Automatic archiving and forgetting control method based on enterprise memory heat and compliance strategy

Abstract

The application discloses an automatic archiving and forgetting control method based on enterprise memory heat and compliance strategy, and comprises the following steps: S1, structurally modeling enterprise memory objects in an enterprise information management system; S2, calculating heat index values of each enterprise memory object; S3, dynamically migrating the life cycle state of the enterprise memory object according to the calculated heat index values; and S4, updating the structural parameters of the enterprise memory object model, and repeatedly executing the heat index value calculation, comparison and migration process according to the updated data to realize continuous dynamic management of the enterprise memory object. The application can objectively quantify the active degree of the enterprise memory object, avoid relying on manual judgment, realize automatic migration of the enterprise memory object through state migration management corresponding to the heat index value, and reduce the workload of manual maintenance.

Description

Technical Field

[0001] This invention relates to the technical field of enterprise information management, specifically to an automatic archiving and forgetting control method based on enterprise memory popularity and compliance strategies. Background Technology

[0002] In enterprise information management systems, the vast amounts of knowledge, documents, business records, project data, and other stored data generated within an enterprise are typically managed in the following ways: Administrators periodically clean, archive, or delete documents manually, or process them uniformly according to fixed time rules (such as three or five years of retention). This method relies on manual judgment, resulting in low management efficiency and difficulty in adapting to the dynamic changes in the value of enterprise stored data. Enterprise stored data is categorized and stored using static tags such as folders, project numbers, and business types. Once categorized, these categories are rarely dynamically adjusted. Under this method, once an enterprise stored data object is classified, its storage location and management strategy are fixed, failing to reflect the actual usage frequency and business value changes of enterprise stored data at different stages. Data is cold-stored or deleted based on single conditions such as "last access time" or "whether it has not been accessed for a set time," leading to inaccurate assessments of the actual value of enterprise stored data. Summary of the Invention

[0003] This invention provides an automatic archiving and forgetting control method based on enterprise memory popularity and compliance strategy to solve the aforementioned technical problems. Specifically, the technical solution is as follows: An automated archiving and forgetting control method based on enterprise memory popularity and compliance strategy includes the following steps: S1: In the enterprise information management system, perform structured modeling of enterprise memory objects; S2: Calculate the popularity index value for each of the enterprise memory objects; S3: Based on the calculated heat index value, dynamically manage the lifecycle status of the enterprise memory object; S4: Update the structured parameters of the enterprise memory object model, and repeatedly execute the heat index value calculation, comparison and migration process according to the updated data to realize continuous dynamic management of enterprise memory objects.

[0004] Furthermore, the structured parameters of each enterprise memory object include at least an object identifier field, a creation time field, an access record field, a reference relationship field, and a lifecycle status field; Based on the above fields, within a fixed statistical period, the access record fields of the enterprise memory objects are statistically analyzed to calculate the access frequency factor. The access frequency factor is then subjected to time decay processing to obtain the time decay factor. By analyzing the reference relationships between the enterprise memory objects, the reference association factor is calculated. The access frequency factor, the time decay factor, and the reference association factor are used as input variables to calculate the popularity index value of each enterprise memory object.

[0005] Furthermore, the method for calculating the access frequency factor includes: Read access type records within the statistical period that include at least view access, edit access, and call access, and perform a weighted count according to the access type to obtain the weighted access value F for the statistical period. type The calculation formula is as follows: , Among them, a view a represents the number of times the view was accessed. edit a is the number of times the edit was accessed. call The number of times the access is called is given, and w1, w2, and w3 are the preset weights of the viewing access, the editing access, and the calling access, respectively, and satisfying w2 > w3 > w1; The statistical period is divided into two consecutive time windows, each with a length of d days, denoted as the near d-day window and the previous d-day window, respectively. The type-weighted access value F for the near d-day window is calculated for each window. recent The type-weighted access value F within the previous d days previous The time distribution adjustment coefficient D is calculated as follows: , Where ε is a preset non-zero smoothing constant; For each access behavior, its corresponding project identifier is extracted, and a project weight coefficient P is pre-set according to the importance of the project. The access frequency factor F is then calculated as follows: .

