Opportunity exchange and conversion analysis method and system based on hierarchical authorization and event graph

Abstract

This invention discloses a method and system for analyzing business opportunity exchange and conversion based on hierarchical authorization and event graphs, relating to the fields of CRM and knowledge graph technology. This invention controls the quality of business opportunity entry points through structured modeling and credibility detection; it achieves controllable release of sensitive customer information through hierarchical information disclosure and two-way exchange of authorization tokens, transforming business exchange behavior into traceable and verifiable technical authorization events; it breaks down data barriers between heterogeneous systems by unifying multi-source data into unified events, constructing a local event graph centered on the target business opportunity and generating conversion paths; simultaneously, it achieves conversion scoring, contribution attribution, and automatic bottleneck identification, and employs a local subgraph incremental update mechanism to reduce computational overhead. This invention can achieve full-link traceability and quantifiable analysis of business opportunity exchange while ensuring customer information security, effectively improving business opportunity conversion efficiency and platform operational efficiency, and reducing system computing resource consumption.

Description

Technical Field

[0001] This invention relates to the fields of customer relationship management (CRM), knowledge graph and data processing technology, specifically to information security control and conversion analysis technology in cross-entity business opportunity exchange scenarios, and particularly to a method and system for business opportunity exchange and conversion analysis based on hierarchical authorization and event graph. Background Technology

[0002] As enterprises deepen their digital transformation, customer relationship management systems have become a core infrastructure for sales operations. The efficiency of acquiring, following up on, and converting business opportunities directly determines a company's sales performance. To expand customer acquisition channels and reduce customer acquisition costs, cross-enterprise and cross-industry business opportunity resource exchange and collaborative conversion are gradually becoming the mainstream business model. This involves different market players exchanging and matching their respective potential business opportunity information to achieve resource complementarity and joint transactions.

[0003] Currently, mainstream business opportunity exchange platforms or CRM extension functions have the following technical shortcomings in practical applications: First, the granularity of information security control is insufficient, leading to a high risk of customer information leakage. Existing solutions mostly employ simple role-based access control, often exposing sensitive business opportunity information (such as customer names, contact information, and detailed addresses) before the exchange is confirmed. There is a lack of a tiered disclosure mechanism based on the exchange process stages. Furthermore, business opportunity exchanges are merely records of business operations, failing to create technical authorization credentials with defined boundaries, time constraints, and revocable verification. This makes it impossible to effectively constrain unauthorized access and information dissemination, posing a risk of customer data leakage and misuse.

[0004] Second, the platform's ability to control the quality of business opportunity submissions is weak, with invalid data wasting platform resources. The existing platform lacks unified structured standards and quality verification mechanisms for business opportunity postings. A large number of low-quality business opportunities with missing fields, false information, and duplicate submissions enter the exchange pool, increasing both user screening costs and the platform's operational and data governance expenses.

[0005] Third, the heterogeneous and fragmented data across multiple systems makes it impossible to fully trace the conversion chain. Data from the entire lifecycle of a business opportunity—including posting, browsing, communication, quoting, contracting, and closing—is scattered across multiple heterogeneous business systems such as CRM systems, front-end mini-programs, instant messaging systems, and order and contract systems. The data formats and identification systems are inconsistent, making it impossible to form an end-to-end conversion chain. At the same time, it is difficult to accurately quantify the contributions of multiple participants in the business opportunity exchange process, such as recommending, matching, and following up, and it is impossible to provide a reliable basis for commission settlement, intelligent recommendation, and resource allocation.

[0006] Fourth, full-scale graph analysis is computationally expensive and its real-time performance is difficult to guarantee. If a full-scale knowledge graph is used to perform correlation analysis on all business opportunities and user data on the platform, the resource overhead of graph retrieval and path calculation increases exponentially with the growth of data scale, which cannot support high-frequency real-time updates and dynamic analysis; moreover, a large amount of irrelevant data in the full-scale graph will interfere with the accuracy of path analysis for individual business opportunities, resulting in a waste of computing resources.

[0007] Fifth, the reliance on manual investigation for conversion stagnation limits operational efficiency and conversion rates. Current solutions rely on manual identification and follow-up by operations personnel to address bottlenecks in opportunity conversion. This results in significant delays, high investigation costs, and insufficient targeted recommendations. Furthermore, the solutions cannot automatically generate precise operational strategies based on the real-time status of opportunities, thus hindering the overall improvement of opportunity conversion efficiency.

