Data processing methods and apparatus, data processing systems

Abstract

This specification provides a data processing method and apparatus, and a data processing system. The method, applied to a scene change recognition module, includes: receiving a target data packet sent by a data change monitoring module; obtaining a target scene model from a scene model set that matches the target data packet; obtaining the corresponding target field and its value from the target data packet according to the scene recognition rules of the target scene model, forming a target feature combination; matching the target feature combination with the feature combination of a scene instance set, and determining the scene recognition result based on the matching result. By receiving the target data packet, accurate capture and structured processing of online data changes are achieved; furthermore, the target scene model and scene model set are intelligently matched and their target feature combinations are extracted; finally, the scene recognition result is quickly output through a locally cached, efficient matching mechanism.

Description

Technical Field

[0001] This specification relates to the field of computer technology, and in particular to a data processing method. This specification also relates to a data processing apparatus, a data processing system, a computing device, a computer-readable storage medium, and a computer program product. Background Technology

[0002] Throughout the software development and testing process, although the testing phase strives to simulate real-world data and requests from the production environment to verify system functionality, discrepancies inevitably exist between simulated data and real traffic. Furthermore, manually constructed test cases cannot fully cover the complex and ever-changing business scenarios online, potentially leading to some issues being overlooked in the production environment. When such defects are triggered by actual business traffic, problems are often only discovered through post-event monitoring, data verification, or other passive means. Although the problem is eventually discovered, it has already had a substantial impact on real business operations, and the extent of this impact directly depends on how quickly the problem is discovered and contained.

[0003] In-depth analysis of historical online issues revealed that most online failures were caused by new scenarios, such as new repayment methods upstream, new business periods, unknown events, unfamiliar payment channels or credit limit types, all of which create previously unexplored scenarios in the downstream business chain. Although a canary release strategy is usually adopted for verification during routine demand deployments, limitations in canary release duration, resource coverage, and product characteristics mean that some new scenarios are difficult to fully expose during the canary release period, and the specific timing and location of such new scenarios are also uncertain. Summary of the Invention

[0004] In view of the above, one or more embodiments of this specification provide a data processing method, a data processing apparatus, a data processing system, a computing device, a computer-readable storage medium, and a computer program product to solve the technical problem that some new scenarios in the prior art are difficult to fully expose during the gray period, and there is uncertainty about when and where such new scenarios will occur.

[0005] According to a first aspect of one or more embodiments of this specification, a data processing method is provided, applied to a scene change recognition module, comprising:

[0006] The data change monitoring module receives a target data packet, wherein the target data packet is a data packet obtained by the data change monitoring module assembling the changed data according to the target data model when it detects a data change in the data table corresponding to the target data model carried in the data acquisition task.

[0007] Obtain the target scene model from the scene model set that matches the target data packet;

[0008] Based on the scene recognition rules of the target scene model, the corresponding target fields and their values ​​are obtained from the target data packet to form a target feature combination;

[0009] The target feature combination is matched with the feature combination of the scene instance set, and the scene recognition result is determined based on the matching result.

[0010] According to a second aspect of one or more embodiments of this specification, a data processing apparatus is provided, applied to a scene change recognition module, wherein...

[0011] The scene change recognition module is configured as follows:

[0012] The data change monitoring module receives a target data packet, wherein the target data packet is a data packet obtained by the data change monitoring module assembling the changed data according to the target data model when it detects a data change in the data table corresponding to the target data model carried in the data acquisition task.

[0013] Obtain the target scene model from the scene model set that matches the target data packet;

[0014] Based on the scene recognition rules of the target scene model, the corresponding target fields and their values ​​are obtained from the target data packet to form a target feature combination;

[0015] The target feature combination is matched with the feature combination of the scene instance set, and the scene recognition result is determined based on the matching result.

[0016] According to a third aspect of one or more embodiments of this specification, a data processing system is provided, including a data change monitoring module and a scene change recognition module, wherein...

[0017] The data change monitoring module is used to receive data collection tasks, and when it detects data changes in the data table corresponding to the target data model carried in the data collection task, it acquires the changed data, assembles the changed data according to the target data model, generates a target data packet, and sends the target data packet to the scene change recognition module.

[0018] The scene change recognition module is used to receive the target data packet, obtain a target scene model that matches the target data packet from the scene model set; obtain the corresponding target field and the field value of the target field from the target data packet according to the scene recognition rules of the target scene model to form a target feature combination; match the target feature combination with the feature combination of the scene instance set, and determine the scene recognition result according to the matching result.

[0019] According to a fourth aspect of one or more embodiments of this specification, a computing device is provided, comprising:

[0020] Memory and processor;

[0021] The memory is used to store computer programs / instructions, and the processor is used to execute the computer programs / instructions, which, when executed by the processor, implement the steps of the above-described data processing method.

[0022] According to a fifth aspect of one or more embodiments of this specification, a computer-readable storage medium is provided that stores a computer program / instructions that, when executed by a processor, implement the steps of the data processing method described above.

[0023] According to a sixth aspect of one or more embodiments of this specification, a computer program product is provided, including a computer program / instructions that, when executed by a processor, implement the steps of the data processing method described above.

[0024] The data processing method provided in one or more embodiments of this specification is applied to a scene change recognition module, comprising: receiving a target data packet sent by a data change monitoring module, wherein the target data packet is a data packet obtained by the data change monitoring module assembling the changed data according to the target data model when it detects a data change in the data table corresponding to the target data model carried in the data acquisition task; obtaining a target scene model from a scene model set that matches the target data packet; obtaining the corresponding target field and the field value of the target field from the target data packet according to the scene recognition rules of the target scene model to form a target feature combination; matching the target feature combination with the feature combination of a scene instance set, and determining the scene recognition result based on the matching result.

[0025] Specifically, the scene change recognition module of this data processing method receives target data packets assembled according to the target data model specifications sent by the data change monitoring module, thereby achieving accurate capture and structured processing of online data changes; then it intelligently matches the stored target scene model and extracts target feature combinations, and finally outputs the scene recognition results quickly through the efficient matching mechanism of local storage. First, it achieves proactive real-time perception and automated identification of new online scenarios, enabling scenario judgment immediately upon data changes triggered by business traffic. This avoids the passive problem discovery mode that relies on post-event monitoring and manual verification, significantly shortening the time for discovering and mitigating unknown scenarios and effectively curbing their negative impact on real business. Second, this module is based on non-intrusive data monitoring of the database, requiring no modification to the business system. It has extremely low access costs and implementation risks, and is widely applicable, covering any system with a database, greatly improving the feasibility of the solution's promotion. Third, through a local caching mechanism and feature combination matching strategy, this module can continuously accumulate and optimize scenario recognition capabilities while ensuring high-concurrency processing performance. It systematically solves the problem of blind spots in scenario coverage caused by time and scope limitations in gray-scale verification, providing a low-cost, low-risk, and highly stable solution for new scenario risk perception and prevention in production environments. Attached Figure Description

[0026] To more clearly illustrate the technical solutions in the embodiments or prior art of this specification, the drawings used in the description of the embodiments or prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments recorded in this specification. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0027] Figure 1 This is a schematic diagram illustrating an application scenario of a data processing method provided in one embodiment of this specification;

[0028] Figure 2 This is a flowchart of a data processing method provided in one embodiment of this specification;

[0029] Figure 3 This is a schematic diagram of the structure of a data processing system provided in one embodiment of this specification;

[0030] Figure 4 This is a schematic diagram of the structure of a data processing device provided in one embodiment of this specification;

[0031] Figure 5 This is a structural block diagram of a computing device provided in one embodiment of this specification. Detailed Implementation

[0032] To enable those skilled in the art to better understand the technical solutions in this specification, the technical solutions in the embodiments of this specification will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this specification, and not all embodiments. Based on the embodiments in this specification, all other embodiments obtained by those skilled in the art without creative effort should fall within the scope of protection of this specification.

[0033] This specification uses specific terms to describe embodiments thereof. Terms such as "an embodiment," "one embodiment," and / or "some embodiments" refer to a particular feature, structure, or characteristic associated with at least one embodiment of this specification. Therefore, it should be emphasized and noted that references to "an embodiment," "one embodiment," or "an alternative embodiment" in different locations throughout this specification do not necessarily refer to the same embodiment. Furthermore, those skilled in the art can combine and integrate the different embodiments or examples described herein, as well as the features of those different embodiments or examples, without contradiction.

[0034] The terminology used in one or more embodiments of this specification is for the purpose of describing particular embodiments only and is not intended to be limiting of the one or more embodiments of this specification. The singular forms “a,” “an,” “an,” “the,” and “the” as used in one or more embodiments of this specification and the appended claims are also intended to include the plural forms unless the context clearly indicates otherwise. It should also be understood that the term “and / or” as used in one or more embodiments of this specification includes any or all possible combinations of one or more associated listed items.

[0035] The terms “comprising,” “including,” or any other variations thereof are intended to cover a non-exclusive inclusion, such that a process, method, product, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, product, or apparatus. Without further limitation, the presence of additional identical or equivalent elements in the process, method, product, or apparatus that includes said elements is not excluded.

[0036] Although the terms "first," "second," etc., may be used to describe various information in one or more embodiments of this specification, this information should not be limited to these terms. These terms are only used to distinguish information of the same type from one another. For example, "first" may also be referred to as "second," and similarly, "second" may also be referred to as "first," without departing from the scope of one or more embodiments of this specification. Ordinal numbers such as "first," "second," etc., do not necessarily indicate order; often they are used to facilitate the distinction of objects. For example, "first server" and "second server" usually refer to two servers. To distinguish these two servers, they are described as "first server" and "second server." Of course, sometimes these two servers may be the same server.