[0006] Furthermore, a time decay process is introduced into the access frequency factor to obtain the time decay factor, and the method includes: Obtain the access time corresponding to each access behavior for each access type, and calculate the time interval Δtk between the access time and the current time, denoted as: , Among them, t k Let t be the access time for the k-th access action. now This is the current calculation time; Based on a preset time decay function, a basic decay calculation is performed on the time interval corresponding to each access behavior to obtain the basic time decay factor T for that access behavior: , Where μ is the basic attenuation coefficient.

[0007] Furthermore, the method for obtaining the time decay factor by introducing a time decay process on the access frequency factor also includes: To obtain the project lifecycle information of the project to which the enterprise memory object belongs, and assuming the project phase adjustment coefficient is S, then the time decay factor T after project phase adjustment is: , Wherein, S is taken as S1 when the project to which the enterprise memory object belongs is in the execution phase, and S is taken as S2 when the project to which the enterprise memory object belongs is in the termination phase, and S1 is satisfied. <S2。

[0008] Furthermore, the method for calculating the reference association factor by analyzing the reference relationships between the enterprise memory objects includes: Read the reference relationship field of the target enterprise's memory object and construct the associated network structure; The set of first-level associated objects that directly reference the target enterprise memory object is identified based on the associated network structure, denoted as: N1(i), j∈N1(i) represents enterprise memory object j that directly references the target enterprise memory object i; Obtain the access frequency factor F corresponding to each first-level associated object j. j and time decay factor T j The two are then combined to obtain the basic association contribution value C of the first associated object. j : , Where η and θ are preset weighting coefficients; Calculate the reference association factor R of the enterprise memory object described in the target: .

[0009] Furthermore, using the access frequency factor, the time decay factor, and the reference association factor as input variables, the popularity index value H for each of the enterprise memory objects is calculated: , Wherein, α, β, and γ are preset weight coefficients, and satisfy α+β+γ=1.

[0010] Furthermore, the dynamic migration management of the lifecycle state of the enterprise memory object includes: The calculated heat index value is compared with a preset first heat threshold and a second heat threshold, wherein the first heat threshold is greater than the second heat threshold; When the popularity index value of the enterprise memory object is greater than the first popularity threshold, it is migrated to the active state; when the popularity index value of the enterprise memory object is less than the first popularity threshold but greater than the second popularity threshold, it is migrated to the cooling state; when the popularity index value of the enterprise memory object is continuously lower than the second popularity threshold within a preset continuous period, it is migrated to the archived state.

[0011] Furthermore, the dynamic migration management of the lifecycle state of the enterprise memory object also includes: when the time the enterprise memory object is in the archived state exceeds a preset value, the lifecycle state of the enterprise memory object is migrated from the archived state to the deleted state.

[0012] Furthermore, the dynamic migration management of the lifecycle state of the enterprise memory object also includes: Adjust the system resource allocation priority of the enterprise memory object according to its status.

[0013] The advantage of this invention lies in the fact that the automatic archiving and forgetting control method based on enterprise memory popularity and compliance strategy can objectively quantify the usage activity of enterprise memory objects, avoid relying on manual judgment, and realize the automatic migration of enterprise memory objects by continuously calculating the popularity index value of enterprise memory objects and the state migration management corresponding to the popularity index value. It can automatically switch enterprise memory objects between different life cycle states, reducing the workload of manual maintenance. Attached Figure Description

[0014] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0015] Figure 1 This is a schematic diagram of the automatic archiving and forgetting control method based on corporate memory popularity and compliance strategy proposed in this application. Detailed Implementation

[0016] Embodiments of the present invention are described in detail below. Examples of these embodiments are illustrated in the accompanying drawings, wherein the same or similar reference numerals denote the same or similar elements or elements having the same or similar functions throughout. The embodiments described below with reference to the accompanying drawings are exemplary and intended to explain the present invention, and should not be construed as limiting the present invention.