[0008] Therefore, there is an urgent need for a business opportunity exchange and conversion analysis technology solution that can take into account information security, quality control, full-chain traceability and efficient analysis, in order to solve the above-mentioned problems in existing technologies. Summary of the Invention

[0009] (a) Technical problems to be solved The purpose of this invention is to address the shortcomings of existing technologies by providing a business opportunity exchange and conversion analysis method and system based on hierarchical authorization and event graphs. This addresses the technical problems in existing business opportunity exchange scenarios, such as high risk of customer information leakage, weak business opportunity quality control, fragmented data from multiple systems making traceability impossible, excessively high cost of full graph calculation, and delayed identification of conversion bottlenecks.

[0010] (II) Technical Solution To achieve the above objectives, the present invention adopts the following technical solution: On the one hand, this invention provides a method for analyzing business opportunity exchange and conversion based on hierarchical authorization and event graphs, including the following steps: 1. Business Opportunity Quality Access Control: Receive business opportunity information submitted by publishers, perform unified structured modeling on the business opportunities, and generate structured business opportunity data containing basic attributes, demand attributes, resource attributes, sensitive attributes, permission attributes, and lifecycle attributes; calculate the business opportunity credibility score based on field completeness, publisher's historical performance, tag consistency, similar business opportunity duplication, and abnormal behavior risk dimensions; impose publishing restrictions or manual review on business opportunities with credibility scores below a preset threshold.

[0011] 2. Tiered Authorization Exchange Control: Based on the access level of the accessing entity and the exchange stage of the business opportunity, a de-identified display view with corresponding disclosure levels is generated, and sensitive customer information is hidden in stages; after receiving the exchange request from the applicant, an exchange item to be confirmed is generated. After mutual confirmation by the issuer and the applicant, an exchange authorization token carrying the scope of authorized fields, validity period, operation permissions and audit mark is issued. Sensitive information within the authorized scope is only opened to the applicant within the validity period of the token.

[0012] 3. Multi-source data event normalization: Collect raw behavioral data of the entire lifecycle of business opportunities from multiple heterogeneous business systems, and convert all heterogeneous raw data into standard event records in a unified format; after desensitizing or tokenizing sensitive fields in the standard event records, they are stored and associated with calculations, and can only be decrypted and displayed by authorized views.

[0013] 4. Local Event Graph Conversion Analysis: Taking the target business opportunity identifier as the entry point, the system retrieves associated standard event records through multi-level indexes to construct a local event graph centered on the target business opportunity. Based on the local event graph, the system performs directed traversal according to time constraints, relationship constraints, and authorization constraints to generate candidate conversion paths, calculate path conversion scores, and calculate and attribute the contribution values ​​of each participating entity in the completed business opportunities.

[0014] 5. Bottleneck Identification and Incremental Update: Based on the dwell time and status transition characteristics of each node in the conversion path, the system automatically identifies business opportunity conversion bottlenecks and outputs corresponding follow-up suggestions. When a new event occurs for a target business opportunity, the system only locates and updates the target business opportunity and the local sub-image, conversion score, and bottleneck status within the preset associated range, without performing a full data recalculation on the platform.

[0015] Furthermore, this method also includes an intelligent matching step for candidate exchange targets: calculating the matching degree based on business opportunity tags, applicant resource capabilities, historical conversion performance and credibility score, generating and pushing candidate exchange targets in a tiered manner.

[0016] On the other hand, the present invention provides a business opportunity exchange and conversion analysis system based on hierarchical authorization and event graph, including a CRM management terminal, a business opportunity exchange front-end, a business opportunity analysis and exchange server, an event index database, a permission and authorization service module, and a management display terminal.

[0017] CRM management interface: used to maintain customer files, business opportunity information, follow-up records, quotation records, contract records, and transaction and payment data; Business opportunity exchange front-end: An interactive entry point for users to publish, browse, search, collect, apply for, confirm, recommend, communicate and provide feedback on business opportunities; Event Index Database: Used to store structured business opportunity data, standard event records, event graph node and edge data, and multi-level search indexes; The Permissions and Authorization Service Module is used to generate hierarchical de-identified views, perform sensitive field encryption and decryption, issue and verify exchange authorization tokens, and record access audit logs. Business opportunity analysis and exchange server: This is the core computing unit, used to execute the above-mentioned business opportunity quality control, hierarchical authorization exchange, multi-source data event normalization, local graph transformation and analysis, bottleneck identification and incremental update logic; Management Display Terminal: Used to output display conversion paths, conversion scores, contribution attribution, bottleneck warnings and operational suggestions to operations and management personnel, and to perform permission field filtering before output.

