An information processing method and device, electronic equipment and storage medium
By filtering, labeling, and evaluating target objects during the target identification process, scheduling tasks are automatically generated, solving the problems of complex and time-consuming target identification in existing technologies, and realizing efficient target object identification and automated evaluation of policy information.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- BEIJING DAJIA INTERNET INFORMATION TECH CO LTD
- Filing Date
- 2023-04-12
- Publication Date
- 2026-07-10
Smart Images
Figure CN116541591B_ABST
Abstract
Description
Technical Field
[0001] This disclosure relates to the Internet field, and more particularly to an information processing method, apparatus, electronic device, and storage medium. Background Technology
[0002] Different strategies are needed to identify target objects in different scenarios. Therefore, different identification strategies need to be developed for different scenarios. Furthermore, the developed identification strategies need to be evaluated to ensure that their identification effect meets expectations.
[0003] In existing technologies, the target identification process based on resource information is relatively complex and lacks an automated mechanism, resulting in low timeliness. Summary of the Invention
[0004] This disclosure provides an information processing method, apparatus, electronic device, and storage medium to at least solve the problems in related technologies where the target identification process based on resource information is complex, lacks an automated mechanism, and therefore has low timeliness. The technical solution of this disclosure is as follows:
[0005] According to a first aspect of the present disclosure, an information processing method is provided, comprising:
[0006] Based on the strategy information corresponding to the preset business scenario and the feature data of the sample objects, target objects are selected from the sample objects;
[0007] Obtain the tagging information of the target object; the tagging information is used to indicate the matching result of the target object and the preset business scenario;
[0008] Based on the marking information, the evaluation information for the target object is determined;
[0009] If the evaluation information meets the preset conditions, a scheduling task corresponding to the strategy information is created, and the scheduling task is executed in response to the scheduling instruction of the scheduling task; the scheduling task is used to filter objects to be processed according to the strategy information.
[0010] Optionally, before filtering target objects from sample objects based on strategy information corresponding to a preset business scenario and feature data of sample objects, the method further includes:
[0011] Obtain offline data of the sample object from a preset database table;
[0012] The offline data is aggregated to obtain an offline feature table, which serves as the feature data for the sample objects.
[0013] Optionally, before filtering target objects from sample objects based on strategy information corresponding to a preset business scenario and feature data of sample objects, the method further includes:
[0014] Obtain real-time data of the sample object;
[0015] The real-time data and the offline feature data are synchronized to obtain a real-time database table, which serves as the feature data of the sample object.
[0016] Optionally, when the feature data is a single item, the step of filtering target objects from sample objects based on strategy information corresponding to a preset business scenario and feature data of sample objects includes:
[0017] Convert the strategy information into processing instructions;
[0018] In the real-time database table, the processing instructions are executed on the feature data of the sample objects to determine the target object from the sample objects.
[0019] Optionally, when there are multiple feature data, the step of filtering target objects from sample objects based on strategy information corresponding to a preset business scenario and feature data of sample objects includes:
[0020] Create the offline task corresponding to the strategy information;
[0021] The offline task is performed in the offline feature table to determine the target object from the sample objects.
[0022] Optionally, after creating the scheduling task corresponding to the strategy information when the evaluation information meets preset conditions, the method further includes:
[0023] If the evaluation information meets the preset conditions, a scheduling task corresponding to the strategy information is created.
[0024] The feasibility of the scheduling task is verified, and the verification result is obtained.
[0025] If the verification result indicates that the scheduling task is feasible, the step of executing the scheduling task in response to the scheduling instruction for the scheduling task is performed.
[0026] Optionally, the scheduling instruction carries a task scheduling frequency; executing the scheduling task in response to the scheduling instruction includes:
[0027] In response to the scheduling instruction for the scheduled task, the scheduled task is executed according to the task scheduling frequency.
[0028] Optionally, after executing the scheduling task in response to the scheduling instruction for the scheduling task, the method further includes:
[0029] Based on the execution results of the scheduled tasks, a result database table is generated; the result database table includes at least one object to be disposed of.
[0030] The results database table is pushed to the business platform so that the business platform can perform preset processing operations on the objects to be processed.
[0031] According to a second aspect of the present disclosure, an information processing apparatus is provided, comprising:
[0032] The filtering unit is configured to filter target objects from the sample objects based on the strategy information corresponding to the preset business scenario and the feature data of the sample objects;
[0033] The tagging unit is configured to acquire tagging information of the target object; the tagging information is used to indicate the matching result between the target object and the preset business scenario;
[0034] The evaluation unit is configured to determine evaluation information for the target object based on the tagging information;
[0035] The scheduling unit is configured to create a scheduling task corresponding to the strategy information when the evaluation information meets preset conditions, and to execute the scheduling task in response to the scheduling instruction of the scheduling task; the scheduling task is used to filter objects to be processed according to the strategy information.
[0036] According to a third aspect of the present disclosure, an information processing electronic device is provided, comprising:
[0037] processor;
[0038] Memory used to store the processor's executable instructions;
[0039] The processor is configured to execute the instructions to implement the information processing method described in the first item above.
[0040] According to a fourth aspect of the present disclosure, a computer-readable storage medium is provided, wherein when instructions in the computer-readable storage medium are executed by a processor of the electronic device, the information processing electronic device is enabled to perform the information processing method described in the first item above.
[0041] According to a fifth aspect of the present disclosure, a computer program product is provided, comprising a computer program, characterized in that, when the computer program is executed by a processor, it implements the information processing method described in the first item above.
