CN115907986B - Asset checking method and device, medium and computer equipment - Google Patents
Asset checking method and device, medium and computer equipment Download PDFInfo
- Publication number
- CN115907986B CN115907986B CN202211431042.1A CN202211431042A CN115907986B CN 115907986 B CN115907986 B CN 115907986B CN 202211431042 A CN202211431042 A CN 202211431042A CN 115907986 B CN115907986 B CN 115907986B
- Authority
- CN
- China
- Prior art keywords
- asset
- model
- target
- tables
- checking
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
Landscapes
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
- Stored Programmes (AREA)
Abstract
Embodiments of the present disclosure provide an asset collation method and apparatus, medium and computer device, the method comprising: determining a target asset model to be checked, wherein the target asset model is used for recording a plurality of target asset tables and asset change information among the target asset tables; acquiring subscription information of a target check model, wherein the target check model is obtained by training based on a plurality of first sample asset tables and asset change information among the plurality of first sample asset tables; and creating a first asset checking task for calling the target checking model to check the target asset model based on the subscription information.
Description
Technical Field
The present disclosure relates to the field of risk control technology, and in particular, to an asset checking method and apparatus, a medium, and a computer device.
Background
In order to ensure the correctness of the asset change information during the transaction or the like by the user, the asset check is required to avoid the problem of the asset loss as much as possible. In the related art, a collation model for performing asset collation is generally generated based on expert experience. However, the number of transactions is often very large, information related to various transactions is very complicated, and collation rules for collation models are also various. Therefore, the method of generating the collation model based on the experience of the expert has a problem of high labor cost.
Disclosure of Invention
In a first aspect, embodiments of the present disclosure provide an asset collation method, the method comprising: determining a target asset model to be checked, wherein the target asset model is used for recording a plurality of target asset tables and asset change information among the target asset tables; acquiring subscription information of a target check model, wherein the target check model is obtained by training based on a plurality of first sample asset tables and asset change information among the plurality of first sample asset tables; and creating a first asset checking task for calling the target checking model to check the target asset model based on the subscription information.
In some embodiments, the determining the target asset model to be reconciled comprises: acquiring search information input by a user and a pre-established topology model; the topology model comprises a plurality of nodes and edges connected with the nodes, each node corresponds to a search object corresponding to one item of search information, and each edge is used for representing the association relation between two search objects or representing the association relation between one search object and the target asset model; and determining a target asset model to be checked based on the search information and the topology model.
In some embodiments, the search information includes at least one of: identification information of asset transfer links corresponding to a plurality of target asset tables in the target asset model; an order number in any one of the target asset tables in the target asset model; identification information of any target asset table in the target asset model; the identification information is used for being related to an application, and the application is used for operating a plurality of target asset tables in the target asset model; identification information of the created target asset model.
In some embodiments, the method further comprises: pushing at least one candidate collation model in the model database to the user; the target collation model is determined from the at least one candidate collation model based on a selection instruction sent by a user.
In some embodiments, the pushing the at least one candidate collation model in the model database to the user comprises: verifying the candidate verification model based on a plurality of second sample asset tables and asset change information among the plurality of second sample asset tables; and pushing the candidate verification model which is verified successfully to the user.
In some embodiments, the method further comprises: generating a first report, the first report comprising at least one of: the number of candidate check models newly added in the model database; the number of newly added target asset models; the total number of the newly added target fields to be checked; the number of updated target asset models; the number of candidate verification models that are successfully verified; the success rate of verification of each candidate verification model; the ratio of the number of candidate verification models which are successfully verified to the total number of candidate verification models; and checking results of checking the candidate checking model through the second sample asset table each time.
In some embodiments, the method further comprises: and under the condition that a preset updating triggering condition is met, updating the target check model.
In some embodiments, the update trigger condition includes at least one of: reaching a preset updating time interval; the checking accuracy of the target asset model is lower than a preset accuracy threshold; and receiving an update instruction input by a user.
In some embodiments, after creating a first asset reconciliation task that invokes the target reconciliation model to reconcile a target asset model based on the subscription information, the method further comprises: and if the checking result of the first asset checking task is abnormal, outputting alarm information.
In some embodiments, the method further comprises: if the checking result of the first asset checking task is abnormal, acquiring first marking information of the target asset model and second marking information of the target checking model; the first marking information is used for representing whether the target asset model is correct or not, and the second marking information is used for representing whether the checking result is correct or not; and if the target asset model and the checking result are both correct, generating a correlation event corresponding to the checking result, wherein the correlation event is used for carrying out exception processing on the checking result.
In some embodiments, the method further comprises: in the asset tables generated in the target asset transfer scene, if the number of the first asset tables is larger than a preset first number threshold, determining the first asset tables as first sample asset tables of the target asset transfer scene; and/or determining the first asset change information as the asset change information among the plurality of first sample asset tables if the number of the first sample asset tables meeting the first asset change information is greater than a preset second number threshold value.
In some embodiments, the method further comprises: creating an empty second asset reconciliation task if the target asset model is not acquired; after the target asset model is acquired, the target asset model is added to the second asset collation task.
In some embodiments, the method further comprises: creating an empty third asset reconciliation task if the target reconciliation model was not obtained; after the target collation model is acquired, the target asset model is added to the third asset collation task.
In some embodiments, the method further comprises: generating a second report; the second report includes at least one of the following information: the number of target asset models being collated; the number of target collation models for asset collation; alarming times when the checking result of the first asset checking task is abnormal; discovery rate of abnormal condition of asset change information.
In some embodiments, the method further comprises: acquiring a configuration instruction input by a user; and configuring parameters of the target checking model and/or target fields to be checked in the target asset tables based on the configuration instruction.
In a second aspect, embodiments of the present disclosure provide an asset collation apparatus, the apparatus comprising: the system comprises a determining module, a verification module and a verification module, wherein the determining module is used for determining a target asset model to be checked, and the target asset model is used for recording a plurality of target asset tables and asset change information among the target asset tables; the acquisition module is used for acquiring subscription information of a target check model, wherein the target check model is obtained by training based on a plurality of first sample asset tables and asset change information among the plurality of first sample asset tables; and the creation module is used for creating a first asset checking task for calling the target checking model to check the target asset model based on the subscription information.
In a third aspect, the disclosed embodiments provide a computer readable storage medium having stored thereon a computer program which, when executed by a processor, implements the method of any of the embodiments.
