Public resource transaction platform data sharing cooperation mechanism operation method and system

Abstract

The invention relates to the technical field of data processing, and discloses a public resource transaction platform data sharing cooperation mechanism operation method and system. The method comprises the steps of collecting data flow records among a data ownership party, a data management party and a data user to construct a data flow direction traceability graph, extracting quality scores, processing durations and use feedback of all collaborative nodes to calculate collaborative value conversion efficiency, screening nodes lower than a failure threshold value to adjust routing weights, and sending the data flow direction traceability graph to the data management party. And updating the routing rule based on the adjusted weight and performing an imbalance prevention operation. According to the method, the problems that a new system bottleneck is easy to introduce and resource allocation is unbalanced in a weight dynamic adjustment process are solved.

Description

Technical Field

[0001] This application relates to the field of data processing technology, and in particular to a method and system for operating a data sharing and collaboration mechanism for a public resource trading platform. Background Technology

[0002] As a crucial bridge connecting supply and demand, public resource trading platforms undertake the function of providing trading services for public resources. Currently, the data sharing and collaboration mechanism of public resource trading platforms mainly adopts the traditional information system docking model. Each level of platform establishes an independent database system to store local transaction data. When data sharing is required, the data owner uploads the data to the central database or data exchange platform maintained by the data manager according to the agreed format. The data manager classifies, organizes, and secures the received data, establishing a data catalog and access permission system. Data users submit requests, undergo identity verification and permission review, and then obtain the required data through API interfaces or file transfer. In terms of vertical collaboration, local platforms regularly report summarized data to the central platform. In terms of horizontal collaboration, different local platforms achieve information exchange through data exchange platforms or point-to-point docking. The entire process relies on technologies such as data encryption, access control, and security auditing to ensure data security.

[0003] Existing data sharing and collaboration mechanisms suffer from several shortcomings. The prominent problem of data silos hinders true interconnectivity between local platforms due to inconsistent standards. Data sharing lacks timeliness and accuracy due to time lags in traditional manual reporting or periodic batch transmission methods, and data is prone to errors during multiple conversions and transfers. The lack of unified collaboration mechanisms and service standards leads to unclear division of responsibilities among participants and blurred boundaries of data ownership, management, and use, resulting in chaotic and inefficient sharing processes. Cross-regional collaboration is difficult because different regions lack unified standards in areas such as digital certificate authentication, expert resource sharing, and mutual recognition of credit data. Data security and privacy protection mechanisms are inadequate, posing a risk of leakage during data transmission, storage, and use, and lacking effective supervision and accountability mechanisms. Furthermore, the lack of intelligent monitoring and early warning capabilities prevents real-time identification and early warning of anomalies and violations during data sharing. Summary of the Invention

[0004] This application provides a method and system for operating a data sharing and collaboration mechanism for a public resource trading platform. It is used to construct a data flow traceability map that includes quality changes and usage feedback, and to evaluate nodes and dynamically adjust weights based on the efficiency of collaborative value conversion. This solves the problems in the prior art where the single evaluation standard for collaborative nodes leads to the inability to accurately identify inefficient nodes and the lack of adaptive capability in data flow adjustment. By establishing a multi-level imbalance prevention mechanism that dually monitors load rate and weight growth rate, it solves the problem of easily introducing new system bottlenecks and resource allocation imbalances during the dynamic adjustment of weights.

[0005] Firstly, this application provides a method for operating a data sharing and collaboration mechanism for a public resource trading platform, the method comprising: Step S1: Collect data flow records between the data owner, data manager, and data user, and construct a data flow traceability map based on the flow path and attribute information; Step S2: Extract the input quality score, output quality score, normalized processing time, user satisfaction score, and duplicate request rate of each collaborative node in the data flow tracing map. Multiply the quality score ratio, the reciprocal of the normalized processing time, the ratio of satisfaction to duplicate request rate, and the usage frequency influence factor by their respective weight coefficients and sum them to calculate the collaborative value conversion efficiency. Step S3: Select nodes whose collaborative value conversion efficiency is lower than the failure threshold as failed nodes, and adjust the routing weights of the failed nodes and their parallel candidate nodes. Step S4: Update the data flow routing rules based on the adjusted routing weights, monitor the operating status of each node, and perform imbalance prevention operations.

[0006] Secondly, this application provides a data sharing and collaboration mechanism operation system for a public resource trading platform, the system comprising: The data acquisition module is used to collect data flow records between data owners, data managers, and data users, and to build a data flow traceability map based on the flow path and attribute information. The calculation module is used to extract the input quality score, output quality score, normalized processing time, user satisfaction score and duplicate request rate of each collaborative node in the data flow tracing map. The module multiplies the quality score ratio, the reciprocal of the normalized processing time, the ratio of satisfaction to duplicate request rate and the usage frequency influence factor by the corresponding weight coefficients and then sums them to calculate the collaborative value conversion efficiency. The adjustment module is used to filter nodes whose collaborative value conversion efficiency is lower than the failure threshold as failure nodes, and adjust the routing weights of the failure nodes and their parallel candidate nodes. The monitoring module is used to update the data flow routing rules based on the adjusted routing weights, monitor the operating status of each node, and perform imbalance prevention operations.

