Training content generation and data processing method, device and equipment, and storage medium

Abstract

Embodiments of the present application provide a training content generation and data processing method, device and equipment, and a storage medium. In the embodiments of the present application, the training task is described from multiple dimensions of task type information, personnel description information and content description information, so that the training task is more diversified and flexible. In the case of receiving the training task, the target personnel required to participate in the training task is determined according to the personnel description information, the task type information and the content description information required by the training task, and the personalized training content is generated for each target personnel by using a machine learning model in combination with the operation behavior data of each target personnel, so that the personalized recommendation of the training content can be achieved in combination with the individual differences of different personnel, which is beneficial to improve the training effect and further shorten the growth cycle of personnel. In addition, the machine learning is combined in the task training, which is beneficial to improve the generation efficiency of the training content, improve the training efficiency and further shorten the growth cycle of personnel.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of Internet, and particularly relates to a training content generation and data processing method and device, equipment and a storage medium. BACKGROUND

[0002] With the continuous pursuit of the spiritual world, the e-commerce field gradually appears a service mode taking multimedia content marketing as the main mode, for example, the most common ones are live broadcast, short video, small video and the like, and such e-commerce can be called content e-commerce. In order to ensure a good network environment, the multimedia content published by the content e-commerce needs to be audited.

[0003] At present, content auditing needs to be manually audited by an auditor with certain auditing skills. Because content auditing is a labor-intensive and mobile work, in order to enable the auditor to master the necessary auditing skills as soon as possible, a large number of and frequent personnel training is needed. The existing training mode is to stratify the auditors and centrally train the personnel of the same level offline. However, the training effect of this training mode is not ideal, resulting in a long growth cycle of the auditors. SUMMARY

[0004] Aspects of the present application provide a training content generation and data processing method, device, equipment and storage medium, to provide personalized training content to the personnel in need of training, improve the training effect and reduce the growth cycle of the personnel.

[0005] The present application provides a training content generation method, comprising: in response to a training task submission operation, obtaining task type information, personnel description information and content description information required by the present training task, the task type information being used to limit the content source type of the present training task; determining at least one target personnel required to participate in the present training task according to the personnel description information required by the present training task and combining portrait data of multiple personnel; generating personalized training content for each target personnel according to the task type information and the content description information required by the present training task and combining operation behavior data of each target personnel by using a machine learning model; wherein the personalized training content comprises content adapted to the content description information obtained from the content source type limited by the task type information.

[0006] The embodiment of the present application further provides a data processing method, comprising: in response to a learning task submission operation, obtaining task type information, personnel description information and content description information required by the current learning task, wherein the task type information is used to limit the content source type of the current learning task; determining at least one target personnel required to participate in the current learning task according to the personnel description information required by the current learning task and in combination with portrait data of multiple personnel; generating personalized learning content for each target personnel respectively by using a machine learning model according to the task type information and the content description information required by the current learning task and in combination with daily behavior data of each target personnel; wherein the personalized learning content comprises content obtained from the content source type limited by the task type information and adapted to the content description information.

[0007] The embodiment of the present application further provides a training content generation device, comprising: an acquisition module, configured to obtain task type information, personnel description information and content description information required by the current training task in response to a training task submission operation, wherein the task type information is used to limit the content source type of the current training task; a determination module, configured to determine at least one target personnel required to participate in the current training task according to the personnel description information required by the current training task and in combination with portrait data of multiple personnel; and a generation module, configured to generate personalized training content for each target personnel respectively by using a machine learning model according to the task type information and the content description information required by the current training task and in combination with operation behavior data of each target personnel; wherein the personalized training content comprises content obtained from the content source type limited by the task type information and adapted to the content description information.

[0008] The embodiment of the present application further provides an electronic device, comprising: a memory and a processor; the memory is configured to store a computer program, and the processor is coupled with the memory and is configured to execute the computer program to implement steps in the training content generation method provided by the embodiment of the present application.

[0009] The embodiment of the present application further provides a computer readable storage medium storing a computer program / instruction, when the computer program / instruction is executed by a processor, the processor can implement steps in the training content generation method provided by the embodiment of the present application.

[0010] In the embodiment of the present application, the training task is described from multiple dimensions of task type information, personnel description information and content description information, which can make the training task more diversified and flexible. When receiving the training task, the target personnel required to participate in the training task is determined according to the personnel description information, the task type information and the content description information required by the training task, and the personalized training content can be generated for each target personnel by using the machine learning model in combination with the operation behavior data of each target personnel, which can realize personalized recommendation of training content in combination with individual differences of different personnel, and is beneficial to improve the training effect and further shorten the growth cycle of personnel. In addition, the machine learning is combined in the training task, which is beneficial to improve the generation efficiency of training content, improve the training efficiency, and further shorten the growth cycle of personnel. BRIEF DESCRIPTION OF DRAWINGS

[0011] The accompanying drawings, which are included to provide a further understanding of the present application, constitute a part of this application and illustrate embodiments of the present application and together with the description serve to explain the present application. In the drawings:

[0012] Figure 1 A structural schematic diagram of a training system provided for an exemplary embodiment of the present application;

[0013] Figure 2a A flowchart of a training content generation method provided for an exemplary embodiment of the present application;

[0014] Figure 2b An interface diagram of a content type and content attribute provided for an exemplary embodiment of the present application;

[0015] Figure 2c An application diagram of an exemplary embodiment of the present application in a content review personnel training scenario;

[0016] Figure 3 A flowchart of a data processing method provided for an exemplary embodiment of the present application;

[0017] Figure 4 A structural schematic diagram of a training content generation device provided for an exemplary embodiment of the present application;

[0018] Figure 5 A structural schematic diagram of an electronic device provided for an exemplary embodiment of the present application. DETAILED DESCRIPTION

[0019] In order to make the purposes, technical solutions and advantages of the present application clearer, the technical solutions of the present application will be described below in connection with specific embodiments of the present application and corresponding drawings. Obviously, the described embodiments are only some of the embodiments of the present application, rather than all the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those of ordinary skill in the art without creative efforts belong to the scope of protection of the present application.

[0020] In view of the technical problem that the training effect is not ideal when the existing centralized offline training is performed on the auditors of the same level, leading to a long growth cycle of the auditors, in some embodiments of the present application, a solution of providing personalized training content to auditors in combination with machine learning and operation behavior data of the auditors is provided. Specifically, the training task is described from multiple dimensions of task type information, personnel description information and content description information, so that the training task is more diversified and flexible. On this basis, when the training task is received, the target personnel required to participate in the training task is determined according to the personnel description information, the task type information and the content description information required by the training task, and the personalized training content is generated for each target personnel by using a machine learning model in combination with the operation behavior data of each target personnel, so that the personalized recommendation of the training content can be achieved in combination with the individual differences of different personnel, which is beneficial to improve the training effect and further shorten the growth cycle of the personnel. In addition, the machine learning is combined in the task training, which is beneficial to improve the generation efficiency of the training content, improve the training efficiency and further shorten the growth cycle of the personnel.

[0021] The technical solutions provided by the embodiments of the present application will be described in detail below in connection with the drawings.

[0022] Figure 1 A structural schematic diagram of a training system provided by an exemplary embodiment of the present application is shown in FIG. 1. Figure 1 As shown in FIG. 1, the system 100 includes a training demand end 101, a training processing end 102 and a training participation end 103.

[0023] The training processing end 102 and the training demand end 101 can be connected by a wired or wireless network, and the training processing end 102 and the training participation end 103 can be connected by a wired or wireless network. Alternatively, the training processing end 102 can be connected to the training demand end 101 or the training participation end 103 through a mobile network, and the network standard of the mobile network can be any one of 2G (GSM), 2.5G (GPRS), 3G (WCDMA, TD-SCDMA, CDMA2000, UTMS), 4G (LTE), 4G+ (LTE+), 5G, WiMax, or a new network standard to be introduced in the future. Alternatively, the training processing end 102 can be connected to the training demand end 101 or the training participation end 103 through Bluetooth, WiFi, infrared, zigbee, or NFC.

[0024] In the embodiment, the training demand end 101 refers to an end that has personnel training demand and needs to initiate a training task, for example, but not limited to, a training manager or an organization personnel in a company, an enterprise, or an application scenario. The training manager or the organization personnel initiates a training task by using a terminal device. Corresponding to the training demand end 101, the training participation end 103 refers to an end that needs to participate in the training task initiated by the training demand end 101, for example, but not limited to, an employee in a company, an enterprise, or an application scenario. The employee participates in the training task by using a terminal device to perform online training.

[0025] The training processing end 102 refers to an end that can provide online training services for the training demand end 101 and the training participation end 103. On the one hand, the training processing end 102 is connected to the training demand end 101 and can provide a service interface for the training demand end 101, so that the training demand end 101 submits a training task through the service interface. The training processing end 102 responds to the training task submission operation initiated by the training demand end 101 through the service interface, and then determines target personnel participating in the training task for the training demand end 101, and generates online training content for the target personnel. On the other hand, the training processing end 102 is connected to the training participation end 103 and can provide the online training content to the training participation end 103, so that the target personnel corresponding to the training participation end 103 participates in the online training and continuously improves their own skills. For the training participation end 103, the online training content provided by the training processing end 102 can be output in the form of display or audio, so that the corresponding target personnel can participate in the online training and improve their own skills.

[0026] In the embodiments of the present application, the product implementation of the training processing end 102, the training demand end 101 and the training participation end 103 is not limited, and the implementation of the service interface provided by the training processing end 102 to the training demand end 101 will be different according to the product implementation. In an optional embodiment, a C / S architecture is adopted, the training processing end 102 is implemented as a service end, the training demand end 101 and the training participation end 103 are respectively implemented as a client or an APP, the service interface is implemented as an application page, and the training demand end 101 submits the training task to the training processing end 102 through the application page. In another optional embodiment, the training processing end 102 is implemented as a web service end, the training demand end 101 and the training participation end 103 are respectively implemented as a browser, the service interface is implemented as a web page, and the training demand end 101 submits the training task to the training processing end 102 through the web page.