[0037] Depending on the context, the word "if" as used here can be interpreted as "when," "when," or "in response to determination."

[0038] In this specification, unless explicitly stated otherwise, "receiving and sending data" does not necessarily mean direct receiving and sending; it can also mean indirect receiving and sending. For example, A receiving data sent by B can be understood as A directly receiving the data sent by B, or it can be understood as A indirectly receiving the data sent by B through other entities such as C. Similarly, B sending data to A can be understood as B sending the data directly to A, or it can be understood as B indirectly sending the data to A through other entities such as C. Here, C can be one entity, or it can be two or more entities.

[0039] In this specification, unless explicitly stated otherwise, the relationships between structures can be direct or indirect. For example, when describing "A is connected to B," unless it is explicitly stated that A and B are directly connected, it should be understood that A can be directly connected to B or indirectly connected to B. Similarly, when describing "A is on top of B," unless it is explicitly stated that A is directly above B (AB is adjacent and A is above B), it should be understood that A can be directly above B or indirectly above B (AB is separated by other elements, and A is above B). And so on.

[0040] The user information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data used for analysis, stored data, displayed data, etc.) involved in one or more embodiments of this specification are all information and data authorized by the user or fully authorized by all parties. The collection, use and processing of related data shall comply with the relevant laws, regulations and standards of the relevant regions, and corresponding operation entry points shall be provided for users to choose to authorize or refuse.

[0041] The following explains the terms and concepts used in one or more embodiments of this specification.

[0042] DB: Database, a repository used for persistent storage of data.

[0043] MSG: Message, a method of data transmission.

[0044] RPC: Remote Procedure Call is a computer communication protocol that allows a program on one computer (client) to call a function or service on another computer (server) as if it were a local function, without the programmer explicitly handling the underlying network communication details.

[0045] TR: A type of RPC.

[0046] RMC: Resource Meta Center, a risk data platform that can obtain database change messages captured and pushed by the data synchronization middleware after the data has finally stabilized.

[0047] SDK: Software Development Kit, is a set of development tools designed for a specific platform, system, hardware, or service, designed to help developers build applications more quickly and easily.

[0048] To address the deficiencies in related technologies, this specification provides a data processing method through one or more embodiments, and also relates to a data processing apparatus, a computing device, a computer-readable storage medium, and a computer program product, which will be described in detail in the following embodiments.

[0049] The technical solutions provided in the various embodiments of this specification are described in detail below with reference to the accompanying drawings.

[0050] See Figure 1 , Figure 1 This is a schematic diagram illustrating an application scenario of a data processing method provided in an embodiment of this specification.

[0051] Figure 1 It includes a data change monitoring module 102, a scene change recognition module 104, a scene recognition software development tool 106, and a data configuration platform 108.

[0052] Specifically, the data change monitoring module 102 is used to monitor database tables related to the RMC data model in real time and capture resource data change events. When the data change monitoring module 102 detects data changes in the relevant database tables according to the pre-configured data acquisition task, it acquires the changed data and assembles the scattered changed data into a target data packet with complete business semantics according to the predefined data model specifications, and sends the target data packet to the scenario change recognition module 104.

[0053] The scene change recognition module 104 is used to perceive scene models and identify new and old scenes, including model access, scene recognition access, scene alarm access, and scene circuit breaker access functions.

[0054] The model access can be understood as the scenario change identification module 104 allowing test engineers or external systems to register, create, modify and manage data models. It belongs to the input configuration link of the scenario change identification module 104. Without data models, the scenario change identification module 104 will not know which tables the data change monitoring module 102 should monitor, how to assemble the data, etc.

[0055] Scene recognition access can be understood as a standardized interface provided by the scene change recognition module 104 to receive business data that needs to be recognized. Specifically, after the data change monitoring module 102 detects a data change and assembles the target data packet, it calls the scene recognition access interface to send the target data packet. Upon receiving the packet, the scene change recognition module 104 calls the scene recognition software development tool 106 to extract the scene based on the target data packet and then judges whether the extracted scene is a new or old scene.

[0056] Scene alarm access can be understood as an scalable, pluggable alarm channel management capability provided by the scene change identification module 104. When the scene change identification module 104 determines that the scene extracted from the target data packet is a new scene through the scene identification software development tool 106, it performs new scene processing, that is, it accesses the external alarm system through the scene alarm access function to send new scene verification alarms.

[0057] Scenario-based circuit breaker access can be understood as a capability provided by the scenario change identification module 104. When a relatively dangerous or abnormal business scenario is identified, it can automatically or manually trigger a predefined "circuit breaker" action, instead of simply issuing alarms and waiting for manual handling. For example, if the scenario change identification module 104 identifies an abnormal repayment scenario caused by a code bug (error or vulnerability) that is being generated in large numbers and repeatedly, and many such scenarios are generated per minute, alarms alone may not be enough to handle them. In this case, "circuit breaker" rules can be pre-configured: if an abnormal scenario occurs more than 1000 times within one minute, a "safety switch" interface of the business system will be automatically invoked to temporarily shut down the problematic function or channel, preventing the problem from escalating. Circuit breaker modes can be understood as the different circuit breaker strategies or mechanisms supported by the scenario change identification module 104. It defines the conditions and methods under which circuit breaker is triggered, such as manual circuit breaker, automatic circuit breaker, and tiered circuit breaker modes. The circuit breaker message queue is an asynchronous and decoupled component. When the scenario change identification module 104 decides to execute circuit breaking, it does not directly call potentially unstable or time-consuming business system interfaces. Instead, it quickly adds the "circuit breaking instruction" as a message to a circuit breaking message queue, enabling passive consumption. Downstream systems retrieve their corresponding circuit breaking instructions from this queue according to their capabilities, preventing the downstream systems from being overwhelmed. Furthermore, the message queue can persist messages to ensure that circuit breaking instructions are not lost. The circuit breaking database or rule base stores all rules and configurations related to "circuit breaking," such as circuit breaking rules (what scenarios and conditions trigger circuit breaking), circuit breaking actions (the specific operations to be performed after triggering circuit breaking), and whitelists (which businesses, users, or channels are exempt from circuit breaking). In practice, the scenario change identification module 104 determines whether to execute circuit breaking based on the circuit breaking database or rule base. When circuit breaking is required, it adds the circuit breaking instruction as a message to a circuit breaking message queue. Business applications requiring this type of circuit breaking can actively query or periodically check with the scenario change identification module 104 or a specific circuit breaking service to see if there are any circuit breaking instructions for them currently being processed.

[0058] The scene recognition software development tool 106, which can be understood as a scene recognition engine, first quickly matches the target scene model corresponding to the target data packet from its recognition model cache. Then, it extracts specific target feature value combinations from the data packet according to the scene recognition rules defined by the model. Subsequently, it can quickly compare this feature combination with existing scene instances in the historical scene cache to achieve new scene recognition. If the match is successful, it is determined to be an old scene. After returning the result to the scene change recognition module 104, the process ends, ensuring high efficiency.

[0059] If no matching feature combination is found in the local cache (historical scene cache) (i.e., the scene recognition SDK initially determines it to be a new scene or cannot be determined), the feature combination is sent to the data configuration platform 108 for secondary arbitration via the TR direct interaction mode, enabling the data configuration platform to extract the feature combination. The data configuration platform 108 then uses its stored full-scale scene instance library to perform the final matching and adjudication, and returns the confirmation result.

[0060] The scene change recognition module 104 can make decisions based on the returned final result: if it is confirmed as a new scene, it will initiate subsequent processes, call a certain detection engine to send an alarm notification, thereby completing the closed loop from new scene discovery to risk warning. The entire process, through the data change monitoring module 102, the scene change recognition module 104, the scene recognition software development tool 106, and the data configuration platform 108, demonstrates the system's scalability in task queue management and external system integration, and together constitutes an efficient and reliable solution for new scene risk perception and prevention in the production environment.

[0061] See Figure 2 , Figure 2 The flowchart of a data processing method provided in the embodiments of this specification specifically includes the following steps.

[0062] Specifically, this data processing method is applied to the scene change recognition module, namely the scene change recognition module 104 in the above embodiment.

[0063] Step 202: Receive the target data packet sent by the data change monitoring module.

[0064] The target data packet is a data packet obtained by the data change monitoring module when it detects data changes in the data table corresponding to the target data model carried in the data acquisition task, based on the target data model.

[0065] The data processing methods provided in one or more embodiments of this specification can be applied to scenarios such as financial risk and security detection (for identifying new types of risky transactions), network security and intrusion detection (for discovering new attack vectors), industrial IoT and predictive maintenance (for discovering new failure modes or new equipment degradation trends), intelligent transportation and urban command (for discovering new traffic bottlenecks), and retail and consumer behavior analysis (for discovering new consumer groups, predicting blockbuster products, etc.). For ease of understanding, the application of the data processing methods provided in one or more embodiments of this specification to the financial risk and security detection scenario will be used as an example in the description.

[0066] The data change monitoring module is the data change monitoring module 102 in the above embodiment, which is used to monitor the data tables corresponding to the target data model carried in the data acquisition task. Any changes such as addition, deletion, and modification of these data tables will be monitored.

[0067] Data acquisition tasks can be understood as predefined monitoring tasks, which guide the data change monitoring module to monitor the corresponding data tables according to the rules defined in the target data model.

[0068] The target data model defines which data tables, fields, and relationships between the data tables need to be monitored.

[0069] Assembly can be understood as piecing together and linking scattered change data into a complete data unit that describes a business event, according to the definition of the target data model. For example, the business event of "user overdue payment" requires seeing both the "overdue status" in the repayment schedule table and the "loan amount" in the loan master table to determine it. Assembly is the purpose of constructing this complete business context.