[0007] The technical solution provided in this application constructs a data flow traceability map containing flow paths and attribute information, transforming the traditional unidirectional data transmission link into a traceable multidimensional collaborative network. This allows for precise location and traceability of every flow between the data owner, data manager, and data user. Based on this map, multi-dimensional indicators such as input quality score, output quality score, normalized processing time, user satisfaction score, and duplicate request rate are extracted for each collaborative node. The collaborative value conversion efficiency is calculated by summing the ratio of quality scores, the reciprocal of normalized processing time, the ratio of satisfaction to duplicate request rate, and the frequency of use factor, each multiplied by its corresponding weight coefficient. This comprehensive evaluation mechanism overcomes the limitations of traditional methods that rely solely on a single timeliness indicator or simple throughput statistics. It comprehensively evaluates collaborative nodes from three core dimensions: data quality value-added capability, processing efficiency, and actual downstream use value. This accurately identifies efficient nodes that truly create value during data flow and inefficient nodes that fail to play their role effectively. By selecting nodes with collaborative value conversion efficiency below the failure threshold as failed nodes and adjusting the routing weights of failed nodes and their parallel alternative nodes, adaptive migration of data traffic from inefficient nodes to efficient nodes is achieved, avoiding resource waste and service quality degradation caused by traditional static routing configurations. Based on the adjusted routing weights, data flow routing rules are updated, and target routing nodes are selected using a probability distribution method, dynamically matching data traffic allocation with the actual processing capacity of nodes. More importantly, by monitoring the operating status of each node and performing imbalance prevention operations, when a node simultaneously meets the conditions of exceeding the load threshold and its weight continuously increasing over multiple adjustment cycles, weight increases are frozen in a timely manner, and emergency weight adjustments for suboptimal nodes are initiated. This forms a multi-layered imbalance prevention mechanism, effectively solving the problem of introducing new system bottlenecks and resource allocation imbalances during dynamic weight adjustments.

[0008] In the specific application area of ​​data sharing and collaboration in public resource trading platforms, the collaborative value conversion efficiency calculation method of this application fully considers the business characteristics of different data types. By setting differentiated weight coefficient combinations for transaction entity data, transaction project data, transaction process data, credit data, and cross-regional transaction data, the algorithm can perform targeted weighted evaluations of quality conversion capability, processing timeliness, user satisfaction, and usage frequency according to the actual needs of different data types. For example, setting the quality conversion weight coefficient to the highest value for credit data reflects the strict requirements of the credit evaluation system for data accuracy; setting the timeliness weight coefficient to the highest value for transaction process data ensures the timely flow of process node information to meet compliance requirements; and setting the quality conversion, timeliness, and satisfaction weight coefficients to equal values ​​for cross-regional transaction data achieves a balance of multi-dimensional needs in cross-domain collaboration scenarios. This adaptive weight configuration mechanism based on business characteristics makes the algorithm evaluation results more closely match the actual operational needs of public resource trading platforms. Compared with the traditional method of using a unified standard to evaluate all nodes, it can more accurately identify the truly ineffective nodes in different data type processing scenarios. Meanwhile, the imbalance prevention mechanism avoids misjudgments that may occur by relying solely on load rate thresholds through the combination of dual monitoring indicators. The coordinated operation of weight freezing and emergency adjustment of suboptimal nodes ensures that the overall service capacity of the system is not affected while preventing overload. This algorithm feature has significant practical value in application scenarios such as public resource trading platforms that need to process multiple types of data and have extremely high requirements for service stability. Attached Figure Description

[0009] To more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings used in the description of the embodiments will be briefly introduced below. Obviously, the drawings described below are some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0010] Figure 1 This is a schematic diagram of an embodiment of the operation method of the data sharing and collaboration mechanism of the public resource trading platform in this application. Figure 2 This is a schematic diagram illustrating the changes in data quality scores in the embodiments of this application; Figure 3 This is a schematic diagram illustrating the effect of the imbalance prevention mechanism in the embodiments of this application. Detailed Implementation

[0011] This application provides a method and system for operating a data sharing and collaboration mechanism for a public resource trading platform. The terms "first," "second," "third," "fourth," etc. (if present)," in the specification, claims, and accompanying drawings of this application are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such data can be interchanged where appropriate so that the embodiments described herein can be implemented in a sequence other than that illustrated or described herein. Furthermore, the terms "comprising" or "having," and any variations thereof, are intended to cover non-exclusive inclusion; for example, a process, method, system, product, or device that includes a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or units not explicitly listed or inherent to such processes, methods, products, or devices.

[0012] For ease of understanding, the specific process of the embodiments of this application is described below. Please refer to [link / reference]. Figure 1 One embodiment of the data sharing and collaboration mechanism operation method of the public resource trading platform in this application includes: Step S1: Collect data flow records between the data owner, data manager, and data user, and construct a data flow traceability map based on the flow path and attribute information; The construction of the data flow traceability graph differs from traditional log recording. It adopts a graph structure storage method, where nodes represent three types of participating entities: data owner, data manager, and data user. Edges represent data flow paths, and the attribute information of the edges includes three core data types: timestamp, quality score, and usage feedback. This forms a complete data flow trajectory traceability system. Each edge in the graph is associated with a unique data identifier and data type label, enabling precise location and backtracking at every stage of the data flow process.

[0013] Step S2: Extract the input quality score, output quality score, normalized processing time, user satisfaction score, and repeat request rate of each collaborative node in the data flow traceability map. Multiply the quality score ratio, the reciprocal of the normalized processing time, the ratio of satisfaction to repeat request rate, and the usage frequency impact factor by their respective weight coefficients and sum them to calculate the collaborative value conversion efficiency. Specifically, the three dimensions of data quality improvement capability, processing timeliness, and downstream actual use value are integrated into a single quantitative indicator. The quality score ratio reflects the node's ability to add value to the data, the reciprocal of the normalized processing time reflects the node's efficiency deviation from the historical baseline, the ratio of satisfaction to duplicate request rate reflects the downstream's recognition of the data processing results, and the frequency of use influence factor avoids extreme value interference through logarithmic transformation. The calculation method of multiplying the four components by weight coefficients and then summing them ensures that different data types obtain differentiated evaluation standards according to business characteristics.

[0014] Step S3: Select nodes whose collaborative value conversion efficiency is lower than the failure threshold as failed nodes, and adjust the routing weights of the failed nodes and their parallel candidate nodes. Among them, the identification of failed nodes is based on the comparison between collaborative value conversion efficiency and failure threshold. The weight adjustment adopts a two-way mechanism. The weight of failed nodes is reduced by a decay factor, and the weight of parallel candidate nodes is enhanced by a boost factor. Parallel candidate nodes refer to nodes in the collaborative network topology that perform the same data processing functions as failed nodes and whose data flow can be interchanged. The calculation of weight adjustment parameters takes into account both the degree of failure and the load status of candidate nodes, avoiding over- or under-adjustment caused by simple fixed ratio adjustment.

[0015] Step S4: Update the data flow routing rules based on the adjusted routing weights, monitor the operating status of each node, and perform imbalance prevention operations.