[0027] In the embodiments of the present application, the deployment implementation of the training processing end 102, the training demand end 101 and the training participation end 103 is not limited. For example, in terms of deployment, the training processing end 102 can be deployed on a terminal device such as a mobile phone, a notebook computer or a desktop computer, or can be deployed on a regular server, a cloud server, a cloud host, a virtual center or a server array. Correspondingly, the training demand end 101 or the training participation end 103 can be deployed on a terminal device such as a mobile phone, a notebook computer or a desktop computer.

[0028] In the embodiments of the present application, the training processing end 102 can provide personalized training content to the target personnel who need to participate in the training task by combining machine learning with personnel operation behavior data. Further, as shown in Figure 1 , the model parameters of the machine learning model used by the training processing end 102 can also be optimized according to the result data of the training participation end 103 in the training learning according to the personalized training content. In addition, the training demand end 101 can also adjust or update the training task according to the performance of the training participation end 103 after the training, so that the personnel can continuously participate in the training task adapted to them and continuously improve their skills. For detailed description of the above process, please refer to the subsequent method embodiments, which will not be described here

[0029] Figure 2a A flowchart of a training content generation method provided by an exemplary embodiment of the present application. The execution subject of the present embodiment can be the training processing end 102 in the system as shown in Figure 1 , but is not limited thereto, and can also be any electronic device with computing and communication capabilities. As shown in Figure 2a , the method comprises the following steps.

[0030] 201. In response to a training task submission operation, obtain task type information, personnel description information, and content description information required for this training task. The task type information is used to limit the content source type of this training task.

[0031] 202. Based on the personnel description information required for this training task and combined with the portrait data of multiple personnel, determine at least one target person who needs to participate in this training task.

[0032] 203. Based on the task type information and content description information required by this training task, combined with the operational behavior data of each target person, a machine learning model is used to generate personalized training content for each target person; wherein, the personalized training content includes content obtained from the content source type specified by the above task type information and adapted to the above content description information.

[0033] In this embodiment, the application scenarios of the training task are not limited. For example, the application scenario may be to provide review skills training for content reviewers in the content e-commerce field, to provide safety training for on-site construction workers in the construction engineering field, to provide online training for manual customer service in the internet field, and so on. Depending on the training scenario, the personnel required to participate in the training task and the required training content will vary.

[0034] For example, in the training scenario for e-commerce content reviewers, the personnel required to participate in the training task may be individual content reviewers. Content reviewers refer to the staff who review multimedia content before it is published on the e-commerce platform. Accordingly, the required training content will mainly focus on various multimedia content. For another example, in the training scenario for construction project construction personnel, the personnel required to participate in the training task may be on-site construction personnel. Accordingly, the required training content will mainly focus on various safety knowledge. For another example, in the training scenario for online customer service, the personnel required to participate in the training task may be individual customer service representatives. Accordingly, the required training content will mainly focus on various customer service Q&A knowledge.

[0035] Regardless of the training scenario, in this embodiment, the training task is described or defined from multiple dimensions: task type information, personnel description information, and content description information. In other words, a training task involves at least three types of information: task type information, personnel description information, and content description information. Personnel description information is used to define the personnel required to participate in the training task and can include various personnel attribute information. Content description information is used to define the content required for the training task from the content dimension, for example, it can include content type and / or content attributes.

[0036] In the embodiment, the information of the task type information dimension is added for the training task, and the training task of the embodiment can support multiple different task type information. The task type information is used to limit the source type of the required training content of the training task, which is referred to as a content source type. Different task type information is used to limit different content source types. For any training task, the task type information used by the training task can be specified, and the content source type of the required training content of the training task is also specified. Different content source types can provide different training content, but it should be noted that the training content provided by different content source types should have or correspond to the same content description information. For example, in the case where the content description information includes the content type, the training content provided by different content source types is not completely the same, but these training content should have the same content type. The task type information and the content description information limit the required training content of the training task from two different dimensions.

[0037] As known from the above, in the embodiment, the training task is described from multiple dimensions of the task type information, the personnel description information and the content description information. The training demand end can limit the training task from multiple dimensions, which can not only limit the personnel required to participate in the training task and the training content required by the training task, but also limit the content source type corresponding to the training task, so that the training task is more diversified and flexible, and is conducive to meeting the diversified training task demand.

[0038] Further, based on the above description or definition of the training task, when the training demand end needs to arrange the training task, the training demand end only needs to configure the task type information, the personnel description information and the content description information required by the training task in the process of submitting the training task according to the above description or definition of the training task, without the need to configure the personnel information required to participate in the training task one by one, and without the need to configure the training content required by the training task one by one, which has the advantages of simple and efficient operation. For the execution subject (for example, the training processing end) of the method, the task type information, the personnel description information and the content description information required by the training task can be obtained in response to the training task submission operation.

[0039] In an optional embodiment, in response to the training task submission operation, the process of obtaining the task type information, the personnel description information and the content description information required by the current training task includes: in response to the training task submission operation, displaying a training task generation page, wherein the training task generation page at least includes a strategy configuration item and a task type option; the strategy configuration item can be one or more, which is configured by the training demand end to configure the personnel description information and the content description information required by the current training task; the task type option can be one or more, which is configured by the training demand end to configure the task type information required by the current training task, and each task type option corresponds to a task type information in the case of multiple task type options. The training demand end can configure the personnel description information and the content description information required by the current training task through the strategy configuration item, and accordingly, the personnel description information and the content description information required by the current training task can be obtained in response to the configuration operation of the strategy configuration item. Similarly, the training demand end can select the task type information required by the current training task through the task type option, and accordingly, the task type information corresponding to the selected task type option is determined as the task type information required by the current training task in response to the selection operation of the task type option.

[0040] Further optionally, the strategy configuration item includes a personnel configuration item and a content configuration item. Based on this, one way of obtaining the personnel description information and the content description information required by the current training task in response to the configuration operation of the strategy configuration item includes: obtaining the personnel description information required by the current training task in response to the configuration operation of the personnel configuration item; and obtaining the content description information required by the current training task in response to the configuration operation of the content configuration item.

[0041] In the embodiments of the present application, the implementation of the personnel description information is not limited, and any information description method that can limit the personnel required to participate in the current training task is applicable to the embodiments of the present application. In an optional embodiment, the personnel description information can be implemented as attribute information of the personnel, that is, the attribute information of the personnel is used to limit the personnel required to participate in the current training task. It should be noted that the attribute information of the personnel will be different according to different training scenarios.

[0042] In an implementation, the personnel required to participate in the current training task is defined by a statistical attribute, which is an attribute counted based on the operation behavior data of the personnel, such as a certain pass rate, error rate, total amount, or proportion, etc. In various training fields, there can be multiple statistical attributes, and in this embodiment, the statistical attribute to be used is referred to as a target statistical attribute, which can be one or more. Based on this, one implementation form of the personnel description information is: at least including the target statistical attribute and its corresponding time range and data condition. The time range is the time attribute of the target statistical attribute, which refers to the target statistical attribute in which time period is required; the data condition is used to limit the data range corresponding to the target statistical attribute, which can be an upper limit value or a lower limit value or a data range.

[0043] In another implementation, the personnel required to participate in the current training task is defined by a basic attribute, which is relative to the statistical attribute, and is an attribute that does not depend on the operation behavior data of the personnel, such as the level, skill level / qualification, type of work responsible for, department belonging to, and other relatively constant or low frequency of change attributes. In various training fields, there can be multiple basic attributes, and in this embodiment, the basic attribute to be used is referred to as a target basic attribute, which can be one or more. Based on this, another implementation form of the personnel description information is: at least including the target basic attribute and its corresponding data condition. Here, the data condition is used to limit the data range corresponding to the target basic attribute, which can be an upper limit value or a lower limit value or a data range.

[0044] In yet another implementation, the personnel required to participate in the current training task is defined by both the statistical attribute and the basic attribute. For related descriptions of the statistical attribute and the basic attribute, see the above. Based on this, another implementation form of the personnel description information is: at least including the target statistical attribute and its corresponding time range and data condition, and the target basic attribute and its corresponding data condition.

[0045] Correspondingly, the above response to the configuration operation of the personnel configuration item acquires one specific implementation of the personnel description information required by the current training task: in response to the configuration operation of the personnel configuration item, the target statistical attribute and its corresponding time range and data condition, and / or the target basic attribute and its corresponding data condition required by the current training task are acquired as the personnel description information.

[0046] In the embodiments of the present application, the implementation of the content description information is not limited, and any information description manner capable of defining the training content required by the training task is applicable to the embodiments of the present application. In an optional embodiment, the content description information can be implemented as a content type, that is, the training content required by the training task is defined by the content type. The content type refers to the category of the content in the field to which the content belongs. According to different training scenarios, the content type will be different. For example, taking the content e-commerce field as an example, multimedia content can be divided into fashion, makeup, home, food, baby, play, digital, pet, sports, etc. Correspondingly, if the content description information limits the makeup type, it means that the multimedia content under the makeup type is required as the training content for the training task; if the content description information limits the food type, it means that the multimedia content under the food type is required as the training content for the training task; if the content description information limits the food type and the baby type, it means that the multimedia content under the food type and the baby type is required as the training content for the training task, and so on. If the content description information includes a content type, it means that the content under the content type is required as the training content for the training task; if the content description information includes two or more content types, it means that the content under the two or more content types is required as the training content for the training task.