[0070] Specifically, receiving the target data packet sent by the data change monitoring module can be understood as follows: the data change monitoring module monitors the data tables corresponding to the target data model specified in the predefined data collection task in real time. If some data tables in these data tables have undergone data changes, the module can assemble the scattered changed data according to the definition of the target data model to generate a target data packet containing complete business information and send it to the scene change identification module. After receiving the target data packet sent by the data change monitoring module, the scene change identification module can perform subsequent identification of new and old scenes.

[0071] Step 204: Obtain the target scene model that matches the target data packet from the scene model set.

[0072] The scenario model set can be understood as a set of scenario models stored in the current machine's memory and cached locally, without needing to be directly queried from a remote server (such as the data configuration platform 108). The scenario model set can be understood as a collection of multiple scenario models, containing multiple scenario models, and each scenario model includes, but is not limited to, metadata (model ID, name, description), data source (which data tables are monitored), and identification criteria (which feature fields need to be extracted from these tables and what rules these fields need to meet).

[0073] Specifically, the target scene model obtained from the scene model set and matched with the target data packet includes:

[0074] Using the scene recognition software development tool, a target data table corresponding to the changed data in the target data packet is obtained, and a target scene model matching the target data table is obtained from the locally cached scene model set according to the target data table. The target data table is one or more data tables in the data table corresponding to the target data model, and the associated data table of the target scene model matches the target data table.

[0075] The scene recognition software development tool, namely the scene recognition software development tool 106 in the above embodiment, can also be called a scene recognition SDK, which is a pre-packaged software toolkit. In practical applications, the scene change recognition module itself does not contain complex recognition algorithms, but rather implements the scene recognition work in the scene recognition SDK through the API (Application Programming Interface) provided by the scene recognition SDK.

[0076] The target data packet is a data unit assembled by the data change monitoring module, containing complete business information. This target data packet includes not only the specific numerical values ​​of the changes but also important metadata, such as metadata indicating which data tables the changed data originates from. In practical applications, the target data model may correspond to five data tables, and the changed data may originate from three of these tables; these three tables can then be considered the target data tables.

[0077] The association data table for each scene model can be understood as the data table corresponding to each scene model, i.e., the predefined data table that each scene model needs to monitor. Matching the association data table of the target scene model with the target data table can be understood as comparing the target data table with the association data table of each scene model in the scene model set. If the association data table of a certain scene model is a subset of the target data table, then a successful match can be determined. The target scene model is one or more scene models selected after the matching operation.

[0078] In practice, when the scene change recognition module receives the target data packet sent by the data change monitoring module, it calls the scene recognition software development tool and sends the target data packet to the scene recognition software development tool. The scene recognition software development tool obtains the target data table corresponding to the changed data in the target data packet and matches the target data table with the associated data table of each scene model in the locally cached scene model set. One or more matching scene models are identified as target scene models, and the new and old scenes are then determined according to the rules of the target scene model.

[0079] The data processing method provided in one or more embodiments of this specification utilizes scene recognition software development tools to process received target data packets to achieve the determination of new and old scenes. First, scene recognition SDKs typically encapsulate the algorithms required for scene recognition, eliminating the need to build models from scratch or handle complex image recognition logic. By calling the APIs provided by the scene recognition SDK, scene recognition functionality can be quickly integrated, significantly shortening the development cycle and time to market. Second, scene recognition may involve complex recognition technologies; scene recognition SDKs abstract these complexities, providing simple and easy-to-use interfaces, enabling even developers without relevant backgrounds to easily implement advanced functions, lowering the technical threshold and reducing labor costs. Furthermore, many scene recognition SDKs support multiple platforms, allowing developers to uniformly implement scene recognition functionality on different devices, ensuring a consistent user experience while reducing repetitive development work.

[0080] Step 206: Based on the scene recognition rules of the target scene model, obtain the corresponding target fields and the field values ​​of the target fields from the target data packet to form a target feature combination.

[0081] Specifically, the implementation of step 206 can be understood as the processing operation using scene recognition software development tools.

[0082] The scenario recognition rules can be understood as the recognition standards defined by the target scenario model. They specify which key database fields are needed to determine the business scenario, the recognition rules for each database field, and the logical combinations of these fields. Each target scenario model's scenario recognition rules are its corresponding features or fingerprints. The difference between each target scenario model lies in these different scenario recognition rules. For example, a target scenario model for an "overdue scenario model" might have the following scenario recognition rules: Field: status (repayment status), rule: equal to 'OVERDUE' (overdue); Field: overdue_days (overdue days), rule: greater than 0; the logical combination is: status = 'OVERDUE' and overdue_days > 0.

[0083] In practical implementation, for each matched target scene model, the scene recognition SDK extracts specific target fields and their values ​​as feature values ​​from the target data packet according to the recognition standards defined for that target scene model. Continuing with the previous example, for the "overdue scene model," it extracts the repayment status and overdue days fields and their values, forming a target feature combination based on the logical combination corresponding to this "overdue scene model." The goal is to extract a concise and standardized data fingerprint from the massive original target data packet according to the target scene model, preparing for subsequent identification of new and old scenes. The same principle applies to other target scene models, and will not be elaborated further here.

[0084] Step 208: Match the target feature combination with the feature combination of the scene instance set, and determine the scene recognition result based on the matching result.

[0085] Specifically, the implementation of step 208 can be understood as the processing operation using scene recognition software development tools.

[0086] The scene instance set can be understood as a collection of all known scene instances stored in the local cache. This scene instance set and / or the scene model set mentioned above can both be understood as being fully retrieved from the data configuration platform by the scene change recognition module or gradually learned and cached during operation.

[0087] In practical applications, a scenario model can be understood as a rule, an abstract definition; a scenario instance can be understood as the specific result produced after rule matching, a unique combination of feature values; for example, (status='OVERDUE', overdue_days='1') and (status='OVERDUE', overdue_days='2') are two different scenario instances. The scenario instance set can be understood as the collection of all different combinations of feature values ​​that have ever appeared in the current system.

[0088] In specific implementation, determining the scene recognition result based on the matching result includes:

[0089] If the matching result is determined to be a match, the scene recognition result is determined to be the scene corresponding to the target feature combination as an existing scene;

[0090] If the matching result is determined to be a mismatch, the scene recognition result is determined to be that the scene corresponding to the target feature combination does not exist.

[0091] or

[0092] The target feature combination is sent to the data configuration platform and matched with the feature combination in the scene instance library of the data configuration platform. The scene recognition result is determined based on the matching result returned by the data configuration platform.

[0093] Specifically, when multiple target feature combinations exist, the scene recognition SDK matches each target feature combination with the feature combinations of each scene instance in the local cached scene instance set, and determines the scene recognition result based on the matching result. That is, if the matching result is a match, the scene recognition result indicates that the scene corresponding to the target feature combination is an existing scene, i.e., an old scene. In this case, the scene recognition result can be returned to the scene change recognition module, ending the process. If the matching result is a mismatch, there are two possible operations: one is to directly determine that the scene corresponding to the target feature combination is a non-existent scene, i.e., a new scene; the other, to ensure scene recognition accuracy, is to send the target feature combination to the data configuration platform, where it matches the target feature combination with the feature combinations of each scene instance in the data configuration platform's scene instance library, and determines the scene recognition result based on the matching result returned by the data configuration platform.

[0094] The data processing method provided in one or more embodiments of this specification first uses a scene recognition SDK to match target feature combinations with feature combinations of scene instance sets in a local cache. Based on the matching results, it initially determines whether the scene corresponding to the target feature combination is a new scene. If the scene recognition SDK determines that the scene corresponding to the target feature combination is an old scene, the process can be quickly terminated, achieving efficient processing of new and old scene judgment. If the scene recognition SDK determines that the scene corresponding to the target feature combination does not belong to an old scene, the target feature combination can be sent to a data configuration platform for secondary judgment, effectively improving the accuracy of scene recognition and the reliability of system decisions. By introducing the centralized analysis capabilities of the backend platform, more comprehensive historical data, stronger computing resources, and more complex model algorithms can be used to verify the initial judgment at the edge, avoiding recognition deviations caused by misjudgment of local features or model limitations.

[0095] When the scene change recognition module determines, using the scene recognition SDK, that the scene corresponding to the target feature combination is a new scene, it will not only send a new scene alarm message via the scene alarm interface, but also, if it determines that the scene corresponding to the target feature combination meets the preset circuit breaker conditions, execute a circuit breaker processing strategy for the scene corresponding to the target feature combination. The specific implementation is as follows:

[0096] After determining that the scene recognition result is that the scene corresponding to the target feature combination is a non-existent scene, the method further includes:

[0097] If the scenario corresponding to the target feature combination is determined to meet the preset circuit breaker conditions, a circuit breaker processing strategy for the scenario corresponding to the target feature combination is executed.

[0098] The preset circuit breaker conditions can be set according to actual needs. For example, the preset circuit breaker condition is the number of times (e.g., 1000 times) that the scenario corresponding to the target feature combination occurs within a preset time period (e.g., 1 hour). The circuit breaker processing strategy can be understood as the circuit breaker processing steps for the scenario corresponding to the target feature combination. For example, the circuit breaker processing strategy is to add the circuit breaker instruction to the corresponding message queue when a circuit breaker instruction for the scenario corresponding to the target feature combination is received, so that the business application that needs to be circuit-broken can listen to and process it from the message queue.

