Method, system, electronic device and computer readable storage medium for dynamically generating points for knowledge contribution based on expert behavior

Abstract

The present application belongs to the technical field of knowledge community operation and expert incentive mechanism, and particularly relates to a knowledge contribution incentive method and system based on dynamic generation of expert behavior points, an electronic device and a computer readable storage medium. The historical behavior data of experts in the knowledge community is collected. Based on a preset fixed rule, the basic points are calculated according to the historical behavior data. Based on the historical behavior data, at least one dynamic coefficient reflecting the contribution quality of the expert is calculated in real time, and the dynamic coefficient includes at least one of the quality coefficient, the difficulty coefficient and the influence coefficient. Through a weighted fusion model, the basic points and the dynamic coefficient are fused to generate the final points for motivating the expert. The core is to introduce the dynamic coefficient calculated in real time based on the historical behavior, thereby breaking the limitation of the traditional fixed point rule "only quantity theory", and laying a foundation for subsequent precise and fair contribution quantification.

Description

Technical Field

[0001] This invention belongs to the technical field of knowledge community operation and expert incentive mechanism, specifically involving a knowledge contribution incentive method, system, electronic device and computer-readable storage medium based on dynamically generated points of expert behavior. Background Technology

[0002] In the operation and maintenance (O&M) of large equipment such as wind turbines, establishing and maintaining a high-quality knowledge base is crucial. The fault solutions and maintenance experience accumulated in this knowledge base can significantly improve O&M efficiency. To encourage domain experts to actively contribute their knowledge, many knowledge communities or O&M platforms have adopted point-based incentive systems.

[0003] Currently, most common expert point systems employ fixed and rigid rules, such as "1 point for publishing an answer" or "5 points for an accepted answer." Once these rules are set, they typically remain unchanged for a long period. Existing technologies suffer from the following main drawbacks: (1) Failure to reflect the true value of contributions: Fixed rules treat high-quality, painstaking solutions the same as low-quality, perfunctory solutions, and measure them only by simple standards such as "number of times" or "whether they are adopted", resulting in a serious disconnect between the scoring results and the actual value of the experts' contributions.

[0004] (2) Ignoring the difference in problem difficulty: Solving a rare and complex system failure requires significantly more intellectual effort and creates vastly different value than answering a common operational question. Existing fixed-integration rules fail to distinguish this difference in difficulty, thus discouraging experts from solving complex and high-value problems.

[0005] (3) Lack of dynamism and adaptability: The construction of knowledge bases is usually divided into different stages (such as encouraging content filling in the early stage and focusing on quality optimization in the later stage). Fixed rules cannot be dynamically adjusted according to changes in the community's development stage and operational goals, resulting in a rigid incentive orientation.

[0006] (4) Ignoring the long-tail impact of knowledge: A knowledge item that is cited multiple times and widely disseminated has a long-term value far greater than an isolated answer. Existing systems lack a quantitative assessment mechanism for the impact or dissemination of knowledge.

[0007] Therefore, there is an urgent need for a new method that can overcome the above-mentioned shortcomings, quantify expert knowledge contributions more fairly, accurately, and intelligently, and achieve effective incentives. Summary of the Invention

[0008] In view of this, the purpose of the present invention is to provide a knowledge contribution incentive method, system, electronic device and computer-readable storage medium based on dynamically generating integrals of expert behavior, so as to solve the technical problems involved in the above background art.

[0009] To achieve the above objectives, the technical solution adopted by the present invention is as follows: A knowledge contribution incentive method based on dynamically generated integrals from expert behavior includes the following steps: Collect historical behavioral data of experts in knowledge communities; Based on preset fixed rules, a basic score is calculated according to the historical behavior data; based on the historical behavior data, at least one dynamic coefficient reflecting the quality of expert contributions is calculated in real time, the dynamic coefficient including at least one of quality coefficient, difficulty coefficient and influence coefficient; The basic integral and the dynamic coefficient are fused using a weighted fusion model to generate the final integral used to incentivize the experts.

[0010] Furthermore, the historical behavior data includes: the number of answers, the number of accepted answers, the number of knowledge edits, user satisfaction rating data, and the number of answers to high-difficulty questions; wherein, high-difficulty questions are determined based on a preset difficulty threshold.