[0017] Current methods for managing enterprise memory objects lack dynamic evaluation mechanisms, often leaving enterprises with a passive choice between the cost pressures of long-term storage and the information risks of premature deletion. Long-term storage of all memory data leads to continuously increasing storage costs, while premature deletion may result in the loss of potentially valuable information. Current technologies cannot differentiate management based on the actual usage frequency of memory objects to achieve a balance between the two. Figure 1 As shown, this solution proposes an automatic archiving and forgetting control method based on enterprise memory popularity and compliance strategy, including the following steps: S1: In the enterprise information management system, perform structured modeling of enterprise memory objects; S2: Calculate the popularity index value for each of the enterprise memory objects; S3: Based on the calculated heat index value, dynamically manage the lifecycle status of the enterprise memory object; S4: Update the structured parameters of the enterprise memory object model, and repeatedly execute the heat index value calculation, comparison and migration process according to the updated data to realize continuous dynamic management of enterprise memory objects.

[0018] Regarding step S1 above, in the enterprise information management system, the system first performs structured modeling of enterprise memory objects to establish a data foundation for subsequent popularity calculation and lifecycle management. Specifically, each enterprise memory object is stored in the form of a structured data object, containing at least an object identifier field, a creation time field, an access record field, a reference relationship field, and a lifecycle status field. The object identifier field uniquely identifies the enterprise memory object; the creation time field records the time when the enterprise memory object was created; the access record field records the access time and access type of each access behavior; the reference relationship field records the relationships between the enterprise memory object and other enterprise memory objects; and the lifecycle status field records the current lifecycle status of the enterprise memory object. These attribute fields serve as the sole data source for subsequent popularity index calculation and lifecycle status migration.

[0019] Regarding step S2 above, after structured modeling is completed, the system, based on the access record field and reference relationship field of the enterprise memory object, performs statistical analysis on the access record field of the enterprise memory object within a fixed statistical period, calculates the access frequency factor, introduces time decay processing on the access frequency factor to obtain the time decay factor, and calculates the reference association factor by analyzing the reference relationships between the enterprise memory objects. The access frequency factor, the time decay factor, and the reference association factor are then used as input variables to calculate the popularity index value of each enterprise memory object. In other words, this popularity index value is calculated by comprehensively considering three determined popularity factors: the access frequency factor, the time decay factor, and the reference association factor. The implementation method is as follows: First, the access frequency factor is calculated by statistically analyzing access records within the statistical period and weighting the counts according to access type. Specifically: The system reads access type records within the statistical period. Access types include at least view access, edit access, and call access. The system performs a weighted count based on the access types to obtain a weighted access value F for the statistical period. type The calculation formula is as follows: , Among them, a view a represents the number of times the view was accessed. edit a is the number of times the edit was accessed. call The number of times the access is called is given, and w1, w2, and w3 are the preset weights of the viewing access, the editing access, and the calling access, respectively, and satisfying w2 > w3 > w1; The statistical period is divided into two consecutive time windows, each with a length of d days, denoted as the near d-day window and the previous d-day window, respectively. The type-weighted access value F for the near d-day window is calculated for each window. recent The type-weighted access value F within the previous d days previous The time distribution adjustment coefficient D is calculated as follows: , Where ε is a preset non-zero smoothing constant; For each access behavior, its corresponding project identifier is extracted, and a project weight coefficient P is pre-set according to the importance of the project. The access frequency factor F is then calculated as follows: .