[0018] (III) Beneficial Effects Compared with the prior art, the present invention has the following technical advantages: 1. Implement information security technology control throughout the entire business opportunity exchange process: Through a tiered information disclosure mechanism, sensitive information is released in stages according to the exchange phase. Coupled with time-sensitive, scope-specific, and revocable exchange authorization tokens, the "business opportunity exchange" at the business level is transformed into a verifiable, traceable, and auditable technical authorization event. This technically prevents unauthorized access and information leakage, thereby enhancing the platform's data security control capabilities.

[0019] 2. Improve the data quality of the business opportunity pool from the entry point: Through unified structured modeling and multi-dimensional credibility detection mechanisms, automatically identify and filter false, duplicate, and incomplete business opportunities, reduce the occupation of platform resources by invalid data, reduce user screening costs, and improve the matching efficiency of business opportunity exchange.

[0020] 3. Break down data barriers between heterogeneous systems and achieve full-chain traceability of conversion: Through unified event processing of multi-source data, heterogeneous data scattered across different business systems is converted into standardized event records, solving the problems of inconsistent data identification and inability to associate data across multiple systems; based on local event graphs, the entire process of business opportunities from release to transaction can be fully traced, clearly presenting the behavior and role of each participating entity.

[0021] 4. Significantly reduce computational resource overhead for graph analysis: By adopting a target opportunity-driven local event graph construction mode, coupled with multi-level indexing and local subgraph incremental update mechanism, full data retrieval and full graph recalculation are avoided, significantly reducing database pressure and server computing overhead, and improving the real-time performance of conversion analysis and the system's concurrent capacity.

[0022] 5. Support refined operations and improve business opportunity conversion efficiency: Through path conversion scoring, multi-entity contribution attribution and automated bottleneck identification, it provides quantitative basis for platform commission settlement, intelligent recommendation and resource allocation, while automatically outputting targeted follow-up suggestions to replace manual investigation and improve operational efficiency and overall business opportunity conversion rate.

[0023] 6. Build a comprehensive risk control and audit system: By combining authorization scope control, full-process access logs and abnormal behavior detection, the platform can support flexible business opportunity exchange while having complete security audit and risk management capabilities to meet data compliance and regulatory requirements. Attached Figure Description

[0024] Figure 1 is a schematic diagram of the overall architecture of the business opportunity exchange CRM system described in this invention; Figure 2 is a schematic diagram of the hierarchical information disclosure and business opportunity exchange authorization process described in this invention; Figure 3 is a schematic diagram of the local business opportunity map and conversion path described in this invention; Figure 4 is a schematic diagram of the overall process of the business opportunity exchange and conversion analysis method described in this invention. Detailed Implementation

[0025] 1. Overall Implementation Environment and System Architecture This invention can be deployed in a cloud-based distributed computer system or in an enterprise local server cluster. The overall system architecture is shown in Figure 1. The core of the system consists of four parts: the interaction layer, the service engine layer, the data layer, and the permission platform.

[0026] The interaction layer comprises three types of terminals: the CRM management terminal, the business opportunity exchange mini-program terminal, and the management display terminal. The CRM management terminal is designed for internal sales and management personnel, providing functions such as customer file maintenance, business opportunity entry, follow-up records, quotation management, contract management, and transaction payment registration. The business opportunity exchange mini-program terminal is designed for registered users of the platform, providing mobile interactive functions such as business opportunity posting, browsing and searching, saving, exchange application, exchange confirmation, business opportunity forwarding and recommendation, online instant communication, and evaluation feedback. The management display terminal is designed for platform operators and enterprise managers, providing functions such as conversion path visualization, contribution attribution reports, bottleneck warnings, and operational suggestion output.

[0027] The service engine layer, namely the business opportunity analysis and exchange server, is the core computing unit of this invention. It has nine built-in functional units: structured modeling unit, credibility detection unit, matching recommendation unit, event standardization unit, graph construction unit, path analysis unit, attribution scoring unit, bottleneck identification unit, and incremental update unit. These units correspond to the various steps of the method of this invention.

[0028] The data layer includes a relational business database and a graph / event index database. The former stores structured business data such as customers, business opportunities, and contracts, while the latter stores standard event records, event graph nodes and edges, multi-level indexes, path caching, and scoring results. The permission middle platform, namely the permission and authorization service module, is independently deployed between the service layer and the data layer. It is responsible for end-to-end permission verification, sensitive field de-identification and encryption / decryption, authorization token issuance and revocation, and access audit log recording.