In a fourth aspect, embodiments of the present disclosure provide a computer device comprising a memory, a processor, and a computer program stored on the memory and executable on the processor, the processor implementing the method of any of the embodiments when the program is executed.
After obtaining subscription information of the target verification model, the first asset verification task for calling the target verification model to verify the target asset model can be automatically created based on the subscription information. The first asset checking task is run to automatically check the target asset model through the target checking model. In addition, the target checking model can be obtained by training based on the plurality of first sample asset tables and asset change information among the plurality of first sample asset tables, so that the target checking model is not required to be generated based on expert experience, the dependence on manpower in the process of generating the target checking model is reduced, and the labor cost is reduced.
It is to be understood that both the foregoing general description and the following detailed description are exemplary and explanatory only and are not restrictive of the disclosure.
Drawings
The accompanying drawings, which are incorporated in and constitute a part of this specification, illustrate embodiments consistent with the disclosure and together with the description, serve to explain the technical aspects of the disclosure.
Fig. 1 is a flowchart of an asset collation method of an embodiment of the present disclosure.
Fig. 2 is a schematic diagram of a topology model of an embodiment of the present disclosure.
Fig. 3 is a general flow chart of an embodiment of the present disclosure.
Fig. 4 is a schematic diagram of a training and freshness process for a target verification model according to an embodiment of the present disclosure.
Fig. 5 is a block diagram of an asset collation apparatus of an embodiment of the present disclosure.
Fig. 6 is a schematic structural diagram of a computer device according to an embodiment of the present disclosure.
Detailed Description
Reference will now be made in detail to exemplary embodiments, examples of which are illustrated in the accompanying drawings. When the following description refers to the accompanying drawings, the same numbers in different drawings refer to the same or similar elements, unless otherwise indicated. The implementations described in the following exemplary examples are not representative of all implementations consistent with the present disclosure. Rather, they are merely examples of apparatus and methods consistent with some aspects of the present disclosure as detailed in the accompanying claims.
The terminology used in the present disclosure is for the purpose of describing particular embodiments only and is not intended to be limiting of the disclosure. As used in this disclosure and the appended claims, the singular forms "a," "an," and "the" are intended to include the plural forms as well, unless the context clearly indicates otherwise. It should also be understood that the term "and/or" as used herein refers to and encompasses any or all possible combinations of one or more of the associated listed items. In addition, the term "at least one" herein means any one of a plurality or any combination of at least two of a plurality.
It should be understood that although the terms first, second, third, etc. may be used in this disclosure to describe various information, these information should not be limited to these terms. These terms are only used to distinguish one type of information from another. For example, first information may also be referred to as second information, and similarly, second information may also be referred to as first information, without departing from the scope of the present disclosure. The word "if" as used herein may be interpreted as "at … …" or "at … …" or "responsive to a determination", depending on the context.
In order to better understand the technical solutions in the embodiments of the present disclosure and make the above objects, features and advantages of the embodiments of the present disclosure more comprehensible, the technical solutions in the embodiments of the present disclosure are described in further detail below with reference to the accompanying drawings.
In order to ensure the correctness of the asset change information during the transaction or the like by the user, the asset check is required to avoid the problem of the asset loss as much as possible. Asset reconciliation is a type of fine-grained monitoring process for asset security scenarios for risk checking of data related to an asset. The object of the asset check is called an asset model, which is an abstraction of a data model in asset loss prevention and control, and the asset model comprises a plurality of asset tables, a trigger table, association relations among the plurality of asset tables and the like. The asset table refers to a table related to asset information, such as an account balance table, a transaction detail table, and the like. The trigger table refers to an asset table with changed asset information, for example, the asset information in an account balance table is account balance, and if the account balance in one account balance table is changed, the account balance table can be used as the trigger table. The association relationship between the asset tables is used for representing asset change information between the asset tables, taking two asset tables as an example, assuming that account A pays account B10 yuan, in the account balance table of account A, the balance of account A is reduced by 10 yuan (namely the balance change value of account A is-10), in the account balance table of account B, the balance of account B is increased by 10 yuan (namely the balance change value of account B is 10), and the association relationship between the account balance table of account A and the account balance table of account B is that the asset of account A is transferred to account B.
In performing asset validation, multiple asset tables in an asset model are typically validated based on a pre-generated validation model. The checking model is specific logic for performing risk checking on the asset model, and comprises target fields to be checked in a plurality of asset tables and relations among the target fields in each asset table. For example, in the example of account a paying account B for 10, the target fields to be checked are account a's balance change value and account B's balance change value. The relationship between the destination fields in each asset table is that the amount of balance decrease for account A is equal to the amount of balance increase for account B (i.e., the sum of the balance change value for account A and the balance change value for account B is equal to 0). Those skilled in the art will appreciate that the above is merely an exemplary illustration, and that in other examples, the target fields to be checked in different asset tables may be different, and that the relationship between the target fields may be other relationships.
In the related art, a collation model for performing asset collation is generally generated based on expert experience. However, the number of transactions is often very large, information related to various transactions is very complicated, and collation rules for collation models are also various. Therefore, the method of generating the collation model based on the experience of the expert has a problem of high labor cost.
Based on this, the present disclosure provides an asset collation method, referring to fig. 1, comprising:
step 102: determining a target asset model to be checked, wherein the target asset model is used for recording a plurality of target asset tables and asset change information among the target asset tables;
step 104: acquiring subscription information of a target check model, wherein the target check model is obtained by training based on a plurality of first sample asset tables and asset change information among the plurality of first sample asset tables;
step 106: and creating a first asset checking task for calling the target checking model to check the target asset model based on the subscription information.
In step 102, the target asset model may be various types of asset models, and the types of asset models may be determined according to the asset transfer scenario. For example, in a transfer scenario, the target asset model is an asset model that includes account balance tables for a plurality of users and balance variation information between the account balance tables for the plurality of users. In another example, in the interest settlement scenario, the target asset model is an asset model including a plurality of interest settlement schedules of accounts of the same user at different times and interest fluctuation information between the plurality of interest settlement schedules. The balance change information and interest change information are asset change information under corresponding scenes.