[0011] Furthermore, the quality coefficient is calculated based on the ratio of the number of adopted answers to the total number of answers, and the user satisfaction rating data; The difficulty coefficient is calculated based on the ratio of the number of answers to the number of responses to the highly difficult questions; The influence coefficient is calculated based on the ratio of the number of times the knowledge entries edited by the expert are cited by other content to the number of knowledge edits.

[0012] Furthermore, the weighted fusion model is as follows: FinalScore=BaseScore×(α+β×Q+γ×D+δ×I); Where FinalScore is the final score, BaseScore is the basic score, Q is the quality coefficient, D is the difficulty coefficient, I is the influence coefficient, and α, β, γ, and δ are preset weight parameters, and α+β+γ+δ=1.

[0013] Furthermore, it also includes applying the final score to at least one of the following scenarios: generating an expert score ranking list, distributing rewards based on the score ladder, using it as a basis for expert level promotion, and granting eligibility to participate in the review of specific projects.

[0014] Furthermore, it also includes: periodically collecting feedback data on the fairness of the integral, and adjusting and optimizing the weight parameters in the weighted fusion model based on the feedback data.

[0015] A knowledge contribution incentive system based on dynamically generated integrals from expert behavior includes: The data acquisition module is configured to collect historical behavioral data of experts in the knowledge community; The basic integral calculation module is configured to calculate the basic integral based on the historical behavior data according to preset fixed rules. The dynamic coefficient calculation module is configured to calculate at least one dynamic coefficient reflecting the quality of expert contributions in real time based on the historical behavior data. The integral fusion generation module is configured to fuse the basic integral with the dynamic coefficient through a weighted fusion model to generate the final integral used to incentivize the experts.

[0016] Furthermore, the dynamic coefficient calculation module is specifically configured to calculate the quality coefficient, difficulty coefficient, and influence coefficient; wherein, the quality coefficient is calculated based on the ratio of the number of adopted answers to the number of answers, and user satisfaction rating data; the difficulty coefficient is calculated based on the ratio of the number of answers to high-difficulty questions to the number of answers; and the influence coefficient is calculated based on the ratio of the number of times the knowledge entry edited by the expert is cited by other content to the number of knowledge edits.

[0017] An electronic device includes a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to implement the steps of the knowledge contribution incentive method based on dynamically generating integrals based on expert behavior.

[0018] A computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the steps of the knowledge contribution incentive method based on dynamically generating integrals based on expert behavior.

[0019] The beneficial effects of this invention are: This invention provides a knowledge contribution incentive method, system, electronic device, and computer-readable storage medium based on dynamically generated scores from expert behavior. The method involves collecting historical behavioral data of experts in knowledge communities; calculating a basic score based on the historical behavioral data according to preset fixed rules; calculating at least one dynamic coefficient reflecting the quality of expert contributions in real time based on the historical behavioral data, whereby the dynamic coefficient includes at least one of a quality coefficient, a difficulty coefficient, and an influence coefficient; and fusing the basic score and the dynamic coefficient through a weighted fusion model to generate a final score for incentivizing the expert. Its core lies in introducing a dynamic coefficient calculated in real time based on historical behavior, thereby breaking the limitations of traditional fixed score rules that rely solely on quantity, and laying the foundation for subsequent accurate and fair contribution measurement. Attached Figure Description

[0020] Figure 1The diagram shows a flowchart of a knowledge contribution incentive method based on dynamic generation of points from expert behavior according to the present invention. Figure 2 The diagram shown is a structural block diagram of a knowledge contribution incentive system based on dynamic generation of integrals from expert behavior according to the present invention. Figure 3 The diagram shown is a structural block diagram of an electronic device according to the present invention; Explanation of icon numbers: 1-Processor; 2-Memory; 101-Data Acquisition Module; 102-Basic Integral Calculation Module; 103-Dynamic Coefficient Calculation Module; 104-Integral Fusion Generation Module; 105-Integral Management and Application Module; 106-Model Configuration and Iteration Module. Detailed Implementation

[0021] The present invention will be further described below with reference to the accompanying drawings and specific embodiments: like Figure 1 As shown, the present invention provides a knowledge contribution incentive method based on dynamically generated integrals of expert behavior, comprising the following steps: Collect historical behavioral data of experts in knowledge communities; Based on preset fixed rules, a basic score is calculated according to the historical behavior data; based on the historical behavior data, at least one dynamic coefficient reflecting the quality of expert contributions is calculated in real time, the dynamic coefficient including at least one of quality coefficient, difficulty coefficient and influence coefficient; The basic integral and the dynamic coefficient are fused using a weighted fusion model to generate the final integral used to incentivize the experts.