[0029] 2. Detailed Implementation Steps of the Method and Process The overall method flow of the present invention is shown in Figure 4, and is specifically executed according to the following steps S100 to S110: Step S100: Business Opportunity Posting and Structured Modeling The publisher submits a business opportunity posting request through the CRM management terminal or the business opportunity exchange mini-program. After receiving the original business opportunity information, the system first generates standard tags for unstructured text such as title, requirement description, and customer information using keyword extraction algorithms, industry rule mapping, or pre-trained natural language processing models. The system then performs consistency verification with the tags manually selected by the publisher. If the tag overlap is lower than a preset threshold, a correction prompt is pushed to the publisher.

[0030] After completing tag verification, the system performs unified structured modeling on the business opportunity, generating six categories of standard fields: Basic fields: Business opportunity unique number, business opportunity title, publisher user ID, affiliated company ID, source terminal, publication timestamp, validity period expiration timestamp; Requirements fields: Industry Level 1 / Level 2 tags, Product / Service Category tags, Region / Province / City tags, Budget Range (Upper and Lower Limits), Required Quantity, Delivery Cycle, and Cooperation Method tags; Resource fields: The publisher can provide resource description tags, expected resource exchange tags, acceptable cooperation conditions, and a set of historical fulfillment tags; Sensitive fields: Customer full name, contact person name, mobile phone number, landline number, detailed address, precise budget amount, internal remarks; Permission fields: Public level, Request visibility level, Confirmation visibility level, Internal enterprise visibility level, Member level permissions, Role permissions; Lifecycle field: The initial status is marked as "Pending review", and it will be updated sequentially as the process progresses to "Published", "Application in progress", "Confirmed", "Under communication", "Quoted", "Transaction completed", and "Closed".

[0031] Step S101: Business opportunity credibility detection and duplicate business opportunity identification The system calculates a credibility score based on the structured business opportunity data. The scoring formula is as follows: In the formula, the values ​​of each parameter are normalized to the interval 0-1, and the weights are... to It can be dynamically optimized through platform operational data. This includes field completeness. Calculated based on the fill ratio of required fields; authentication level. Calculated based on the publisher's enterprise certification status, contact information verification status, and customer lead verification status; historical performance. Data is taken from the publisher's historical business opportunity conversion rate, complaint rate, and user rating scores; tag consistency is maintained. The degree of overlap between automatically generated tags and manually selected tags for text; risk of duplication. Calculated based on the number of similar business opportunities and their similarity; risk of abnormal behavior. Calculated based on the number of times the publisher publishes in batches within a short period of time, cancels frequently, and is reported.

[0032] Simultaneously, the system performs duplicate opportunity identification: extracting six key features from new business opportunities—title text vector, industry tag set, region code, de-identified hash value of customer name, budget range, and posting time window—and calculating a weighted comprehensive similarity with historical business opportunities on the platform within the past 180 days. When the comprehensive similarity exceeds 70% (first threshold), it is marked as a suspected duplicate business opportunity, and its recommendation weight on the platform is reduced; when the comprehensive similarity exceeds 90% (second threshold) and the posting entities do not belong to the same company, the business opportunity status is automatically set to "awaiting manual review," restricting its public visibility.

[0033] Step S102: Hierarchical Information Disclosure and De-identification View Generation The permissions and authorization service module dynamically generates corresponding display views based on the user ID, enterprise affiliation, role permissions, membership level of the accessing subject, and the current lifecycle status of the target business opportunity, as shown in Figure 2. There are five levels in total: 1. Public Summary View: Applicable to logged-in users who are not logged in or have not submitted an exchange request. It only displays industry category, geographical scope, demand category, budget range, validity period, and anonymized summary of business opportunities. All sensitive customer fields are hidden. 2. Exchange Application View: Applicable to applicants who have submitted an exchange application. It additionally displays a summary of the publisher's company resources, an explanation of the exchange conditions, a business opportunity credibility score range, and an exchange application button, while still hiding detailed customer information and internal notes. 3. Pending Confirmation View: Applicable to both parties in the exchange, it displays a summary of the resources of both companies, supply and demand matching scores, a description of the scope of the exchange, and confirmation / rejection buttons, while hiding the other party's sensitive third-party customer information; 4. Post-authorization view: Applicable to both parties who have completed mutual confirmation and whose authorization tokens are valid. It displays sensitive information such as customer name, contact person, and contact information according to the set of authorized fields agreed upon by the token. Fields outside the scope of authorization are still hidden. 5. Management View: Suitable for platform administrators and enterprise administrators, it displays the complete conversion path, scoring results, risk indicators, operation logs, bottleneck warnings, and operational suggestions, while performing secondary field filtering according to data sensitivity levels.