[0020] The above scheme extracts the temporal distribution characteristics of access types, identifies the distribution of each access type within the statistical period, and calculates the access frequency factor by combining the temporal distribution adjustment coefficient and the item weight coefficient. In other words: Regarding access type weighting, the system reads access type records within the statistical period. Access types at least include view access, edit access, and call access. The system then performs a weighted count based on the access type to obtain the access type weighted access value F. type Among them, the preset weight w2 for editing access is the highest, followed by the preset weight w3 for calling access, and the preset weight w1 for viewing access is the lowest, thus reflecting the different contributions of different access behaviors to the activity level of enterprise memory objects.

[0021] Regarding time distribution adjustment, the system divides the statistical period into two consecutive time windows, each with a length of d days. It calculates the type-weighted access values ​​within the window of the most recent d days and the window of the previous d days, and calculates the time distribution adjustment coefficient D based on the ratio of the two: when D is large, it indicates that access behavior is more concentrated within the most recent d days, indicating that the recent activity of the enterprise's memory objects is relatively high; when D is close to 1, it indicates that access is relatively balanced within the two time windows; when D is small, it indicates that recent access activity has decreased.

[0022] Regarding project weighting, the system extracts the corresponding project identifier for each access behavior and pre-sets a project weight coefficient P based on the project's importance; the higher the project's importance, the larger the corresponding weight coefficient. Finally, the type-weighted access value F is calculated. type The access frequency factor F is obtained by comprehensively calculating the time distribution adjustment coefficient D and the project weight coefficient P.

[0023] This solution differentiates between different access types and assigns them varying weights, enabling the popularity assessment to reflect the actual value differences in access behaviors. Editing and calling accesses contribute more to popularity than viewing accesses, better aligning with the actual usage scenarios of enterprise-recognized objects. Furthermore, by introducing a time distribution adjustment coefficient, this solution identifies the distribution of access behaviors within the statistical period, distinguishing between recent activity levels and avoiding assessment biases caused by relying solely on cumulative access counts. Finally, by introducing a project weight coefficient, this solution reflects the importance of the project corresponding to the access behavior in the popularity calculation, ensuring that enterprise-recognized objects in important projects receive higher popularity assessments. Therefore, the access frequency factor not only reflects differences in the number of accesses and access types but also the time distribution characteristics of access behaviors and the importance of the projects corresponding to those behaviors.

[0024] Secondly, a time decay process is introduced into the access frequency factor to obtain the time decay factor. This is achieved by applying a basic decay calculation based on the time interval for each access behavior, and a project stage adjustment coefficient based on the project's current stage. Specifically: Obtain the access time corresponding to each access behavior for each access type, and calculate the time interval Δtk between the access time and the current time, denoted as: , where t kis the access time of the k-th access behavior, t now is the current calculation time. The larger the time interval, the longer ago the access behavior is from the current moment. Based on a preset time decay function, a basic decay calculation is performed on the time interval corresponding to each access behavior to obtain the basic time decay factor T of this access behavior: , where μ is the basic decay coefficient, used to control the decay speed. The output of the time decay function is the time contribution weight of the access behavior rather than the decay amount, and this weight decreases as the time interval increases, thus achieving a natural decay of the influence of historical access behaviors.

[0025] Through the above processing, this solution makes the access behavior with an access time closer to the current moment contribute more to the current popularity, and the access behavior with an earlier access time contribute less. That is to say, by introducing a basic decay calculation based on the time interval for each access behavior of each access type, the influence of historical access behaviors on the current popularity is gradually reduced, avoiding the continuous high popularity evaluation of memory objects that were frequently accessed in the early stage but rarely used recently, so that the popularity index value can more accurately reflect the current actual usage activity of the enterprise's memory objects, avoiding the long-term occupation of system resources by memories with low actual value, and improving the timeliness and accuracy of popularity evaluation.