[0016] Specifically, the data flow routing rule update adopts a probability distribution method, normalizes the weight of each node into a routing probability, selects the target node by matching random numbers with the cumulative probability interval, the imbalance prevention operation identifies potential overload risks through dual monitoring of load rate and weight growth rate, the weight freezing mechanism blocks further traffic tilting to high-load nodes, the emergency weight adjustment of suboptimal nodes diverts data traffic from overloaded nodes, and the multi-level prevention mechanism ensures that the dynamic weight adjustment process does not introduce new system bottlenecks.

[0017] In one specific embodiment, step S1 includes: The unique identifier, data type label, transfer start timestamp, and data source node identifier of the data when the data is transferred from the data owner to the data management party are recorded as the first transfer attribute. The processing completion timestamp, target node identifier, input quality score, and output quality score of the data user in the data management direction are obtained and recorded as the second flow attribute. The frequency of data usage, the rate of repeated requests, and the satisfaction rating of the users of the collected data are recorded as usage feedback attributes. Using data owners, data managers, and data users as nodes, data flow paths as edges, and the first flow attribute, second flow attribute, and usage feedback attribute as the attribute set of the edges, a data flow tracing graph is constructed.

[0018] Specifically, the first flow attribute records key information about the data flow from the originator to the manager. A unique data identifier assigns a globally unique code to each piece of flow data; this code remains unchanged throughout the data's lifecycle and serves as an index key for subsequent tracing. The data type label identifies which category the data belongs to: transaction entity data, transaction project data, transaction process data, credit data, or cross-regional transaction data. The flow start timestamp records the precise moment the data leaves the originator and is received by the manager. The data source node identifier records which originator node the data originates from. The second flow attribute records key information about the data flow from the manager to the user. The processing completion timestamp records the number of transactions completed by the manager. The data is processed and prepared for output to the user at the moment of processing. The target node identifier records which user node the data will flow to. The input quality score records the quality score value calculated based on the integrity score, accuracy score, and timeliness score when the data enters the management node. The output quality score records the processed quality score value when the data leaves the management node. The feedback attribute records the user's actual use of the received data. The usage frequency counts the total number of times the user accesses the data within the statistical period. The duplicate request rate calculates the ratio of the number of times the user repeatedly requests the same data because the data does not meet their needs to the total number of requests. The satisfaction score records the user's overall satisfaction with the data quality and timeliness.

[0019] The data flow traceability graph maps three main entities—owner, manager, and user—as nodes in the graph. Data flow paths are mapped as directed edges connecting these nodes. The direction of each edge indicates the direction of data flow. An edge from the owner to the manager carries a first flow attribute as its edge attribute, while an edge from the manager to the user carries a second flow attribute and a usage feedback attribute as its edge attributes. The complete flow of the same data is represented by two consecutive edges in the graph. The first edge connects the owner node and the manager node and includes a unique data identifier, data type label, flow start timestamp, and data source node identifier. The second edge connects the manager node and the user node and includes... The graph includes processing completion timestamp, target node identifier, input quality score, output quality score, usage frequency, duplicate request rate, and satisfaction score. Each management node in the graph acts as a collaborative node. Its incoming edges provide the initial state information when the node receives data, and its outgoing edges provide the processing result information and downstream feedback information when the node outputs data. By querying all incoming and outgoing edges of a management node in the graph, the graph extracts the input quality score, output quality score, flow start timestamp, processing completion timestamp, usage frequency, duplicate request rate, and satisfaction score from the edge attributes, providing all the necessary data to calculate the collaborative value conversion efficiency of that collaborative node.

[0020] In one specific embodiment, step S2 includes: Extract the set of edge attributes of each collaborative node in the data flow tracing graph within the statistical period, obtain the input quality score and output quality score from the second flow attribute, and calculate the ratio of the two to obtain the quality score ratio. Obtain the flow start timestamp from the first flow attribute and the processing completion timestamp from the second flow attribute. Calculate the time difference between the two to obtain the actual processing time. Query the historical average processing time baseline for the corresponding data type of the corresponding collaborative node. Divide the actual processing time by the historical average processing time baseline to obtain the normalized processing time. Obtain the user satisfaction score and repeat request rate from the user feedback attributes, calculate the ratio of the two to obtain the satisfaction repeat rate ratio, obtain the usage frequency, add one to the usage frequency and take the logarithm to obtain the usage frequency influence factor. Based on the data type label currently being processed, obtain the quality conversion weight coefficient, timeliness weight coefficient, satisfaction weight coefficient, and impact factor weight coefficient from the weight configuration table. Multiply the quality score ratio by the quality conversion weight coefficient, multiply the reciprocal of the normalized processing time by the timeliness weight coefficient, multiply the satisfaction repetition rate ratio by the satisfaction weight coefficient, and multiply the usage frequency impact factor by the impact factor weight coefficient. Sum the results of the four products to obtain the collaborative value conversion efficiency.

[0021] Specifically, the quality score ratio is calculated by dividing the output quality score by the input quality score. This ratio reflects the value-added processing capability of the collaborative node. A ratio greater than 1 indicates that the data quality has improved after processing by this node, a ratio equal to 1 indicates that the data quality has not changed, and a ratio less than 1 indicates that the data quality has decreased. The normalized processing time is calculated by dividing the actual processing time by the historical average processing time baseline. The historical average processing time baseline is stored in a historical statistics database, and an independent baseline value is maintained for each collaborative node and each data type. This baseline value is derived from the average processing time of the node for the same type of data in the past. The reciprocal of the normalized processing time is used to calculate the collaborative value conversion efficiency. The reciprocal processing ensures that the shorter the processing time, the greater the contribution value. The satisfaction repetition rate ratio is calculated by dividing the satisfaction score by the repetition request rate. To avoid division by zero, the minimum value is set when the repetition request rate is zero. This ratio reflects the downstream user's acceptance of the data processing results. A high satisfaction rate combined with a low repetition request rate produces a large ratio, indicating that the data processing meets the user's needs in one go. The usage frequency impact factor is calculated by adding one to the usage frequency and taking the common logarithm. The addition operation ensures that the logarithmic operation is effective when the usage frequency is zero. The logarithmic transformation converts the linearly increasing usage frequency into a logarithmically increasing impact factor, avoiding the excessive influence of extremely high usage frequency on the calculation of collaborative value conversion efficiency.