In some embodiments, the target asset model may be searched through search information entered by the user. The search information may be identification information of the created target asset model itself (e.g., model ID of the target asset model) or identification information of other objects (referred to as search objects) related to the target asset model. The search object may include, but is not limited to, at least one of an asset transfer link corresponding to a plurality of target asset tables in a target asset model, a target asset table, an application for operating a plurality of target asset tables in the target asset model, and the like, and accordingly, the search information may include identification information (denoted trace) of an asset transfer link corresponding to a plurality of target asset tables in a target asset model, a order number of any target asset table in a target asset model, identification information (e.g., a form ID of any target asset table) of any target asset table in a target asset model, or identification information related to an application for operating a plurality of target asset tables in the target asset model. The following describes a specific manner of determining a target asset model.
In the case of using the identification information of the target asset model itself as search information, one or more target asset models may be created manually or automatically, each target asset model that is successfully created may correspond to a model ID, and the target asset model may be searched by inputting the model ID.
In the case where the identification information of the asset transfer link is used as the search information, the asset output path from one asset transfer-out party to one asset transfer-in party is referred to as one asset transfer link. For example, in transferring money to account B through account A, the transfer path from account A to account B is an asset transfer link; the transfer path from account a to account C is another asset transfer link when transferring money to account C through account a. Each asset transfer link may correspond to an identification information (referred to as a link ID) for identifying the asset transfer link. The corresponding asset transfer link can be searched by using the link ID as search information, thereby acquiring a target asset model related to the asset transfer link.
In the case of using the order number of any one of the target asset tables in the target asset model as the search information, one of the order numbers may be corresponding to one of the transactions for uniquely identifying the one of the transactions. Since at least two asset tables related to the same transaction share the same order number, each asset table sharing the order number can be searched by using the order number as search information, and thus the target asset model related to each asset table can be searched.
In the case where identification information of any one of the target asset tables in the target asset model is used as search information, each of the target asset tables may include an identification information (referred to as a form ID) for uniquely identifying the target asset table. By using the form ID as search information, a corresponding target asset table can be searched out, thereby searching out a target asset model including the target asset table.
In the case of using the identification information related to the application as search information, a target asset table may be operated by one or more applications, one application may operate one or more target asset tables, and the operation result is used to update field values of target fields related to the asset in the target asset table. For example, the transfer application may operate the user balance table to update the balance field in the user balance table. The identification information related to the application may be identification information of the application itself (i.e., application ID). Since each application may include one or more databases, each database included by an application may store a set of target asset tables that the application is capable of operating on, and different databases may store different target asset tables that the application is capable of operating on. Thus, the identification information related to the application may also be identification information of a database comprised by the application. By taking the identification information related to the application as search information, each target asset table which can be operated by the application can be searched, and further, a target asset model related to each target asset table is searched.
Those skilled in the art will appreciate that the examples set forth above are illustrative only. In addition to the search information and search objects listed above, embodiments of the present disclosure may employ other information as search information and correspondingly employ other objects as search objects, which are not listed here.
In some embodiments, a topology model may be pre-established, where the topology model includes a plurality of nodes, and edges connecting the plurality of nodes, each node corresponding to a search object corresponding to a piece of search information, and each edge is used to characterize an association relationship between two search objects or an association relationship between one search object and the target asset model. Referring to FIG. 2, the search object may include a manual model, an automatic model, an application, a physical database list, a physical database, a logical database list, a logical database, and a target asset table. The artificial model is a target asset model created manually, the automatic model is a target asset model obtained by training based on a plurality of first sample asset tables and asset change information among the plurality of first sample asset tables, and the application can comprise one or more physical databases and one or more logical databases. The relationship between the target asset table and the artificial model may be maintained in the model node.
After the topology model is built, the target asset model may be determined together based on the search information entered by the user and the topology model. Taking the searching object as an application as an example, a physical database list corresponding to the application can be searched based on the identification information of the physical database included in the application, then the corresponding physical database is searched from the physical database list, and then the target asset table in the physical database is searched. The automation model may then be searched based on the relationships between the target asset table stored in the model node and the automation model. The above list examples of searching for the target asset model by applying the related search information, and the manner of searching for the target asset model by other search information is similar to the above manner, and will not be repeated here. It should be noted that, in the figure, the case where the application includes a physical database and a logical database is enumerated, and in practical application, the application may include only the physical database, and not include the logical database.
According to the embodiment, the target asset model can be searched based on the search information of various search objects by establishing the topology model, and the user can acquire the target asset model only by inputting the search information, so that the efficiency and the flexibility of acquiring the target asset model are improved, and the operation complexity of acquiring the target asset model is reduced.
In step 104, subscription information for the target collation model may be obtained, and the subscription information may include identification information of the target collation model, and may further include information indicating a subscription manner, such as a subscription start time and a subscription end time.
The target checking models can be various types of checking models, the number of the target checking models can be more than or equal to 1, and different target checking models can adopt different checking rules to check the target asset models or check different target fields in a plurality of target asset tables. For example, in the foregoing example of transferring to account B through account a, the target field of one target collation model collation may be the balance change value in the plurality of target asset tables, and the collation rule is that the sum of the balance change values in the plurality of target asset tables is equal to 0. In other examples, other target collation models may use other collation rules to collate target fields other than balance change values.
The target collation model may be trained based on asset change information between a plurality of first sample asset tables. The target checking model can automatically identify target fields to be checked in the plurality of first sample asset tables, and automatically learn the association relation between the target fields.
In some embodiments, the plurality of first sample asset tables may be determined based on: and if the number of the first asset tables is larger than a preset first number threshold value in the asset tables generated in the target asset transfer scene, determining the first asset tables as first sample asset tables of the target asset transfer scene. Wherein the asset table generated in the target asset transfer scenario may be an asset table generated over a historical period of time (e.g., last week or last month). For example, in the asset tables generated in the transfer scenario in the last week, the number of account balance tables is more than 90% of the total number of first sample asset tables, and thus, the account balance tables may be determined as the first sample asset tables of the transfer scenario. Of course, the above is merely exemplary, and the first number threshold used in this embodiment may be set to other values according to actual needs, and the target asset transfer scenario in this embodiment may be other than the transfer scenario.