[0026] Furthermore, the system also obtains the project cycle information of the project to which the enterprise memory object belongs. Let the project stage adjustment coefficient be S, then the time decay factor T after project stage adjustment is: , where S takes S1 when the project to which the enterprise memory object belongs is in the execution stage, and S takes S2 when the project to which the enterprise memory object belongs is in the end stage, and S1 < S2 is satisfied. In this way, on the basis of the basic decay calculation, the system further obtains the project cycle information of the project to which the enterprise memory object belongs, and adjusts the decay coefficient according to the project stage. Let the project stage adjustment coefficient be S. When the project to which the enterprise memory object belongs is in the execution stage, take S1; when it is in the end stage, take S2, and S1 < S2 is satisfied. That is, a lower stage adjustment coefficient is used in the project execution stage to reduce the decay intensity, so that the influence of the historical access behavior of the enterprise memory object in the project execution stage on the current popularity decays more slowly; a higher stage adjustment coefficient is used in the project end stage to increase the decay intensity, so that the influence of the historical access behavior of the enterprise memory object after the project ends on the current popularity decays faster. The time decay factor and the access frequency factor after project stage adjustment respectively characterize the enterprise memory object from the time dimension and the behavior frequency dimension, and the two are jointly used as input variables for calculating the popularity index.

[0027] The above solution introduces a project phase adjustment coefficient, enabling time decay processing to be differentiated according to the actual progress stage of the project to which the enterprise memory object belongs. During the project execution phase, the decay intensity is reduced to retain the contribution of historical visits to popularity, ensuring that enterprise memory objects associated with ongoing projects do not prematurely lose popularity evaluation over time. During the project completion phase, the decay intensity is increased to accelerate the decay of the impact of historical visits, allowing enterprise memory objects in completed projects to lose popularity evaluation more quickly. This better aligns with the actual business rhythm of enterprises and improves the rationality and accuracy of lifecycle state migration.

[0028] Finally, by analyzing the reference relationships between the enterprise memory objects, the reference association factor is calculated. This is achieved by constructing a directed graph with enterprise memory objects as nodes and reference relationships as directed edges, identifying the first-level associated objects that directly reference the target enterprise memory object, and then calculating the factor through a weighted sum based on the access frequency factor and the access value after time decay of the first-level associated objects. Specifically: The system reads the reference relationship field of the target enterprise memory object and constructs an association network structure. Here, the system uses enterprise memory objects as nodes and reference relationships as directed edges to form a directed graph: G = (V, E), where: V represents the set of enterprise memory objects; E represents the set of reference relationships. Based on the association network structure, the system identifies the set of first-level associated objects that directly reference the target enterprise memory object, denoted as: N1(i), j∈N1(i) represents enterprise memory object j that directly references the target enterprise memory object i. The system obtains the access frequency factor F corresponding to each first-level associated object j. j and time decay factor T j The two are then combined to obtain the basic association contribution value C of the first associated object. j : Where η and θ are preset weighting coefficients. Finally, the system calculates the reference association factor R of the target enterprise memory object: .

[0029] In this solution, the system first reads the reference relationship field of the target enterprise memory object and constructs an association network structure, using enterprise memory objects as nodes and reference relationships as directed edges to form a directed graph. Then, based on the association network structure, it identifies the set of first-level associated objects that directly reference the target enterprise memory object i. Next, the system obtains the access frequency factor F corresponding to each first-level associated object j. j and time decay factor T j The two are then combined according to preset weight coefficients η and θ to obtain the basic association contribution value C of the first-level associated object. jFinally, the basic association contribution values ​​of all first-level related objects are weighted and accumulated to calculate the reference association factor R of the target enterprise memory object. This process ensures that the reference association factor reflects not only the number of times it is referenced, but also the activity level and time decay of other memory objects referencing the enterprise memory object. This allows for a more accurate measurement of the importance and structural value of the enterprise memory object within the overall enterprise knowledge system, avoiding evaluation biases caused by relying solely on simple reference counting.