[0022] The weighting configuration table sets four weight coefficients for each of the five data types. Transaction entity data emphasizes quality conversion capability, transaction project data emphasizes processing timeliness, transaction process data emphasizes timeliness, credit data emphasizes quality conversion capability, and cross-regional transaction data maintains a balance across quality conversion, timeliness, and satisfaction. The calculation of collaborative value conversion efficiency involves multiplying the quality score ratio by the quality conversion weight coefficient to obtain the quality conversion contribution item, multiplying the reciprocal of the normalized processing time by the timeliness weight coefficient to obtain the timeliness contribution item, multiplying the satisfaction repetition rate ratio by the satisfaction weight coefficient to obtain the satisfaction contribution item, and using frequency... The frequency of use contribution is obtained by multiplying the impact factor by the impact factor weight coefficient. The sum of the four contribution values ​​is the collaborative value conversion efficiency of the collaborative node for the data flow record. The average collaborative value conversion efficiency of all data flow records processed by the collaborative node within the statistical period is taken as the comprehensive collaborative value conversion efficiency of the node in the current statistical period. This efficiency value comprehensively reflects the node's performance in four dimensions: data quality improvement capability, processing timeliness, downstream satisfaction, and actual use value. Compared with the traditional method that only evaluates data transmission speed, the collaborative value conversion efficiency evaluates the actual contribution of the collaborative node from the perspective of data value creation.

[0023] In one specific embodiment, the calculation process for the input quality score and the output quality score includes: Obtain data integrity score, accuracy score, and timeliness score, where integrity score is the data field completeness rate, accuracy score is the data validation pass rate, and timeliness score is the data timeliness score. The quality score is obtained by multiplying the completeness score by the first weight value, the accuracy score by the second weight value, and the timeliness score by the third weight value, and then summing the three products. The quality score before the data enters the collaborative node is calculated as the input quality score, and the quality score after the data is output to the collaborative node is calculated as the output quality score.

[0024] Specifically, the data field completeness rate is calculated by the ratio of the number of filled fields in the statistical data record to the total number of standard fields required for that data type. Standard fields for transaction entity data include mandatory fields such as company name, unified social credit code, company qualification, and legal representative. Standard fields for transaction project data include mandatory fields such as project number, project name, budget amount, and bidding method. A data record is considered incomplete if any standard field is empty or missing. The completeness score ranges from 0 to 100. The data validation pass rate is calculated by the ratio of the number of fields that pass format validation, range validation, and logical validation after performing format validation, range validation, and logical validation on each field in the data record to the total number of fields. The calculation shows that format validation checks whether field values ​​conform to predefined data formats such as date and amount formats; range validation checks whether numeric fields are within a reasonable range; and logic validation checks whether the logical relationships between multiple fields are consistent, such as the bid opening time being later than the tender announcement release time. The accuracy score ranges from 0 to 100. The data timeliness score is calculated based on the time difference between the data generation time and the current time. The smaller the time difference, the fresher the data and the higher the timeliness score. The specific calculation method is to set a timeliness threshold. When the time difference is less than the threshold, a full score of 100 points is obtained. When the time difference exceeds the threshold, points are deducted linearly according to the degree of timeout. The timeliness score ranges from 0 to 100.

[0025] The first, second, and third weight values ​​correspond to the relative importance of completeness, accuracy, and timeliness in the quality score calculation, respectively. The sum of the three weight values ​​equals 1. The completeness score is multiplied by the first weight value to obtain the completeness contribution score; the accuracy score is multiplied by the second weight value to obtain the accuracy contribution score; and the timeliness score is multiplied by the third weight value to obtain the timeliness contribution score. The sum of the three contribution scores yields the comprehensive quality score. The input quality score is calculated when the data first enters the collaborative node and is received by the management. At this time, the data records are extracted, the completeness score, accuracy score, and timeliness score are calculated, and then the weighted sum is obtained to obtain the input quality score. The output quality score is calculated when the data is ready to be output to the user after being processed by the collaborative nodes. At this time, the collaborative nodes have performed data cleaning, completion, and correction operations, and recalculated the integrity score, accuracy score, and timeliness score of the processed data. These scores are then weighted and summed to obtain the output quality score. The difference between the input quality score and the output quality score reflects the degree of improvement in data quality by the collaborative nodes. If the output quality score is higher than the input quality score, it means that the collaborative nodes have improved the data quality through data processing. The ratio of the two is the quality score ratio, which is one of the core indicators for calculating the collaborative value conversion efficiency.

[0026] Figure 2 This is a schematic diagram illustrating the changes in data quality scores in an embodiment of this application. Figure 2This diagram illustrates the changes in input and output quality scores for eight collaborative nodes during data processing. The horizontal axis represents different collaborative nodes (nodes A to H), and the vertical axis represents the quality score values ​​(ranging from 60 to 100). Solid circles indicate the input quality score before the data enters the collaborative node, while dashed squares indicate the output quality score after processing by the collaborative node. The value next to each node in the diagram represents the quality score ratio (output quality score / input quality score), which reflects the value-added processing capability of the collaborative node. The diagram shows that the output quality score of all collaborative nodes is higher than the input quality score, with quality score ratios ranging from 1.20 to 1.27, and an average quality improvement of 16.6 points. This verifies that the method described in this application can effectively improve data quality through collaborative node processing, and its quality transformation efficiency is significantly higher than that of traditional methods.

[0027] In one specific embodiment, the weight configuration table sets different combinations of weight coefficients for different data type labels: For transaction entity data, the quality conversion weight coefficient is set higher than the timeliness weight coefficient and higher than the impact factor weight coefficient; For transaction project data, the timeliness weight coefficient is set higher than the quality conversion weight coefficient and higher than the satisfaction weight coefficient; For transaction process data, the timeliness weight coefficient is set to the highest value among the four weight coefficients; Set the quality conversion weight coefficient for credit data to the highest value among the four weight coefficients; For cross-regional transaction data, the weight coefficients for quality conversion, timeliness, and satisfaction are set to be equal and all higher than the weight coefficients for influencing factors.