In some embodiments, asset transition information between the plurality of first sample asset tables may be obtained based on: and if the number of the first sample asset tables meeting the first asset change information is larger than a preset second number threshold value in the plurality of first sample asset tables, determining the first asset change information as asset change information among the plurality of first sample asset tables. For example, in the transfer scenario, if the number of the first sample asset tables satisfying the sum of the balance change value of the account and the balance change value of the account of the other first sample asset table is equal to 0 is greater than 95% of the total number of the first sample asset tables, the first asset change information is determined as the sum of the balance change values in the two first sample asset tables is equal to 0. The second number threshold may be equal to or different from the first number threshold, and a specific value of the second number threshold may be set according to actual needs.
One or more candidate collation models may be trained and stored in the model database in the manner described above. Wherein different candidate collation models may be trained using different first sample asset tables. At least one candidate collation model in the database may be pushed to the user when required by the user, and the target collation model is determined from the at least one candidate collation model based on a selection instruction sent by the user.
The trained candidate collation model may have some noise, i.e., the collation accuracy of some of the candidate collation models may be low. In some embodiments, the candidate collation model may be checked based on the plurality of second sample asset tables and asset change information between the plurality of second sample asset tables, and the successfully checked candidate collation model may be pushed to the user. If the verification fails, the candidate verification model is retrained. The above-described process is called gradation management of the candidate collation model. Through gray level management, only the candidate verification model passing verification is pushed to the user, so that the disturbance to the user can be reduced. In some embodiments, if the number of second sample asset tables satisfying the collation rules corresponding to the candidate collation model is greater than a preset third number threshold, it is determined that the verification is successful. For example, assuming that the collation rule of one candidate collation model is that the sum of the balance change values in the two asset tables is equal to 0, and that the number of second sample asset tables satisfying the sum of the balance change values of the account and the balance change values of the account of another second sample asset table is equal to 0 is greater than 90% of the total number of second sample asset tables, the candidate collation model is determined to be successful in authentication; otherwise, determining that the candidate verification model fails to verify. Wherein the third number threshold may be the same as or different from any one of the first number threshold and the second number threshold in the foregoing embodiments.
In some embodiments, during the gray scale management process, a first report may also be generated to determine the progress of the gray scale management process. Optionally, the first report includes at least one of the following information:
the number of candidate collation models newly added in the model database. Assuming that the number of candidate check models in the model database is N1 when the previous statistics is performed and the number of candidate check models in the model database is N2 when the current statistics is performed, the number of newly added candidate check models in the model database is equal to N2-N1. In some embodiments, statistics may be performed on the number of candidate collation models newly added to the model database at preset time intervals, or under the triggering of preset statistics instructions, or based on other conditions.
Number of new target asset models. The number of newly added target asset models is statistically similar to the number of newly added candidate collation models in the model database. Assuming that the number of the target asset models in the last statistics is M1 and the number of the target asset models in the current statistics is M2, the number of the newly added target asset models is equal to M2-M1.
Total number of newly added target fields. Assuming that the number of target fields in the target asset table of each target asset model is K1 when the last statistics is performed and the number of target fields in the target asset table of each target asset model is K2 when the current statistics is performed, the total number of the newly added target fields is equal to K2-K1.
Number of updated target asset models. Since the target asset model itself may be erroneous, e.g., the target fields of multiple target asset tables in the target asset model may be selected incorrectly, the target asset model may be updated. The number of updated target asset models may be increased by 1 for each update of one target asset model.
The number of candidate collation models that succeed in the collation. Alternatively, each candidate collation model that is successful in verification may be pushed to the user, so that the number of candidate collation models that are successful in verification is equal to the number of candidate collation models that are pushed to the user. The noise amount of the candidate collation model can be determined by counting the number of candidate collation models for which the verification is successful. The larger the amount of noise, the fewer the number of candidate collation models that are successful in verification.
The success rate of verification of each candidate verification model. The verification success rate of one candidate collation model may be determined based on the ratio of the number of second sample asset tables satisfying the collation rules corresponding to the candidate collation model to the total number of second sample asset tables.
The ratio of the number of candidate collation models that are successfully checked to the total number of candidate collation models. Assuming that the number of candidate collation models which are successful in verification is M, and the total number of candidate collation models is N, the ratio is equal to M/N.
And checking results of checking the candidate checking model through the second sample asset table each time. And if one second sample asset table meets the check rule corresponding to the candidate check model, judging that the check result is successful, otherwise, judging that the check result is failed.
In addition to the various information listed above, other information may be included in the first report, which is not listed here.
In step 106, a first asset reconciliation task may be created. And by running the first asset checking task, the target asset model can be checked through the target checking model to obtain a checking result. The stage of running the first asset reconciliation task may also be referred to as the online stage of the target reconciliation model. A second report may be generated during the online phase; the second report includes at least one of the following information: the number of the target asset models to be checked, the number of the target check models for asset check, the number of alarms when the check result of the first asset check task is abnormal, the discovery rate of the abnormal condition of the asset change information. By generating the second report, the user is facilitated to grasp the asset collation condition of the on-line stage of the object collation model.
Further, if the target asset model is not acquired in step 102, an empty second asset reconciliation task may be created in this step and the target asset model may be added to the second asset reconciliation task after the target asset model is acquired. In practical applications, because of the large number of target asset tables, asset change information between some target asset tables and other target asset tables may be difficult to mine out in time, resulting in portions of the target asset tables that may not be included in any target asset model. With the continuous creation of the target asset model, the mined asset change information is gradually increased, so that more target asset models can be built. The above process is also referred to as a training or derivation process for the target asset model. That is, aspects of embodiments of the present disclosure may train the target asset model and the target collation model simultaneously. Thus, in the event that the target asset model is not acquired, an empty second asset reconciliation task may be created first, which does not include the target asset model. After the target asset model is established, the target asset model may be automatically added to the second asset reconciliation task.
Similarly, if the target collation model is not obtained in step 104, then an empty third asset collation task may be created in this step. The third asset checking task does not include the target checking model, but may include the target asset model or may not include the target asset model. After the target collation model is obtained, the target asset model may be added to a third asset collation task.
In this way, the user can create an asset reconciliation task regardless of whether the target asset model and/or the target reconciliation model is present. In this way, the user can generate an asset collation task for a risk object (e.g., a target asset table) of his own attention at any time, and immediately after the target asset model and/or the target collation model are created, start the asset collation task, thereby improving asset collation efficiency. In addition, the user does not need to pay attention to whether the target asset model and/or the target check model are generated or not at all, so that the operation convenience of the user is improved.