[0047] In another optional embodiment, the content also has one or more attribute information, referred to as content attribute, under the same content category. The content attribute refers to the attribute information that the content has under the same content category, and the content can be further classified according to the content attribute. Taking the content e-commerce field as an example, under the content type of pets, the pets are further divided into dogs, cats, aquatic animals, pet food, etc. based on the attributes of the pets. Based on this, another implementation manner of the content description information can include the content type and the content attribute, that is, the training content required by the current training task is defined by the content type and the content attribute. For example, taking the content e-commerce field as an example, the multimedia content can be divided into fashion, makeup, home furnishing, food, mother and baby, play, digital, pet, sports, etc.; further, under the content type of pets, the pets are further divided into dogs, cats, aquatic animals, pet food, etc. based on the attributes of the pets; based on this, if the content description information includes the "dog" attribute under the content type of pets, it means that the multimedia content related to "dog" under the content type of pets is required as the training content for the current training task; if the content description information includes the "dog" attribute and the "pet food" attribute under the content type of pets, it means that the multimedia content related to "dog" and "pet food" under the content type of pets is required as the training content for the current training task. It should be noted that the content description information can include one or more content types, and each content type can include one or more content attributes. Of course, the content description information can also include one or more content attributes alone without including the content type, and through the belonging relationship between the content attribute and the content type, the content type to which the content attribute belongs can be directly inferred.

[0048] Correspondingly, the above response to the configuration operation of the content configuration item acquires one specific implementation manner of the content description information required by the current training task, that is, the content type and / or the content attribute required by the current training task are acquired as the content description information in response to the configuration operation of the content configuration item. Further optionally, as shown in Figure 2b illustrated, acquiring the content type and the content attribute required by the current training task includes: in response to the configuration operation of the content configuration item, displaying a content type list, in response to the selection operation of the content type in the content type list, determining the content type required by the current training task, and displaying the content attribute list associated with the content type; and in response to the selection operation of the content attribute list, determining the content attribute under the content type required by the current training task.

[0049] After obtaining the task type information, the personnel description information and the content description information of the current training task requirement, at least one target personnel who needs to participate in the current training task can be determined according to the personnel description information of the current training task requirement and in combination with the portrait data of multiple personnel. In the embodiment, the personnel range to which the multiple personnel belong is not limited, for example, the multiple personnel can be all personnel involved in the training scene, or part of the personnel involved in the training scene. Taking the content e-commerce personnel training scene as an example, the multiple personnel can be all content review personnel, part of the content review personnel in a department, a group or responsible for a vertical content field, content review personnel of a certain level or within a certain level range, and the like.

[0050] In the embodiment of the present application, the portrait data of the personnel refers to data for describing the personnel in various ways, data capable of presenting various characteristics of the personnel, and data capable of distinguishing the personnel from other personnel. For example, the portrait data of the personnel can include but is not limited to operation behavior data of the personnel, basic attributes of the personnel, and statistical attributes of the personnel. It should be noted that in different training scenes, the operation behavior data, the basic attributes and the statistical attributes of the personnel will be different, and accordingly, the portrait data of the personnel will be different in specific implementation. The operation behavior data refers to data generated by the personnel in the application scene where the personnel is located when performing daily operations, which is referred to as operation behavior data. Taking the content e-commerce field as an example, the content review personnel needs to review various multimedia content in daily life, and then the multimedia content reviewed by the content review personnel, the attribute information of the multimedia content, and the review detail data of the content review personnel on the multimedia content can form the operation behavior data of the content review personnel. The review detail data of the content review personnel on the multimedia content includes but is not limited to the information of the content review personnel, the review time of the content review personnel on the multimedia content, the review result, and the tagging information corresponding to the review result. The tagging information can be the reason corresponding to the review result, for example, the reason for not passing or the reason for passing the review.

[0051] In the embodiment of the present application, the portrait data of the personnel at least includes data corresponding to the personnel description information, and therefore, based on the portrait data of the personnel, it can be determined whether the personnel meets the requirements of the personnel description information. If yes, it is determined that the personnel needs to participate in the current training task; if not, it is determined that the personnel does not need to participate in the current training task. It should be noted that according to the difference of the personnel description information in specific implementation, according to the personnel description information of the current training task requirement and in combination with the portrait data of multiple personnel, the way of determining at least one target personnel who needs to participate in the current training task will also be different. The implementation of determining the target personnel will be exemplarily described below in combination with the examples of the personnel description information given above.

[0052] In the above example, the personnel description information only includes one target statistical attribute, i.e., the audit pass rate, but is not limited thereto. For example, the personnel description information can also include two target statistical attributes, i.e., the audit pass rate and the audit volume, wherein the audit pass rate corresponds to a time range of the last week and a data condition of being lower than a first threshold, e.g., 80%; and the audit volume corresponds to a time range of the last week and a data condition of being higher than a second threshold, e.g., 300. Then, attribute statistics can be performed on the operation behavior data of the multiple content auditors in the last week (at least including multimedia contents audited in the last week and whether the multimedia contents pass the audit), to obtain the content audit pass rate and the audit volume of the multiple content auditors in the last week. Then, the content audit pass rate of the multiple content auditors in the last week is compared with the first threshold (e.g., 80%), and the audit volume of the multiple content auditors in the last week is compared with the second threshold (e.g., 300), to obtain content auditors whose content audit pass rate in the last week is lower than the first threshold (e.g., 80%) and whose audit volume in the last week is higher than the second threshold (e.g., 300), and these content auditors are taken as target personnel who need to participate in the current training task.

[0053] In the field of content e-commerce, assuming that the target statistical attribute is the audit pass rate, the corresponding time range is the last week, and the corresponding data condition is lower than a first threshold, e.g., 80%, attribute statistics can be performed on the operation behavior data of the multiple content auditors in the last week (at least including multimedia contents audited in the last week and whether the multimedia contents pass the audit), to obtain the content audit pass rate of the multiple content auditors in the last week. Then, the content audit pass rate of the multiple content auditors in the last week is compared with the first threshold (e.g., 80%), to obtain content auditors whose content audit pass rate in the last week is lower than the first threshold (e.g., 80%), and these content auditors are taken as target personnel who need to participate in the current training task.

[0054] In the above example, the personnel description information only includes one target statistical attribute, i.e., the audit pass rate, but is not limited thereto. For example, the personnel description information can also include two target statistical attributes, i.e., the audit pass rate and the audit volume, wherein the audit pass rate corresponds to a time range of the last week and a data condition of being lower than a first threshold, e.g., 80%; and the audit volume corresponds to a time range of the last week and a data condition of being higher than a second threshold, e.g., 300. Then, attribute statistics can be performed on the operation behavior data of the multiple content auditors in the last week (at least including multimedia contents audited in the last week and whether the multimedia contents pass the audit), to obtain the content audit pass rate and the audit volume of the multiple content auditors in the last week. Then, the content audit pass rate of the multiple content auditors in the last week is compared with the first threshold (e.g., 80%), and the audit volume of the multiple content auditors in the last week is compared with the second threshold (e.g., 300), to obtain content auditors whose content audit pass rate in the last week is lower than the first threshold (e.g., 80%) and whose audit volume in the last week is higher than the second threshold (e.g., 300), and these content auditors are taken as target personnel who need to participate in the current training task.

[0055] Example A2: The personnel description information includes a target basic attribute and a corresponding data condition. On this basis, a manner of determining at least one target personnel who needs to participate in the training task according to the personnel description information required by the training task and in combination with the portrait data of multiple personnel includes: obtaining actual data of the multiple personnel under the target basic attribute from the portrait data of the multiple personnel; and then selecting at least one target personnel from the multiple personnel according to the actual data of the multiple personnel under the target basic attribute and the data condition corresponding to the target basic attribute. The actual data of the personnel under the target basic attribute can be directly obtained from the portrait data.

[0056] Taking the content e-commerce field as an example, assuming that the target basic attribute is the qualification level, and the corresponding data condition is lower than a specified qualification level, for example, a five-level, then the qualification levels of multiple content review personnel can be obtained from the portrait data of the multiple content review personnel; then, the qualification levels of the multiple content review personnel are compared with the specified qualification level (such as the five-level), and the content review personnel whose qualification level is lower than the specified qualification level (such as the five-level) are obtained, and these content review personnel are taken as target personnel who need to participate in the training task.

[0057] In an example A3, the personnel description information includes not only the target statistical attribute and the corresponding time range and data condition, but also the target basic attribute and the corresponding data condition. On this basis, a manner of determining at least one target personnel required to participate in the current training task according to the personnel description information required by the current training task and in combination with the portrait data of the plurality of personnel includes: selecting at least one candidate personnel from the plurality of personnel according to the actual data of the plurality of personnel in the first attribute and the data condition corresponding to the first attribute; and selecting at least one target personnel from the at least one candidate personnel according to the actual data of the at least one candidate personnel in the second attribute and the data condition corresponding to the second attribute. Wherein, any one of the target statistical attribute and the target basic attribute is the first attribute, and the other is the second attribute. That is, at least one candidate personnel can be selected from the plurality of personnel according to the actual data of the plurality of personnel in the target statistical attribute and the data condition corresponding to the target statistical attribute; and then at least one target personnel is selected from the at least one candidate personnel according to the actual data of the at least one candidate personnel in the target basic attribute and the data condition corresponding to the target basic attribute. Alternatively, at least one candidate personnel can be selected from the plurality of personnel according to the actual data of the plurality of personnel in the target basic attribute and the data condition corresponding to the target basic attribute; and then at least one target personnel is selected from the at least one candidate personnel according to the actual data of the at least one candidate personnel in the target statistical attribute and the data condition corresponding to the target statistical attribute. In the foregoing two embodiments, before the actual data of the plurality of personnel in the target statistical attribute is used, attribute statistics needs to be performed on the operation behavior data generated by the plurality of personnel in the time range corresponding to the target statistical attribute to obtain the actual data of the plurality of personnel in the target statistical attribute.