[0042] The technical solutions provided by the embodiments of this disclosure have at least the following beneficial effects:
[0043] Based on the strategy information corresponding to the preset business scenario and the feature data of the sample objects, target objects are selected from the sample objects; the marking information of the target objects is obtained; the marking information is used to indicate whether the target objects match the preset business scenario; the evaluation information of the target objects is determined based on the marking information; if the evaluation information meets the preset conditions, a scheduling task corresponding to the strategy information is created, and the scheduling task is executed in response to the scheduling instructions of the scheduling task; the scheduling task is used to filter the objects to be processed according to the strategy information.
[0044] In this way, target objects can be selected from sample objects based on the strategy information corresponding to the preset business scenario. Then, the execution of the strategy information can be evaluated through the evaluation information of the target objects. If the evaluation is passed, the scheduling task corresponding to the strategy information can be automatically generated to complete the strategy deployment. This automates the target object identification, simplifies the operation process, enables efficient evaluation and iteration of the strategy information corresponding to the preset business scenario, and improves the timeliness of target object identification.
[0045] It should be understood that the above general description and the following detailed description are exemplary and explanatory only, and are not intended to limit this disclosure. Attached Figure Description
[0046] The accompanying drawings, which are incorporated in and form part of this specification, illustrate embodiments consistent with this disclosure and, together with the description, serve to explain the principles of this disclosure, and are not intended to unduly limit this disclosure.
[0047] Figure 1 This is a schematic diagram illustrating an information processing method according to an exemplary embodiment.
[0048] Figure 2 This is an architectural diagram illustrating an information processing method according to an exemplary embodiment.
[0049] Figure 3 This is a flowchart illustrating an information processing method according to an exemplary embodiment.
[0050] Figure 4 This is a block diagram illustrating an information processing apparatus according to an exemplary embodiment.
[0051] Figure 5 A block diagram of an information processing electronic device is shown according to an exemplary embodiment.
[0052] Figure 6 This is a block diagram illustrating an apparatus for information processing according to an exemplary embodiment. Detailed Implementation
[0053] To enable those skilled in the art to better understand the technical solutions of this disclosure, the technical solutions in the embodiments of this disclosure will be clearly and completely described below with reference to the accompanying drawings.
[0054] It should be noted that the terms "first," "second," etc., used in the specification, claims, and accompanying drawings of this disclosure are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such data can be interchanged where appropriate so that the embodiments of this disclosure described herein can be implemented in orders other than those illustrated or described herein. The embodiments described in the following exemplary embodiments do not represent all embodiments consistent with this disclosure. Rather, they are merely examples of apparatuses and methods consistent with some aspects of this disclosure as detailed in the appended claims.
[0055] First, the implementation environment involved in the embodiments of this application will be described. The implementation environment includes a terminal and a server, and the terminal and the server are connected through a communication network.
[0056] The terminal interacts with the user, obtains policy information configured by the user for a preset business scenario, and sends the obtained policy information to the server. The server then filters target objects from the sample objects based on the policy information corresponding to the preset business scenario and the feature data of the sample objects.
[0057] Then, the server can obtain the tagging information of the target object. The tagging information is used to indicate the matching result between the target object and the preset business scenario. The tagging information can be obtained by manual matching. For example, the server can display or push the target object and let the user tag the target object to obtain the tagging information. Alternatively, it can be obtained by machine matching. The target object and the preset business scenario are matched and calculated through a preset model to determine the tagging information, and so on.
[0058] Furthermore, the server can determine the evaluation information for the target object based on the tagging information. If the evaluation information meets the preset conditions, a scheduling task corresponding to the policy information is created, and the scheduling task is launched. After the scheduling task is launched, users can schedule the launched scheduling task through the terminal and initiate scheduling commands. The server responds to the scheduling command and can execute the scheduling task, filtering the objects to be processed according to the policy information.
[0059] The aforementioned terminal can be various forms of terminal devices such as mobile phones, tablets, desktop computers, and portable laptops, and this application embodiment does not limit this to any particular type.
[0060] It is worth noting that the aforementioned servers can be independent physical servers, server clusters or distributed systems composed of multiple physical servers, or cloud servers that provide basic cloud computing services such as cloud services, cloud databases, cloud computing, cloud functions, cloud storage, network services, cloud communication, middleware services, domain name services, security services, content delivery networks (CDN), and big data and artificial intelligence platforms.
[0061] Cloud technology refers to a hosting technology that unifies hardware, software, and network resources within a wide area network (WAN) or local area network (LAN) to achieve data computation, storage, processing, and sharing. Based on the cloud computing business model, cloud technology encompasses network technology, information technology, integration technology, management platform technology, and application technology. It can form resource pools, providing flexible and convenient on-demand access. Cloud computing technology will become a crucial support. Backend services of technical network systems require substantial computing and storage resources, such as video websites, image websites, and many portal websites. With the rapid development and application of the internet industry, every item may have its own identification mark in the future, requiring transmission to backend systems for logical processing. Data at different levels will be processed separately, and various industry data will require robust system support, which can only be achieved through cloud computing.
[0062] In some embodiments, the server described above can also be implemented as a node in a blockchain system. Blockchain is a novel application model of computer technologies such as distributed data storage, peer-to-peer transmission, consensus mechanisms, and cryptographic algorithms. Essentially, a blockchain is a decentralized database, a chain of data blocks linked together using cryptographic methods. Each data block contains information about a batch of network transactions, used to verify the validity of the information (anti-counterfeiting) and generate the next block. A blockchain can include a blockchain underlying platform, a platform product service layer, and an application service layer.
[0063] In light of the above implementation environment, the information processing method provided in this application will be described below. Figure 1 This is a flowchart illustrating an information processing method according to an exemplary embodiment, applied to a server, the information processing method including the following steps.