[0028] Specifically, transaction entity data includes basic entity attributes such as enterprise qualification information, personnel qualification information, and registration information. The core value of this type of data lies in its accuracy and completeness rather than its timeliness. Verifying the validity of enterprise qualification certificates and the authenticity of personnel professional qualifications requires in-depth data cleaning and verification by collaborative nodes. The quality conversion weight coefficient is set higher than the timeliness weight coefficient to reflect the emphasis on data value-added processing capabilities. The impact factor weight coefficient is set to the lowest value because the transaction entity data, as basic archival data, has a relatively stable usage frequency. Transaction project data includes dynamically changing project attributes such as basic project information, bidding and procurement information, and project progress information. The timeliness of this type of data directly affects the transaction entity. In bidding decisions and resource allocation, the time window from project information release to bid opening is often short. Therefore, the timeliness weight coefficient is set higher than the quality conversion weight coefficient to reflect the requirement for rapid processing and timely delivery. The satisfaction weight coefficient is set relatively low because project data is highly standardized and processing procedures are relatively rigid. Transaction process data records node information for transaction stages such as the release of bidding announcements, submission of bid documents, bid opening and evaluation, and announcement of winning bids. Process data must be completed within the specified time limit; otherwise, it will affect the legality and compliance of the entire transaction activity. The timeliness weight coefficient is set to the highest value among the four coefficients to ensure that the processing time of process data is minimized. The quality conversion weight coefficient and satisfaction weight coefficient are next, and the impact factor weight coefficient is the lowest.

[0029] Credit data includes evaluative information such as the transaction entity's performance records, violation penalty records, and credit scores. The accuracy and completeness of credit data directly affect the credibility of the credit rating system. Collaborative nodes need to perform complex quality improvement processes on multi-source credit data, such as cross-validation, deduplication and merging, and score calculation. The quality conversion weight coefficient is set to the highest value to emphasize the importance of credit data quality. The timeliness weight coefficient is second because credit records need to be updated promptly but are not as sensitive as process data. The satisfaction weight coefficient and the impact factor weight coefficient are set relatively low. Cross-regional transaction data involves data exchange and sharing between platforms in different regions, including cross-regional project information. Information such as remote bidding experts and cross-regional credit mutual recognition information requires high-quality transformation to eliminate differences in data standards across regions, high timeliness to support real-time business such as cross-regional remote bidding, and high satisfaction to promote inter-regional collaboration. The weighting coefficients for quality transformation, timeliness, and satisfaction are set to equal values ​​to reflect a balanced strategy that emphasizes all three. The weighting coefficients for influencing factors are set lower than the other three because the frequency of cross-regional transactions is relatively low. The differentiated design of the weighting configuration for the five data types allows the calculation of collaborative value transformation efficiency to be specifically evaluated based on data characteristics, avoiding evaluation bias caused by using a uniform standard.

[0030] In one specific embodiment, step S3 includes: Based on the data type label, the corresponding collaboration failure threshold is obtained from the configuration parameter table. The collaboration value conversion efficiency of each collaboration node is compared with the collaboration failure threshold, and nodes with collaboration value conversion efficiency lower than the collaboration failure threshold are selected as failure nodes. Query the current routing weight of the failed node, calculate the weight decay factor based on the difference between the collaborative failure threshold and the collaborative value conversion efficiency, and multiply the current routing weight by the weight decay factor to obtain the new weight of the failed node. From the data flow tracing graph, query the parallel candidate nodes that are at the same collaborative level as the failed node, obtain the current data processing load rate of each parallel candidate node, filter the parallel candidate nodes whose load rate is lower than the load threshold, calculate the weight boosting factor based on the difference between the load threshold and the load rate, and multiply the current routing weight of the parallel candidate node by the weight boosting factor to obtain the new weight of the parallel candidate node.

[0031] Specifically, the configuration parameter table sets collaboration failure thresholds for five data types. The failure threshold for transaction entity data is relatively low because of its high processing complexity; the failure threshold for transaction project data is higher because of its standardized processing flow; the failure threshold for transaction process data is highest because its processing steps are fixed; the failure threshold for credit data is lowest because of its stringent quality requirements; and the failure threshold for cross-regional transaction data is moderately high because of its moderate processing difficulty. The collaboration value conversion efficiency of a collaborative node is compared with the corresponding data type's collaboration failure threshold. When the collaboration value conversion efficiency is less than the failure threshold, the node is considered a failed node. A failed node indicates that the collaborative node failed when processing that type of data. The value conversion capability of the node did not meet the expected standard. The weight decay factor is calculated by subtracting the collaborative value conversion efficiency from the collaborative failure threshold. The larger the difference, the more severe the failure. The difference is divided by the collaborative failure threshold to obtain the failure ratio. The weight decay factor is obtained by subtracting the failure ratio from the fixed decay upper limit and multiplying it by the decay change range. The weight decay factor ranges from 0.7 to 0.9. The more severe the failure, the smaller the decay factor. The current routing weight of the failed node is stored in the collaborative node weight table. The current routing weight reflects the probability base of the node being selected during routing. The current routing weight is multiplied by the weight decay factor to obtain the new weight of the failed node. The new weight is less than the current routing weight, which reduces the probability of the failed node being selected in subsequent data flow routes.