[0058] Taking the field of content e-commerce as an example, the target statistical attribute is the pass rate, the corresponding time range is the last week, the corresponding data condition is lower than the first threshold, and the target basic attribute is the qualification level, and the corresponding data condition is lower than the specified qualification level, for example, the five-level level, for example, 80%. According to the operation behavior data (at least including multimedia content audited in the last week and whether the audit result is passed) of the plurality of content auditors in the last week, the content audit pass rate of the plurality of content auditors in the last week can be obtained. Then, the content audit pass rate of the plurality of content auditors in the last week is compared with the first threshold (such as 80%), and the content auditors whose content audit pass rate in the last week is lower than the first threshold (such as 80%) are obtained. These content auditors are the candidate personnel; then, from the portrait data of the plurality of content auditors, the qualification level of the candidate personnel is obtained, and the qualification level of the candidate personnel is compared with the specified qualification level (such as the five-level level), and the candidate personnel whose qualification level is lower than the specified qualification level (such as the five-level level) are obtained. These candidate personnel are the target personnel who need to participate in the training task. The implementation process of obtaining the target personnel here is only an example and is not limited thereto.

[0059] After determining the at least one target personnel who needs to participate in the training task, the personalized training content can be generated for each target personnel according to the task type information and content description information required by the training task, combined with the operation behavior data of each target personnel, by using a machine learning model. In this embodiment, the personalized training content is generated for each target personnel combined with the operation behavior data of each target personnel, which can achieve personalized recommendation of training content according to individual differences of different personnel, realize thousand faces for thousand people in the training scene, and be beneficial to improve the training effect, and thus shorten the growth cycle of personnel. In addition, in this embodiment, machine learning is introduced into the training scene, which is beneficial to improve the generation efficiency of the training content, improve the training efficiency, and be beneficial to further shorten the growth cycle of personnel.

[0060] In the embodiments of the present application, a machine learning model can be pre-trained, which can generate personalized training content for each target person according to the task type information and content description information of the current training task requirements, in combination with the operation behavior data of each target person. The machine learning model is obtained by iterative model training using a large amount of sample data, and each sample data includes sample task type information, sample content description information and sample personnel operation behavior data that have been labeled, and corresponding sample training content. The model structure of the machine learning model can include but is not limited to Convolutional Neural Networks (CNN), Recurrent Neural Networks (RNN) and Long Short-Term Memory (LSTM).

[0061] In an optional embodiment, the manner of generating personalized training content for each target person by using the machine learning model includes: for each target person, inputting the task type information and content description information of the current training task requirements and the operation behavior data of the target person into the machine learning model; in the machine learning model, determining available content that is adapted to the content source type defined by the task type information based on the operation behavior data of the target person; and generating personalized training content corresponding to the target person according to the content in the available content that is adapted to the content description information. It should be noted that the operation behavior data of the target person and other personnel contains content data that can be used as training content, and these content data will be different according to different training scenarios. For example, in the content e-commerce field, the operation behavior data of each person includes multimedia content that each person has reviewed, which can be used as training content. For example, in the construction field, the operation behavior data of each person includes construction information of each person, which belongs to a topic in one or more construction safety files, and the topic can be used as training content.

[0062] It should be noted that for different task type information, the content source type defined by the task type information is different, and based on this, the manner of determining available content based on the operation behavior data of the target person and the determined available content will also be different. The following is an example:

[0063] In Example B1, the task type information required by the current training task is first task type information, and the content source type defined by the first task type information is a first content source type. The first task type information refers to a task type that requires intensive training on content that the target personnel is not good at or has made mistakes on. Accordingly, the first content source type requires obtaining content that the target personnel needs to learn intensively from the operation behavior data of the target personnel as training content. In this case, one way of determining available content based on the operation behavior data of the target personnel includes: for each target personnel, at least obtaining content corresponding to the first content source type from the operation behavior data of the target personnel as available content. For example, the first content source type is a wrong question type, and for each target personnel, the content in which the target personnel has made mistakes is obtained from the operation behavior data of the target personnel as available content.

[0064] Further optionally, in addition to obtaining available content from the operation behavior data of the target personnel, other personnel with the same or similar attributes as the target personnel can be determined according to the portrait data of the target personnel, and available content can be obtained from the operation behavior data of the other personnel with the same or similar attributes as the target personnel, which can improve the richness of available content. Optionally, the other personnel with the same or similar attributes as the target personnel can be other personnel belonging to the same group, department, or having the same level as the target personnel. In the case where the first content source type is a wrong question type, for each target personnel, the content in which the target personnel has made mistakes is obtained from the operation behavior data of the target personnel as available content, and the content in which other personnel has made mistakes is obtained from the operation behavior data of other personnel belonging to the same group, department, or having the same level as the target personnel as available content. Further optionally, if there is a repetition between the available content obtained from the operation behavior data of the target personnel and the available content obtained from the operation behavior data of the other personnel, a deduplication process can be performed, that is, only one copy of the same available content is retained, which not only enriches the available content, but also reduces the storage space occupied by the repeated available content and reduces the computational load.

[0065] Further, in the embodiments of the present application, in response to the training task submission operation, in addition to the task type information, the personnel description information and the content description information required by the training task, the content quantity, the total score, the passing score and the name of the training content required by the training task and other basic information of the training content can also be obtained. Based on this, for each target personnel, before determining other personnel having the same or similar attributes as the target personnel according to the portrait data of the target personnel, it can also be judged whether the available content meeting the content quantity can be obtained from the operation behavior data of the target personnel according to the content quantity required by the training task; if not enough available content can be obtained from the operation behavior data of the target personnel, other personnel having the same or similar attributes as the target personnel are determined according to the portrait data of the target personnel, and available content is continuously obtained from the operation behavior data of the other personnel until the requirement of the content quantity is met. Of course, if enough available content can be obtained from the operation behavior data of the target personnel, only the available content from the operation behavior data of the target personnel is needed.

[0066] Example B2: The task type information required by the training task is the second task type information, and the content source type defined by the second task type information is the second content source type. The second task type information refers to the task type that needs to be improved for the content that the target personnel has not contacted, and accordingly, the second content source type requires the content that is helpful for the target personnel to improve as the training content. In this case, one way of determining the available content based on the operation behavior data of the target personnel includes: for each target personnel, the content that does not appear in the operation behavior data of the target personnel is obtained from the available operation behavior data as the available content. For example, the second content source type is the new question type, and for each target personnel, the content that the target personnel has not contacted can be obtained from the available operation behavior data as the available content. The available operation behavior data can be other operation behavior data in the entire database except the operation behavior data of the target personnel, which is not limited.

[0067] Example B3: the task type information required by the current training task is third task type information, and the content source type defined by the third task type information is a third content source type. The third task type information refers to a task type that requires comprehensive training on content that the target personnel is not good at or makes mistakes on and content that the target personnel has not touched. Accordingly, the third content source type is a comprehensive content source type that contains the first content source type and the second content source type. In this case, one way of determining the available content based on the operation behavior data of the target personnel includes: for each target personnel, on the one hand, at least from the operation behavior data of the target personnel, obtaining content corresponding to the first content source type as part of the available content, and on the other hand, from the available operation behavior data, obtaining content that does not appear in the operation behavior data of the target personnel as another part of the available content. This example B3 is a fusion of the above-mentioned examples B1 and B2. For details of the implementation of at least obtaining content corresponding to the first content source type from the operation behavior data of the target personnel, please refer to the above. Similarly, for details of the implementation of obtaining content that does not appear in the operation behavior data of the target personnel from the available operation behavior data, please refer to the above. No further description is given.

[0068] For each target personnel, based on the available content corresponding to the target personnel, the content description information required by the current training task can be used to select content that matches the content description information from the available content to generate personalized training content corresponding to the target personnel. Alternatively, the machine learning model can maintain one or more content templates, fill the content that matches the content description information into a certain content template, and obtain the personalized training content. Alternatively, the machine learning model can also maintain content organization rules, and organize the content that matches the content description information according to the content organization rules to obtain the personalized training content.

[0069] In an optional embodiment, the above-mentioned personalized training content can be realized as an electronic test paper in terms of presentation form. The machine learning model can generate an electronic test paper carrying the personalized training content according to the test paper template or the test question organization rule according to the content that matches the content description information. Further optionally, taking the content e-commerce field as an example, the above-mentioned content description information is a content type, and the content that matches the content description information can be a certain type of content, such as a food video, a makeup video, a digital video, etc., or a food text, a sports text, etc. The electronic test paper contains food videos, makeup videos, digital videos, food texts, or sports texts for the target personnel to learn content auditing skills.

[0070] In another optional embodiment of the present application, during the use of the machine learning model, the model parameters used by the machine learning model can also be continuously optimized to improve the quality and accuracy of the personalized training content generated based on the machine learning model, thereby improving the training effect based on the personalized training content. Based on this, the result data obtained by each target person according to the corresponding personalized training content can also be collected; and the model parameters used by the machine learning model can be continuously optimized according to these result data. The optimization process of the model parameters can be automatically executed, or can be realized in a semi-automatic manner by manual intervention.

[0071] In an optional embodiment, in addition to the current model parameters used by the machine learning model, at least one candidate model parameter can be determined in advance, the at least one candidate model parameter being different from the current model parameters, and there may be optimized parameters in the at least one candidate model parameter. For example, the at least one candidate model parameter can be determined according to artificial experience, or the current model parameters can be adjusted appropriately to obtain the at least one candidate model parameter. In order to facilitate the machine learning model to use more optimized model parameters, during the process of generating personalized training content for each target person by using the machine learning model, if there are multiple target persons, the multiple target persons can be divided into at least two groups, including a first group and other groups; the first group corresponds to the current model parameters used by the machine learning model, and the other groups correspond to different candidate model parameters, and the other groups can be one or more, which is not limited; for each group, the machine learning model adopts the model parameters corresponding to the group, and then according to the task type information and content description information required by the current training task, combined with the operation behavior data of each target person in the group, the machine learning model generates personalized training content for each target person in the group. Optionally, the number of target persons in the first group is more than the number of target persons in the other groups, so as to ensure that most target persons can obtain relatively good personalized training content in quality.