[0064] In step S11, the target object is selected from the sample objects based on the strategy information corresponding to the preset business scenario and the feature data of the sample objects.
[0065] In some scenarios, it is often necessary to identify some target objects that meet certain conditions from multiple sample objects for further processing. Different identification strategies can be formulated for different business scenarios to identify different target objects. Furthermore, it is necessary to evaluate the formulated identification strategies to ensure that their identification effect meets expectations.
[0066] In this application, the sample object refers to the object that needs to be screened and identified. Each sample object can represent a product or merchant, or correspond to an account or terminal, etc., without any specific limitation. The feature data of the sample object can correspond to a variety of different feature dimensions. For example, if the sample object is a product, then the feature dimensions can include its merchant, category, purchase history, and orders, etc., without any specific limitation.
[0067] Strategy information refers to the strategies used to filter and identify sample objects. Different strategies can be formulated and different strategy information can be obtained in different business scenarios. Based on the obtained strategy information, the server can execute subsequent steps to filter sample objects. For example, continuing the previous example, the strategy information may include a list of merchants and a list of categories. Then, among all the products to be processed, only those whose merchants and categories are included in the merchant and category lists will be considered as target objects. The filtering rules included in the strategy information can correspond to all feature dimensions of the sample objects, or only one or some feature dimensions; there is no specific limitation.
[0068] In one implementation, the feature data of the sample objects comes from the offline data of the sample objects that have been acquired and stored in advance. Therefore, before filtering target objects from the sample objects according to the strategy information corresponding to the preset business scenario and the feature data of the sample objects, the following steps can be taken: obtain the offline data of the sample objects from the preset database table; perform aggregation processing on the offline data to obtain an offline feature table, which serves as the feature data of the sample objects.
[0069] Retrieving offline data of sample objects from a pre-defined database table can be achieved by using a data warehousing tool. For example, Hive can be used, as it maps structured data files to a database table and provides SQL (Structured Query Language) query functionality. Alternatively, Spark or Flink tools can also be used, among others. Data warehousing tools enable rapid and visual storage and maintenance of massive amounts of offline data, allowing servers to easily query and access the offline data of sample objects, thereby improving task execution speed and efficiency.
[0070] Offline data refers to the unprocessed historical business data of the sample objects. Hive can map this data to database tables, that is, pre-set database tables. Through the aggregation processing of offline data, the feature data of the sample objects can be extracted from it. This feature data can also be stored in the database table, that is, the offline feature table.
[0071] In other words, offline data of sample objects can be stored in a preset database table in advance. Then, the server can call the data storage tool to query the preset database table through SQL statements to obtain the offline data of the sample objects, and further aggregate the offline data to obtain an offline feature table, which serves as the feature data of the sample objects.
[0072] In one implementation, the feature data of the sample objects comes from the near-line real-time data of the sample objects. Therefore, before filtering target objects from the sample objects according to the strategy information corresponding to the preset business scenario and the feature data of the sample objects, the following steps can be taken: obtain the real-time data of the sample objects; synchronize the real-time data with the offline feature data to obtain a real-time database table, which serves as the feature data of the sample objects.
[0073] Obtaining real-time data of sample objects can be achieved by calling a message queue, such as Kafka. Kafka is a container that stores messages during transmission. Utilizing message queues, it acts as an intermediary between message producers and consumers, offering high scalability and high throughput. The main purpose of a message queue is to provide routing and guarantee message delivery. If the receiver is unavailable when a message is sent, the message queue will retain the message until it can be successfully delivered. Of course, message queues also have a time limit for storing messages.
[0074] Alternatively, message middleware can also use ActiveMQ (Active Message Queue), RocketMQ (Rocket Message Queue), or other message service interface protocols, etc., without any specific restrictions.
[0075] The real-time database table can be an Elasticsearch table. Elasticsearch is a distributed, highly scalable, and real-time search and data analysis engine with the ability to search, analyze, and explore data. Alternatively, other database tables can also be used as real-time database tables; there are no specific restrictions.
[0076] In this way, the server can obtain real-time data of sample objects in an orderly manner under high concurrency, and synchronize the real-time data with offline feature data to obtain a real-time database table, which serves as the feature data of the sample objects.
[0077] In this application, the strategy information includes filtering rules for at least one feature dimension of the sample object. The feature dimension corresponding to the strategy information can be set by the user in advance through a configuration file. For example, each product must have its own merchant and category. Therefore, the corresponding filtering rules can be configured according to its own merchant and category.
[0078] In another implementation, obtaining the feature data and strategy information of the sample object includes: obtaining the feature data of the sample object; and in response to a configuration operation on a preset visualization interface, determining at least one target feature dimension and a filtering rule for the target feature dimension as strategy information.
[0079] In other words, after acquiring the feature data of the sample objects, the feature data is visualized. Users can then configure a preset visualization interface to determine the parts of the feature data to be filtered and configure the corresponding filtering conditions. The server generates policy information based on the user's configuration. This makes the configuration of policy information more flexible, applicable to various scenarios, and facilitates the updating of policy information.
[0080] As mentioned above, in this step, the server can filter the sample objects based on their feature data according to the obtained policy information, and determine the target object from the sample objects. In other words, the feature data of the target object in each feature dimension all meet the filtering rules for that feature dimension included in the policy information.
[0081] Before filtering objects, the strategy information needs to be converted into a processing language that the server can recognize and execute. For example, SQL statements can be executed in both Hive and Elasticsearch tables. Therefore, the filtering rules included in the strategy information can be converted into corresponding SQL statements. The server can then execute these SQL statements to determine the target objects from the sample objects, making the task more efficient and accurate compared to manually writing processing statements. Different conversion methods can be used for different types of strategy information.