[0030] Based on the above factor calculations, the access frequency factor, time decay factor, and reference association factor are comprehensively calculated according to preset weight coefficients to generate a popularity index value for the target enterprise memory object. This popularity index value is used to quantitatively describe the activity level of the enterprise memory object in the current stage. Specifically: the popularity index value H for each enterprise memory object: Where α, β, and γ are preset weighting coefficients, and satisfy α+β+γ=1. This popularity index value serves as the sole criterion for determining lifecycle state migration in the system. Through this comprehensive calculation method, the popularity index value quantifies the activity level of enterprise memory objects from three dimensions: access frequency, time decay, and reference association. This allows the popularity index value to comprehensively and objectively reflect the actual activity level and structural importance of enterprise memory objects in the enterprise information management system, providing a unique and clear criterion for subsequent lifecycle state migration and ensuring the scientific nature and consistency of state migration decisions.

[0031] The above steps, by clarifying the fixed attribute structure of enterprise memory objects, provide a unified and standardized data foundation for the calculation of popularity indicators. Then, by decomposing the popularity indicator value into three dimensions—access frequency factor, time decay factor, and reference association factor—the popularity assessment of enterprise memory objects becomes more comprehensive and objective. It not only reflects the differences in access frequency and access type, but also the time distribution characteristics of access behavior, the importance of the corresponding items, and the structural importance of enterprise memory objects in the overall knowledge network.

[0032] Regarding step S3 above, the system dynamically manages the lifecycle states of enterprise memory objects based on the calculated heat index values. The system presets a first heat threshold and a second heat threshold, where the first heat threshold is greater than the second heat threshold. The calculated heat index values ​​are compared with the preset thresholds: when the heat index value is greater than the first heat threshold, the enterprise memory object migrates to the active state; when the heat index value is less than the first heat threshold but greater than the second heat threshold, it migrates to the cooling state; when the heat index value of an enterprise memory object in the cooling or archived state rises above the first heat threshold again, it migrates back to the active state. In this way, by setting two heat thresholds and establishing clear state migration rules, enterprise memory objects can automatically migrate between active, cooling, and archived states based on objective heat index values. This avoids the extensive management problems caused by relying on manual judgment or fixed time rules in existing technologies, achieving refined and hierarchical management of enterprise memory objects, reducing manual maintenance workload, and improving management efficiency.

[0033] Furthermore, when the time an enterprise memory object remains in the archived state exceeds a preset value, its lifecycle state is transitioned from archived to deleted. In other words, when an enterprise memory object is in the archived state, the system continuously monitors the duration of this state and determines whether a preset deletion condition is met. If the preset deletion condition includes the archived duration, the lifecycle state of the enterprise memory object is transitioned from archived to deleted. In this scheme, when the system receives a compliance deletion command, it also transitions the lifecycle state of the enterprise memory object from archived to deleted. Thus, by setting duration monitoring and deletion conditions for enterprise memory objects in the archived state, the system can automatically identify and clean up enterprise memory objects that have not been used for a long time, further reducing storage costs and operational pressure. Simultaneously, by introducing a compliance deletion command, it ensures that deletion operations respond to the enterprise's compliance policy requirements, controlling information risks while reducing costs, and achieving a balance between cost, efficiency, and risk.

[0034] Furthermore, the system applied in this solution adjusts the system resource allocation priority of the enterprise memory object based on its state. Specifically, the system adjusts the system resource allocation priority of the enterprise memory object according to its state: when the enterprise memory object migrates from the active state to the cooldown state, its system resource allocation priority is reduced; when the enterprise memory object migrates from the cooldown state to the archived state, it is transferred to a low-cost storage medium; when the enterprise memory object migrates from the archived state to the deleted state, a data clearing operation is performed and storage resources are released; when the enterprise memory object migrates back from the cooldown state or the archived state to the active state, its system resource allocation priority is restored and it is transferred to a high-performance storage medium. In this way, by dynamically adjusting the system resource allocation priority and storage medium according to the lifecycle status of enterprise memory objects, the system resource allocation is matched with the actual activity level of enterprise memory objects. That is, highly active enterprise memory objects can obtain higher resource priority and faster access performance, while inactive enterprise memory objects use lower-cost storage resources, thereby optimizing the overall system resource utilization efficiency, reducing storage and operation and maintenance costs, and ensuring that the access performance of active enterprise memory objects is not affected.