[0099] Specifically, when the scene change recognition module determines that the scene corresponding to the target feature combination meets the preset circuit breaker conditions, it sends a query request to the circuit breaker processing module, carrying the scene corresponding to the target feature combination and the current frequency of that scene. Upon receiving a circuit breaker instruction from the circuit breaker processing module in response to the query request, the module adds the circuit breaker instruction to the circuit breaker message queue so that the corresponding business application can listen for and process it. After receiving the query request, the circuit breaker processing module performs a match in the rule base. If a match is successful, it generates and sends a circuit breaker instruction to the scene change recognition module.

[0100] In practice, when the scene change recognition module receives a new scene corresponding to the target feature combination from the scene recognition SDK, it sends a new scene alarm and continuously monitors the frequency of the same new scene appearing in a short period of time. For example, the scene change recognition module sets up a counter to record the number of times each new scene appears within one minute. When the frequency of a new scene meets the threshold (e.g., 100 times within one minute), it actively queries the circuit breaker database or rule base (as in the circuit breaker processing module mentioned above), sending a query request: "Scene ID **, current frequency 100 times / minute, please provide instructions according to the rules." Upon receiving the query request, the database or rule base matches it in its rule base. Example rule: IF Scene X AND Frequency > 50 times / minute THEN Execute circuit breaker action: Call the "Close payment channel" interface of the payment application. After the database or rule base matches the rule, it generates a clear and executable circuit breaker instruction and returns it to the scene change recognition module.

[0101] After receiving the circuit breaker instruction, the scenario change recognition module adds the circuit breaker instruction to the circuit breaker message queue. Business applications that need to be circuit-broken (such as payment services) will listen to the circuit breaker message queue in real time, retrieve the circuit breaker instruction from the circuit breaker message queue, parse the circuit breaker instruction and execute the internal circuit breaker logic, such as calling a certain method to close the switch of the payment channel in memory.

[0102] The data processing method provided in one or more embodiments of this specification involves a scene change identification module that, upon receiving a scene identification result from the SDK indicating a new scene, not only issues an alarm but also simultaneously triggers a circuit breaker mechanism. First, the alarm information promptly notifies maintenance personnel or relevant modules, enabling rapid perception and response to abnormal or unknown scenes, facilitating subsequent analysis and processing. Second, the circuit breaker mechanism immediately blocks subsequent processes that may cause system instability or erroneous decisions, preventing functional abnormalities, data pollution, or user experience degradation due to misjudgment of a new scene, effectively ensuring system stability and security.

[0103] In the data processing method provided in one or more embodiments of this specification, the scene change recognition module receives the target data packet, which is assembled according to the target data model specification, sent by the data change monitoring module, and achieves accurate capture and structured processing of online data changes; then, it uses scene recognition software development tools to intelligently match the target scene model and extract the target feature combination, and finally outputs the scene recognition result quickly through the efficient matching mechanism of local caching. First, it achieves proactive real-time perception and automated identification of new online scenarios, enabling scenario judgment immediately upon data changes triggered by business traffic. This avoids the passive problem discovery mode that relies on post-event monitoring and manual verification, significantly shortening the time for discovering and mitigating unknown scenarios and effectively curbing their negative impact on real business. Second, this module is based on non-intrusive data monitoring of the database, requiring no modification to the business system. It has extremely low access costs and implementation risks, and is widely applicable, covering any system with a database, greatly improving the feasibility of the solution's promotion. Third, through a local caching mechanism and feature combination matching strategy, this module can continuously accumulate and optimize scenario recognition capabilities while ensuring high-concurrency processing performance. It systematically solves the problem of blind spots in scenario coverage caused by time and scope limitations in gray-scale verification, providing a low-cost, low-risk, and highly stable solution for new scenario risk perception and prevention in production environments.

[0104] See Figure 3 , Figure 3 A schematic diagram of the structure of a data processing system provided in one or more embodiments of this specification is shown. Figure 3 As shown, the system includes:

[0105] Data change monitoring module 302 and scene change recognition module 304, wherein,

[0106] The data change monitoring module 302 is used to receive a data acquisition task, and when it detects a data change in the data table corresponding to the target data model carried in the data acquisition task, it acquires the changed data, assembles the changed data according to the target data model, generates a target data packet, and sends the target data packet to the scene change recognition module.

[0107] The scene change recognition module 304 is used to receive the target data packet, obtain a target scene model that matches the target data packet from the scene model set; obtain the corresponding target field and the field value of the target field from the target data packet according to the scene recognition rules of the target scene model to form a target feature combination; match the target feature combination with the feature combination of the scene instance set, and determine the scene recognition result according to the matching result.

[0108] The data change monitoring module 302 is the same as the data change monitoring module 102 in the above embodiment, and the scene change recognition module 304 is the same as the scene change recognition module 104 in the above embodiment. The only difference is that different reference numerals are used because the accompanying drawings are different.

[0109] Optionally, the scene change recognition module 304 is further configured to:

[0110] Using the scene recognition software development tool, a target data table corresponding to the changed data in the target data packet is obtained, and a target scene model matching the target data table is obtained from the locally cached scene model set according to the target data table. The target data table is one or more data tables in the data table corresponding to the target data model, and the data table associated with the target scene model matches the target data table.

[0111] Optionally, the scene change recognition module 304 is further configured to:

[0112] If the matching result is determined to be a match, the scene recognition result is determined to be the scene corresponding to the target feature combination as an existing scene;

[0113] If the matching result is determined to be a mismatch, the scene recognition result is determined to be that the scene corresponding to the target feature combination does not exist.

[0114] or

[0115] The target feature combination is sent to the data configuration platform and matched with the feature combination in the scene instance library of the data configuration platform. The scene recognition result is determined based on the matching result returned by the data configuration platform.

[0116] Optionally, the scene change recognition module 304 is further configured to:

[0117] Receive the scene recognition results returned by the scene recognition software development tool;

[0118] If the scene recognition result indicates that the scene corresponding to the target feature combination does not exist, a scene security prompt is sent.

[0119] Scenario-based security alerts can be understood as the aforementioned alarm messages, which can be sent via email, SMS, or pop-up notifications. In practical applications, after receiving an alarm notification, test engineers can query the corresponding business data, perform relevant checks, and if no problems are found, close the alarm; if problems are found, take appropriate action.

[0120] Optionally, the scene change recognition module 304 is further configured to:

[0121] If the scenario corresponding to the target feature combination is determined to meet the preset circuit breaker conditions, a circuit breaker processing strategy for the scenario corresponding to the target feature combination is executed.

[0122] For details on the specific implementation steps of the scene change recognition module 304, please refer to the specific implementation of the data processing method described above, which will not be repeated here.

[0123] In practice, before the data change monitoring module 302 and the scene change recognition module 304 combine to identify new and old scenes, a scene model needs to be created on the data configuration platform, a data model needs to be created in the scene change recognition module 304, and a data collection task subsequently sent to the data change monitoring module 302 needs to be created based on the created data model. The specific implementation method is as follows:

[0124] The data processing system further includes a data configuration platform, which is the same as the data configuration platform 108 in the above embodiment.

[0125] The data configuration platform is used to respond to a scene model creation request and create a scene model based on the scene model attribute information, associated data tables, and scene recognition rules carried in the scene model creation request.

[0126] Based on the scene model and the target database, create scene instances and store the scene instances in the scene instance library;

[0127] The scene model and the scene instance are sent to the scene change recognition module, or the scene recognition software development tool called by the scene change recognition module is cached locally.

[0128] A scenario model creation request can be understood as a request generated by a developer or test engineer clicking the "Create Scenario Model" button on the graphical user interface (such as a web-based management interface) provided by the data configuration platform. Of course, it can also be a scenario model creation request generated by developers or test engineers through other means (such as voice commands).

[0129] Scene model attribute information includes, but is not limited to, scene model name, scene description, associated system / highlighting, etc.

[0130] Related data tables can be understood as the data tables that the scene model to be created depends on.

[0131] Scene recognition rules can be understood as a unique standard for recognizing a scene model.

[0132] The following are examples of the steps involved in creating a scene model:

[0133] Suppose that the data configuration platform provides a web-based management interface, and test engineers can create scenario models on this web-based management interface.

[0134] For example, when a test engineer clicks the "Create New Scene Model" button, they fill in basic information, define data sources, and configure recognition standards on the pop-up scene model creation page. After clicking the "Confirm" or "Save" button, a corresponding scene model creation request is generated. Once the data configuration platform receives the scene model creation request, it can complete the scene model creation based on the content carried in the request.

[0135] The basic information to be filled in includes, but is not limited to: Scenario Model Name: For example, enter a name with business significance: "Installment Loan Overdue Scenario". Scenario Description: Describe in detail the business meaning of this scenario model, such as "referring to the behavior of a user failing to repay the principal and interest of an installment loan in full on the due date". Associated System / Project: Select the business area to which this scenario belongs, such as "Consumer Credit System".

[0136] Defining data sources includes, but is not limited to: selecting data tables, i.e., choosing the data tables that this scenario model depends on from the database table list; for example, checking the `loan_installment_log` (loan installment repayment log table). If multiple data tables are selected, the relationships between them can also be defined. For example, if a scenario model requires multiple data tables to determine its structure (e.g., not only the "loan installment repayment log table" but also the main table `loan_main` to retrieve the loan amount), then these two data tables need to be added here, and the join relationship between them needs to be defined (e.g., `loan_installment_log.loan_id = loan_main.id`).