[0082] In one implementation, when there is only one feature data item, the target object is selected from the sample objects based on the strategy information corresponding to the preset business scenario and the feature data of the sample objects. This includes: converting the strategy information into processing instructions; executing processing instructions on the feature data of the sample objects in the real-time database table to determine the target object from the sample objects.
[0083] In other words, if the feature data is a single item, then the strategy information only contains filtering rules for a single-dimensional feature. For example, it might only include filtering rules for any one dimension such as merchant, product, category, or order. In this case, the amount of data processed for filtering is relatively small, and the execution process of the processing instructions is also relatively simple. Therefore, the processing instructions can be executed directly in the real-time database table to determine the target object from the sample objects. This can improve the real-time performance and execution efficiency of the task. For example, if the real-time database table is an Elasticsearch table, then the processing instructions can be SQL statements.
[0084] In one implementation, when there are multiple feature data, the target object is selected from the sample objects based on the strategy information corresponding to the preset business scenario and the feature data of the sample objects. This includes: creating an offline task corresponding to the strategy information; executing the offline task in the offline feature table; and determining the target object from the sample objects.
[0085] In other words, if the feature data consists of multiple elements, then the strategy information contains filtering rules for features across multiple dimensions. In this case, the amount of data to be processed for filtering is large, and the execution of processing instructions is also quite complex. If the processing instructions are executed directly in the real-time database table, it may cause the real-time database to become sluggish or malfunction. Therefore, the target object can be determined from the sample objects by creating an offline task.
[0086] In this way, the execution of the target task can be synchronous or asynchronous, without the need for real-time queries and feedback in the real-time database table. If there are multiple offline tasks, each offline task is independent of the others, and a task queue can be used to reduce concurrency and improve system stability.
[0087] In step S12, the tagging information of the target object is obtained; the tagging information is used to indicate the matching result of the target object and the preset business scenario.
[0088] As mentioned above, the target object is a sample object that meets the conditions and is selected based on the strategy information. The strategy information corresponds to the preset business scenario. In other words, ideally, the feature data of the target object should meet the preset business scenario, but the actual selection effect still needs to be further evaluated based on the labeling information.
[0089] In this step, after identifying the target object, we can further obtain its tagging information. This tagging information indicates the matching result between the target object and the preset business scenario. For example, the tagging information can be "yes" or "no," indicating whether the target object matches or does not match the preset business scenario, respectively. Alternatively, the tagging information can include the degree of matching between the target object and the preset business scenario, or it can include the degree of matching between the target object and the preset business scenario in each feature dimension, and so on. In this way, subsequent steps can evaluate the selected target objects based on the tagging information.
[0090] The tagging information can be obtained through manual matching, such as displaying or pushing target objects and having users tag the target objects to obtain tagging information. Alternatively, it can be obtained through machine matching, which uses a preset model to match and calculate the target objects with preset business scenarios to determine the tagging information, and so on. There are no specific limitations.
[0091] In step S13, the evaluation information for the target object is determined based on the marking information.
[0092] In this step, based on the tagging information, the execution accuracy of the strategy information can be analyzed to obtain evaluation information, which facilitates subsequent strategy deployment or iterative debugging. The evaluation information can indicate the execution accuracy of the strategy information and may include the precision and recall rates of the target objects, as well as other metrics for evaluating the effectiveness or execution accuracy of the strategy; specific details are not limited.
[0093] Specifically, accuracy is the ratio of the predicted accuracy to the recalled quantity. For example, if 50 target objects are found, but the tagging information indicates that only 40 of the found target objects are actually marked as matching the preset business scenario, then the accuracy is 40 / 50, or 80%. Accuracy can be used to measure the accuracy of the found portion.
[0094] Recall is the ratio of the exact quantity to the objectively correct quantity in the recall. For example, if there are a total of 60 sample objects and only 50 target objects are found, the recall rate is 50 / 60, or 83.3%. It is used to measure the coverage of the found part to the actual true part.
[0095] In step S14, if the evaluation information meets the preset conditions, a scheduling task corresponding to the strategy information is created, and the scheduling task is executed in response to the scheduling instruction on the scheduling task; the scheduling task is used to filter objects to be processed according to the strategy information.
[0096] In this step, if the evaluation information meets the preset conditions, it can be assumed that the selected target objects meet the business requirements of the preset business scenario. Then, offline large-scale data processing capabilities can be used to automatically create scheduling tasks corresponding to the strategy information, facilitating subsequent execution of the strategy information. The preset conditions can refer to both the accuracy and recall rates of the strategy information being higher than corresponding preset thresholds, or simply one of the accuracy and recall rates being higher than the corresponding preset threshold. These conditions can be set according to the actual business scenario and are not specifically limited.
[0097] Furthermore, a one-click entry point for deploying strategies can be provided. Through this entry point, users can schedule deployed strategies, send corresponding scheduling commands, and the server can respond to the scheduling commands to execute the corresponding scheduling tasks and apply them to specific business scenarios.
[0098] In one implementation, when the evaluation information does not meet preset conditions, the method further includes: responding to an update instruction for the strategy information, iteratively updating the strategy information, and returning a step of filtering target objects from sample objects based on the strategy information and feature data. The update instruction for the strategy information can be issued by the user or automatically generated by the server based on the analysis of the target objects and tagging information; the specific method is not limited.
[0099] It is understandable that if the evaluation information does not meet the preset conditions, then the target objects selected based on the current strategy information do not fully meet the preset business scenario. It can be considered that the formulation of the strategy information is unreasonable. Therefore, the strategy information can be adjusted and recalculated, and the target objects can be screened and evaluated to achieve iterative adjustment of the strategy information until the formulated strategy information can meet the business needs of the preset business scenario.