[0072] Further, after generating the personalized training content for each group of target persons, the personalized training content can be provided to the corresponding target persons, and the target persons can learn and train according to the corresponding personalized training content; further, the result data obtained by each target person according to the corresponding personalized training content can be acquired; and the model parameters used by the machine learning model can be optimized according to the result data of each group. Specifically, the result data of each group can be compared, and a second group with result data better than that of the first group can be obtained from other groups; since the result data of the second group is better than that of the first group, it indicates that the candidate model parameters corresponding to the second group are of better quality, and therefore, the current model parameters used by the machine learning model can be optimized according to the candidate model parameters corresponding to the second group, for example, the current model parameters can be replaced by the candidate model parameters, or the current model parameters can be adjusted in the direction of the candidate model parameters.

[0073] In the foregoing, when comparing the result data of two groups, the result data of each group can be normalized and compared according to the normalized result data; or, the result data higher than the upper limit of the quality requirement in each group can be selected, and the result data higher than the upper limit of the quality requirement in two groups can be compared; or, the result data lower than the lower limit of the quality requirement in each group can be selected, and the result data lower than the lower limit of the quality requirement in two groups can be compared; or, the result data higher than the upper limit of the quality requirement and the result data lower than the lower limit of the quality requirement in each group can be selected, the average of the result data higher than the upper limit of the quality requirement and the result data lower than the lower limit of the quality requirement in each group can be calculated, and the two averages can be compared, and the like. The comparison manner of the result data of two groups is not limited in the embodiments of the present application, and can be determined according to the form of the result data. In an optional embodiment, the result data obtained by each target person according to the corresponding personalized training content can be a score value, for example, 80 points, 95 points, and the like. In another optional embodiment, the result data obtained by each target person according to the corresponding personalized training content can be grade information, for example, excellent, good, or poor, and the like.

[0074] It is explained herein that, in the above-mentioned embodiments of the present application, the training content can be used for training purposes or for examination purposes, and is not limited.

[0075] In the above embodiments of the present application, the training task is described from multiple dimensions of task type information, personnel description information and content description information, which can make the training task more diversified and flexible. When receiving the training task, the target personnel required to participate in the training task is determined according to the personnel description information, task type information and content description information required by the training task, and personalized training content can be generated for each target personnel by combining the operation behavior data of each target personnel and using a machine learning model, which can make personalized recommendation of training content according to individual differences of different personnel, and is beneficial to improve the training effect and shorten the growth cycle of personnel. In addition, the combination of machine learning in the training task is beneficial to improve the generation efficiency of training content, improve the training efficiency, and further shorten the growth cycle of personnel.

[0076] In the above embodiments of the present application, the training scene is not limited, and in an optional embodiment, the training scene can be a training and examination scene for content review personnel in the content e-commerce field. The application of the technical solutions provided by the embodiments of the present application in this specific scene will be described in detail below with reference to the flowchart shown in the figure: Figure 2c

[0077] In the content e-commerce field, the content publishing platform will put content demand to the content providing end (such as each host, video author, etc.), for example, content demand with the theme of Children's Day, content demand with the theme of a certain food, etc. The video providing end records multimedia content such as live video, short video and text information that meets the video demand and uploads it to the content publishing platform. The content publishing platform will audit the quality and compliance of the multimedia content before publishing the multimedia content. Optionally, the auditing process includes a preliminary auditing stage based on machine and a manual auditing stage based on content review personnel. The preliminary auditing based on machine mainly audits objective indicators of the multimedia content such as format, size, resolution and whether it contains specific content. Manual auditing mainly audits the theme style (referred to as tonality) and quality of the multimedia content, whether the theme style of the multimedia content meets the theme requirements of the platform, and of course, some objective indicators missed by machine auditing can also be audited.

[0078] ​In the artificial auditing stage, some marking standards can be preset for each vertical content field, for example, the video only has advertisements, lacks theme content, or the video has no high-quality content, or the video involves sensitive content, or the video involves yellow content, or the content does not belong to the vertical content field, etc. During the quality auditing process of the multimedia content, the auditor can determine whether the multimedia content passes the audit; if it is determined that the multimedia content cannot pass the quality audit, the reason why it cannot pass the audit can be selected from the above marking standards, so as to complete the quality marking. In addition, the auditor can also perform theme style auditing on the multimedia content, and can determine whether the multimedia content is consistent with the theme style required by the platform. If it is considered to be inconsistent, it is determined that the multimedia content cannot pass the theme style audit; if it is considered to be consistent, it is determined that the multimedia content passes the theme style audit. The multimedia content, attribute information of the multimedia content, auditing time, various information of the auditor, auditing result and reason information when the auditing result does not pass, etc. in the above auditing process will form the auditing operation data of the content auditor, that is, a specific implementation of the operation behavior data in the above.

[0079] Regarding the artificial auditing part, it usually depends on the understanding and mastery of the industry auditing standards by the content auditors, and the content auditors perform artificial auditing on the multimedia content according to the industry auditing standards. However, the qualification levels of the content auditors are different, and the degrees of understanding and mastery of the industry auditing standards are different. If the industry auditing standards cannot be well understood and mastered, misjudgment may occur, that is, the multimedia content that meets the requirements is judged as not meeting the requirements, and the multimedia content that does not meet the requirements is judged as meeting the requirements. In order to improve the qualification level of the content auditors, it is necessary to train the content auditors.

[0080] Referring to Figure 2c In this embodiment, the basic information of the content auditors, the auditing operation data and the result data of the previous training or examination can be obtained to form the portrait data of the content auditors; and the personal auditing operation data of the content auditors is collected. Optionally, the auditing operation data of each content auditor can be pre-processed, such as classification, processing, noise reduction, etc. to obtain various statistical attribute information of each content auditor, and pre-stored in the portrait data of the content auditors.

[0081] Further referring to Figure 2cIn this embodiment, training organizers are allowed to configure online training task rules according to training needs and submit training tasks. The training task rules include basic rules related to the test paper, including but not limited to: test paper name, total score, passing score, test time, etc., and also include dynamic rules related to the training task, including but not limited to: task type, time range, personnel attributes, and content type and / or content attributes. After the training organizer submits the training task, the aforementioned training task rules can be obtained and parsed using a rule parser. Based on the parsed time range and personnel attributes, the audit operation data of each content reviewer can be classified, processed, denoised, and otherwise processed to obtain the actual data of each content reviewer on the personnel attribute dimension required by the aforementioned rules. Alternatively, the actual data of each content reviewer on the personnel attribute dimension required by the aforementioned rules can be directly obtained from the profile data of each content reviewer, and combined with other personnel attribute information in the profile data, the target content reviewer participating in this training task can be determined. Furthermore, based on the parsed task type and content type and / or content attributes, the personal audit operation data of the target content reviewer can be combined to generate a training test paper corresponding to the target content reviewer using a machine learning model. The training test paper can then be pushed to the target content reviewer for training and learning, and training or examination result data can be generated. This result data can further improve the profile data of the target content reviewer.

[0082] In addition, if Figure 2c As shown in the figure, after content reviewers participate in training, further statistical data on their review indicators, such as review pass rate and number of reviews, can be collected. These data reflect whether the reviewers' qualifications have improved after training. Training organizers can reassign new training tasks based on the reviewers' new qualifications, thereby continuously providing personalized training for content reviewers, improving their development progress, and enhancing the efficiency and accuracy of content review.

[0083] In this embodiment, on the one hand, the training and examination of auditors can be personalized. Based on user portraits and personal statistical data, on-demand learning can be achieved, and the learning and examinations are "different for each person"; on the other hand, real-time training effects can be achieved. Based on online audit operation data, if the rules are matched and training is required, training papers will be pushed in real time to allow content auditors to learn and avoid similar problems from happening again; furthermore, based on the effects of online feedback, the model parameters used by the machine learning model can also be optimized so that training papers that meet quality requirements can be given every time to achieve the maximum training effect.

[0084] In the above embodiments, the technical solutions of the embodiments of the present application are described by taking the training scenario as an example, but the technical solutions provided by the embodiments of the present application are not limited to the training scenario, and can be extended to various learning-related scenarios, for example, can also be applied to a student examination scenario, etc. Based on this, the following embodiments of the present application also provide a data processing method, as shown in Figure 3 includes the following steps:

[0085] 21. In response to a learning task submission operation, obtaining task type information, personnel description information and content description information required by the current learning task, wherein the task type information is used to limit the content source type of the current learning task;

[0086] 22. According to the personnel description information required by the current learning task, in combination with the portrait data of multiple personnel, at least one target personnel who needs to participate in the current learning task is determined;

[0087] 23. According to the task type information and the content description information required by the current learning task, in combination with the daily behavior data of each target personnel, a personalized learning content is generated for each target personnel by using a machine learning model; wherein the personalized learning content includes content adapted to the content description information obtained from the content source type limited by the task type information.

[0088] In the present embodiment, the learning scenario is not limited, which can be a training scenario or any other learning-related scenario. The difference between the present embodiment and the embodiment shown in Figure 2a is that the application scenarios are different, and the detailed implementation processes of the steps in the present embodiment are the same as or similar to the related steps in the embodiment shown in Figure 2a , and will not be described here again. Please refer to the description of the embodiment shown in Figure 2a .

[0089] It should be noted that the execution subject of each step of the method provided in the above embodiments can be the same device, or the method can also be executed by different devices as the execution subject. For example, the execution subject of steps 21 to 23 can be device A; for example, the execution subject of steps 21 and 22 can be device A, and the execution subject of step 23 can be device B; and so on.