[0034] Step S103: Two-way confirmation and generation of authorization tokens When the applicant clicks the "Apply for Exchange" button, the system extracts the applicant's enterprise resource tags, industry capability tags, historical conversion performance data, and current permission level, and performs a weighted matching calculation with the target business opportunity's demand tags and desired exchange resources. This generates a matching score and an exchange request to be confirmed, which is then pushed to the publisher's message center.

[0035] After the publisher clicks "Confirm Exchange," the system sends a secondary confirmation notification to the applicant. Once the applicant completes the secondary confirmation, the Permissions and Authorization Service module issues a unique exchange authorization token. The token uses JWT or distributed unique ID format, and its core fields include: unique token number, target opportunity number, publisher user ID, applicant user ID, authorized field set, visibility period (default 72 hours), operable actions (view / export / communicate), generation timestamp, expiration timestamp, log hash value, and revocation status flag.

[0036] The exchange authorization token is revocable, auditable, and tamper-proof. The system controls access permissions for sensitive information based on the token. After the token expires or is revoked, the applicant is automatically restored to the request exchange view.

[0037] Step S104: Multi-source data acquisition and unified event processing The business opportunity analysis and exchange server collects raw data from the CRM management system, business opportunity exchange mini-program, instant messaging system, order system, contract system, and external business interfaces through three methods: API interface, message queue, and database log collection. The raw data covers all behaviors throughout the entire business opportunity lifecycle: posting, browsing, saving, inquiring, exchange application, exchange confirmation, recommendation and forwarding, online communication, customer follow-up, quoting, contract signing, closing, evaluation, and payment.

[0038] The system uniformly converts all heterogeneous raw data into standard event records. Each record contains fixed fields: unique event ID, target business opportunity ID, operating entity user ID, associated customer ID, associated enterprise ID, event type code, event occurrence timestamp, source system code, event status code, permission level, associated authorization token ID, and extended tag set.

[0039] For sensitive fields such as customer name, contact person, and contact information, the system uses SHA-256 hash identifiers or de-identified tokens to replace the original plaintext in the event storage and calculation process. The display end only decrypts and displays the original plaintext fields through the permission service when the access subject passes the token permission verification.

[0040] Step S105: Construction of Local Event Graph The system uses the target business opportunity ID as the entry point and retrieves all associated standard event records from the event index database through multi-level indexes to construct a local event graph corresponding to the business opportunity, as shown in Figure 3.

[0041] The graph node types include: opportunity nodes, publisher user nodes, applicant user nodes, referrer user nodes, customer nodes, enterprise nodes, event nodes, authorization token nodes, quotation nodes, contract nodes, and transaction nodes. All relationship edges in the graph are directed, and each edge carries the following attributes: event time, event type, participating entity ID, source system, permission level, authorization scope, and edge weight. Typical relationship edges include: publisher user node — publishing relationship edge — opportunity node; applicant user node — application exchange relationship edge — opportunity node; authorization token node — authorized access relationship edge — sensitive field set node; opportunity node — transaction relationship edge — transaction node.

[0042] To improve retrieval efficiency, the system pre-establishes five levels of independent indexes: a primary index for business opportunity numbers, an index for de-identified customer identifiers, an index for user identifiers, an index for authorization token identifiers, and an index for event time windows. When constructing a partial event graph, only related nodes are located through the indexes, without traversing the entire platform's business opportunity and user data.

[0043] Step S106: Candidate exchange object generation and recommendation matching The system calculates a supply-demand matching score based on the demand tags and desired resource tags of business opportunities, combined with the resource capability tags, industry coverage, regional coverage, historical conversion rate, credibility score, average response time, and permission level of candidate users. Based on the matching score, candidates are categorized into priority recommendation targets (matching score ≥ 80), ordinary applicants (matching score 60-80), and risk-restricted targets (matching score < 60 or exhibiting abnormal behavior), and then pushed to the publisher or platform recommendation pool in priority order.

[0044] Step S107: Conversion path generation, scoring, and contribution attribution The system starts with the business opportunity posting node and performs a depth-first directed traversal of the local event graph according to the order of event occurrence from earliest to latest, generating one or more candidate conversion paths. The path generation process follows four constraints: 1. Time constraint: The event time of a subsequent node in the path must not be earlier than that of its predecessor node; 2. Relationship Constraints: Adjacent nodes must have a business opportunity association, user association, customer association, enterprise association, or authorization token association; 3. Authorization Constraints: Access events to sensitive fields without authorization tokens will not be counted as valid conversion paths; 4. Deduplication constraint: When there are repeated loops in the path and no new business events are added, the loop part is removed.