Further, in this step, the life cycle of the target collation model may also be managed. The lifecycle may include, among other things, a start time, an end time, and/or an update time period (also referred to as a freshness time) for the target collation model to perform asset collation. Wherein the start time is the time when the object collation model starts asset collation, i.e. the time when the first asset collation task starts running. The end time is the time when the object collation model terminates the asset collation, i.e., the time when the first asset collation task ends running. The starting time and ending time of the asset collation by the object collation model may be determined based on the subscription starting time and subscription ending time acquired in step 104, respectively. In addition, since the asset transition scene may be changed, the asset change information is changed, and thus the verification result of the originally accurate target verification model becomes inaccurate, it is also possible to set an update time period in advance, and update the target verification model based on the update time period in this step. After the updating is completed, the updated target checking model can be called to check the target asset model in the first asset checking task, so that the accuracy of asset checking is improved.
After the first asset collation task is created, if the collation result of the first asset collation task is abnormal, alarm information may be output. Taking the account transfer scenario of the user a to the user B as an example, assuming that the checking rule of the target checking model is that the sum of the balance change values in the two account balance tables is equal to 0, when the account balance table of the user a and the account balance table of the user B are checked by the target checking model, it is found that the sum of the balance change values in the two account balance tables is not equal to 0, for example, the balance change value in the account balance table of the user a is-10, and the balance change value in the account balance table of the user B is 9, at this time, it can be determined that the checking result is abnormal, and alarm information is output.
Further, when the checking result is abnormal, an associated event corresponding to the checking result is generated, and the associated event is used for carrying out abnormal processing on the checking result. For example, the associated events may be stored in a log and actively sent to the associated processor or the log may be actively acquired by the processor so that the processor can obtain the anomaly information and perform anomaly processing on the verification results, e.g., contact the asset owner for asset loss reimbursement, query for anomaly causes, etc. Wherein the association event may include, but is not limited to, part or all of information in a time when the verification result is detected to be abnormal, a target asset table to which the abnormality verification result relates, a category of the abnormality verification result, an asset amount to which the abnormality verification result relates, an abnormality level, and the like.
However, the reasons for the abnormality of the collation result may be: (1) an asset change process is subject to error; (2) an error occurs in the target collation model; (3) an error occurs in the target asset model. That is, the verification result abnormality may be a misjudgment and not cause a loss of the asset owner. In order to reduce the above-described erroneous determination, in the case where the collation result of the first asset collation task is abnormal, first flag information for the target asset model and second flag information for the target collation model may be acquired. Wherein the first tag information is used for indicating whether the target asset model is correct or not, and the second tag information is used for indicating whether the checking result is correct or not. And if the target asset model and the checking result are both correct, generating an associated event corresponding to the checking result. In this embodiment, if both the target asset model and the collation result are correct, the cause of the abnormality of the collation result is that the asset change process is in error in general, and therefore, the association event is generated only when both the target asset model and the collation result are correct. Thus, erroneous judgment can be effectively reduced.
In some embodiments, the target collation model may be updated as described above. The update process may be performed in case that a preset update trigger condition is satisfied. Wherein the update trigger condition includes at least one of: reaching a preset updating time interval; the checking accuracy of the target asset model is lower than a preset accuracy threshold; and receiving an update instruction input by a user. The process of updating the target collation model may also be referred to as a freshness retaining process of the target collation model. Through the preservation treatment, the target checking model can adapt to the change of the asset transfer scene, so that the checking accuracy of the target checking model is ensured.
In the fresh-keeping processing process, a configuration instruction input by a user can be obtained, and parameters of the target checking model and/or target fields to be checked in the target asset tables are configured based on the configuration instruction. The parameters may include, but are not limited to, a lifecycle of the target collation model (e.g., start time, end time and/or update time period), the first number threshold described above, the second number threshold, and/or the third number threshold.
The asset collation flow of the embodiment of the present disclosure is explained below with reference to fig. 3 and 4. FIG. 3 is an overall flow chart of an implementation of the present disclosure, an asset reconciliation process of an embodiment of the present disclosure comprising the steps of:
(1) And acquiring a risk search example, namely acquiring a target asset model to be checked. The risk object (e.g., asset table) of interest to the user may be searched for through trace, order number, identification information of the application, form ID of the asset table, model ID, etc. information.
(2) And (5) preventing subscription. I.e., selecting a target asset model and a target collation model. First, it is determined whether the risk object has a corresponding model (including a target asset model and a target collation model). If both the target asset model and the target collation model exist, then the target asset model and the target collation model are selected. One or more candidate collation models in the model database may be recommended to the user to enable the user to determine a target collation model from the candidate collation models and subscribe.
(3) An asset reconciliation task is generated. If the target asset model or the target collation model does not exist in the step (2), generating an empty asset collation task; and (3) if the target asset model and the target checking model exist in the step (2), generating a non-empty asset checking task. The asset reconciliation task is then lifecycle managed. If the target asset model is not present, a risk object of interest (e.g., an asset table) may also be entered by the user in order to generate the target asset model.
(4) Exception/event management. If the asset checking result is abnormal, an abnormality alarm is output, and an associated event is generated based on the first tag information and the second tag information by user marking (i.e., tagging the first tag information and the second tag information).
The training and fresh-keeping process of the target verification model can refer to fig. 4. First, an initialization is performed. After initialization, no automatic model (i.e., the training-generated target collation model) is included in the asset collation task. Thus, at least one candidate collation model may be generated through training. For an empty asset reconciliation task (e.g., an asset reconciliation task that does not include any target asset models), an in-training state may be entered where the target asset models and target reconciliation models are continually automatically derived based on risk objects of interest to the asset reconciliation task, and if new asset models and new reconciliation models are produced, the user may be notified to subscribe to the asset models and new reconciliation models.
The asset collation task subscribed to the asset model and the collation model enters a gray level in-state to verify the validity of the subscribed collation model, realize noise reduction of the collation model, and produce a gray level operation report (i.e., first report).
After the gray level is finished or the user triggers, the checking model can enter an upper line state, and the user starts to normally receive abnormal alarm information and operation reports (namely, second reports) generated by the asset checking task.
The asset checking task freshness keeping can be designated by a user or triggered at fixed time, so that the asset model and the checking model are automatically deduced and updated, and the timeliness of the asset model and the checking model is ensured. The updated asset model and the verification model can be subjected to gray level management again, and the asset model and the verification model are online again after the verification of the gray level management process is successful.