[0035] Regarding step S4 above, after completing one calculation of the popularity index value and the lifecycle state migration, the system updates the structured parameters of the enterprise memory object model, including updating the access record field, reference relationship field, and lifecycle state field, and repeats the popularity index value calculation, comparison, and migration process based on the updated data. Through this cyclical execution mechanism, the system achieves continuous dynamic management of enterprise memory objects, ensuring that the lifecycle state of enterprise memory objects continuously adapts to changes in actual usage needs.

[0036] The proposed automatic archiving and forgetting control method based on enterprise memory popularity and compliance strategies achieves refined and automated management of enterprise memory objects by performing structured modeling of enterprise memory objects in the enterprise information management system and calculating popularity index values ​​based on access records and reference relationships. It then dynamically manages the lifecycle states of enterprise memory objects based on these popularity index values. Compared with existing technologies, this invention can objectively quantify the usage activity of enterprise memory objects, avoiding reliance on manual judgment or single time rules. Furthermore, it achieves automatic migration of enterprise memory objects between active, cool-down, archived, and deleted states through popularity thresholds and state transition rules, reducing manual maintenance workload. During the calculation of popularity index values, a time decay process is introduced to gradually reduce the impact of historical access on current popularity, preventing low-value memories from occupying system resources for extended periods. Simultaneously, it supports automatic migration back to the active state when cooled-down or archived objects are frequently accessed again, ensuring that management results continuously align with actual usage needs. This reduces storage and maintenance costs and improves the overall system operation and management efficiency.

[0037] The foregoing has shown and described the basic principles, main features, and advantages of the present invention. Those skilled in the art should understand that the above embodiments do not limit the present invention in any way, and all technical solutions obtained by equivalent substitution or equivalent transformation fall within the protection scope of the present invention.

Claims

1. An automatic archiving and forgetting control method based on enterprise memory popularity and compliance strategy, characterized in that, Includes the following steps: S1: In the enterprise information management system, perform structured modeling of enterprise memory objects; S2: Calculate the popularity index value for each of the enterprise memory objects; S3: Based on the calculated heat index value, dynamically manage the lifecycle status of the enterprise memory object; S4: Update the structured parameters of the enterprise memory object model, and repeatedly execute the heat index value calculation, comparison and migration process according to the updated data to realize continuous dynamic management of enterprise memory objects.

2. The automatic archiving and forgetting control method based on enterprise memory popularity and compliance strategy as described in claim 1, characterized in that, Each of the enterprise memory objects has structured parameters that include at least an object identifier field, a creation time field, an access record field, a reference relationship field, and a lifecycle status field. Based on the above fields, within a fixed statistical period, the access record fields of the enterprise memory objects are statistically analyzed to calculate the access frequency factor. The access frequency factor is then subjected to time decay processing to obtain the time decay factor. By analyzing the reference relationships between the enterprise memory objects, the reference association factor is calculated. The access frequency factor, the time decay factor, and the reference association factor are used as input variables to calculate the popularity index value of each enterprise memory object.