[0032] Parallel alternative nodes refer to other collaborative nodes in the collaborative network topology of the data flow tracing graph that are at the same level as the failed node and perform the same data processing functions. The collaborative level is determined by the node's position in the data flow path. Nodes at the same level receive data from the same upstream node and output to the same downstream node. Parallel alternative nodes can substitute for each other in the data flow direction. The data processing load rate is calculated by dividing the number of data items currently being processed by the collaborative node by the node's maximum concurrent processing capacity. The maximum concurrent processing capacity is pre-configured based on the node's hardware resources and processing performance. The load rate ranges from 0 to 1, with a higher load rate indicating a busier node. A load threshold of 0.85 indicates that when a node's load rate exceeds 85%, it is considered a high-load state and further traffic should not be increased. Nodes with load rates lower than the load threshold are selected. The parallel candidate nodes with the threshold serve as candidate nodes that can take over the traffic of the failed nodes. The weight boosting factor is calculated by subtracting the load rate from the load threshold to obtain the load margin. The larger the load margin, the higher the idle level of the node. The load margin is divided by the load threshold to obtain the idle ratio. The weight boosting factor is obtained by adding the idle ratio to the fixed boosting lower limit and multiplying it by the boosting range. The weight boosting factor ranges from 1.1 to 1.3. The lower the load, the larger the boosting factor. The current routing weight of the parallel candidate node is multiplied by the weight boosting factor to obtain the new weight of the parallel candidate node. The new weight is greater than the current routing weight, which increases the probability of the parallel candidate node being selected in the subsequent data flow routing. Through the two-way adjustment mechanism of reducing the weight of failed nodes and increasing the weight of candidate nodes, the automatic migration of data traffic from inefficient nodes to efficient nodes is realized.

[0033] In one specific embodiment, step S4 includes: Update the new weights of the failed nodes and the new weights of the parallel alternative nodes to the cooperative node weight table, extract the updated weights of all processable nodes corresponding to each data type, calculate the total weights, and divide the weights of each node by the total weights to obtain the routing probability distribution. When a new data transfer request arrives, the corresponding route probability distribution is obtained according to the data type label, a random number is generated, the cumulative probability of each node is calculated, and the node whose cumulative probability interval contains the random number is selected as the target route node. Monitor the current load rate and routing weight of each collaborative node, and determine whether the node simultaneously meets the conditions that the load rate exceeds the load threshold and the weight continues to rise in multiple consecutive adjustment cycles. If the conditions are met, mark it as a warning node and freeze the weight increase operation of the corresponding node. Among the parallel candidate nodes corresponding to the early warning node, they are sorted from high to low according to the collaborative value conversion efficiency. The node with the second-best collaborative value conversion efficiency and the load rate is lower than the load threshold is selected as the second-best node. The emergency weight improvement factor is calculated and the routing weight of the second-best node is updated.

[0034] Specifically, the collaborative node weight table stores the current routing weight values ​​of all collaborative nodes. The weight table uses the node identifier as the index key. The new weights of failed nodes and parallel alternative nodes are written to the corresponding node's weight field to complete the weight update operation. All processable nodes for each data type refer to the set of all collaborative nodes capable of processing that data type. The updated weight values ​​of all nodes in this set are extracted from the weight table, and these weight values ​​are summed to obtain the total weight. The weight value of each node is divided by the total weight to obtain the routing probability of that node. The sum of the routing probabilities of all nodes equals 1. The routing probability distribution is stored in the data flow routing table using the data type label as the index key. Each data type corresponds to a set of routing probability distributions. When a new data flow request carrying a data type label arrives at the collaborative network, the routing table... The system queries the routing probability distribution corresponding to the data type, generates a random number between 0 and 1, and accumulates the probabilities of each node in the routing probability distribution according to their order of arrangement to obtain a cumulative probability sequence. The cumulative probability of the first node is equal to its routing probability, the cumulative probability of the second node is equal to the sum of the routing probabilities of the first two nodes, and so on, with the cumulative probability of the last node being equal to 1. The system then finds which node in the cumulative probability sequence has a cumulative probability interval that contains the generated random number. The cumulative probability interval is defined as the numerical range from the cumulative probability of the previous node to the cumulative probability of the current node. This node is selected as the target routing node, and the data flow request is assigned to the target routing node for processing. This probability distribution-based routing selection method ensures that nodes with high weights are selected more likely to be selected, while nodes with low weights are selected less likely to be selected, thus achieving a tilted distribution of data traffic to efficient nodes.

[0035] The monitoring module polls all nodes in the collaborative node weight table at fixed time intervals to obtain the current load rate and routing weight of each node. The current load rate reflects the node's data processing workload in real time, and the routing weight reflects the node's priority in route allocation. It records the routing weight change trajectory of each node over multiple consecutive adjustment periods. The adjustment period refers to the time interval between weight adjustments. The module determines whether a node meets two conditions: first, the current load rate is greater than the load threshold of 0.85; second, the routing weight in each of the last three consecutive adjustment periods is higher than the weight in the previous period. The cumulative growth rate exceeds 30%. The cumulative growth rate is calculated by subtracting the weight from three periods ago from the current period's weight, and then dividing by the weight three periods ago. When a node meets both conditions simultaneously, it indicates that although its weight is continuously increasing, it is approaching its load limit. Further weight increases will lead to overload. This node is marked as a warning node, and its freeze flag is set to true in the weight freeze flag table. Nodes with a true freeze flag are not allowed to perform weight increases during subsequent weight adjustments. The set of parallel candidate nodes corresponding to the warning node is queried from the data flow source map, and the collaborative value of each parallel candidate node in the current statistical period is extracted. The efficiency value is used to rank parallel candidate nodes from highest to lowest, with the node having the highest efficiency ranked first, the node with the second highest efficiency ranked second, and the second-ranked node selected as the second-best candidate. The current load rate of this second-best node is checked against a load threshold. If it is below the threshold, it is confirmed as the second-best node; otherwise, the next node in the ranking is checked until a second-best node that meets the load condition is found. The emergency weight enhancement factor is calculated with a larger enhancement margin than the regular enhancement factor, by adding a negative value to a fixed emergency enhancement lower limit. The emergency weight increase factor is obtained by multiplying the load margin ratio by the emergency increase change range. The value range of the emergency weight increase factor is between 1.2 and 1.3, which is higher than the range of 1.1 to 1.3 of the regular increase factor. The new routing weight of the suboptimal node is obtained by multiplying the current routing weight of the suboptimal node by the emergency weight increase factor. The weight value of the suboptimal node in the cooperative node weight table is updated. By freezing the weight increase of the warning node and the weight increase of the suboptimal node in the emergency, the overload of the warning node due to continuous weight increase is avoided. At the same time, the data traffic that should be allocated to the warning node is diverted to the suboptimal node, so as to achieve dynamic balance of the cooperative network load.