[0022] As can be seen from the above description, the present invention has the following beneficial effects: This invention provides a knowledge contribution incentive method based on dynamically generated points for expert behavior. The method involves collecting historical behavioral data of experts in knowledge communities; calculating basic points based on the historical behavioral data according to preset fixed rules; calculating at least one dynamic coefficient reflecting the quality of expert contributions in real time based on the historical behavioral data, whereby the dynamic coefficient includes at least one of a quality coefficient, a difficulty coefficient, and an influence coefficient; and fusing the basic points and the dynamic coefficients through a weighted fusion model to generate a final score for incentivizing the experts. Its core lies in introducing dynamic coefficients calculated in real time based on historical behavior, thereby breaking the limitations of traditional fixed point rules that rely solely on quantity, and laying the foundation for subsequent accurate and fair contribution measurement.

[0023] Furthermore, the historical behavior data includes: the number of answers, the number of accepted answers, the number of knowledge edits, user satisfaction rating data, and the number of answers to high-difficulty questions; wherein, high-difficulty questions are determined based on a preset difficulty threshold.

[0024] As described above, the data sources for evaluating expert contributions are clearly defined, providing a basis for subsequent evaluation calculations. In particular, the introduction of the dimension of "answering difficult questions" can effectively distinguish between routine contributions and high-value contributions.

[0025] Furthermore, the quality coefficient is calculated based on the ratio of the number of adopted answers to the total number of answers, and the user satisfaction rating data; The difficulty coefficient is calculated based on the ratio of the number of answers to the number of responses to the highly difficult questions; The influence coefficient is calculated based on the ratio of the number of times the knowledge entries edited by the expert are cited by other content to the number of knowledge edits.

[0026] As described above, specific quantitative calculation methods for each dynamic coefficient are defined. The quality coefficient integrates objective adoption rate and subjective satisfaction, the difficulty coefficient focuses on the ability to solve complex problems, and the influence coefficient measures the dissemination and reuse value of knowledge. Together, they constitute a comprehensive and three-dimensional evaluation system for expert contributions.

[0027] Furthermore, the weighted fusion model is as follows: FinalScore=BaseScore×(α+β×Q+γ×D+δ×I); Where FinalScore is the final score, BaseScore is the basic score, Q is the quality coefficient, D is the difficulty coefficient, I is the influence coefficient, and α, β, γ, and δ are preset weight parameters, and α+β+γ+δ=1.

[0028] As described above, a preferred fusion model is disclosed. This model uses a basic integral as a benchmark and adjusts the weights through dynamic coefficients, resulting in a simple, efficient, and highly interpretable computation. The design of the weight parameters allows the system to flexibly adjust the incentive orientation according to the objectives of different operational stages.

[0029] Furthermore, it also includes applying the final score to at least one of the following scenarios: generating an expert score ranking list, distributing rewards based on the score ladder, using it as a basis for expert level promotion, and granting eligibility to participate in the review of specific projects.

[0030] As can be seen from the above description, by closely integrating the calculated points with actual incentive measures, a complete closed loop of "contribution-points-incentive" is formed, so that the points are no longer a virtual number, but a tool that can effectively drive expert behavior, improve community activity and knowledge quality.

[0031] Furthermore, it also includes: periodically collecting feedback data on the fairness of the integral, and adjusting and optimizing the weight parameters in the weighted fusion model based on the feedback data.

[0032] As described above, the introduction of feedback and iteration mechanisms enables the incentive method of this invention to continuously evolve, adapt to changes in community development and expert groups, and ensure the long-term effectiveness and fairness of the incentive system.

[0033] See Figure 2 The present invention also provides a knowledge contribution incentive system based on dynamically generated integrals of expert behavior, comprising: The data acquisition module is configured to collect historical behavioral data of experts in the knowledge community; The basic integral calculation module is configured to calculate the basic integral based on the historical behavior data according to preset fixed rules. The dynamic coefficient calculation module is configured to calculate at least one dynamic coefficient reflecting the quality of expert contributions in real time based on the historical behavior data. The integral fusion generation module is configured to fuse the basic integral with the dynamic coefficient through a weighted fusion model to generate the final integral used to incentivize the experts.

[0034] As can be seen from the above description, a system architecture for implementing the above method is provided, the division of each functional module is clearly defined, and a clear technical solution is provided for the specific implementation of the system.