[0045] For each candidate conversion path, the system calculates a path conversion score using the following formula: In the formula To determine the match between business opportunity needs and the applicant's resources; The intensity of effective interaction within the path is calculated comprehensively based on the number of communications, the number of follow-ups, and the depth of dwell time. This represents the historical transformation of the participating entities; The weighting of key business events such as quotations, contracts, and transactions is determined by the order of events; the closer to the transaction, the higher the weighting. The credibility of the recommendation or authorization relationship; Rate the credibility of the business opportunity itself; The longer a node stays, the more points are deducted due to the risk of exceeding the time limit. to Configurable weights.

[0046] For business opportunities that have been marked as completed, the system further calculates the contribution value of each participating entity. The contribution value is calculated comprehensively based on the event's position in the path, the event type, the time distance from the completion node, the number of subsequent conversion nodes, and the weight of the authorization relationship. For example, users who recommend a business opportunity for the first time receive a referral contribution value, while users who complete the quotation and contract signing receive a completion contribution value. The platform can use the contribution value for commission settlement, service evaluation, and resource allocation.

[0047] Step S108: Bottleneck Identification and Automatic Follow-up Suggestions Based on the number of events, the time interval between events, the current path score, the authorization status, and the status transition, the system automatically identifies seven types of conversion bottlenecks and outputs corresponding actionable operational suggestions: Step S109: Local subgraph incremental update When a target business opportunity generates new views, favorites, applications, confirmations, authorizations, recommendations, communications, follow-ups, quotations, contracts, or transactions, the system does not perform a full-platform graph recalculation, but only a partial incremental update: 1. Locate the opportunity node using the main index based on the target opportunity ID in the newly added event; 2. Retrieve all associated nodes within a preset 2-hop range using the user index, customer de-identification index, and token index; 3. Add relationship edges corresponding to the new events and update node attributes; 4. Only recalculate the affected candidate conversion paths, conversion scores, contribution values, and bottleneck status; 5. Update path cache and rating results.

[0048] By using a local incremental update mechanism, the computation time for a single business opportunity event can be reduced from seconds in full mode to milliseconds, significantly reducing database retrieval pressure and server computing resource consumption.

[0049] Step S110: Analysis Result Output and Access Control The server pushes analysis results, including conversion path, current conversion stage, key stakeholders, path score, conversion probability, contribution attribution, bottleneck nodes, and follow-up suggestions, to the management display or mini-program management page. Before output, the permissions and authorization service module performs field filtering based on the access subject's enterprise affiliation, role permissions, and authorization token status: unauthorized entities only display de-identified identifiers and scope information; authorized entities can only view sensitive fields within the token validity period and authorization scope; all access operations are recorded in audit logs throughout the process.

[0050] 3. Specific Application Examples Example 1: Smart Hardware Business Opportunity Exchange and End-to-End Conversion User A from a smart hardware company in Zhejiang province posted a "smart toy OEM manufacturing opportunity" through a business opportunity exchange mini-program. The user selected industry tags: smart hardware, AI toys, and electronic mechanisms; region: Zhejiang and Guangdong; budget range: 500,000-2,000,000 RMB; quantity requested: 100,000 sets; and set the customer name, contact person, and contact information to "visible only after confirmation." The system completed structured modeling and credibility testing, achieving a credibility score of 92, marking it as a high-quality business opportunity, and generating a public summary view for external display.

[0051] User B, from a toy factory in Guangdong, browsed the business opportunity and submitted an exchange application, uploading a summary of their factory's production capacity, qualifications, and past cases. The system calculated a matching score of 86 and pushed the application to User A. User A clicked to confirm the exchange, and the system sent a secondary confirmation notification to User B. After User B completed the secondary confirmation, the system generated an exchange authorization token valid for 72 hours, granting User B access to the customer contact's name and phone number.

[0052] User B, realizing their production capacity was insufficient, recommended the business opportunity to User C, who possessed OEM capabilities, through the platform. After reviewing the customer information, User C entered customer follow-up records, created a quotation, and ultimately signed a contract and marked the transaction as completed in the CRM management system. The system converts all actions—posting, browsing, applying, confirming, authorizing, recommending, following up, quoting, signing contracts, and completing transactions—into standard event records, constructs a partial event graph, generates a complete conversion path, and automatically calculates User B's recommendation contribution value as 35% and User C's transaction-driving contribution value as 65%, which serves as the basis for subsequent commission distribution and evaluation.