[0036] Figure 3 This is a schematic diagram illustrating the effect of the imbalance prevention mechanism in the embodiments of this application. Figure 3The comparison illustrates the trend of collaborative node load rate over time with and without an imbalance prevention mechanism. The horizontal axis represents time nodes (0-12 adjustment cycles), and the vertical axis represents node load rate (range 0.4-1.0). Solid circle curves represent load rate changes without an imbalance prevention mechanism, while dashed square curves represent load rate changes with the imbalance prevention mechanism. The dotted line in the figure marks the load threshold (0.85), and areas exceeding this threshold are overload risk zones (gray grid filled areas). As can be seen from the figure, without an imbalance prevention mechanism, the node load rate continuously rises after time node 6 and exceeds the load threshold, eventually reaching an overload state of 1.0. With the imbalance prevention mechanism, when the node load rate approaches the threshold, an early warning is triggered, and the weight increase operation is frozen. Simultaneously, an emergency weight adjustment for suboptimal nodes is initiated, causing the load rate to peak at time node 8 and then begin to fall back and stabilize within the safe range of 0.78-0.88, effectively preventing node overload and verifying the effectiveness of the imbalance prevention mechanism proposed in this application.

[0037] The above describes the operation method of the data sharing and collaboration mechanism of the public resource trading platform in the embodiments of this application. The following describes the operation system of the data sharing and collaboration mechanism of the public resource trading platform in the embodiments of this application. One embodiment of the operation system of the data sharing and collaboration mechanism of the public resource trading platform in the embodiments of this application includes: The data acquisition module is used to collect data flow records between data owners, data managers, and data users, and to build a data flow traceability map based on the flow path and attribute information. The calculation module is used to extract the input quality score, output quality score, normalized processing time, user satisfaction score and duplicate request rate of each collaborative node in the data flow tracing map. The module multiplies the quality score ratio, the reciprocal of the normalized processing time, the ratio of satisfaction to duplicate request rate and the usage frequency influence factor by the corresponding weight coefficients and then sums them to calculate the collaborative value conversion efficiency. The adjustment module is used to filter nodes whose collaborative value conversion efficiency is lower than the failure threshold as failure nodes, and adjust the routing weights of the failure nodes and their parallel candidate nodes. The monitoring module is used to update the data flow routing rules based on the adjusted routing weights, monitor the operating status of each node, and perform imbalance prevention operations.

[0038] The above embodiments are only used to illustrate the technical solutions of the present invention, and are not intended to limit it. Although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features. Such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A method for operating a data sharing and collaboration mechanism on a public resource trading platform, characterized in that, The method includes: Step S1: Collect data flow records between the data owner, data manager, and data user, and construct a data flow traceability map based on the flow path and attribute information; Step S2: Extract the input quality score, output quality score, normalized processing time, user satisfaction score, and duplicate request rate of each collaborative node in the data flow tracing map. Multiply the quality score ratio, the reciprocal of the normalized processing time, the ratio of satisfaction to duplicate request rate, and the usage frequency influence factor by their respective weight coefficients and sum them to calculate the collaborative value conversion efficiency. Step S3: Select nodes whose collaborative value conversion efficiency is lower than the failure threshold as failed nodes, and adjust the routing weights of the failed nodes and their parallel candidate nodes. Step S4: Update the data flow routing rules based on the adjusted routing weights, monitor the operating status of each node, and perform imbalance prevention operations.

2. The method for operating the data sharing and collaboration mechanism of the public resource trading platform according to claim 1, characterized in that, Step S1 includes: The unique identifier, data type label, transfer start timestamp, and data source node identifier of the data when the data is transferred from the data owner to the data management party are recorded as the first transfer attribute. The processing completion timestamp, target node identifier, input quality score, and output quality score of the data user in the data management direction are obtained and recorded as the second flow attribute. The frequency of data usage, the rate of repeated requests, and the satisfaction rating of the users of the collected data are recorded as usage feedback attributes. Using the data owner, data manager, and data user as nodes, the data flow path as edges, and the first flow attribute, the second flow attribute, and the usage feedback attribute as the attribute set of the edges, a data flow tracing graph is constructed.

3. The method for operating the data sharing and collaboration mechanism of the public resource trading platform according to claim 2, characterized in that, Step S2 includes: Extract the set of edge attributes of each collaborative node in the data flow tracing graph within the statistical period, obtain the input quality score and output quality score from the second flow attribute, and calculate the ratio of the two to obtain the quality score ratio. Obtain the flow start timestamp from the first flow attribute, obtain the processing completion timestamp from the second flow attribute, calculate the time difference between the two to obtain the actual processing time, query the historical average processing time baseline of the corresponding data type of the corresponding collaborative node, and divide the actual processing time by the historical average processing time baseline to obtain the normalized processing time. From the usage feedback attributes, obtain the usage satisfaction score and repeat request rate, calculate the ratio of the two to obtain the satisfaction repeat rate ratio, obtain the usage frequency, add one to the usage frequency and take the logarithm to obtain the usage frequency influence factor. Based on the data type label currently being processed, obtain the quality conversion weight coefficient, timeliness weight coefficient, satisfaction weight coefficient, and impact factor weight coefficient from the weight configuration table. Multiply the quality score ratio by the quality conversion weight coefficient, multiply the reciprocal of the normalized processing time by the timeliness weight coefficient, multiply the satisfaction repetition rate ratio by the satisfaction weight coefficient, and multiply the usage frequency impact factor by the impact factor weight coefficient. Sum the results of the four products to obtain the collaborative value conversion efficiency.

4. The method for operating the data sharing and collaboration mechanism of the public resource trading platform according to claim 3, characterized in that, The calculation process for input quality scores and output quality scores includes: Obtain data integrity score, accuracy score, and timeliness score, where integrity score is the data field completeness rate, accuracy score is the data validation pass rate, and timeliness score is the data timeliness score. The quality score is obtained by multiplying the completeness score by a first weight value, the accuracy score by a second weight value, and the timeliness score by a third weight value, and then summing the three products. The quality score before the data enters the collaborative node is calculated as the input quality score, and the quality score after the data is output to the collaborative node is calculated as the output quality score.