[0035] Furthermore, the dynamic coefficient calculation module is specifically configured to calculate the quality coefficient, difficulty coefficient, and influence coefficient; wherein, the quality coefficient is calculated based on the ratio of the number of adopted answers to the number of answers, and user satisfaction rating data; the difficulty coefficient is calculated based on the ratio of the number of answers to high-difficulty questions to the number of answers; and the influence coefficient is calculated based on the ratio of the number of times the knowledge entry edited by the expert is cited by other content to the number of knowledge edits.

[0036] As described above, the specific calculation logic of the dynamic coefficients is further defined at the system level, ensuring that the system can automatically and accurately perform multi-dimensional contribution assessment.

[0037] like Figure 3As shown, the present invention also provides an electronic device, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to implement the steps of the knowledge contribution incentive method based on expert behavior dynamic generation of integrals.

[0038] As can be seen from the above description, extending the method of the present invention to hardware entities clarifies that it can be implemented through general-purpose computing devices, thereby enhancing the practicality and coverage of the protection scope.

[0039] A computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the steps of the knowledge contribution incentive method based on dynamically generating integrals based on expert behavior.

[0040] As can be seen from the above description, a software carrier for carrying the method of the present invention is provided, which protects the program code itself that implements the method, thus forming a complete patent protection chain from the method to the device to the medium.

[0041] The following are several preferred embodiments or application embodiments to help those skilled in the art better understand the technical content of the present invention and the technical contributions made by the present invention compared with the prior art: Preferred embodiment 1: refer to Figure 1 This embodiment provides a knowledge contribution incentive method based on dynamically generated integrals from expert behavior. The specific steps are as follows: S101: Collect historical behavioral data of experts.

[0042] The system continuously records and updates the activity logs of each expert in the wind turbine knowledge community. Key data collected includes: answer_count (total number of answers), adopted_count (number of adopted answers), edit_count (number of edited knowledge entries), satisfaction_avg (average user satisfaction rating for their answers, 1-5 points), and hard_q_count (number of difficult questions answered). Difficult questions are defined as new questions within the community with a confidence level below 0.3 (a preset threshold).

[0043] S102: Calculate the basic integral.

[0044] BaseScore is calculated according to pre-set fixed rules. For example: +1 point for each answer, +2 points for each knowledge edit, +5 points for each accepted answer, and +0.5 points for each user like. This step retains the simplicity and transparency of traditional point systems.

[0045] S103: Calculate the dynamic coefficients.

[0046] Based on the data collected by S101, three normalized (e.g., to the [0,1] interval) dynamic coefficients are calculated: (1) Quality coefficient Q: measures the reliability and popularity of expert answers.

[0047] Q=0.6×(adopted_count / (answer_count+ε))+0.4×(satisfaction_avg / 5); Here, ε is a very small number (e.g., 1e-6) to prevent division by zero errors. The adoption rate weight (0.6) and satisfaction weight (0.4) in the formula can be adjusted as needed.

[0048] (2) Difficulty coefficient D: measures the expert’s tendency and ability to solve complex problems.

[0049] D=hard_q_count / (answer_count+ε); (3) Influence coefficient I: measures the dissemination value and long-term utility of the knowledge contributed by experts.

[0050] I = Σ(citation count) / (edit_count + ε); Wherein, Σ (citation count) refers to the total number of times the knowledge entry created or edited by the expert is cited by other questions and answers or knowledge entries.

[0051] S104: Weighted fusion generates the final integral.

[0052] A weighted fusion model is used to combine the basic integral with the dynamic coefficients: FinalScore=BaseScore×(0.5+0.3×Q+0.1×D+0.1×I); In this example, the weighting is as follows: baseline 0.5, quality coefficient 0.3, difficulty coefficient 0.1, and influence coefficient 0.1. This weighting indicates that while acknowledging basic contributions, the system prioritizes incentivizing high-quality contributions. Operators can dynamically adjust these weights based on different objectives such as "encouraging exploration" (increasing the weight of D) or "encouraging the accumulation of high-quality content" (increasing the weight of I).

[0053] S105: Application and incentives of points.

[0054] At the beginning of each month, the final points for the previous month are calculated and applied to: (a) Generate a top 20 leaderboard of expert points across the entire site and award them with honors and incentives; (b) Cash rewards will be distributed based on a points tier system (e.g., 1000 points can be redeemed for 100 yuan). (c) Set a points threshold as the basis for automatic promotion of expert levels (junior, intermediate, senior, chief); (d) Experts with high scores will automatically be given priority to participate in major project reviews or internal technical seminars.