[0053] Example 2: Bottleneck Identification and Operational Intervention for Unsold Business Opportunities After a business opportunity for a certain enterprise service was posted, it received 18 views within 72 hours, but no exchange requests were received. Based on view and request thresholds, the system automatically identified this as a "viewed but not requested" bottleneck and provided operational suggestions: optimize the business opportunity title, add specific demand scenario tags, adjust the budget range from a vague range to a specific numerical range, and target the opportunity to 10 highly relevant enterprise service providers.

[0054] After the operations staff optimized according to the suggestions, they received 3 exchange requests the next day. Subsequently, one of the requests was not processed by the publisher for more than 30 hours. The system identified this as a "request unconfirmed" bottleneck, sent an SMS reminder to the publisher, and displayed the requests in sorted order by matching score. The publisher confirmed the request on the same day. All subsequent new events only update a partial sub-map of that business opportunity, without affecting other business opportunity data on the platform.

[0055] Example 3: Risk Control of Duplicate Opportunities and Unauthorized Access User D posted a new business opportunity for building materials cooperation. The system comparison found that its title vector, industry tags, region, and customer name hash value were 92% similar to a business opportunity posted by another user 3 days ago, and the posting entities were different. The system automatically marked it as a highly suspected duplicate business opportunity, placed it in a manual review status, and restricted public recommendation.

[0056] User E attempted to access the contact information page of a business opportunity directly via a shared link without submitting an exchange request. The access control system verified that E lacked a valid authorization token, returned a public summary view, and logged unauthorized access. After detecting that User E attempted to access sensitive fields of 8 different business opportunities within 1 hour, the system automatically increased E's abnormal behavior risk score and restricted E's exchange request privileges for 24 hours.

[0057] 4. Description of Alternative Implementation Methods The above description is merely a preferred embodiment of the present invention and does not limit the scope of protection of the present invention. For example, the weight parameters of the credibility scoring formula and conversion scoring formula in the present invention can be dynamically adjusted according to industry characteristics; the hop count range of the local event graph can be configured according to the data scale; the default validity period of the authorization token can be customized according to business needs; in addition to the mini-program terminal, the business opportunity exchange front-end can also be replaced with other terminal forms such as H5 web page terminal and PC client. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the scope of protection of the present invention.

Claims

1. A method for analyzing business opportunity exchange and conversion based on hierarchical authorization and event graphs, characterized in that, Includes the following steps: S1 performs structured modeling and credibility calculation on published business opportunities, and imposes access restrictions on business opportunities with credibility below a preset threshold; S2 generates a de-identified display view at the corresponding level based on the access subject's permissions and the stage of business opportunity exchange, hiding sensitive customer information; Upon receiving an exchange request and after mutual confirmation by both parties, an exchange authorization token is generated, carrying the scope of authorization, validity period, and audit identifier. Sensitive information within the corresponding scope is then granted based on the token. S3 collects raw data from the entire lifecycle of business opportunities from multiple heterogeneous systems, converts it into standard event records, and stores and performs correlation calculations after desensitizing sensitive fields. S4 retrieves relevant standard events with the target business opportunity as the core, constructs a local event graph, generates candidate conversion paths based on graph traversal, calculates path conversion scores, and attributes the contributions of participating entities to the completed business opportunities. S5 identifies bottleneck nodes in the conversion path and generates follow-up suggestions; when a new event occurs in the target business opportunity, it only updates the target business opportunity and the local sub-graph and analysis results within the preset related scope, without performing a full data recalculation.

2. The business opportunity exchange and conversion analysis method based on hierarchical authorization and event graph as described in claim 1, characterized in that, In step S1, the formula for calculating the business opportunity credibility score is as follows: in, For target business opportunities, For field completeness, To ensure the degree of verification of identity and customer leads, Based on the publisher's historical performance in fulfilling its obligations, For label consistency, To mitigate the risk of repetition, For abnormal behavior risks, to Weight parameters that can be preset or dynamically adjusted; Step S1 further includes duplicate business opportunity identification: extracting the title vector, industry tag, region, customer de-identification identifier, budget range and posting time features of new business opportunities, and calculating the comprehensive similarity with historical business opportunities; when the similarity exceeds the first threshold, it is marked as a suspected duplicate business opportunity and the recommendation weight is reduced; when it exceeds the second threshold and the posting entity is different, manual review or restriction of visibility is triggered.