[0100] In one implementation, after creating a scheduling task corresponding to the strategy information when the evaluation information meets preset conditions, the method further includes: creating a scheduling task corresponding to the strategy information when the evaluation information meets preset conditions; performing a feasibility check on the scheduling task and obtaining a check result; and, if the check result indicates that the scheduling task is feasible, executing a scheduling task in response to a scheduling instruction on the scheduling task.
[0101] In other words, after generating the scheduling task with strategy information, the feasibility of the scheduling task can be verified. Only the feasible scheduling task can be used for subsequent task scheduling, so that each scheduling task that the user can schedule has a high feasibility. This can improve the success rate of scheduling task execution and reduce the occurrence of scheduling task execution failure.
[0102] In one implementation, the scheduling instruction carries a task scheduling frequency; in response to the scheduling instruction for the scheduled task, executing the scheduled task includes: in response to the scheduling instruction for the scheduled task, executing the scheduled task according to the task scheduling frequency.
[0103] In other words, the scheduling task supports specifying a periodic scheduling frequency and executing it periodically, making the execution of the scheduling task more flexible and applicable to various business scenarios. The task scheduling frequency can be weekly, daily, minute-by-minute, or second-by-second; there are no specific limitations.
[0104] In one implementation, after executing the scheduling task in response to the scheduling instruction for the scheduling task, when the evaluation information meets the preset conditions, the method further includes: generating a result database table based on the execution result of the scheduling task; the result database table includes at least one object to be disposed of; and pushing the result database table to the business platform so that the business platform can perform preset disposal operations on the object to be disposed of.
[0105] In other words, after the scheduled task is completed, the execution result can be automatically pushed to the business platform for consumption. The business platform then processes the objects to be processed according to the pre-configured handling template. During the process of pushing the result database table to the business platform, different delivery methods can be selected by calling a message middleware, such as internal message notification, manual accountability confirmation, automatic system handling, or delivery to the merchant's end. The message middleware can be Kafka, ActiveMQ, or RocketMQ, etc., with no specific limitation. This further expands the system's functionality.
[0106] In one implementation, after executing the scheduling task in response to the scheduling instruction, the method further includes: generating an alarm message and pushing the alarm message to the business platform if the scheduling task fails. In this way, if the scheduling task fails, an alarm mechanism can be set up to promptly push alarm information to the business platform, enabling users to promptly identify and fix the problem, thus maintaining system operation.
[0107] As can be seen from the above, the technical solution provided by the embodiments of this disclosure can filter target objects from sample objects based on strategy information corresponding to a preset business scenario. Then, through the evaluation information of the target objects, the execution of the strategy information can be evaluated. If the evaluation is passed, the scheduling task corresponding to the strategy information can be automatically generated to complete the strategy deployment. This achieves the automation of target object identification, simplifies the operation process, realizes efficient evaluation and iteration of strategy information corresponding to the preset business scenario, and improves the timeliness of target object identification.
[0108] like Figure 2The diagram shown is an architecture diagram of an information processing method according to an embodiment of this application. Taking the two feature dimensions of product and merchant as an example, this information processing method is executed by a strategy platform and includes the following modules:
[0109] ① Strategy Data Acquisition: The feature data of the sample objects can come from offline business data Hive tables. This data is processed and aggregated into Hive wide tables corresponding to each feature dimension (offline feature tables). Then, using Hive2es data synchronization technology, it is stored in the Elasticsearch tables (ES tables) for each feature dimension via the feature write interface (real-time data tables). Alternatively, it can come from near-line real-time data, utilizing Kafka and accessing the ES tables via the feature write interface. The ES tables also provide a feature query interface, supporting real-time queries of a single dimension, which can improve the efficiency of the calculation.
[0110] ② Strategy Rule Configuration: The management terminal of the strategy platform displays the feature dimensions of sample objects through a visual interface. It allows operations personnel to select different feature dimensions for combination, configure different filtering rules, and obtain strategy information. Furthermore, the strategy information can be reconfigured to achieve strategy iteration. One scenario can correspond to multiple strategy information entries, one strategy information entry can contain multiple filtering rules, and one filtering rule can contain multiple features. In addition, the strategy platform management terminal also provides users with access points for strategy calculation and execution, calculation effect evaluation, and strategy log analysis.
[0111] ③ Strategy Calculation and Performance Evaluation: Structured strategy information is converted into a processing language that the system can recognize and execute, such as Elasticsearch (ES) SQL queries or Hive SQL queries. The system automatically identifies the feature dimensions covered by the strategy information. If it contains only a single-dimensional feature, real-time data query calculation is performed through an ES table. If it contains multiple-dimensional features, offline tasks are created through the Internal Development Platform (IDP), and strategy query calculation is performed using a task queue approach. After identifying the target object, it can be fed back to the user, allowing the user to quickly mark the accuracy of the calculation results. After marking, the precision and recall of the strategy can be automatically calculated as evaluation information to determine the effectiveness of the strategy.
[0112] ④ Strategy Scheduling and Execution: After evaluating the effectiveness of the strategy, indicating its feasibility for production use, IDP can be used to create executable scheduling tasks corresponding to the strategy information, and the availability of the execution statements for these tasks can be verified. Furthermore, the management platform provides a one-click entry point for strategy deployment. After deployment, the strategy caller can use the execution end of the strategy platform to start the strategy execution engine, synchronize the strategy information configuration, and query feature data in the Elasticsearch table. The execution end also includes modules such as an audit center, a judgment center, and an identification center, used to call the scheduling tasks corresponding to the strategy and to deliver the strategy results. The scheduling tasks can support periodic scheduling frequencies, such as weekly or daily execution. If a task fails, an alarm mechanism can be set to detect the problem promptly. Based on the execution results of the scheduling tasks, Hive result tables (i.e., result database tables) corresponding to each feature dimension can be generated for subsequent strategy result delivery.