[0090] In addition, in some of the processes described in the above embodiments and the accompanying drawings, a plurality of operations are included in a specific order, but it should be clear that these operations can be executed in the order in which they appear in this document or in parallel, and the serial numbers of the operations such as 21, 22, etc. are only used to distinguish different operations, and the serial numbers themselves do not represent any execution order. In addition, these processes can include more or fewer operations, and the operations can be executed in sequence or in parallel. It should be noted that the "first", "second", etc. described herein are used to distinguish different messages, devices, modules, etc., and do not represent the order of precedence, nor do "first" and "second" represent different types.

[0091] Figure 4 A schematic structural diagram of a training content device is provided for an exemplary embodiment of the present application. As shown in the figure, the device includes an acquisition module 41, a determination module 42, and a generation module 43. Figure 4

[0092] The acquisition module 41 is configured to, in response to a training task submission operation, acquire task type information, personnel description information, and content description information required by the current training task, wherein the task type information is used to limit the content source type of the current training task.

[0093] The determination module 42 is configured to determine at least one target personnel required to participate in the current training task according to the personnel description information required by the current training task, in combination with portrait data of a plurality of personnel.

[0094] The generation module 43 is configured to generate personalized training content for each target personnel according to the task type information and the content description information required by the current training task, in combination with operation behavior data of each target personnel, using a machine learning model. The personalized training content includes content obtained from the content source type limited by the task type information and adapted to the content description information.

[0095] In an optional embodiment, the personnel description information at least includes a target statistical attribute, a corresponding time range, and a data condition. When determining at least one target personnel required to participate in the current training task, the determination module 42 is specifically configured to: perform attribute statistics on operation behavior data generated by a plurality of personnel in the above time range to obtain actual data of the plurality of personnel under the target statistical attribute, wherein the actual data under the target statistical attribute belongs to the portrait data; and select at least one target personnel from the plurality of personnel according to the actual data of the plurality of personnel under the target statistical attribute and the data condition corresponding to the target statistical attribute.

[0096] ​Further optionally, the personnel description information further comprises: a target basic attribute and a corresponding data condition thereof. Based on this, the determining module 42, when selecting at least one target personnel from the plurality of personnel according to the actual data of the plurality of personnel under the target statistical attribute and the data condition corresponding to the target statistical attribute, is specifically configured to: select at least one candidate personnel from the plurality of personnel according to the actual data of the plurality of personnel under a first attribute and the data condition corresponding to the first attribute; select at least one target personnel from the at least one candidate personnel according to the actual data of the at least one candidate personnel under a second attribute and the data condition corresponding to the second attribute; wherein the first attribute is any one of the target statistical attribute and the target basic attribute, and the second attribute is the other attribute, and the actual data under the target basic attribute belongs to the portrait data.

[0097] In an optional embodiment, the generating module 43, when generating the personalized training content for each target personnel by using the machine learning model, is specifically configured to: input the task type information, the content description information, and the operation behavior data of the target personnel into the machine learning model for each target personnel; determine available content that is adapted to the content source type defined by the task type information based on the operation behavior data of the target personnel; and generate the personalized training content corresponding to the target personnel according to the content in the available content that is adapted to the content description information.

[0098] Further optionally, the generating module 43, when determining the available content that is adapted to the content source type defined by the task type information, is specifically configured to:

[0099] In a case where the content source type defined by the task type information is a first content source type, at least the content corresponding to the first content source type is obtained from the operation behavior data of the target personnel as the available content;

[0100] Or

[0101] In a case where the content source type defined by the task type information is a second content source type, the content that does not appear in the operation behavior data of the target personnel is obtained from the available operation behavior data as the available content;

[0102] Or

[0103] In a case where the content source type defined by the task type information is a third content source type including the first content source type and the second content source type, at least the content corresponding to the first content source type is obtained from the operation behavior data of the target personnel as a part of the available content, and the content that does not appear in the operation behavior data of the target personnel is obtained from the available operation behavior data as another part of the available content.

[0104] Further optionally, the generating module 43 is specifically configured to: determine other personnel having the same or similar attributes as the target personnel according to the portrait data of the target personnel; and obtain the content corresponding to the first content source type from the operation behavior data of the target personnel and the other personnel.

[0105] Further optionally, before determining other personnel having the same or similar attributes as the target personnel according to the portrait data of the target personnel, the generating module 43 is further configured to: determine whether the available content meeting the content quantity can be obtained from the operation behavior data of the target personnel according to the content quantity required by the current training task; and perform the operation of determining other personnel having the same or similar attributes as the target personnel according to the portrait data of the target personnel in the case of determining that the available content meeting the content quantity cannot be obtained from the operation behavior data of the target personnel. Further, in the case of determining that the available content meeting the content quantity can be obtained from the operation behavior data of the target personnel, the content corresponding to the first content source type is directly obtained from the operation behavior data of the target personnel.

[0106] In an optional embodiment, when the generating module 43 generates the personalized training content for each target personnel by using the machine learning model, the generating module 43 is specifically configured to: divide the multiple target personnel into at least two groups, wherein the first group of the at least two groups corresponds to the current model parameter used by the machine learning model, and each of the other groups corresponds to a different candidate model parameter; for each group, set the machine learning model to use the model parameter corresponding to the group, and generate the personalized training content for each target personnel in the group by using the machine learning model according to the task type information and the content description information required by the current training task and the operation behavior data of each target personnel in the group.

[0107] Further optionally, as shown in Figure 4 the training content apparatus further includes a parameter optimization module 44 configured to: obtain result data obtained by the target personnel in each group after training and learning according to the corresponding personalized training content; and in the case that there is a second group in which the result data is better than the result data of the first group among the other groups, optimize the current model parameter used by the machine learning model according to the candidate model parameter corresponding to the second group.

[0108] In an optional embodiment, when acquiring the task type information, personnel description information and content description information required for this training task, the acquisition module 41 is specifically used to: respond to the training task submission operation, display the training task generation page, and the training task generation page includes at least policy configuration items and task type options; respond to the configuration operation of the policy configuration items, acquire the personnel description information and content description information required for this training task; respond to the selection operation of the task type option, determine the task type information corresponding to the selected task type option as the task type information required for this training task.

[0109] Further optionally, when acquiring the task type information, personnel description information and content description information required for this training task, the acquisition module 41 is specifically used to: respond to the configuration operation of the personnel configuration item, acquire the target statistical attributes required for this training task and their corresponding time range and data conditions and / or target basic attributes and their corresponding data conditions as personnel description information; respond to the configuration operation of the content configuration item, acquire the content type and / or attributes required for this training task as content description information.

[0110] Further optionally, when acquiring the content type and / or attributes required for this training task, the acquisition module 41 is specifically used to: respond to the configuration operation of the content configuration item, display the content type list, respond to the selection operation of the content type in the content type list, determine the content type required for this training task, and display the content attribute list associated with the content type; respond to the selection operation of the content attribute list, and determine the content attributes under the content type required for this training task.

[0111] Further optionally, as Figure 4 As shown, the training content device further includes: a providing module 45, which is used to provide the personalized training content corresponding to each target person to the corresponding target person, so that the target person can perform training and learning according to the personalized training content.

[0112] Figure 4 The training content generation device shown can execute the training content generation method of the above-mentioned method embodiment, and its implementation principle and technical effects are not described in detail here. The specific manner in which each module of the training content generation device in the above-mentioned embodiment performs operations has been described in detail in the relevant method embodiment and will not be elaborated on here.

[0113] In addition, the embodiment of the present application further provides a data processing apparatus, comprising: an obtaining module, a determining module and a generating module. The obtaining module is configured to obtain task type information, personnel description information and content description information required by a learning task in response to a learning task submission operation, wherein the task type information is used to limit the content source type of the learning task. The determining module is configured to determine at least one target personnel required to participate in the learning task according to the personnel description information required by the learning task and in combination with portrait data of multiple personnel. The generating module is configured to generate personalized learning content for each target personnel by using a machine learning model according to the task type information and the content description information required by the learning task and in combination with operation behavior data of each target personnel. The personalized learning content comprises content adapted to the content description information obtained from the content source type limited by the task type information. The specific operation modes of the modules of the data processing apparatus can refer to the detailed description of the foregoing method embodiments, and will not be described in detail here.

[0114] The exemplary embodiments of the present application also provide an electronic device, as shown in the accompanying drawings, which comprises a memory 51 and a processor 52. Figure 5

[0115] The memory 51 is configured to store computer programs and can be configured to store various data to support operations on the electronic device. Examples of the data include instructions of any application program or method for operating on the electronic device, contact data, phonebook data, messages, pictures, videos, etc.

[0116] The memory 51 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 memory, flash memory, magnetic disk or optical disk.

[0117] ​The processor 52 is coupled with the memory 51 and is configured to execute a computer program in the memory 51 to: in response to a training task submission operation, acquire task type information, personnel description information and content description information required by the current training task, the task type information being used to limit the content source type of the current training task; determine at least one target personnel required to participate in the current training task according to the personnel description information required by the current training task and in combination with portrait data of multiple personnel; and generate personalized training content for each target personnel according to the task type information and the content description information required by the current training task and in combination with operation behavior data of each target personnel by using a machine learning model, wherein the personalized training content includes content adapted to the content description information and acquired from the content source type limited by the task type information.

[0118] In an optional embodiment, the personnel description information at least includes a target statistical attribute and a corresponding time range and data condition. Based on this, the processor 52, when determining at least one target personnel required to participate in the current training task, is specifically configured to: perform attribute statistics on operation behavior data generated by multiple personnel in the time range to obtain actual data of multiple personnel under the target statistical attribute, the actual data under the target statistical attribute belonging to the portrait data; and select at least one target personnel from the multiple personnel according to the actual data of multiple personnel under the target statistical attribute and the data condition corresponding to the target statistical attribute.