[0137] Configuration identification criteria: The data configuration platform will list all fields of the selected data table. Test engineers need to select those fields that can uniquely define a scenario as features. For each feature field, its identification rules also need to be specified. For example: field: status (repayment status), rule: equal to 'OVERDUE' (overdue); field: overdue_days (overdue days), rule: greater than 0. In addition, for multiple fields, their logical combinations also need to be set: that is, multiple conditions can be defined and combined using logical relationships (AND, OR); for example: status = 'OVERDUE' and overdue_days>0.

[0138] Specifically, a scenario model can be understood as a rule template that abstractly defines the common data features of a class of scenarios. For example, an "overdue scenario model" defines common rules for all overdue behaviors, rather than a specific overdue payment. Identification criteria, on the other hand, can be understood as a set of feature extraction rules and judgment logic.

[0139] In practical applications, when a data processing system goes live for the first time, the scenario instance library of the data configuration platform is empty. At this point, any incoming business traffic will not be found in the local cache by the scenario recognition SDK, nor in the scenario instance library of the data configuration platform. Consequently, the data processing system will misclassify all business traffic as new scenarios, generating a massive number of meaningless alerts. Therefore, to avoid this situation, before connecting the data processing system to online traffic, it is necessary to pre-store historically existing, known, and valid scenario models in the scenario instance library of the data configuration platform. This will allow the data configuration platform to accurately identify new scenarios.

[0140] Specifically, the scene instance library stores scene instances, not scene models. A scene model is a rule, an abstract definition. A scene instance, on the other hand, is the concrete result of rule matching, a unique combination of feature values; for example, (status='OVERDUE', overdue_days='1') and (status='OVERDUE', overdue_days='2') are two different instances. The scene instances in the scene instance library represent all the different combinations of feature values ​​that have ever appeared in the current system.

[0141] Therefore, after creating the scenario model, scenario instances are created based on the created scenario model and the target database, and then the scenario instances are stored in the scenario instance library. The target database can be understood as the database corresponding to a previously existing, known business system. Furthermore, the creation of scenario instances can be achieved offline, online, or through other methods.

[0142] The creation of scene instances is illustrated using offline, online, or other methods, with the specific implementation steps described below:

[0143] Offline Method: First, determine the data source: for example, query directly from the production environment database. Second, write SQL (Structured Query Language) queries. For example, test engineers can write corresponding SQL statements based on the scenario model created above to query all unique feature value combinations from relevant data tables, forming scenario instances. Then, generate a data file: export the SQL query results as a structured file. Finally, push this structured file to the data configuration platform: use a data push tool (or the API provided by the data configuration platform) to batch import this structured file into the scenario instance library of the data configuration platform.

[0144] Online approach: First, determine the data source: obtain data from logs or message queues of the real-time business system. Second, within the business system, at the locations where business data is generated (or by parsing logs), embed tracking points to record data conforming to the scenario model definition, forming scenario instances. Then, send these scenario instances to the scenario instance library of the data configuration platform.

[0145] Other methods: Creating scenario instances through log file analysis: Collect log files printed by business systems, parse and deduplicate them using scripts or log processing tools, extract multiple feature values ​​to form scenario instances, and then import the scenario instances into the scenario instance library of the data configuration platform. Creating scenario instances during backup data recovery: If this is an upgrade of an old system, the scenario instance identification records in the old system can be directly migrated to the scenario instance library of the new data configuration platform.

[0146] After creating the scene model and scene instances based on the scene model, the scene model and scene instances can be sent to the scene recognition software development tool for local caching. When the data processing system is actually running, the scene recognition SDK can make a preliminary determination of new and old scenes based on the local cache.

[0147] The data processing system provided in one or more embodiments of this specification pre-creates a scene model on a data configuration platform and generates specific scene instances based on the model before determining whether a scene is new or old. Simultaneously, the scene model and scene instance information are synchronized to the local cache of the scene recognition SDK. This method first ensures that the scene recognition SDK has complete scene recognition criteria and metadata support at runtime, avoiding delays or network unavailability caused by real-time requests, significantly improving recognition efficiency and response speed. Second, the scene model and scene instances cached locally by the scene recognition SDK support offline recognition, enhancing the robustness and availability of the data processing system, especially suitable for scenarios with weak networks or edge devices. Furthermore, centralized configuration enables unified management, rapid iteration, and canary release of scene models, helping to improve the accuracy and consistency of scene recognition, providing a reliable data foundation for subsequent comparisons of new and old scenes and dynamic updates, thereby achieving efficient, stable, and scalable intelligent recognition capabilities.

[0148] After creating the scene model and scene instances, a corresponding data model can be created in the scene change recognition module. Based on this data model, a data acquisition task can be subsequently issued to the data change monitoring module 302. The specific implementation method is as follows:

[0149] The scene change recognition module 304 is also used for:

[0150] In response to a data model creation request, a data model is created based on the data model attribute information, associated scene models, and associated data tables of the associated scene models carried in the data model creation request.

[0151] Based on the data model, a data acquisition task is created and sent to the data change monitoring module. The target data model is one or more of the data models, and the data table corresponding to the target data model is one or more of the associated data tables of the scene model associated with the target data model.

[0152] The data model creation request can be understood as a request sent by the test engineer to the scene change recognition module 304 via the graphical user interface provided by the scene change recognition module 304. This data model tells the data change monitoring module 302 what data needs to be monitored and how to assemble this data, so that the scene change recognition module 304 can provide the target data packet it receives to the scene recognition SDK for new and old scene recognition.

[0153] Data model attribute information includes, but is not limited to, data model name and data model description; associated scenario models can be understood as the scenario models created in the data configuration platform; associated data tables of associated scenario models can be understood as the data tables selected for the scenario models created in the data configuration platform.

[0154] The creation of a data model is illustrated with an example, and the specific implementation steps are as follows:

[0155] Suppose that the scene change recognition module 304 provides a web-based management interface, then test engineers can create data models on this web-based management interface.

[0156] For example, when a test engineer clicks the "Create Data Model" button, they fill in basic information, define data entities, and define entity attributes on the pop-up data model creation page. If the scenario model corresponding to this data model involves multiple data tables, they define entity relationships. After clicking the "Confirm" or "Save" button, a corresponding data model creation request is generated. After receiving the data model creation request, the scenario change recognition module 304 can complete the data model creation based on the content carried in the data model creation request.

[0157] The process of filling in basic information includes, but is not limited to, filling in the data model name, i.e., entering a name with business significance, such as "Loan Installment Repayment Data Model" which corresponds to the scenario model in the data configuration platform; defining the data model description information: briefly explaining its purpose, such as "data used to assemble the loan installment repayment table for overdue scenario identification"; and setting the associated scenario model: selecting the scenario model previously created in the data configuration platform from a drop-down list, such as "installment loan overdue scenario", to establish the binding relationship between the data model and the scenario model.

[0158] Defining a data entity can be understood as test engineers being able to click "Add Entity" or "Add Data Table" and set the table name in the pop-up page: that is, enter the actual table name in the database, such as loan_installation_log; they can also set an alias: for example, set a more easily understood alias: "Repayment Plan Table"; and determine the data source: that is, select the database connection information where this data table is located (such as the address, account, etc. of the test database and production database). This connection information is usually pre-configured in the scenario change recognition module 304.

[0159] Defining entity attributes can be understood as adding fields to a newly added data table, such as the `loan_installation_log` table. Clicking "Add Fields" allows you to add the fields used in the corresponding scenario model's identification criteria and the fields required for relationships. Aliases for fields (such as status or overdue days) can also be added. Furthermore, you can mark primary keys and foreign keys; for example, mark `id` (identifier) ​​as the primary key and `loan_id` as a foreign key.

[0160] Continuing with the previous example, defining the entity relationship can be understood as follows: if the "installment loan overdue scenario" still requires the `loan_amount` field from the main table `loan_main`, then the steps of defining the data entity and defining entity attributes need to be repeated, adding the `loan_main` table. Then click "Add Relationship". At this point, the main entity is the `loan_installment_log` data table, and the associated field is `loan_id`; the secondary entity is the `loan_main` data table, and the associated field is `id`; and the association type is set; subsequently, the data change monitoring module 302 can combine the changed data from these two data tables based on this entity relationship.

[0161] After creating the data model, a specific and executable data monitoring and collection task can be started based on the created data model; that is, a data collection task is created according to the data model and sent to the data change monitoring module 302, so that the data change monitoring module 302 can monitor data changes in the data tables associated with the data model in each data collection task.

[0162] Following the previous example, test engineers can click the "Create Data Acquisition Task" button based on the data model. In the pop-up task creation page, they can fill in the task configuration, trigger conditions, and scheduling strategy. The task configuration includes, but is not limited to, the task name (e.g., "Production Database Loan Repayment Data Acquisition"), the selected data model (e.g., selecting the previously created "Loan Installment Repayment Data Model" from a dropdown list), and the execution mode (e.g., full acquisition: fetching all existing data from the table at once (for initialization), or incremental monitoring: monitoring the database's binary log (or other change logs) and capturing incremental data in real time). Trigger conditions can be understood as the triggering rules for the data acquisition task, such as "acquire only when the status field changes." This trigger condition can be set or not, depending on specific needs. The scheduling strategy can be understood as the execution frequency of the data acquisition task. For example, for full acquisition, a scheduled task (e.g., executing once daily at 3 AM) can be set to trigger the data acquisition task. For incremental monitoring, the data acquisition task can be triggered whenever incremental data is available. In practical applications, the scheduling strategy for data acquisition tasks can also be set according to actual needs; no restrictions are imposed here. After filling in the above information, click the save or publish button to complete the creation of the data collection task.