[0113] ⑤ Strategy Result Delivery: The Hive result tables corresponding to each feature dimension can be transformed into message streams through Hive2Kafka technology, and pushed to the identification center of the business platform for consumption via Kafka. The identification center selects different delivery methods according to the configured handling template, such as internal message notification, push to the accountability center for manual accountability confirmation, push to the handling center for control and handling, and delivery to the merchant, and automatically pushes the strategy results downstream.
[0114] In one implementation, such as Figure 3 The diagram shown is a flowchart illustrating an information processing method according to an embodiment of this application. First, strategies can be configured according to different scenarios, creating strategy information corresponding to each scenario. After obtaining the strategy information, the server can perform calculations, experiments, and deployment of the strategy information.
[0115] During the calculation process, the strategy information is parsed. If the strategy information corresponds to single-dimensional feature data, a corresponding JSON script is generated by parsing and assembling the strategy information. By executing the JSON script, an Elasticsearch wide table can be queried, which is the real-time data table storing the feature data of sample objects. Then, the number of strategy hit results and sample results, i.e., the number of target objects and their feature data, are returned in real time. If the strategy information corresponds to multi-dimensional feature data, an corresponding SQL statement is generated by parsing and assembling the strategy information. The SQL statement is used to query an offline Hive table, which is the offline feature table of the data to be processed. Then, the SQL statements corresponding to the strategy information are combined into an offline task and inserted into the task table on the server side. The server can periodically perform task checks, query the number of strategy hit results and sample results, i.e., the number of target objects and their feature data, and notify the download.
[0116] During the experiment, if the policy information corresponds to an offline scenario and does not need to be processed in the real-time data table, then the offline task can be inserted into the server's task table. The server can periodically execute and submit the offline task. IDP creates and executes the offline task, and then the append function can be called to write the task execution result to the end of the Hive result table. The server can periodically check the task table to query whether the task has been executed successfully. If successful, it notifies the Hive result table to be updated. If unsuccessful, it notifies the user of the failure and checks whether the policy is already online. If it is online, it alerts the user and administrator.
[0117] During the deployment process, if the strategy information corresponds to an offline scenario and does not require processing in the real-time data table, it is necessary to determine whether the strategy information has undergone experimental testing. If not, the experimental process must be executed first. If it has been tested, the server can periodically check the task table, filter out the scheduling tasks corresponding to strategies that are offline and already deployed, and execute the corresponding SQL for the selected scheduling tasks. The IDP creates and publishes the scheduling task. The tasks in the IDP can be executed periodically on a daily basis, calling the overwrite function to write to the Hive result table, which is the result database corresponding to the scheduling task. The Hive result table can include data such as strategy identifier, rule identifier, product identifier, and rule feature information. Then, it is written to Kafka through Hive2Kafka technology and sent to the recognition center with message notification. At the recognition center, the Hive result table can be analyzed to determine the recognition results, and then the recognition results are consumed and stored, with the recognition results stored in a MySQL database. Furthermore, based on the rule metadata and policy publisher information provided by the policy center, the system periodically queries the MySQL database during task scheduling to obtain identification data, and then aggregates and assembles the data as the policy execution result. The rule metadata corresponds to the filtering rules for feature data of a certain feature dimension contained in the policy information.
[0118] As can be seen from the above, the technical solution provided by the embodiments of this disclosure can filter target objects from sample objects based on strategy information corresponding to a preset business scenario. Then, through the evaluation information of the target objects, the execution of the strategy information can be evaluated. If the evaluation is passed, the scheduling task corresponding to the strategy information can be automatically generated to complete the strategy deployment. This achieves the automation of target object identification, simplifies the operation process, realizes efficient evaluation and iteration of strategy information corresponding to the preset business scenario, and improves the timeliness of target object identification.
[0119] Figure 4 This is a block diagram of an information processing apparatus according to an exemplary embodiment, applied to a server, the apparatus comprising:
[0120] The filtering unit 201 is configured to filter target objects from the sample objects based on the strategy information corresponding to the preset business scenario and the feature data of the sample objects.
[0121] The tagging unit 202 is configured to acquire tagging information of the target object; the tagging information is used to indicate the matching result of the target object and the preset business scenario;
[0122] Evaluation unit 203 is configured to perform evaluation of the target object based on the tagging information;
[0123] The scheduling unit 204 is configured to create a scheduling task corresponding to the strategy information when the evaluation information meets preset conditions, and to execute the scheduling task in response to the scheduling instruction of the scheduling task; the scheduling task is used to filter objects to be processed according to the strategy information.
[0124] As can be seen from the above, the technical solution provided by the embodiments of this disclosure can filter target objects from sample objects based on strategy information corresponding to a preset business scenario. Then, through the evaluation information of the target objects, the execution of the strategy information can be evaluated. If the evaluation is passed, the scheduling task corresponding to the strategy information can be automatically generated to complete the strategy deployment. This achieves the automation of target object identification, simplifies the operation process, realizes efficient evaluation and iteration of strategy information corresponding to the preset business scenario, and improves the timeliness of target object identification.
[0125] Regarding the apparatus in the above embodiments, the specific manner in which each module performs its operation has been described in detail in the embodiments related to the method, and will not be elaborated upon here.
[0126] Figure 5 This is a block diagram illustrating an information processing electronic device according to an exemplary embodiment.