5. The method for operating the data sharing and collaboration mechanism of the public resource trading platform according to claim 4, characterized in that, In the weight configuration table, different weight coefficient combinations are set for tags of different data types: For transaction entity data, the quality conversion weight coefficient is set higher than the timeliness weight coefficient and higher than the impact factor weight coefficient; For transaction project data, the timeliness weight coefficient is set higher than the quality conversion weight coefficient and higher than the satisfaction weight coefficient; For transaction process data, the timeliness weight coefficient is set to the highest value among the four weight coefficients; Set the quality conversion weight coefficient for credit data to the highest value among the four weight coefficients; For cross-regional transaction data, the weight coefficients for quality conversion, timeliness, and satisfaction are set to be equal and all higher than the weight coefficients for influencing factors.

6. The method for operating the data sharing and collaboration mechanism of the public resource trading platform according to claim 5, characterized in that, Step S3 includes: The corresponding collaboration failure threshold is obtained from the configuration parameter table based on the data type label. The collaboration value conversion efficiency of each collaboration node is compared with the collaboration failure threshold, and nodes with collaboration value conversion efficiency lower than the collaboration failure threshold are selected as failure nodes. Query the current routing weight of the failed node, calculate the weight decay factor based on the difference between the collaborative failure threshold and the collaborative value conversion efficiency, and multiply the current routing weight by the weight decay factor to obtain the new weight of the failed node; From the data flow tracing graph, query the parallel candidate nodes that are at the same collaborative level as the failed node, obtain the current data processing load rate of each parallel candidate node, filter the parallel candidate nodes whose load rate is lower than the load threshold, calculate the weight boosting factor based on the difference between the load threshold and the load rate, and multiply the current routing weight of the parallel candidate node by the weight boosting factor to obtain the new weight of the parallel candidate node.

7. The method for operating the data sharing and collaboration mechanism of the public resource trading platform according to claim 6, characterized in that, Step S4 includes: The new weights of the failed nodes and the new weights of the parallel alternative nodes are updated to the collaborative node weight table. The updated weights of all processable nodes corresponding to each data type are extracted, the total weights are calculated, and the weights of each node are divided by the total weights to obtain the routing probability distribution. When a new data transfer request arrives, the corresponding routing probability distribution is obtained according to the data type label, a random number is generated, the cumulative probability of each node is calculated, and the node whose cumulative probability interval contains the random number is selected as the target routing node. Monitor the current load rate and routing weight of each collaborative node, and determine whether the node simultaneously meets the conditions that the load rate exceeds the load threshold and the weight continues to rise in multiple consecutive adjustment cycles. If the conditions are met, mark it as a warning node and freeze the weight increase operation of the corresponding node. Among the parallel candidate nodes corresponding to the early warning node, they are sorted from high to low according to their collaborative value conversion efficiency. The node with the second-best collaborative value conversion efficiency and a load rate lower than the load threshold is selected as the second-best node. The emergency weight enhancement factor is calculated and the routing weight of the second-best node is updated.

8. A data sharing and collaboration mechanism operation system for a public resource trading platform, characterized in that, For implementing the operation method of the public resource trading platform data sharing and collaboration mechanism as described in any one of claims 1-7, the public resource trading platform data sharing and collaboration mechanism operation system includes: The data acquisition module is used to collect data flow records between data owners, data managers, and data users, and to build a data flow traceability map based on the flow path and attribute information. The calculation module is used to extract the input quality score, output quality score, normalized processing time, user satisfaction score and duplicate request rate of each collaborative node in the data flow tracing map. The module multiplies the quality score ratio, the reciprocal of the normalized processing time, the ratio of satisfaction to duplicate request rate and the usage frequency influence factor by the corresponding weight coefficients and then sums them to calculate the collaborative value conversion efficiency. The adjustment module is used to filter nodes whose collaborative value conversion efficiency is lower than the failure threshold as failure nodes, and adjust the routing weights of the failure nodes and their parallel candidate nodes. The monitoring module is used to update the data flow routing rules based on the adjusted routing weights, monitor the operating status of each node, and perform imbalance prevention operations.

9. The system according to claim 8, characterized in that, Collect data flow records between data owners, data managers, and data users, and construct a data flow traceability map based on the flow path and attribute information, including: The unique identifier, data type label, transfer start timestamp, and data source node identifier of the data when the data is transferred from the data owner to the data management party are recorded as the first transfer attribute. The processing completion timestamp, target node identifier, input quality score, and output quality score of the data user in the data management direction are obtained and recorded as the second flow attribute. The frequency of data usage, the rate of repeated requests, and the satisfaction rating of the users of the collected data are recorded as usage feedback attributes. Using the data owner, data manager, and data user as nodes, the data flow path as edges, and the first flow attribute, the second flow attribute, and the usage feedback attribute as the attribute set of the edges, a data flow tracing graph is constructed.

10. The system according to claim 8, characterized in that, The data flow routing rules are updated based on the adjusted routing weights, and the operational status of each node is monitored and imbalance prevention operations are performed, including: The new weights of the failed nodes and the new weights of the parallel alternative nodes are updated to the collaborative node weight table. The updated weights of all processable nodes corresponding to each data type are extracted, the total weights are calculated, and the weights of each node are divided by the total weights to obtain the routing probability distribution. When a new data transfer request arrives, the corresponding routing probability distribution is obtained according to the data type label, a random number is generated, the cumulative probability of each node is calculated, and the node whose cumulative probability interval contains the random number is selected as the target routing node. Monitor the current load rate and routing weight of each collaborative node, and determine whether the node simultaneously meets the conditions that the load rate exceeds the load threshold and the weight continues to rise in multiple consecutive adjustment cycles. If the conditions are met, mark it as a warning node and freeze the weight increase operation of the corresponding node. Among the parallel candidate nodes corresponding to the early warning node, they are sorted from high to low according to their collaborative value conversion efficiency. The node with the second-best collaborative value conversion efficiency and a load rate lower than the load threshold is selected as the second-best node. The emergency weight enhancement factor is calculated and the routing weight of the second-best node is updated.

Images