[0163] The data processing system provided in one or more embodiments of this specification creates a data model in the scene change recognition module 304 and creates corresponding data acquisition tasks based on the data model. Firstly, the creation of the data model enables the abstraction and centralized management of monitoring rules. Test engineers do not need to delve into the underlying details of the database; they only need to define the monitoring objects (such as "loan overdue transaction records"), key fields (such as "status" and "amount"), and inter-table relationships using a visual interface in business language. This "what you see is what you get" configuration method transforms scattered technical fields into data entities with clear business semantics, greatly reducing the configuration threshold and maintenance costs, while ensuring the accuracy and consistency of the monitoring scope. Secondly, by creating data acquisition tasks based on data models, the data capture process is automated and standardized. After a data acquisition task is released, the data change monitoring module automatically listens for changes in the target data tables associated with the target data model carried in the data acquisition task, and assembles data packets in real time according to the rules defined in the target data model. This avoids the incompleteness and bias of manually simulated data, and ensures that every real production business transaction is seamlessly captured and transformed into identifiable events, achieving comprehensive coverage of online business scenarios and preventing leaks caused by test omissions at the source. Furthermore, a predefined data model can be reused by multiple data acquisition tasks to monitor similar businesses in different environments; when the business table structure changes, only adjustments need to be made at the data model level, and all related data acquisition tasks will automatically take effect.

[0164] Once the data acquisition task is created, the scene change recognition module 304 can automatically send the data acquisition task to the data change monitoring module 302 for execution. After receiving the data acquisition task, the data change monitoring module 302 will parse the data model in the data acquisition task, establish a connection with the target data table in the target database, start the listener, and begin to monitor data changes in the target data table according to the requirements of the data model.

[0165] When a business application processes a user request—for example, a user has an overdue loan repayment, and the system inserts a new record into the `loan_installment_log` table or updates the `status` field of a record to 'OVERDUE'—the data change monitoring module 302 will obtain all the data required for a complete business event based on the relationships defined in the target data model carried in the data collection task, not just the changed data. In practical applications, if a corresponding data collection task is not issued to the data change monitoring module 302 in advance, after receiving a change event, it will determine which table the change occurred in (e.g., the `loan_installment_log` table) and which of its registered data models monitors this table (e.g., the "loan installment repayment data model" monitors this table). Only after confirming that a change has occurred in a data table monitored by a specific data model will it proceed to the next step.

[0166] For data assembly, the data change monitoring module 302 will handle two scenarios. In one scenario, the scenario model involves only one data table. For example, if the data change monitoring module 302 detects a change in a single record in the `loan_installation_log` table (e.g., the record with id=1024 has been updated), since the data model only defines this one table, the data change monitoring module 302 will directly query this table, obtain the latest values ​​of all fields for this record, and then assemble it into a JSON object (target data packet) containing all fields of this record. In the other scenario, the scenario model involves multiple data tables, assuming the scenario model also needs the `loan_amount` field from the `loan_main` table. The assembly process of the data change monitoring module 302 is as follows:

[0167] The data change monitoring module 302 captures data changes in the `loan_installment_log` table (e.g., records with `loan_id=888`). Based on the predefined table relationships in the data model (`loan_installment_log.loan_id = loan_main.id`), the module 302 automatically executes a join query, using the `loan_id` value of the changed record (i.e., 888) as a condition to retrieve the corresponding loan information from the `loan_main` table, obtaining fields such as `loan_amount`. These fields are then assembled into a nested JSON object (target data packet), containing complete, joined data from multiple tables.

[0168] After generating the target data packet, the data change monitoring module 302 sends the assembled target data packet to the scene change recognition module 304 through an asynchronous message queue or an HTTP (Hypertext Transfer Protocol) API call.

[0169] Based on the above description, the specific operation steps of the scene change recognition module 304 are as follows:

[0170] Step 1: Scene change recognition module 304 calls scene recognition SDK for initial recognition: After receiving the assembled target data packet sent by data change monitoring module 302, scene change recognition module 304 calls the recognition interface of its built-in scene recognition SDK and passes the target data packet as a parameter so as to delegate the work of recognizing the new and old scenes to the scene recognition SDK.

[0171] Step 2: Scene Recognition SDK Model Filtering and Feature Extraction: After receiving the target data packet, the scene recognition SDK filters out matching target scene models from all scene models in its locally cached scene model set based on key information such as `table_name` in the target data packet. For example, if the data comes from the `loan_installment_log` table, the scene recognition SDK will look for scene models that define which table to monitor. For each matching target scene model, the scene recognition SDK extracts specific feature values ​​from the target data packet according to the recognition criteria defined for that scene model. For example, for the "overdue scene model", the values ​​of `status` and `overdue_days` are extracted and combined into a target feature combination: `('OVERDUE', 2)`.

[0172] Step 3: Scene recognition SDK local cache judgment: That is, the scene recognition SDK compares this target feature combination ('OVERDUE', 2) with the feature combinations of scene instances in the locally cached scene instance set.

[0173] If the cache is hit, meaning that the feature combination already exists in the local cache, it means that the exact same scenario has been processed before. At this point, the scenario recognition SDK determines that this is an old scenario and returns the recognition result (such as scenario ID and scenario name) to the scenario change recognition module 304, and the process ends here.

[0174] If the cache is not hit, meaning the target feature combination cannot be found in the local cache, the scene recognition SDK can initially determine that this may be a new scene. Since the scene recognition SDK's local cache may have a delay (not containing all the latest scenes), to ensure the accuracy of scene recognition, it can send the target feature combination to the data configuration platform, initiating a secondary recognition request.

[0175] Step 4: Platform Secondary Recognition and Final Scene Recognition: The scene recognition SDK sends the target feature combination to the data configuration platform via a single TR call. Upon receiving the target feature combination, the data configuration platform performs the operations described in Step 3 above, matching the target feature combination with the combined features of all scene instances in the full scene instance library. If the scene instance exists in the library, the scene recognition result for the old scene is returned; otherwise, the scene recognition result for the new scene is returned. Simultaneously, the new scene is persistently stored in the scene instance library and can also be pushed to the scene recognition SDK's local cache for caching. The specific implementation method is as follows:

[0176] The data configuration platform is also used for:

[0177] Receive the target feature combination sent by the scene recognition software development tool, and match the target feature combination with the feature combination of the scene instance library;

[0178] If the matching result is determined to be a match, the scene corresponding to the target feature combination is sent as an existing scene to the scene recognition software development tool as the scene recognition result;

[0179] If the matching result is determined to be a mismatch, the scene corresponding to the target feature combination is a non-existent scene, which is sent to the scene recognition software development tool as the scene recognition result.

[0180] Specifically, if the matching result is determined to be a match, the scene corresponding to the target feature combination is sent to the scene recognition software development tool as an existing scene as the scene recognition result; that is, if the scene exists in the above-mentioned scene instance library, the scene recognition result of the old scene is returned and sent to the scene recognition SDK.

[0181] If the matching result is determined to be a mismatch, the scene corresponding to the target feature combination is a non-existent scene, which is sent to the scene recognition software development tool as the scene recognition result; that is, it can be understood that the scene does not exist in the above-mentioned scene instance library, so the scene recognition result of the new scene is returned and sent to the scene recognition SDK.

[0182] The data processing system provided in one or more embodiments of this specification, after the scene recognition SDK initially determines a new scene based on its local cache, sends its target feature combination to the data configuration platform for secondary judgment, thus forming a dual insurance mechanism for the system's accuracy and reliability. First, this method effectively overcomes the inherent limitations of local caching, ensuring the accuracy of scene identification. Due to performance considerations, the scene recognition SDK's local cache cannot and should not store all scene instances. By submitting suspected new scenes to the data configuration platform, which possesses a complete scene library, for final judgment, misjudgments caused by scene recognition SDK cache update delays, data asynchrony, or local node cache anomalies can be effectively avoided. This ensures that each new scene identification is confirmed by a single trusted source (the data configuration platform), fundamentally maintaining the accuracy of the recognition results. Second, it achieves centralized management and dynamic evolution of recognition capabilities. The data configuration platform, as the repository of all scene rules, is the only platform qualified to perform the final judgment. All suspected new scenes from all scene recognition SDK nodes are gathered here, enabling the data configuration platform to continuously collect new situations encountered by edge nodes and transform them into new scene knowledge. This approach not only ensures a high degree of uniformity in recognition standards within a distributed environment but also enables the entire system to collectively learn and self-optimize. Over time, its recognition accuracy and coverage continuously improve without requiring manual upgrades to each scene recognition SDK node. Ultimately, this method achieves a balance between performance and reliability. Through a hierarchical strategy of "local rapid judgment - remote authoritative judgment," most older scenes are rapidly filtered at the local layer of the scene recognition SDK, ensuring high throughput and low latency. Only a small number of suspected new scenes trigger judgment from the data configuration platform. This approach trades a relatively small performance overhead for a high level of reliability, providing a solid data-driven decision-making foundation for stable operation in the production environment.

[0183] The data processing system provided in one or more embodiments of this specification introduces a data change monitoring module to asynchronously perceive changed data in the database, achieving non-intrusive data collection of business systems. The solution is standardized and can be used by any system with a database, fundamentally eliminating business risks arising from system modifications. After accurate scenario identification by the scenario change recognition module, comparison with the existing scenario database efficiently discovers new scenarios. The scenario change recognition module then provides unified alarm and maintenance, forming a complete closed loop from data perception to risk prevention. This data processing system implements a low-cost, low-risk, and highly stable solution for new scenario risk perception and prevention in production environments. Specifically, this data processing system accesses data through the data change monitoring module, asynchronously perceiving all business data without requiring any modifications to business systems, posing no intrusive risks, and enabling unified deployment across the consumer finance dimension, significantly improving the coverage and consistency of risk management. This data processing system, through the organic integration of data change monitoring module, scenario change identification module, and data configuration platform, can achieve real-time alarms for new scenarios in the production environment at low cost, without business risk, and with high stability. It can effectively prevent unknown risks brought about by new scenarios and can be promoted and deployed on a large scale within the consumer finance system without having to consider business risks. It realizes the ability to perceive production environment risks across the entire domain and process, providing a reliable guarantee for the stable operation of business.