[0127] In an exemplary embodiment, a computer-readable storage medium including instructions is also provided, such as a memory including instructions, which can be executed by a processor of an electronic device to perform the above-described method. Optionally, the computer-readable storage medium may be a ROM, random access memory (RAM), CD-ROM, magnetic tape, floppy disk, and optical data storage device, etc.
[0128] In an exemplary embodiment, a computer program product is also provided, which, when run on a computer, enables the computer to perform the above-described information processing method.
[0129] As can be seen from the above, the technical solution provided by the embodiments of this disclosure can filter target objects from sample objects based on strategy information corresponding to a preset business scenario. Then, through the evaluation information of the target objects, the execution of the strategy information can be evaluated. If the evaluation is passed, the scheduling task corresponding to the strategy information can be automatically generated to complete the strategy deployment. This achieves the automation of target object identification, simplifies the operation process, realizes efficient evaluation and iteration of strategy information corresponding to the preset business scenario, and improves the timeliness of target object identification.
[0130] Figure 6 This is a block diagram illustrating an apparatus 800 for information processing according to an exemplary embodiment.
[0131] For example, device 800 can be a mobile phone, computer, digital broadcasting electronic device, messaging device, game console, tablet device, medical device, fitness equipment, personal digital assistant, etc.
[0132] Reference Figure 6 The device 800 may include one or more of the following components: a processing component 802, a memory 804, a power component 806, a multimedia component 808, an audio component 810, an input / output (I / O) interface 812, a sensor component 814, and a communication component 816.
[0133] Processing component 802 typically controls the overall operation of device 800, such as operations associated with display, telephone calls, data communication, camera operation, and recording. Processing component 802 may include one or more processors 820 to execute instructions to perform all or part of the steps of the methods described above. Furthermore, processing component 802 may include one or more modules to facilitate interaction between processing component 802 and other components. For example, processing component 802 may include a multimedia module to facilitate interaction between multimedia component 808 and processing component 802.
[0134] Memory 804 is configured to store various types of data to support the operation of device 800. Examples of this data include instructions for any application or method operating on device 800, contact data, phonebook data, messages, pictures, videos, etc. Memory 804 can be implemented by any type of volatile or non-volatile storage device or a combination thereof, such as static random access memory (SRAM), electrically erasable programmable read-only memory (EEPROM), erasable programmable read-only memory (EPROM), programmable read-only memory (PROM), read-only memory (ROM), magnetic storage, flash memory, magnetic disk, or optical disk.
[0135] Power supply component 807 provides power to various components of device 800. Power supply component 807 may include a power management system, one or more power supplies, and other components associated with generating, managing, and distributing power to device 800.
[0136] Multimedia component 808 includes a screen that provides an output interface between the device 800 and the user. In some embodiments, the screen may include a liquid crystal display (LCD) and a touch panel (TP). If the screen includes a touch panel, the screen may be implemented as a touchscreen to receive input signals from the user. The touch panel includes one or more touch sensors to sense touches, swipes, and gestures on the touch panel. The touch sensors may sense not only the boundaries of the touch or swipe action but also the duration and pressure associated with the touch or swipe operation. In some embodiments, multimedia component 808 includes a front-facing camera and / or a rear-facing camera. When the device 800 is in an operating mode, such as a shooting mode or a video mode, the front-facing camera and / or the rear-facing camera may receive external multimedia data. Each front-facing camera and rear-facing camera may be a fixed optical lens system or have focal length and optical zoom capabilities.
[0137] Audio component 810 is configured to output and / or input audio signals. For example, audio component 810 includes a microphone (MIC) configured to receive external audio signals when device 800 is in an operating mode, such as call mode, recording mode, and voice recognition mode. The received audio signals may be further stored in memory 404 or transmitted via communication component 816. In some embodiments, audio component 810 also includes a speaker for outputting audio signals.
[0138] I / O interface 812 provides an interface between processing component 802 and peripheral interface modules, such as keyboards, click wheels, buttons, etc. These buttons may include, but are not limited to, home buttons, volume buttons, power buttons, and lock buttons.
[0139] Sensor assembly 814 includes one or more sensors for providing status assessments of various aspects of device 800. For example, sensor assembly 814 may detect the on / off state of device 800, the relative positioning of components such as the display and keypad of device 800, changes in the position of device 800 or a component of device 800, the presence or absence of user contact with device 800, the orientation or acceleration / deceleration of device 800, and temperature changes of device 800. Sensor assembly 814 may include a proximity sensor configured to detect the presence of nearby objects without any physical contact. Sensor assembly 814 may also include a light sensor, such as a CMOS or CCD image sensor, for use in imaging applications. In some embodiments, sensor assembly 814 may also include an accelerometer, a gyroscope, a magnetometer, a pressure sensor, or a temperature sensor.
[0140] Communication component 816 is configured to facilitate wired or wireless communication between device 800 and other devices. Device 800 can access wireless networks based on communication standards, such as WiFi, carrier networks (such as 2G, 3G, 4G, or 5G), or combinations thereof. In one exemplary embodiment, communication component 816 receives broadcast signals or broadcast-related information from an external broadcast management system via a broadcast channel. In one exemplary embodiment, communication component 816 also includes a near-field communication (NFC) module to facilitate short-range communication. For example, the NFC module may be implemented based on radio frequency identification (RFID) technology, Infrared Data Association (IrDA) technology, ultra-wideband (UWB) technology, Bluetooth (BT) technology, and other technologies.
[0141] In an exemplary embodiment, the apparatus 800 may be implemented by one or more application-specific integrated circuits (ASICs), digital signal processors (DSPs), digital signal processing devices (DSPDs), programmable logic devices (PLDs), field-programmable gate arrays (FPGAs), controllers, microcontrollers, microprocessors, or other electronic components to perform the methods described in the first and second aspects.