3. The business opportunity exchange and conversion analysis method based on hierarchical authorization and event graphs according to claim 1, characterized in that, The anonymized display view in step S2 includes five levels: public summary view, which is applicable to entities that have not submitted an exchange application and only displays the industry, region, demand category, budget range and anonymized summary; The request exchange view is for applicants who have submitted exchange requests, and displays a summary of the resources from the publisher, exchange conditions, and a confidence range. The pending confirmation view is applicable to both parties in the exchange, displaying a summary of resources, matching scores, and the entry point for the exchange operation. The authorized view is applicable to entities with mutual confirmation and valid tokens, displaying sensitive fields such as customer name and contact information according to the scope of authorization. The management view, suitable for platform or enterprise administrators, displays conversion paths, ratings, risks, and operational recommendations. The exchange authorization token in step S2 includes at least the token number, target business opportunity number, publisher identifier, applicant identifier, authorization field set, visibility period, operable actions, generation time, expiration time, log hash value, and revocation status.

4. The business opportunity exchange and conversion analysis method based on hierarchical authorization and event graph as described in claim 1, characterized in that, In step S4, the nodes of the local event graph include business opportunity nodes, participating entity nodes, customer nodes, enterprise nodes, event nodes, authorization token nodes, quotation nodes, contract nodes, and transaction nodes; the relationship edges of the graph are directed edges, carrying event time, event type, participating entity, source system, permission level, authorization scope, and weight parameters.

5. The business opportunity exchange and conversion analysis method based on hierarchical authorization and event graphs according to claim 4, characterized in that, In step S4, the generation of candidate conversion paths follows four constraints: time constraint, that is, the event time of subsequent nodes is not earlier than that of previous nodes; relationship constraint, that is, adjacent nodes have business opportunities, entities, customers or authorization associations; authorization constraint, that is, unauthorized access events to sensitive fields are not included in the valid conversion path; and deduplication constraint, that is, low-relevance cyclic nodes in the path are removed. The system establishes independent indexes based on business opportunity number, customer anonymization identifier, user identifier, authorization token identifier, and event time. When constructing a local event graph, it retrieves related nodes through the indexes without traversing the entire business opportunity data.

6. The business opportunity exchange and conversion analysis method based on hierarchical authorization and event graph as described in claim 1, characterized in that, In step S4, the formula for calculating the conversion path score is as follows: in, As candidate conversion paths, To ensure a good match between business opportunities and the applicant's resources, For the effective interaction strength of the path, As a manifestation of the historical transformation of the participating entities, Weighting of quotes, contracts, or transaction events. To ensure the credibility of the authorization relationship, To ensure the credibility of business opportunities, For node dwell time and risk of exceeding the time limit, to These are preset or dynamically updated weight parameters.

7. The business opportunity exchange and conversion analysis method based on hierarchical authorization and event graph as described in claim 1, characterized in that, The conversion bottlenecks identified in step S5 include seven categories: browsing without application, application without confirmation, confirmation without authorization, authorization without communication, communication without quotation, quotation without transaction, and transaction without feedback. The system outputs corresponding operational follow-up suggestions for each type of bottleneck. The specific logic of incremental update of local subgraph in step S5 is as follows: based on the business opportunity identifier, associated entity identifier and authorization token identifier of the new event, locate the target business opportunity node and associated nodes within the preset number of hops, update only the relationship edge corresponding to the new event, and recalculate the affected conversion path, score, attribution and bottleneck status.

8. A business opportunity exchange and conversion analysis system based on hierarchical authorization and event graphs, characterized in that, It includes a CRM management terminal, a business opportunity exchange front-end, a business opportunity analysis and exchange server, an event index database, a permissions and authorization service module, and a management and display terminal; The CRM management interface is used to maintain customer, business opportunity, follow-up, quotation, contract and transaction data; The business opportunity exchange front-end serves as an interactive entry point for providing business opportunity posting, browsing, exchange application confirmation, recommendation, and communication; The event index database is used to store structured business opportunity data, standard event records, graph data, and multi-level indexes; The permission and authorization service module is used to generate hierarchical de-identified views, encrypt and decrypt sensitive fields, issue and verify authorization tokens, and record access logs. The business opportunity analysis and exchange server is used to perform the steps of the method described in any one of claims 1-7; The management display terminal is used to output conversion analysis results and perform field filtering based on the access subject's permissions.

9. The business opportunity exchange and conversion analysis system based on hierarchical authorization and event graphs according to claim 8, characterized in that, The event index database includes built-in business opportunity number index, customer anonymization identifier index, user identifier index, authorization token index, and time window index, which are used to support rapid retrieval and incremental updates of local graphs.

10. The business opportunity exchange and conversion analysis system based on hierarchical authorization and event graphs according to claim 8, characterized in that, The permission and authorization service module logs all access to sensitive fields throughout the process. When it detects that a subject has accessed the field multiple times without authorization in a short period of time, it automatically raises the risk level of the abnormal behavior and restricts the permission to request exchange.

Images