditions into basic evidence items; hierarchically associating multiple basic evidence items according to subject relationships, time relationships, and business affiliation relationships; and excluding computational conclusions derived from the basic evidence items that are not used as verification criteria from the evidence item model.

4. The service solution matching method based on customer profile according to claim 1, characterized in that, The conversion of the constraint verification template includes converting the admission conditions, exclusion conditions, exception conditions and dependency conditions in the adaptation condition set into corresponding verification nodes, configuring the verification nodes with subordinate relationships as parent-child constraint structures, and writing the parent-child constraint structures into the service semantic graph after performing consistency verification.

5. The service solution matching method based on customer profile according to claim 1, characterized in that, Generating the customer profile map includes: querying the data source directory according to the evidence item model, determining the data system that stores each basic evidence item, requesting the minimum evidence data related to the service plan to be matched from the data system, and writing the minimum evidence data into the customer profile map according to the evidence item model.

6. The service solution matching method based on customer profile according to claim 1, characterized in that, The business potential indicators are generated by forming potential levels based on changes in customer business volume, frequency of changes in customer service content, and development status of the customer's industry. The performance cost indicators are generated by forming cost levels based on service reach frequency, service performance distance, and service resource occupancy status. The relationship resilience indicators are generated by forming resilience levels based on contract continuity status, stable interaction status, and historical service continuity status.

7. The service solution matching method based on customer profile according to claim 1, characterized in that, Generating the profile partition includes comparing the business potential index, the performance cost index, and the relationship resilience index with the corresponding adaptive boundary conditions to obtain the level results of each profile dimension, and determining the profile partition in which the customer is located according to the combination of the level results.

8. The service solution matching method based on customer profile according to claim 1, characterized in that, Generating the service solution matching result includes: if the verification result shows that the target customer meets the set of matching conditions, determining the recommended order of the service solution based on the profile partition, the business potential index, the performance cost index, and the relationship resilience index; if the verification result shows that the target customer does not meet the set of matching conditions, generating the corresponding actions among the evidence completion action, manual review action, and solution adjustment action as the coordination action.

9. The service solution matching method based on customer profile according to claim 1, characterized in that, The method further includes recording changes in customer behavior, service performance, and evidence gaps after the service scheme matching result is adopted; updating adaptive boundary conditions and recommendation ranking rules based on the recording results; and, when an update to the rule text is detected, re-segmenting the updated rule text semantically and identifying candidate rules, and updating the constraint verification template and the service semantic graph.

10. A service solution matching system based on customer profiles, the system comprising: The data acquisition module is used to obtain the solution description text, rule text, and customer business information of the target customer corresponding to the service solution to be matched; The semantic parsing module is used to perform semantic segmentation and candidate rule identification on the rule text to obtain a set of adaptation conditions related to the service scheme to be matched; The evidence item modeling module is used to construct an evidence item model based on the set of adaptation conditions and the preset customer profile dimensions. The evidence item model includes semantic associations between customer subjects, business activities, service consumption status, relationship persistence status and development potential status. The service graph construction module is used to convert the evidence item model and the adaptation condition set into a constraint verification template, and associate the service scheme to be matched, the adaptation condition set, the evidence item model and the constraint verification template into the service semantic graph; The customer profile building module is used to generate a customer profile map based on the customer business data. The customer profile map includes business potential indicators, performance cost indicators and relationship resilience indicators. The business potential indicators are generated based on changes in customer business activities and the development status of the customer's field. The performance cost indicators are generated based on service reach frequency and service delivery distance. The relationship resilience indicators are generated based on the continuity of the relationship between the customer and the service provider. The graph mapping module is used to semantically align the customer profile graph with the service semantic graph, so that the mapping evidence items in the customer profile graph are mapped to the objects to be verified in the constraint verification template. The condition verification module is used to generate the profile partition of the target customer based on the business potential index, the performance cost index and the relationship resilience index, and execute the constraint verification template to perform condition verification on the evidence items in the customer profile map to obtain the verification result. The matching decision module is used to generate service solution matching results based on the profile partitioning and the verification results. The verification results are used to determine whether the service solution to be matched is suitable for the target customer. The profile partitioning is used to determine the recommendation order or coordination action type, and outputs recommendation reasons and coordination actions based on the suitability conditions met or not met in the verification results.

Images