[0055] S106: Model Iterative Optimization.

[0056] Quarterly feedback on the fairness of the scoring system is collected from experts through questionnaires or interviews. Simultaneously, A / B testing is conducted, for example, by applying different weight combinations to two groups of experts (e.g., one group focusing on quality, the other on difficulty), comparing changes in their behavior and knowledge output. Based on the feedback and test results, the weights of the fusion model in S104 are optimized and updated, and a new version of the rules is released.

[0057] The advantages of this preferred embodiment are as follows: 1. By introducing multi-dimensional dynamic coefficients such as quality, difficulty, and influence, the evaluation of experts' contributions is transformed from a single "quantity" evaluation to a three-dimensional evaluation of "quantity + quality + value," realizing "more work, more pay; better work, better pay," and making incentives more fair and precise.

[0058] 2. By adjusting the weight parameters in the weighted fusion model (such as initially increasing the weight of the difficulty coefficient D to encourage solving difficult problems), the points system can flexibly adapt to the operational needs of different stages of knowledge base construction and guide expert behavior toward the expected goals.

[0059] 3. The introduction of high difficulty and impact coefficients enables higher returns for solving complex problems and producing knowledge with dissemination value, fundamentally motivating experts to dare to challenge difficult problems and be happy to share high-quality knowledge.

[0060] 4. This method combines short-term behavioral incentives with long-term contribution value assessment, which helps to build a healthy and sustainable knowledge co-creation ecosystem, thereby continuously improving the quality and usability of the entire knowledge base.

[0061] Preferred embodiment two: See Figure 2 The present invention also provides a knowledge contribution incentive system based on dynamically generated integrals of expert behavior, comprising: The data acquisition module 101 is connected to the knowledge community database and is used to automatically capture and update the behavioral data of experts.

[0062] The basic score calculation module 102 is configured with a fixed score rule library and is used to calculate basic scores based on behavioral data.

[0063] The dynamic coefficient calculation module 103 includes built-in sub-modules for calculating quality, difficulty, and influence coefficients, used for real-time calculation of various dynamic coefficients. Specifically, it is configured to calculate the quality coefficient, difficulty coefficient, and influence coefficient; wherein, the quality coefficient is calculated based on the ratio of the number of accepted answers to the total number of answers, and user satisfaction rating data; the difficulty coefficient is calculated based on the ratio of the number of answers to high-difficulty questions to the total number of answers; and the influence coefficient is calculated based on the ratio of the number of times the knowledge entry edited by the expert is cited by other content to the total number of knowledge edits.

[0064] The integral fusion generation module 104 has a built-in configurable weighted fusion model for calculating the final integral.

[0065] The points management and application module 105 is used to store points, generate leaderboards, trigger reward distribution, and handle level changes.

[0066] The model configuration and iteration module 106 provides a management interface for operators to adjust weights and collect feedback data to drive model optimization.

[0067] The beneficial effects of this preferred embodiment two are as follows: Example 2 provides a system architecture for implementing the above method, clearly defining the division of each functional module and providing a clear technical solution for the specific implementation of the system. At the system level, the specific calculation logic of the dynamic coefficients is further defined, ensuring that the system can automatically and accurately perform multi-dimensional contribution assessment.

[0068] Preferred embodiment three: like Figure 3 As shown, this embodiment provides an electronic device, including a memory 2 and a processor 1. The memory 2 stores a computer program, which, when executed by the processor 1, can implement all or part of the steps S101 to S106 described in Embodiment 1 above. This electronic device can be a backend server for a knowledge community.

[0069] The beneficial effects of this preferred embodiment three are as follows: Example 3 extends the method of the present invention to hardware entities, clarifying that it can be implemented through general computing devices, thereby enhancing the practicality and coverage of the protection scope.

[0070] Preferred embodiment four: This embodiment provides a computer-readable storage medium, such as a USB flash drive, hard disk, or cloud storage space. The medium stores a computer program, which, when executed by a processor, can perform all or part of the steps S101 to S106 described in Embodiment 1 above.

[0071] The advantages of this preferred embodiment four are as follows: Example 4 provides a software carrier for the method of the present invention, which protects the program code that implements the method itself, thus forming a complete patent protection chain from the method to the device to the medium.