[0184] Corresponding to the above method embodiments, this specification also provides embodiments of a data processing apparatus. Figure 4 A schematic diagram of the structure of a data processing apparatus according to one embodiment of this specification is shown. Figure 4 As shown, this device is applied to the scene change recognition module 402, wherein,

[0185] The scene change recognition module 402 is configured as follows:

[0186] The data change monitoring module receives a target data packet, wherein the target data packet is a data packet obtained by the data change monitoring module assembling the changed data according to the target data model when it detects a data change in the data table corresponding to the target data model carried in the data acquisition task.

[0187] Obtain the target scene model from the scene model set that matches the target data packet;

[0188] Based on the scene recognition rules of the target scene model, the corresponding target fields and their values ​​are obtained from the target data packet to form a target feature combination;

[0189] The target feature combination is matched with the feature combination of the scene instance set, and the scene recognition result is determined based on the matching result.

[0190] Optionally, the scene change recognition module 402 is further configured to:

[0191] Using the scene recognition software development tool, a target data table corresponding to the changed data in the target data packet is obtained, and a target scene model matching the target data table is obtained from the locally cached scene model set according to the target data table. The target data table is one or more data tables in the data table corresponding to the target data model, and the associated data table of the target scene model matches the target data table.

[0192] Optionally, the scene change recognition module 402 is further configured to:

[0193] If the matching result is determined to be a match, the scene recognition result is determined to be the scene corresponding to the target feature combination as an existing scene;

[0194] If the matching result is determined to be a mismatch, the scene recognition result is determined to be that the scene corresponding to the target feature combination does not exist.

[0195] or

[0196] The target feature combination is sent to the data configuration platform and matched with the feature combination in the scene instance library of the data configuration platform. The scene recognition result is determined based on the matching result returned by the data configuration platform.

[0197] Optionally, the scene change recognition module 402 is further configured to:

[0198] If the scenario corresponding to the target feature combination is determined to meet the preset circuit breaker conditions, a circuit breaker processing strategy for the scenario corresponding to the target feature combination is executed.

[0199] Specifically, the scene change recognition module 402 is the same as the scene change recognition modules 104 and 304 in the above embodiments, except that different reference numerals are used due to the different accompanying drawings.

[0200] It is understood that the modules mentioned above refer to computer programs or program segments used to perform one or more specific functions. Furthermore, the distinction between these modules does not imply that the actual program code must also be separate.

[0201] For ease of description, the above devices are described by dividing them into various modules or units based on their functions. Of course, when implementing one or more of these specifications, the functions of each module or unit can be implemented in the same or different software and / or hardware, or a module that performs the same function can be implemented by a combination of multiple sub-modules or sub-units, etc. The device embodiments described above are merely illustrative. For example, the division of units is only a logical functional division; in actual implementation, there may be other division methods. For example, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed.

[0202] The above is an illustrative scheme of a data processing apparatus according to this embodiment. It should be noted that the technical solution of this data processing apparatus and the technical solution of the data processing method described above belong to the same concept. For details not described in detail in the technical solution of the data processing apparatus, please refer to the description of the technical solution of the data processing method described above.

[0203] See Figure 5 , Figure 5 A structural block diagram of a computing device 500 according to one or more embodiments of this specification is shown. The components of the computing device 500 include, but are not limited to, a memory 510 and a processor 520. The processor 520 is connected to the memory 510 via a bus 530, and a database 550 is used to store data.

[0204] The computing device 500 also includes an access device 540, which enables the computing device 500 to communicate via one or more networks 560. Examples of these networks include Public Switched Telephone Network (PSTN), Local Area Network (LAN), Wide Area Network (WAN), Personal Area Network (PAN), or combinations of communication networks such as the Internet. The access device 540 may include one or more of any type of wired or wireless network interface (e.g., a network interface card (NIC)), such as an IEEE 802.11 Wireless Local Area Network (WLAN) wireless interface, a Wi-MAX (Worldwide Interoperability for Microwave Access) interface, an Ethernet interface, a Universal Serial Bus (USB) interface, a cellular network interface, a Bluetooth interface, a Near Field Communication (NFC) interface, and so on.

[0205] In one embodiment of this application, the aforementioned components of the computing device 500 and Figure 5 Other components, not shown, can also be connected to each other, for example, via a bus. It should be understood that... Figure 5 The block diagram of the computing device shown is for illustrative purposes only and is not intended to limit the scope of this application. Those skilled in the art can add or replace other components as needed.

[0206] The computing device 500 can be any type of stationary or mobile computing device, including mobile computers or mobile computing devices (e.g., tablet computers, personal digital assistants, laptop computers, notebook computers, netbooks, etc.), mobile phones (e.g., smartphones), wearable computing devices (e.g., smartwatches, smart glasses, etc.) or other types of mobile devices, or stationary computing devices such as desktop computers or personal computers (PCs). The computing device 500 can also be a mobile or stationary server.

[0207] The processor 520 is used to execute the following computer program / instructions, which, when executed by the processor, implement the steps of the above-described data processing method.

[0208] The above is an illustrative scheme of a computing device according to this embodiment. It should be noted that the technical solution of this computing device and the technical solution of the data processing method described above belong to the same concept. For details not described in detail in the technical solution of the computing device, please refer to the description of the technical solution of the data processing method described above.

[0209] An embodiment of this specification also provides a computer-readable storage medium storing a computer program / instructions that, when executed by a processor, implement the steps of the above-described data processing method.

[0210] The various embodiments in this specification are described in a progressive manner. Similar or identical parts between embodiments can be referred to mutually. Each embodiment focuses on describing the differences from other embodiments. In particular, the computer-readable storage medium embodiments are basically similar to the data processing method embodiments, so the description is relatively simple; relevant parts can be referred to the descriptions of the data processing method embodiments.

[0211] An embodiment of this specification also provides a computer program product, including a computer program / instructions that, when executed by a processor, implement the steps of the above-described data processing method.

[0212] The above is an illustrative scheme of a computer program product according to this embodiment. It should be noted that the technical solution of this computer program product and the technical solution of the data processing method described above belong to the same concept. For details not described in detail in the technical solution of the computer program product, please refer to the description of the technical solution of the data processing method described above.

[0213] The various embodiments in this specification are described in a progressive manner, and the same or similar parts between the embodiments can be referred to mutually. Each embodiment focuses on describing the differences from other embodiments. In particular, for the embodiments of [apparatus, device, system], since they are basically similar to the method embodiments, the description is relatively simple, and the relevant parts can be referred to the description of the method embodiments. The [apparatus, device, system] provided in the embodiments of this specification correspond to the methods, and therefore the [apparatus, device, system] also has similar beneficial technical effects as the corresponding methods. Since the beneficial technical effects of the methods have been described in detail above, the beneficial technical effects of the corresponding [apparatus, device, system] will not be repeated here.

[0214] The foregoing has described specific embodiments of this specification. Other embodiments are within the scope of the appended claims. In some cases, the actions or steps recited in the claims may be performed in a different order than that shown in the embodiments and may still achieve the desired result. Furthermore, the processes depicted in the drawings do not necessarily require a specific or sequential order to achieve the desired result. In some embodiments, multitasking and parallel processing are possible or may be advantageous.

[0215] In the 1990s, improvements to a technology could be clearly distinguished as either hardware improvements (e.g., improvements to the circuit structure of diodes, transistors, switches, etc.) or software improvements (improvements to methodology). However, with technological advancements, many methodological improvements today can be considered direct improvements to hardware circuit structures. Designers almost always obtain the corresponding hardware circuit structure by programming the improved methodology into the hardware circuit. Therefore, it cannot be said that a methodological improvement cannot be implemented using hardware physical modules. For example, a Programmable Logic Device (PLD) (such as a Field Programmable Gate Array (FPGA)) is such an integrated circuit whose logic function is determined by the user programming the device. Designers can program a digital system themselves to "integrate" it onto a PLD, without needing chip manufacturers to design and manufacture dedicated integrated circuit chips. Furthermore, nowadays, instead of manually manufacturing integrated circuit chips, this programming is mostly implemented using "logic compiler" software. Similar to the software compiler used in program development, the original code before compilation must also be written in a specific programming language, called a Hardware Description Language (HDL). There are many HDLs, such as ABEL (Advanced Boolean Expression Language), AHDL (Altera Hardware Description Language), Confluence, CUPL (Cornell University Programming Language), HDCal, JHDL (Java Hardware Description Language), Lava, Lola, MyHDL, PALASM, and RHDL (Ruby Hardware Description Language). Currently, the most commonly used are VHDL (Very-High-Speed ​​Integrated Circuit Hardware Description Language) and Verilog. Those skilled in the art should also understand that by simply performing some logic programming on the method flow using one of these hardware description languages ​​and programming it into an integrated circuit, the hardware circuit implementing the logical method flow can be easily obtained.