[0119] Further optionally, the personnel description information further includes a target basic attribute and a corresponding data condition. Based on this, when the processor 52 selects at least one target personnel from the multiple personnel according to the actual data of multiple personnel under the target statistical attribute and the data condition corresponding to the target statistical attribute, the processor 52 is specifically configured to: select at least one candidate personnel from the multiple personnel according to the actual data of multiple personnel under a first attribute and the data condition corresponding to the first attribute; and select at least one target personnel from the at least one candidate personnel according to the actual data of the at least one candidate personnel under a second attribute and the data condition corresponding to the second attribute, wherein any one of the target statistical attribute and the target basic attribute is the first attribute, and the other attribute is the second attribute, and the actual data under the target basic attribute belongs to the portrait data.

[0120] In an optional embodiment, when the processor 52 generates personalized training content for each target personnel by using a machine learning model, the processor 52 is specifically configured to: for each target personnel, input the task type information, the content description information and operation behavior data of the target personnel into the machine learning model; determine available content adapted to the content source type limited by the task type information based on the operation behavior data of the target personnel; and generate personalized training content corresponding to the target personnel according to content adapted to the content description information in the available content.

[0121] Further optionally, the processor 52, in determining the available content adapted to the content source type defined by the task type information, is specifically configured to: in a case where the content source type defined by the task type information is a first content source type, obtain, as the available content, the content corresponding to the first content source type at least from the operation behavior data of the target personnel; or in a case where the content source type defined by the task type information is a second content source type, obtain, as the available content, the content not appearing in the operation behavior data of the target personnel from the available operation behavior data; or in a case where the content source type defined by the task type information is a third content source type including the first content source type and the second content source type, obtain, as a part of the available content, the content corresponding to the first content source type at least from the operation behavior data of the target personnel, and obtain, as another part of the available content, the content not appearing in the operation behavior data of the target personnel from the available operation behavior data.

[0122] In an optional embodiment, the processor 52, in obtaining, as the available content, the content corresponding to the first content source type at least from the operation behavior data of the target personnel, is specifically configured to: determine, according to the portrait data of the target personnel, other personnel having the same or similar attributes as the target personnel; and obtain, as the available content, the content corresponding to the first content source type from the operation behavior data of the target personnel and the other personnel.

[0123] Further optionally, the processor 52 is further configured to: before determining, according to the portrait data of the target personnel, other personnel having the same or similar attributes as the target personnel, determine, according to the content quantity required by the current training task, whether the available content meeting the content quantity can be obtained from the operation behavior data of the target personnel; if not, perform the operation of determining, according to the portrait data of the target personnel, other personnel having the same or similar attributes as the target personnel; and if yes, directly obtain, as the available content, the content corresponding to the first content source type from the operation behavior data of the target personnel.

[0124] In an optional embodiment, the processor 52, in generating, for each target personnel, the personalized training content by using the machine learning model, is specifically configured to: in a case where the target personnel are multiple, divide the multiple target personnel into at least two groups, wherein a first group of the at least two groups corresponds to current model parameters used by the machine learning model, and other groups correspond to different candidate model parameters; for each group, set the machine learning model to use the model parameters corresponding to the group, and generate, for each target personnel in the group, the personalized training content by using the machine learning model according to the task type information and the content description information required by the current training task and in combination with the operation behavior data of each target personnel in the group.

[0125] In an optional embodiment, the processor 52 is further configured to provide the personalized training content corresponding to each target person to the corresponding target person, so that the target person learns and trains according to the personalized training content.

[0126] In an optional embodiment, the processor 52 is further configured to obtain result data of the target persons in each group learning and training according to the corresponding personalized training content, and if there is a second group in which the result data is better than the result data of the first group, optimize the current model parameters used by the machine learning model according to the candidate model parameters corresponding to the second group.

[0127] In an optional embodiment, when the processor 52 obtains the task type information, the personnel description information and the content description information required by the current training task, the processor 52 is specifically configured to: in response to a training task submission operation, display a training task generation page, wherein the training task generation page at least includes a strategy configuration item and a task type option; in response to a configuration operation on the strategy configuration item, obtain the personnel description information and the content description information required by the current training task; and in response to a selection operation on the task type option, determine the task type information corresponding to the selected task type option as the task type information required by the current training task.

[0128] Further optionally, when the processor 52 obtains the personnel description information and the content description information required by the current training task, the processor 52 is specifically configured to: in response to a configuration operation on the personnel configuration item, obtain the target statistical attribute and its corresponding time range and data condition and / or the target basic attribute and its corresponding data condition required by the current training task as the personnel description information; and in response to a configuration operation on the content configuration item, obtain the content type and / or attribute required by the current training task as the content description information.

[0129] Further optionally, when the processor 52 obtains the content type and / or attribute required by the current training task, the processor 52 is specifically configured to: in response to a configuration operation on the content configuration item, display a content type list, in response to a selection operation on a content type in the content type list, determine the content type required by the current training task, and display a content attribute list associated with the content type; and in response to a selection operation on the content attribute list, determine the content attribute under the content type required by the current training task.

[0130] For detailed implementation process of the processor 52 performing each action, refer to the related description in the foregoing method embodiments, which will not be repeated here.

[0131] It should be noted that in addition to the above-mentioned actions, the processor 52 can also perform the following actions in another optional embodiment: in response to a learning task submission operation, obtaining task type information, personnel description information and content description information required by the current learning task, the task type information being used to limit the content source type of the current learning task; determining at least one target person who needs to participate in the current learning task according to the personnel description information required by the current learning task, in combination with the portrait data of multiple persons; and generating personalized learning content for each target person according to the task type information and the content description information required by the current learning task, in combination with the operation behavior data of each target person, by using a machine learning model. The personalized learning content includes content obtained from the content source type limited by the task type information and adapted to the content description information.

[0132] Further, as shown in Figure 5 , the electronic device further includes a communication component 53, a display 54, a power supply component 55, an audio component 56, and other components. Figure 5 Some components are only schematically shown in the electronic device, and it does not mean that the electronic device only includes Figure 5 the components shown. In addition, Figure 5 the components in the dashed box are optional components, not mandatory components, and the specific implementation depends on the product form of the electronic device. The electronic device of the present embodiment can be implemented as a terminal device such as a desktop computer, a notebook computer, a smart phone or an IOT device, or a server device such as a conventional server, a cloud server or a server array. If the electronic device of the present embodiment is implemented as a terminal device such as a desktop computer, a notebook computer or a smart phone, it can include Figure 5 the components in the dashed box; if the electronic device of the present embodiment is implemented as a server device such as a conventional server, a cloud server or a server array, it can not include Figure 5 the components in the dashed box.

[0133] Correspondingly, the present embodiment also provides a computer readable storage medium storing computer programs / instructions, which, when executed by a processor, enable the processor to implement each step in the above-mentioned method embodiment.

[0134] Correspondingly, the present embodiment also provides a computer program product, which includes computer programs / instructions, which, when executed by a processor, enable the processor to implement each step in the above-mentioned method embodiment.

[0135] The communication component in the above-described diagram embodiments is configured to facilitate wired or wireless communication between the device in which the communication component is located and other devices. The device in which the communication component is located can access a wireless network based on a communication standard, such as WiFi, 2G, 3G, 4G / LTE, 5G, or the like, or a combination thereof. In an example embodiment, the communication component receives a broadcast signal or broadcast-related information from an external broadcast managing system via a broadcast channel. In an example embodiment, the communication component further includes a Near Field Communication (NFC) module to facilitate short-range communication. For example, the NFC module can be implemented based on Radio Frequency Identification (RFID) technology, infrared Data Association (IrDA) technology, Ultra-WideBand (UWB) technology, Bluetooth (BT) technology, and other technologies.

[0136] The display in the above-described diagram embodiments includes a screen, which can include a Liquid Crystal Display (LCD) and a Touch Panel (TP). If the screen includes a touch panel, the screen can be implemented as a touch screen to receive an input signal from a user. The touch panel includes one or more touch sensors to sense a touch, a slide, and a gesture on the touch panel. The touch sensor can not only sense a boundary of a touch or a slide, but also detect a duration and a pressure associated with the touch or slide operation.

[0137] The power supply component in the above-described diagram embodiments provides power to various components of the device in which the power supply component is located. The power supply component can include a power management system, one or more power sources, and other components associated with generating, managing, and distributing power to the device in which the power supply component is located.

[0138] The audio component in the above-described diagram embodiments can be configured to output and / or input audio signals. For example, the audio component includes a microphone (MIC) that is configured to receive an external audio signal when the device in which the audio component is located in an operational mode, such as a call mode, a recording mode, and a voice recognition mode. The received audio signal can be further stored in a memory or transmitted via the communication component. In some embodiments, the audio component also includes a speaker to output audio signals.

[0139] Those skilled in the art will appreciate that embodiments of the present application can be supplied as a method, a system, or a computer program product. Accordingly, the present application can take the form of an entirely hardware embodiment, an entirely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, the present application can take the form of a computer program product on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROMs, optical storage media, etc.) embodying computer-readable program code.

[0140] The computer program instructions can also be loaded onto a computer or other programmable data processing apparatus to cause a series of operational steps to be performed on the computer or other programmable apparatus to produce a computer-implemented process such that the instructions which execute on the computer or other programmable apparatus provide steps for implementing the functions specified in the flowchart block or blocks. Figure 1 one or more flowcharts and / or blocks in the flowcharts and / or combination thereof. Figure 1 one or more flowcharts and / or blocks in the flowcharts and / or combination thereof.

[0141] The computer program instructions can also be loaded onto a computer or other programmable data processing apparatus to cause a series of operational steps to be performed on the computer or other programmable apparatus to produce a computer-implemented process such that the instructions which execute on the computer or other programmable apparatus provide steps for implementing the functions specified in the flowchart block or blocks. Figure 1 one or more flowcharts and / or blocks in the flowcharts and / or combination thereof. Figure 1 one or more flowcharts and / or blocks in the flowcharts and / or combination thereof.