According to the embodiment of the disclosure, training of the checking model and the asset model is performed in advance, and the collection of the checking model and the asset model is deposited and provided for a user to select and subscribe, so that the cost of the user for familiarizing with related services of the asset from the beginning is reduced, and the burden of manually creating the checking model and the asset model is reduced; meanwhile, the asset checking task can automatically keep the checking model and the asset model fresh, capture changes caused by business and code changes in real time, and ensure timeliness of asset checking.
It will be appreciated by those skilled in the art that in the above-described method of the specific embodiments, the written order of steps is not meant to imply a strict order of execution but rather should be construed according to the function and possibly inherent logic of the steps.
Referring to fig. 5, an embodiment of the present disclosure further provides an asset collation apparatus, the apparatus comprising:
a determining module 502, configured to determine a target asset model to be checked, where the target asset model is used to record a plurality of target asset tables and asset change information between the plurality of target asset tables;
an obtaining module 504, configured to obtain subscription information for a target collation model, where the target collation model is obtained based on a plurality of first sample asset tables and asset change information between the plurality of first sample asset tables;
a creating module 506, configured to create, based on the subscription information, a first asset checking task that invokes the target checking model to check the target asset model.
In some embodiments, functions or modules included in an apparatus provided by the embodiments of the present disclosure may be used to perform a method described in the foregoing method embodiments, and specific implementations thereof may refer to descriptions of the foregoing method embodiments, which are not repeated herein for brevity.
The embodiments of the present disclosure also provide a computer device at least including a memory, a processor, and a computer program stored on the memory and executable on the processor, wherein the processor implements the method of any of the preceding embodiments when executing the program.
FIG. 6 illustrates a more specific hardware architecture diagram of a computing device provided by embodiments of the present description, which may include: a processor 602, a memory 604, an input/output interface 606, a communication interface 608, and a bus 610. Wherein the processor 602, the memory 604, the input/output interface 606, and the communication interface 608 enable communication connections within the device between each other via a bus 610.
The processor 602 may be implemented by a general-purpose CPU (Central Processing Unit ), microprocessor, application specific integrated circuit (Application Specific Integrated Circuit, ASIC), or one or more integrated circuits, etc. for executing relevant programs to implement the technical solutions provided in the embodiments of the present disclosure. The processor 602 may also include a graphics card, which may be an Nvidia titanium X graphics card, a 1080Ti graphics card, or the like.
The Memory 604 may be implemented in the form of ROM (Read Only Memory), RAM (Random Access Memory ), static storage device, dynamic storage device, or the like. Memory 604 may store an operating system and other application programs, and when the techniques provided by the embodiments of the present disclosure are implemented in software or firmware, the associated program code is stored in memory 604 and executed by processor 602.
The input/output interface 606 is used to connect with an input/output module to achieve information input and output. The input/output module may be configured as a component in a device (not shown) or may be external to the device to provide corresponding functionality. Wherein the input devices may include a keyboard, mouse, touch screen, microphone, various types of sensors, etc., and the output devices may include a display, speaker, vibrator, indicator lights, etc.
The communication interface 608 is used to connect a communication module (not shown) to enable communication interaction between the device and other devices. The communication module may implement communication through a wired manner (such as USB, network cable, etc.), or may implement communication through a wireless manner (such as mobile network, WIFI, bluetooth, etc.).
Bus 610 includes a path to transfer information between components of the device (e.g., processor 602, memory 604, input/output interface 606, and communication interface 608).
It should be noted that although the above-described device only shows the processor 602, the memory 604, the input/output interface 606, the communication interface 608, and the bus 610, in a specific implementation, the device may also include other components necessary to achieve proper operation. Furthermore, it will be understood by those skilled in the art that the above-described apparatus may include only the components necessary to implement the embodiments of the present description, and not all the components shown in the drawings.
The disclosed embodiments also provide a computer readable storage medium having stored thereon a computer program which, when executed by a processor, implements the method of any of the previous embodiments.
Computer readable media, including both non-transitory and non-transitory, removable and non-removable media, may implement information storage by any method or technology. The information may be computer readable instructions, data structures, modules of a program, or other data. Examples of storage media for a computer include, but are not limited to, phase change memory (PRAM), static Random Access Memory (SRAM), dynamic Random Access Memory (DRAM), other types of Random Access Memory (RAM), read Only Memory (ROM), electrically Erasable Programmable Read Only Memory (EEPROM), flash memory or other memory technology, compact disc read only memory (CD-ROM), digital Versatile Discs (DVD) or other optical storage, magnetic cassettes, magnetic tape magnetic disk storage or other magnetic storage devices, or any other non-transmission medium, which can be used to store information that can be accessed by a computing device. Computer-readable media, as defined herein, does not include transitory computer-readable media (transmission media), such as modulated data signals and carrier waves.
From the foregoing description of embodiments, it will be apparent to those skilled in the art that the present embodiments may be implemented in software plus a necessary general purpose hardware platform. Based on such understanding, the technical solutions of the embodiments of the present specification may be embodied in essence or what contributes to the prior art in the form of a software product, which may be stored in a storage medium, such as a ROM/RAM, a magnetic disk, an optical disk, etc., including several instructions for causing a computer device (which may be a personal computer, a server, or a network device, etc.) to execute the method described in the embodiments or some parts of the embodiments of the present specification.
The system, apparatus, module or unit set forth in the above embodiments may be implemented in particular by a computer chip or entity, or by a product having a certain function. A typical implementation device is a computer, which may be in the form of a personal computer, a laptop computer, a cellular telephone, an image capture device telephone, a smart phone, a personal digital assistant, a media player, a navigation device, an email device, a game console, a tablet computer, a wearable device, or a combination of any of these devices.
In this specification, each embodiment is described in a progressive manner, and identical and similar parts of each embodiment are all referred to each other, and each embodiment mainly describes differences from other embodiments. In particular, for the device embodiments, since they are substantially similar to the method embodiments, the description is relatively simple, and reference is made to the description of the method embodiments for relevant points. The apparatus embodiments described above are merely illustrative, in which the modules illustrated as separate components may or may not be physically separate, and the functions of the modules may be implemented in the same piece or pieces of software and/or hardware when implementing the embodiments of the present disclosure. Some or all of the modules may be selected according to actual needs to achieve the purpose of the solution of this embodiment. Those of ordinary skill in the art will understand and implement the present invention without undue burden.