[0216] The controller can be implemented in any suitable manner. For example, it can take the form of a microprocessor or processor and a computer-readable medium storing computer-readable program code (e.g., software or firmware) executable by the (micro)processor, logic gates, switches, application-specific integrated circuits (ASICs), programmable logic controllers, and embedded microcontrollers. Examples of controllers include, but are not limited to, the following microcontrollers: ARC 625D, Atmel AT91SAM, Microchip PIC18F26K20, and Silicon Labs C8051F320. A memory controller can also be implemented as part of the control logic of the memory. Those skilled in the art will also recognize that, in addition to implementing the controller in purely computer-readable program code form, the same functionality can be achieved by logically programming the method steps to make the controller take the form of logic gates, switches, application-specific integrated circuits, programmable logic controllers, and embedded microcontrollers. Therefore, such a controller can be considered a hardware component, and the means included therein for implementing various functions can also be considered as structures within the hardware component. Alternatively, the means for implementing various functions can be considered as both software modules implementing the method and structures within the hardware component.

[0217] The systems, devices, modules, or units described in the above embodiments can be implemented by computer chips or entities, or by products with certain functions. A typical implementation device is a computer. Specifically, a computer can be, for example, a personal computer, laptop computer, cellular phone, camera phone, smartphone, personal digital assistant, media player, navigation device, email device, game console, tablet computer, wearable device, or any combination of these devices.

[0218] For ease of description, the above devices are described separately by function as various units. Of course, in implementing this application, the functions of each unit can be implemented in one or more software and / or hardware.

[0219] Those skilled in the art will understand that one or more embodiments of this specification can be provided as a method, system, or computer program product. Therefore, the invention can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, the invention can take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0220] This invention is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of the invention. It will be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, generate instructions for implementing the flowchart illustrations and / or block diagrams. Figure 1 One or more processes and / or boxes Figure 1 A device that provides the functions specified in one or more boxes.

[0221] These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to function in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means, which are implemented in a process Figure 1 One or more processes and / or boxes Figure 1 The function specified in one or more boxes.

[0222] These computer program instructions may also be loaded onto a computer or other programmable data processing equipment to cause a series of operational steps to be performed on the computer or other programmable equipment to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable equipment for implementing the process. Figure 1 One or more processes and / or boxes Figure 1 The steps of the function specified in one or more boxes.

[0223] In a typical configuration, a computing device includes one or more processors (CPU), input / output interfaces, network interfaces, and memory.

[0224] Memory may include non-persistent storage in computer-readable media, such as random access memory (RAM) and / or non-volatile memory, such as read-only memory (ROM) or flash RAM. Memory is an example of computer-readable media.

[0225] Computer-readable media includes both permanent and non-permanent, removable and non-removable media that can store information using any method or technology. Information can be computer-readable instructions, data structures, modules of programs, or other data. Examples of computer storage media 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 technologies, CD-ROM, digital character versatile optical disc (DVD) or other optical storage, magnetic tape, magnetic disk storage or other magnetic storage devices, or any other non-transferable medium that can be used to store information accessible by a computing device. As defined herein, computer-readable media does not include transient computer-readable media, such as modulated data signals and carrier waves.

[0226] This application can be described in the general context of computer-executable instructions, such as program modules, that are executed by a computer. Generally, program modules include routines, programs, objects, components, data structures, etc., that perform a specific task or implement a specific abstract data type. This application can also be practiced in distributed computing environments where tasks are performed by remote processing devices connected via a communication network. In distributed computing environments, program modules can reside in local and remote computer storage media, including storage devices.

[0227] The above description is merely an embodiment of this application and is not intended to limit this application. Various modifications and variations can be made to this application by those skilled in the art. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principle of this application should be included within the scope of the claims of this application.

Claims

1. A data processing method, characterized in that, Applied to the scene change recognition module, including: The data change monitoring module receives a target data packet, wherein the target data packet is a data packet obtained by the data change monitoring module assembling the changed data according to the target data model when it detects a data change in the data table corresponding to the target data model carried in the data acquisition task. Obtain the target scene model from the scene model set that matches the target data packet; Based on the scene recognition rules of the target scene model, the corresponding target fields and their values ​​are obtained from the target data packet to form a target feature combination; The target feature combination is matched with the feature combination of the scene instance set, and the scene recognition result is determined based on the matching result.

2. The data processing method according to claim 1, characterized in that, The target scene model obtained from the scene model set and matched with the target data packet includes: Using scene recognition software development tools, a target data table corresponding to the changed data in the target data packet is obtained, and a target scene model matching the target data table is obtained from the locally cached scene model set according to the target data table. The target data table is one or more data tables in the data table corresponding to the target data model, and the associated data table of the target scene model matches the target data table.

3. The data processing method according to claim 1, characterized in that, The step of determining the scene recognition result based on the matching result includes: If the matching result is determined to be a match, the scene recognition result is determined to be the scene corresponding to the target feature combination as an existing scene; If the matching result is determined to be a mismatch, the scene recognition result is determined to be that the scene corresponding to the target feature combination does not exist. or The target feature combination is sent to the data configuration platform and matched with the feature combination in the scene instance library of the data configuration platform. The scene recognition result is determined based on the matching result returned by the data configuration platform.

4. The data processing method according to claim 3, characterized in that, After determining that the scene recognition result is that the scene corresponding to the target feature combination is a non-existent scene, the method further includes: If the scenario corresponding to the target feature combination is determined to meet the preset circuit breaker conditions, a circuit breaker processing strategy for the scenario corresponding to the target feature combination is executed.

5. A data processing system, characterized in that, This includes a data change monitoring module and a scene change recognition module, among which... The data change monitoring module is used to receive data collection tasks, and when it detects data changes in the data table corresponding to the target data model carried in the data collection task, it acquires the changed data, assembles the changed data according to the target data model, generates a target data packet, and sends the target data packet to the scene change recognition module. The scene change recognition module is used to receive the target data packet, obtain a target scene model that matches the target data packet from the scene model set; obtain the corresponding target field and the field value of the target field from the target data packet according to the scene recognition rules of the target scene model to form a target feature combination; match the target feature combination with the feature combination of the scene instance set, and determine the scene recognition result according to the matching result.

6. The data processing system according to claim 5, characterized in that, The scene change recognition module is also used for: Using scene recognition software development tools, a target data table corresponding to the changed data in the target data packet is obtained, and a target scene model matching the target data table is obtained from the locally cached scene model set according to the target data table. The target data table is one or more data tables in the data table corresponding to the target data model, and the data table associated with the target scene model matches the target data table.

7. The data processing system according to claim 5, characterized in that, Also includes: Data configuration platform; The data configuration platform is used to respond to a scene model creation request and create a scene model based on the scene model attribute information, associated data tables, and scene recognition rules carried in the scene model creation request. Based on the scene model and the target database, create scene instances and store the scene instances in the scene instance library; The scene model and the scene instance are sent to the scene change recognition module or the scene recognition software development tool called by the scene change recognition module for local caching.

8. The data processing system according to claim 7, characterized in that, The scene change recognition module is also used for: In response to a data model creation request, a data model is created based on the data model attribute information, associated scene models, and associated data tables of the associated scene models carried in the data model creation request. Based on the data model, a data acquisition task is created and sent to the data change monitoring module. The target data model is one or more of the data models, and the data table corresponding to the target data model is one or more of the associated data tables of the scene model associated with the target data model.

9. The data processing system according to claim 5 or 8, characterized in that, The scene change recognition module is also used for: If the matching result is determined to be a match, the scene recognition result is determined to be the scene corresponding to the target feature combination as an existing scene; If the matching result is determined to be a mismatch, the scene recognition result is determined to be that the scene corresponding to the target feature combination does not exist. or The target feature combination is sent to the data configuration platform and matched with the feature combination in the scene instance library of the data configuration platform. The scene recognition result is determined based on the matching result returned by the data configuration platform.

10. The data processing system according to claim 9, characterized in that, The data configuration platform is also used for: The system receives the target feature combination sent by the scene recognition software development tool and matches the target feature combination with the feature combination of the scene instance library. If the matching result is determined to be a match, the scene corresponding to the target feature combination is sent as an existing scene to the scene recognition software development tool as the scene recognition result; If the matching result is determined to be a mismatch, the scene corresponding to the target feature combination is a non-existent scene, which is sent to the scene recognition software development tool as the scene recognition result.

11. The data processing system according to claim 10, characterized in that, The scene change recognition module is also used for: Receive the scene recognition results returned by the scene recognition software development tool; If the scene recognition result indicates that the scene corresponding to the target feature combination does not exist, a scene security prompt is sent.

12. The data processing system according to claim 10, characterized in that, The scene change recognition module is also used for: If the scenario corresponding to the target feature combination is determined to meet the preset circuit breaker conditions, a circuit breaker processing strategy for the scenario corresponding to the target feature combination is executed.

13. A data processing apparatus, characterized in that, It is applied to the scene change recognition module, where, The scene change recognition module is configured as follows: The data change monitoring module receives a target data packet, wherein the target data packet is a data packet obtained by the data change monitoring module assembling the changed data according to the target data model when it detects a data change in the data table corresponding to the target data model carried in the data acquisition task. Obtain the target scene model from the scene model set that matches the target data packet; Based on the scene recognition rules of the target scene model, the corresponding target fields and their values ​​are obtained from the target data packet to form a target feature combination; The target feature combination is matched with the feature combination of the scene instance set, and the scene recognition result is determined based on the matching result.

14. A computing device, characterized in that, include: Memory and processor; The memory is used to store computer programs / instructions, and the processor is used to execute the computer programs / instructions, which, when executed by the processor, implement the steps of the data processing method according to any one of claims 1 to 4.

15. A computer-readable storage medium, characterized in that, It stores a computer program / instruction that, when executed by a processor, implements the steps of the data processing method according to any one of claims 1 to 4.

16. A computer program product, characterized in that, Includes a computer program / instructions that, when executed by a processor, implement the steps of the data processing method according to any one of claims 1 to 4.

Images