[0142] The computer program instructions can also be loaded onto a computer or other programmable data processing apparatus to cause a series of operational steps to be performed on the computer or other programmable apparatus to produce a computer-implemented process such that the instructions which execute on the computer or other programmable apparatus provide steps for implementing the functions specified in the flowchart block or blocks. Figure 1 one or more flowcharts and / or blocks in the flowcharts and / or combination thereof. Figure 1 one or more flowcharts and / or blocks in the flowcharts and / or combination thereof.

[0143] In one typical configuration, the computing device includes one or more processors (CPUs), input / output interfaces, network interfaces, and memory.

[0144] The memory can include non-persistent memory and / or volatile memory, such as random access memory (RAM) and / or cache memory, non-volatile memory, such as read-only memory (ROM), EPROM, and / or flash memory. The memory is an example of computer-readable media.

[0145] Computer-readable media includes permanent and non-permanent, movable and non-movable media that can implement information storage by any method or technology. The information can be computer-readable instructions, data structures, program modules 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, compact disc read-only memory (CD-ROM), digital versatile disc (DVD) or other optical storage, magnetic cassette, magnetic tape disk storage or other magnetic storage devices, or any other non-transmission medium that can be used to store information accessible to a computing device. According to the definition herein, computer-readable media does not include transitory media such as modulated data signals and carriers.

[0146] It should also be noted that the terms "comprising", "containing", or any other variant thereof are intended to cover a non-exclusive inclusion, such that a process, method, article or apparatus that comprises a list of elements does not only include those elements, but can also include other elements not expressly listed or inherent to such process, method, article or apparatus. Without more limitations, the element defined by the statement "comprising a" does not exclude the presence of additional identical elements in the process, method, article or apparatus that includes the element.

[0147] The above only describes the embodiments of the present application and is not intended to limit the present application. For those skilled in the art, the present application can have various changes and modifications. Any modification, equivalent replacement, improvement, etc. within the spirit and principle of the present application shall be included in the scope of claims of the present application.

Claims

1. A training content generation method, characterized in that: include: In response to the training task submission operation, the task type information, personnel description information and content description information required by the training task are obtained. The task type information is used to define the content source type of the training task; Based on the personnel description information required for this training task and combined with the portrait data of multiple personnel, determine at least one target person who needs to participate in this training task; Based on the task type and content description information required by this training task, combined with the operational behavior data of each target person, a machine learning model is used to generate personalized training content for each target person, including: For each target person, input the task type information, content description information, and the target person's operation behavior data into a machine learning model; Based on the operational behavior data of the target person, determining available content that is compatible with the content source type defined by the task type information, including: judging whether, based on the content quantity required by this training task, available content that satisfies the required content quantity can be obtained from the operational behavior data of the target person; if a sufficient amount of available content cannot be obtained from the operational behavior data of the target person, determining other persons having the same or similar attributes as the target person based on the portrait data of the target person, and continuing to obtain available content from the operational behavior data of the other persons until the required content quantity is met; generating personalized training content corresponding to the target person according to the content in the available content that is adapted to the content description information; Among them, operational behavior data refers to the data generated by personnel performing daily operations in their application scenarios. The operational behavior data of target personnel and other personnel contain content data that can be used as training content; Wherein, in the case where there are multiple target persons, the multiple target persons are divided into at least two groups, and different model parameters are used for different groups; The personalized training content includes content obtained from a content source type defined by the task type information and adapted to the content description information.

2. The method according to claim 1, characterized in that The personnel description information includes at least target statistical attributes and their corresponding time range and data conditions. Based on the personnel description information required by this training task and combined with the portrait data of multiple personnel, at least one target person who needs to participate in this training task is determined, including: Performing attribute statistics on the operation behavior data generated by multiple personnel within the time range to obtain actual data of the multiple personnel under the target statistical attributes, where the actual data under the target statistical attributes belong to the portrait data; At least one target person is selected from the multiple persons according to actual data of the multiple persons under the target statistical attribute and a data condition corresponding to the target statistical attribute.

3. The method according to claim 2, characterized in that The personnel description information further includes: target basic attributes and corresponding data conditions. Then, based on the actual data of the multiple personnel under the target statistical attributes and the data conditions corresponding to the target statistical attributes, at least one target person is selected from the multiple personnel, including: selecting at least one candidate from the plurality of persons based on actual data of the plurality of persons under the first attribute and a data condition corresponding to the first attribute; selecting at least one target person from the at least one candidate person according to actual data of the at least one candidate person under the second attribute and a data condition corresponding to the second attribute; Among them, any one of the target statistical attribute and the target basic attribute is the first attribute, the other attribute is the second attribute, and the actual data under the target basic attribute belongs to the portrait data.

4. The method according to claim 1, wherein Based on the target person's operational behavior data, available content that is compatible with the content source type defined by the task type information is determined, including: In a case where the content source type defined by the task type information is a first content source type, obtaining content corresponding to the first content source type as available content from at least the operation behavior data of the target person; or In a case where the content source type defined by the task type information is the second content source type, obtaining, from the available operation behavior data, content that does not appear in the operation behavior data of the target person as available content; or In a case where the content source type defined by the task type information is a third content source type including a first content source type and a second content source type, at least the content corresponding to the first content source type is obtained from the operation behavior data of the target person as a part of the available content, and the content that does not appear in the operation behavior data of the target person is obtained from the available operation behavior data as another part of the available content.

5. The method according to any one of claims 1 to 4, characterized in that Based on the task type and content description information required by this training task, combined with the operational behavior data of each target person, a machine learning model is used to generate personalized training content for each target person, including: In the case where there are multiple target persons, the multiple target persons are divided into at least two groups, a first group of the at least two groups corresponds to current model parameters used by the machine learning model, and the other groups correspond to different candidate model parameters; For each group, a machine learning model is set using model parameters corresponding to the group. Based on the task type information and content description information required by this training task, combined with the operational behavior data of each target person in the group, the machine learning model is used to generate personalized training content for each target person in the group.

6. The method according to claim 5, characterized in that Also includes: Obtain the result data of the training and learning of the target personnel in each group according to the corresponding personalized training content; If there is a second group in the other groups whose result data is better than the result data of the first group, the current model parameters used by the machine learning model are optimized according to the candidate model parameters corresponding to the second group.

7. The method according to any one of claims 1 to 4, characterized in that Respond to the training task submission operation and obtain the task type information, personnel description information, and content description information required by this training task, including: In response to the training task submission operation, a training task generation page is displayed, wherein the training task generation page includes at least a policy configuration item and a task type option; In response to the configuration operation of the policy configuration item, the personnel description information and content description information required by the training task are obtained; In response to a selection operation on the task type option, the task type information corresponding to the selected task type option is determined as the task type information required for this training task.

8. A data processing method, characterized in that: include: In response to the learning task submission operation, the task type information, personnel description information and content description information required by the learning task are obtained. The task type information is used to define the content source type of the learning task; Based on the personnel description information required for this learning task and combined with the portrait data of multiple personnel, determine at least one target person who needs to participate in this learning task; Based on the task type and content description information required by this learning task, combined with the daily behavior data of each target person, a machine learning model is used to generate personalized learning content for each target person, including: For each target person, input the task type information, content description information, and the target person's operation behavior data into a machine learning model; Based on the operational behavior data of the target person, determining available content that is compatible with the content source type defined by the task type information, including: judging whether, based on the content quantity required by this training task, available content that satisfies the required content quantity can be obtained from the operational behavior data of the target person; if a sufficient amount of available content cannot be obtained from the operational behavior data of the target person, determining other persons having the same or similar attributes as the target person based on the portrait data of the target person, and continuing to obtain available content from the operational behavior data of the other persons until the required content quantity is met; generating personalized training content corresponding to the target person according to the content in the available content that is adapted to the content description information; Among them, operational behavior data refers to the data generated by personnel performing daily operations in their application scenarios. The operational behavior data of target personnel and other personnel contain content data that can be used as training content; Wherein, in the case where there are multiple target persons, the multiple target persons are divided into at least two groups, and different model parameters are used for different groups; The personalized learning content includes content obtained from a content source type defined by the task type information and adapted to the content description information.

9. A training content generating device, characterized in that: include: An acquisition module is used to respond to a training task submission operation and obtain the task type information, personnel description information, and content description information required for this training task. The task type information is used to define the content source type of this training task; The determination module is used to determine at least one target person who needs to participate in this training task based on the personnel description information required by this training task and the portrait data of multiple personnel; A generation module is used to generate personalized training content for each target person based on the task type information and content description information required by this training task, combined with the operational behavior data of each target person, using a machine learning model, including: for each target person, inputting the task type information, content description information and the operational behavior data of the target person into the machine learning model; based on the operational behavior data of the target person, determining available content that is adapted to the content source type specified by the task type information, including: judging whether available content that meets the content quantity required by this training task can be obtained from the operational behavior data of the target person; if a sufficient amount of available content cannot be obtained from the operational behavior data of the target person, determining other persons with the same or similar attributes as the target person based on the portrait data of the target person, and continuing to obtain available content from the operational behavior data of the other persons until the content quantity requirement is met; generating personalized training content corresponding to the target person based on the content in the available content that is adapted to the content description information; Among them, operational behavior data refers to the data generated by personnel performing daily operations in their application scenarios. The operational behavior data of target personnel and other personnel contain content data that can be used as training content; Wherein, in the case where there are multiple target persons, the multiple target persons are divided into at least two groups, and different model parameters are used for different groups; The personalized training content includes content obtained from a content source type defined by the task type information and adapted to the content description information.

10. An electronic device, characterized in that: include: memory and processor; The memory is used to store a computer program, and the processor is coupled to the memory and is used to execute the computer program to implement the steps in the method according to any one of claims 1 to 8.

11. A computer-readable storage medium storing a computer program / instruction, characterized in that: When the computer program / instructions are executed by a processor, the processor is enabled to implement the steps of the method according to any one of claims 1 to 9.

Images