The foregoing is merely a specific implementation of the embodiments of this disclosure, and it should be noted that, for a person skilled in the art, several improvements and modifications may be made without departing from the principles of the embodiments of this disclosure, and these improvements and modifications should also be considered as protective scope of the embodiments of this disclosure.
Claims (16)
1. An asset collation method, the method comprising:
determining a target asset model to be checked, wherein the target asset model is used for recording a plurality of target asset tables and asset change information among the target asset tables;
acquiring subscription information of a target check model, wherein the target check model is obtained by training based on a plurality of first sample asset tables and asset change information among the plurality of first sample asset tables;
creating a first asset checking task for calling the target checking model to check the target asset model based on the subscription information;
creating an empty second asset reconciliation task if the target asset model is not acquired; the second asset checking task does not include a target asset model; after the target asset model is acquired, adding the target asset model to the second asset verification task; and/or
Creating an empty third asset reconciliation task if the target reconciliation model was not obtained; the third asset checking task does not comprise a target checking model; after the target collation model is acquired, the target asset model is added to the third asset collation task.
2. The method of claim 1, the determining a target asset model to be reconciled comprising:
acquiring search information input by a user and a pre-established topology model; the topology model comprises a plurality of nodes and edges connected with the nodes, each node corresponds to a search object corresponding to one item of search information, and each edge is used for representing the association relation between two search objects or representing the association relation between one search object and the target asset model;
and determining a target asset model to be checked based on the search information and the topology model.
3. The method of claim 2, the search information comprising at least one of:
identification information of asset transfer links corresponding to a plurality of target asset tables in the target asset model;
an order number in any one of the target asset tables in the target asset model;
identification information of any target asset table in the target asset model;
the identification information is used for being related to an application, and the application is used for operating a plurality of target asset tables in the target asset model;
identification information of the created target asset model.
4. The method of claim 1, the method further comprising:
pushing at least one candidate collation model in the model database to the user;
the target collation model is determined from the at least one candidate collation model based on a selection instruction sent by a user.
5. The method of claim 4, the pushing at least one candidate collation model in a model database to a user comprising:
verifying the candidate verification model based on a plurality of second sample asset tables and asset change information among the plurality of second sample asset tables;
and pushing the candidate verification model which is verified successfully to the user.
6. The method of claim 5, the method further comprising:
generating a first report, the first report comprising at least one of:
the number of candidate check models newly added in the model database;
the number of newly added target asset models;
the total number of the newly added target fields to be checked;
the number of updated target asset models;
the number of candidate verification models that are successfully verified;
the success rate of verification of each candidate verification model;
the ratio of the number of candidate verification models which are successfully verified to the total number of candidate verification models;
And checking results of checking the candidate checking model through the second sample asset table each time.
7. The method of claim 1, the method further comprising:
and under the condition that a preset updating triggering condition is met, updating the target check model.
8. The method of claim 7, the update trigger condition comprising at least one of:
reaching a preset updating time interval;
the checking accuracy of the target asset model is lower than a preset accuracy threshold;
and receiving an update instruction input by a user.
9. The method of claim 1, after creating a first asset reconciliation task that invokes the target reconciliation model to reconcile a target asset model based on the subscription information, the method further comprises:
and if the checking result of the first asset checking task is abnormal, outputting alarm information.
10. The method of claim 9, the method further comprising:
if the checking result of the first asset checking task is abnormal, acquiring first marking information of the target asset model and second marking information of the target checking model; the first marking information is used for representing whether the target asset model is correct or not, and the second marking information is used for representing whether the checking result is correct or not;
And if the target asset model and the checking result are both correct, generating a correlation event corresponding to the checking result, wherein the correlation event is used for carrying out exception processing on the checking result.
11. The method of claim 1, the method further comprising:
in the asset tables generated in the target asset transfer scene, if the number of the first asset tables is larger than a preset first number threshold, determining the first asset tables as first sample asset tables of the target asset transfer scene; and/or
And if the number of the first sample asset tables meeting the first asset change information is larger than a preset second number threshold value in the plurality of first sample asset tables, determining the first asset change information as asset change information among the plurality of first sample asset tables.
12. The method of claim 1, the method further comprising:
generating a second report; the second report includes at least one of the following information:
the number of target asset models being collated;
the number of target collation models for asset collation;
alarming times when the checking result of the first asset checking task is abnormal;
discovery rate of abnormal condition of asset change information.
13. The method of claim 1, the method further comprising:
acquiring a configuration instruction input by a user;
and configuring parameters of the target checking model and/or target fields to be checked in the target asset tables based on the configuration instruction.
14. An asset collation apparatus, the apparatus comprising:
the system comprises a determining module, a verification module and a verification module, wherein the determining module is used for determining a target asset model to be checked, and the target asset model is used for recording a plurality of target asset tables and asset change information among the target asset tables;
the acquisition module is used for acquiring subscription information of a target check model, wherein the target check model is obtained by training based on a plurality of first sample asset tables and asset change information among the plurality of first sample asset tables;
the creation module is used for creating a first asset checking task for calling the target checking model to check the target asset model based on the subscription information;
the creation module is further configured to create an empty second asset verification task if the target asset model is not acquired; the second asset checking task does not include a target asset model; after the target asset model is acquired, adding the target asset model to the second asset verification task; and/or
The creation module is further configured to create an empty third asset verification task if the target verification model is not acquired; the third asset checking task does not comprise a target checking model; after the target collation model is acquired, the target asset model is added to the third asset collation task.
15. A computer readable storage medium having stored thereon a computer program which when executed by a processor implements the method of any of claims 1 to 13.