3. The automatic archiving and forgetting control method based on enterprise memory popularity and compliance strategy as described in claim 2, characterized in that, The method for calculating the access frequency factor includes: Read access type records within the statistical period that include at least view access, edit access, and call access, and perform a weighted count according to the access type to obtain the weighted access value F for the statistical period. type The calculation formula is as follows: ， Among them, a view a represents the number of times the view was accessed. edit a is the number of times the edit was accessed. call The number of times the access is called is given, and w1, w2, and w3 are the preset weights of the viewing access, the editing access, and the calling access, respectively, and w2 > w3 > w1. The statistical period is divided into two consecutive time windows, each with a length of d days, denoted as the near d-day window and the previous d-day window, respectively. The type-weighted access value F for the near d-day window is calculated for each window. recent The type-weighted access value F within the previous d days previous The time distribution adjustment coefficient D is calculated as follows: ， Where ε is a preset non-zero smoothing constant; For each access behavior, its corresponding project identifier is extracted, and a project weight coefficient P is pre-set according to the importance of the project. The access frequency factor F is then calculated as follows: 。 4. The automatic archiving and forgetting control method based on enterprise memory popularity and compliance strategy as described in claim 3, characterized in that, The method for obtaining the time decay factor by introducing time decay processing on the access frequency factor includes: Obtain the access time corresponding to each access behavior for each access type, and calculate the time interval Δtk between the access time and the current time, denoted as: , Among them, t k Let t be the access time for the k-th access action. now This is the current calculation time; Based on a preset time decay function, a basic decay calculation is performed on the time interval corresponding to each access behavior to obtain the basic time decay factor T for that access behavior: ， Where μ is the basic attenuation coefficient.

5. The automatic archiving and forgetting control method based on enterprise memory popularity and compliance strategy according to claim 4, characterized in that, The method for obtaining the time decay factor by introducing time decay processing on the access frequency factor further includes: To obtain the project lifecycle information of the project to which the enterprise memory object belongs, and assuming the project phase adjustment coefficient is S, then the time decay factor T after project phase adjustment is: ， Wherein, S is taken as S1 when the project to which the enterprise memory object belongs is in the execution phase, and S is taken as S2 when the project to which the enterprise memory object belongs is in the termination phase, and S1 is satisfied. <S2。 6. The automatic archiving and forgetting control method based on enterprise memory popularity and compliance strategy according to claim 4, characterized in that, The method for calculating the reference association factor by analyzing the reference relationships between the enterprise memory objects includes: Read the reference relationship field of the target enterprise's memory object and construct the associated network structure; The set of first-level associated objects that directly reference the target enterprise memory object is identified based on the associated network structure, denoted as: N1(i), j∈N1(i) represents enterprise memory object j that directly references the target enterprise memory object i; Obtain the access frequency factor F corresponding to each first-level associated object j. j and time decay factor T j The two are then combined to obtain the basic association contribution value C of the first associated object. j : ， Where η and θ are preset weighting coefficients; Calculate the reference association factor R of the enterprise memory object described in the target: 。 7. The automatic archiving and forgetting control method based on enterprise memory popularity and compliance strategy as described in claim 6, characterized in that, Using the access frequency factor, the time decay factor, and the reference association factor as input variables, calculate the popularity index value H for each of the enterprise memory objects: ， Wherein, α, β, and γ are preset weight coefficients, and satisfy α+β+γ=1.

8. The automatic archiving and forgetting control method based on enterprise memory popularity and compliance strategy according to claim 1, characterized in that, The dynamic migration management of the lifecycle state of the enterprise memory object includes: The calculated heat index value is compared with a preset first heat threshold and a second heat threshold, wherein the first heat threshold is greater than the second heat threshold; When the popularity index value of the enterprise memory object is greater than the first popularity threshold, it is migrated to the active state; when the popularity index value of the enterprise memory object is less than the first popularity threshold but greater than the second popularity threshold, it is migrated to the cooling state; when the popularity index value of the enterprise memory object is continuously lower than the second popularity threshold within a preset continuous period, it is migrated to the archived state.

9. The automatic archiving and forgetting control method based on enterprise memory popularity and compliance strategy according to claim 8, characterized in that, The dynamic migration management of the lifecycle state of the enterprise memory object also includes: when the time the enterprise memory object is in the archived state exceeds a preset value, the lifecycle state of the enterprise memory object is migrated from the archived state to the deleted state.

10. The automatic archiving and forgetting control method based on enterprise memory popularity and compliance strategy according to claim 1, characterized in that, The dynamic migration management of the lifecycle state of the enterprise memory object also includes: Adjust the system resource allocation priority of the enterprise memory object according to its status.

Images