[0072] It should be noted that the ε, the calculation formulas of each coefficient, and the weight values ​​in the fusion model described in the above embodiments are all examples. Those skilled in the art can make adjustments and substitutions according to actual business logic, and these adjustments and substitutions all fall within the protection scope of this invention.

[0073] The present invention has been described with reference to the foregoing embodiments and accompanying drawings; however, the foregoing embodiments are merely examples for implementing the present invention. It must be noted that the disclosed embodiments do not limit the scope of the present invention. Conversely, modifications and equivalents included within the spirit and scope of the claims are all included within the scope of the present invention.

Claims

1. A knowledge contribution incentive method based on dynamically generated integrals from expert behavior, characterized in that, Includes the following steps: Collect historical behavioral data of experts in knowledge communities; Based on preset fixed rules, the basic score is calculated according to the historical behavior data; Based on the historical behavioral data, at least one dynamic coefficient reflecting the quality of expert contributions is calculated in real time, and the dynamic coefficient includes at least one of the quality coefficient, difficulty coefficient, and influence coefficient. The basic integral and the dynamic coefficient are fused using a weighted fusion model to generate the final integral used to incentivize the experts.

2. The knowledge contribution incentive method based on dynamically generated integrals of expert behavior according to claim 1, characterized in that, The historical behavior data includes: the number of answers, the number of accepted answers, the number of knowledge edits, user satisfaction rating data, and the number of answers to high-difficulty questions; among which, high-difficulty questions are determined based on a preset difficulty threshold.

3. The knowledge contribution incentive method based on dynamically generated integrals of expert behavior according to claim 2, characterized in that, The quality coefficient is calculated based on the ratio of the number of adopted answers to the total number of answers, and the user satisfaction rating data. The difficulty coefficient is calculated based on the ratio of the number of answers to the number of responses to the highly difficult questions; The influence coefficient is calculated based on the ratio of the number of times the knowledge entries edited by the expert are cited by other content to the number of knowledge edits.

4. The knowledge contribution incentive method based on dynamically generated integrals of expert behavior according to claim 3, characterized in that, The weighted fusion model is as follows: FinalScore=BaseScore×(α+β×Q+γ×D+δ×I); Where FinalScore is the final score, BaseScore is the basic score, Q is the quality coefficient, D is the difficulty coefficient, I is the influence coefficient, and α, β, γ, and δ are preset weight parameters, and α+β+γ+δ=1.

5. The knowledge contribution incentive method based on dynamically generated integrals of expert behavior according to claim 1, characterized in that, Also includes: The final score can be applied to at least one of the following scenarios: generating an expert score leaderboard, distributing rewards based on the score ladder, serving as a basis for expert level promotion, and granting eligibility to participate in the review of specific projects.

6. The knowledge contribution incentive method based on dynamically generated integrals of expert behavior according to claim 1, characterized in that, Also includes: Feedback data on the fairness of the integral is collected periodically, and the weight parameters in the weighted fusion model are adjusted and optimized based on the feedback data.

7. A knowledge contribution incentive system based on dynamically generated integrals from expert behavior, characterized in that, include: The data acquisition module is configured to collect historical behavioral data of experts in the knowledge community; The basic integral calculation module is configured to calculate the basic integral based on the historical behavior data according to preset fixed rules. The dynamic coefficient calculation module is configured to calculate at least one dynamic coefficient reflecting the quality of expert contributions in real time based on the historical behavior data. The integral fusion generation module is configured to fuse the basic integral with the dynamic coefficient through a weighted fusion model to generate the final integral used to incentivize the experts.

8. The knowledge contribution incentive system based on dynamically generated integrals of expert behavior according to claim 7, characterized in that, The dynamic coefficient calculation module is specifically configured to calculate the quality coefficient, difficulty coefficient, and influence coefficient. The quality coefficient is calculated based on the ratio of the number of adopted answers to the total number of answers, as well as user satisfaction rating data. The difficulty coefficient is calculated based on the ratio of the number of answers to high-difficulty questions to the total number of answers. The influence coefficient is calculated based on the ratio of the number of times the knowledge entry edited by the expert is cited by other content to the total number of knowledge edits.

9. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that, When the processor executes the computer program, it implements the steps of the knowledge contribution incentive method based on dynamic generation of integrals based on expert behavior as described in any one of claims 1 to 6.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by the processor, it implements the steps of the knowledge contribution incentive method based on dynamic generation of integrals based on expert behavior as described in any one of claims 1 to 6.

Images