[0142] In an exemplary embodiment, a non-transitory computer-readable storage medium including instructions is also provided, such as a memory 804 including instructions, which can be executed by a processor 820 of the device 800 to perform the above-described method. Optionally, for example, the storage medium may be a non-transitory computer-readable storage medium, such as a ROM, random access memory (RAM), CD-ROM, magnetic tape, floppy disk, and optical data storage device.
[0143] In an exemplary embodiment, a computer program product containing instructions is also provided, which, when run on a computer, causes the computer to perform the information processing method described in the above embodiments.
[0144] As can be seen from the above, the technical solution provided by the embodiments of this disclosure can filter target objects from sample objects based on strategy information corresponding to a preset business scenario. Then, through the evaluation information of the target objects, the execution of the strategy information can be evaluated. If the evaluation is passed, the scheduling task corresponding to the strategy information can be automatically generated to complete the strategy deployment. This achieves the automation of target object identification, simplifies the operation process, realizes efficient evaluation and iteration of strategy information corresponding to the preset business scenario, and improves the timeliness of target object identification.
[0145] Other embodiments of this disclosure will readily occur to those skilled in the art upon consideration of the specification and practice of the invention disclosed herein. This application is intended to cover any variations, uses, or adaptations of this disclosure that follow the general principles of this disclosure and include common knowledge or customary techniques in the art not disclosed herein. The specification and examples are to be considered exemplary only, and the true scope and spirit of this disclosure are indicated by the following claims.
[0146] It should be understood that this disclosure is not limited to the precise structures described above and shown in the accompanying drawings, and various modifications and changes can be made without departing from its scope. The scope of this disclosure is limited only by the appended claims.
Claims
1. An information processing method, characterized in that, include: Based on the strategy information corresponding to the preset business scenario and the feature data of the sample objects, target objects are selected from the sample objects; Obtain the tagging information of the target object; the tagging information is used to indicate the matching result of the target object and the preset business scenario; Based on the marking information, the evaluation information for the target object is determined; If the evaluation information meets the preset conditions, a scheduling task corresponding to the strategy information is created, and the scheduling task is executed in response to the scheduling instruction of the scheduling task; the scheduling task is used to filter objects to be processed according to the strategy information.
2. The information processing method according to claim 1, characterized in that, Before filtering target objects from the sample objects based on strategy information corresponding to the preset business scenario and feature data of the sample objects, the process further includes: Obtain offline data of the sample object from a preset database table; The offline data is aggregated to obtain an offline feature table, which serves as the feature data for the sample objects.
3. The information processing method according to claim 2, characterized in that, Before filtering target objects from the sample objects based on strategy information corresponding to the preset business scenario and feature data of the sample objects, the process further includes: Obtain real-time data of the sample object; The real-time data and the offline feature data are synchronized to obtain a real-time database table, which serves as the feature data of the sample object.
4. The information processing method according to claim 3, characterized in that, When the feature data is a single item, the step of filtering target objects from the sample objects based on the strategy information corresponding to the preset business scenario and the feature data of the sample objects includes: Convert the strategy information into processing instructions; In the real-time database table, the processing instructions are executed on the feature data of the sample objects to determine the target object from the sample objects.
5. The information processing method according to any one of claims 2 or 3, characterized in that, When there are multiple feature data, the step of filtering target objects from the sample objects based on the strategy information corresponding to the preset business scenario and the feature data of the sample objects includes: Create an offline task corresponding to the strategy information; The offline task is performed in the offline feature table to determine the target object from the sample objects.
6. The information processing method according to claim 1, characterized in that, After creating the scheduling task corresponding to the strategy information when the evaluation information meets preset conditions, the process further includes: The feasibility of the scheduling task is verified, and the verification result is obtained. If the verification result indicates that the scheduling task is feasible, the step of executing the scheduling task in response to the scheduling instruction for the scheduling task is performed.
7. The information processing method according to any one of claims 1 or 6, characterized in that, The scheduling instruction carries a task scheduling frequency; the execution of the scheduling task in response to the scheduling instruction includes: In response to the scheduling instruction for the scheduled task, the scheduled task is executed according to the task scheduling frequency.
8. The information processing method according to claim 1, characterized in that, After executing the scheduling task in response to the scheduling instruction for the scheduling task, the method further includes: Based on the execution results of the scheduled tasks, a result database table is generated; the result database table includes at least one object to be disposed of. The results database table is pushed to the business platform so that the business platform can perform preset processing operations on the objects to be processed.
9. An information processing device, characterized in that, include: The filtering unit is configured to filter target objects from the sample objects based on the strategy information corresponding to the preset business scenario and the feature data of the sample objects; The tagging unit is configured to acquire tagging information of the target object; the tagging information is used to indicate the matching result between the target object and the preset business scenario; The evaluation unit is configured to determine evaluation information for the target object based on the tagging information; The scheduling unit is configured to create a scheduling task corresponding to the strategy information when the evaluation information meets preset conditions, and to execute the scheduling task in response to the scheduling instruction of the scheduling task; the scheduling task is used to filter objects to be processed according to the strategy information.
10. An information processing electronic device, characterized in that, include: processor; Memory used to store the processor's executable instructions; The processor is configured to execute the instructions to implement the information processing method as described in any one of claims 1 to 8.
11. A computer-readable storage medium, wherein instructions in the computer-readable storage medium, when executed by a processor of an information processing electronic device, enable the information processing electronic device to perform the information processing method as described in any one of claims 1 to 8.