16. A computer device comprising a memory, a processor and a computer program stored on the memory and executable on the processor, the processor implementing the method of any one of claims 1 to 13 when the program is executed.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202211431042.1A CN115907986B (en) | 2022-11-15 | 2022-11-15 | Asset checking method and device, medium and computer equipment |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202211431042.1A CN115907986B (en) | 2022-11-15 | 2022-11-15 | Asset checking method and device, medium and computer equipment |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN115907986A CN115907986A (en) | 2023-04-04 |
| CN115907986B true CN115907986B (en) | 2023-08-29 |
Family
ID=86486420
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202211431042.1A Active CN115907986B (en) | 2022-11-15 | 2022-11-15 | Asset checking method and device, medium and computer equipment |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN115907986B (en) |
Citations (12)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110060164A (en) * | 2019-01-28 | 2019-07-26 | 阿里巴巴集团控股有限公司 | Accounting data processing method, device and equipment |
| CN110175916A (en) * | 2019-04-26 | 2019-08-27 | 阿里巴巴集团控股有限公司 | Cash flow checking method and device |
| CN110288452A (en) * | 2019-05-31 | 2019-09-27 | 北京随信云链科技有限公司 | A kind of asset data management system, method and calculate equipment |
| CN111784468A (en) * | 2020-07-01 | 2020-10-16 | 支付宝(杭州)信息技术有限公司 | Account association method and device and electronic equipment |
| CN112381647A (en) * | 2020-10-26 | 2021-02-19 | 支付宝(杭州)信息技术有限公司 | Method, device, equipment and readable medium for transferring funds |
| CN113256404A (en) * | 2021-06-16 | 2021-08-13 | 浙江网商银行股份有限公司 | Data processing method and device |
| WO2021169169A1 (en) * | 2020-02-26 | 2021-09-02 | 深圳壹账通智能科技有限公司 | Resource transfer data checking method and apparatus, computer device, and storage medium |
| CA3139451A1 (en) * | 2020-11-17 | 2022-05-17 | 10353744 Canada Ltd. | Payment anti-transfer method, device and system |
| CN114661740A (en) * | 2022-03-25 | 2022-06-24 | 中国建设银行股份有限公司 | Data processing method, device, equipment, computer storage medium and program product |
| CN114722146A (en) * | 2022-03-22 | 2022-07-08 | 未鲲(上海)科技服务有限公司 | Supply chain asset checking method, device, equipment and medium based on artificial intelligence |
| CN114723206A (en) * | 2021-01-06 | 2022-07-08 | 腾讯科技（深圳）有限公司 | Asset data processing method, computer equipment and readable storage medium |
| CN114968726A (en) * | 2022-06-23 | 2022-08-30 | 北京天融信网络安全技术有限公司 | Method and system for monitoring system asset change, electronic device and storage medium |
Family Cites Families (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20140025414A1 (en) * | 2012-07-20 | 2014-01-23 | Bret Dwayne Worden | Health assessment method and system for assets |
-
2022
- 2022-11-15 CN CN202211431042.1A patent/CN115907986B/en active Active
Patent Citations (12)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110060164A (en) * | 2019-01-28 | 2019-07-26 | 阿里巴巴集团控股有限公司 | Accounting data processing method, device and equipment |
| CN110175916A (en) * | 2019-04-26 | 2019-08-27 | 阿里巴巴集团控股有限公司 | Cash flow checking method and device |
| CN110288452A (en) * | 2019-05-31 | 2019-09-27 | 北京随信云链科技有限公司 | A kind of asset data management system, method and calculate equipment |
| WO2021169169A1 (en) * | 2020-02-26 | 2021-09-02 | 深圳壹账通智能科技有限公司 | Resource transfer data checking method and apparatus, computer device, and storage medium |
| CN111784468A (en) * | 2020-07-01 | 2020-10-16 | 支付宝(杭州)信息技术有限公司 | Account association method and device and electronic equipment |
| CN112381647A (en) * | 2020-10-26 | 2021-02-19 | 支付宝(杭州)信息技术有限公司 | Method, device, equipment and readable medium for transferring funds |
| CA3139451A1 (en) * | 2020-11-17 | 2022-05-17 | 10353744 Canada Ltd. | Payment anti-transfer method, device and system |
| CN114723206A (en) * | 2021-01-06 | 2022-07-08 | 腾讯科技（深圳）有限公司 | Asset data processing method, computer equipment and readable storage medium |
| CN113256404A (en) * | 2021-06-16 | 2021-08-13 | 浙江网商银行股份有限公司 | Data processing method and device |
| CN114722146A (en) * | 2022-03-22 | 2022-07-08 | 未鲲(上海)科技服务有限公司 | Supply chain asset checking method, device, equipment and medium based on artificial intelligence |
| CN114661740A (en) * | 2022-03-25 | 2022-06-24 | 中国建设银行股份有限公司 | Data processing method, device, equipment, computer storage medium and program product |
| CN114968726A (en) * | 2022-06-23 | 2022-08-30 | 北京天融信网络安全技术有限公司 | Method and system for monitoring system asset change, electronic device and storage medium |
Also Published As
| Publication number | Publication date |
|---|---|
| CN115907986A (en) | 2023-04-04 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US10698795B2 (en) | Virtual payments environment | |
| US11132624B2 (en) | Model integration method and device | |
| CN111815169B (en) | Service approval parameter configuration method and device | |
| CN109976998A (en) | A kind of Software Defects Predict Methods, device and electronic equipment | |
| CN110650531B (en) | Base station coordinate calibration method, system, storage medium and equipment | |
| CN114116496A (en) | Automatic testing method, device, equipment and medium | |
| CN111506580B (en) | Transaction storage method based on centralized block chain type account book | |
| CN110046944A (en) | Invoice creation method and device, electronic equipment based on block chain | |
| CN109740352A (en) | A kind of account processing method, device and electronic equipment | |
| CN113094284A (en) | Application fault detection method and device | |
| CN110188793A (en) | Data exception analysis method and device | |
| CN112101939A (en) | Node management method and system based on block chain | |
| EP3131041A1 (en) | Testing of enterprise resource planning systems | |
| CN112669134A (en) | Method, equipment and medium for realizing auditing intellectualization through auditing rule machine learning | |
| CN112286790A (en) | Full link test method, device, equipment and storage medium | |
| CN113935788B (en) | Model evaluation method, device, equipment and computer readable storage medium | |
| CN111784246B (en) | Logistics path estimation method | |
| CN117033417A (en) | Service data processing method, medium and computer equipment | |
| CN110278524B (en) | User position determining method, graph model generating method, device and server | |
| CN115907986B (en) | Asset checking method and device, medium and computer equipment | |
| CN113836154B (en) | Database switching method and device | |
| CN111144987A (en) | Abnormal shopping behavior limiting method, limiting assembly and shopping system | |
| CN110796178A (en) | Decision model training method, sample feature selection method, device and electronic equipment | |
| CN109597706A (en) | It is a kind of to check the detection method of variance data, apparatus and system | |
| CN112200711B (en) | Training method and system of watermark classification model |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |