Methods, apparatus, devices and storage media for information processing

Abstract

According to embodiments of this disclosure, a method, apparatus, device, and storage medium for information processing are provided. The method includes: for a workflow in at least one workflow, in response to identifying at least one event related to the workflow, storing a corresponding event record for the at least one event; in response to the storage of the corresponding event record, updating event record information for the workflow, the event record information indicating event occurrence information or event record usage information of the workflow from at least one dimension; and using the event record stored for the at least one workflow based on the corresponding event record information for the at least one workflow to determine an action for the at least one workflow. This ensures that the event information provided to subsequent nodes (e.g., machine learning models) is sufficient and useful, thereby avoiding waste of information transmission resources. In this way, the efficiency of intelligent workflow processing can be improved.

Description

Technical Field

[0001] The exemplary embodiments disclosed herein generally relate to the field of computers, and more particularly to methods, apparatus, devices and computer-readable storage media for information processing. Background Technology

[0002] With the development of information technology, various terminal devices can provide people with a variety of services in work and life. Applications providing these services can be deployed on these terminal devices. How to utilize terminal devices or applications to provide users with more convenient services is a technical issue that needs to be explored. Summary of the Invention

[0003] In a first aspect of this disclosure, an information processing method is provided. The method includes: for a workflow in at least one workflow, in response to identifying at least one event associated with the workflow, storing a corresponding event record for the at least one event; in response to the storage of the corresponding event record, updating event record information for the workflow, the event record information indicating event occurrence information or event record usage information of the workflow from at least one dimension; and using the event record stored for the at least one workflow based on the corresponding event record information for the at least one workflow to determine an action for the at least one workflow.

[0004] In a second aspect of this disclosure, an apparatus for information processing is provided. The apparatus includes: an event log storage module configured to, for a workflow in at least one workflow, store a corresponding event log for the at least one event in response to identifying at least one event associated with the workflow; a log information update module configured to, in response to the storage of the corresponding event log, update event log information for the workflow, the event log information indicating event occurrence information or event log usage information of the workflow from at least one dimension; and an event log usage module configured to, based on the corresponding event log information for the at least one workflow, use the event logs stored for the at least one workflow to determine an action for a target workflow.

[0005] In a third aspect of this disclosure, an electronic device is provided. The device includes at least one processing unit; and at least one memory coupled to the at least one processing unit and storing instructions for execution by the at least one processing unit. When executed by the at least one processing unit, the instructions cause the device to perform the method of the first aspect.

[0006] In a fourth aspect of this disclosure, a computer-readable storage medium is provided. The computer-readable storage medium stores a computer program that can be executed by a processor to implement the method of the first aspect.

[0007] It should be understood that the content described in this content section is not intended to limit the key or essential features of the embodiments of this disclosure, nor is it intended to restrict the scope of this disclosure. Other features of this disclosure will become readily apparent from the following description. Attached Figure Description

[0008] The above and other features, advantages, and aspects of the embodiments of this disclosure will become more apparent from the accompanying drawings and the following detailed description. In the drawings, the same or similar reference numerals denote the same or similar elements, wherein:

[0009] Figure 1 A schematic diagram of an example environment in which embodiments of the present disclosure can be implemented is shown;

[0010] Figure 2 A schematic diagram of an example architecture for information processing according to some embodiments of the present disclosure is shown;

[0011] Figure 3 A flowchart is shown illustrating an example process for information processing according to some embodiments of this disclosure;

[0012] Figure 4 A schematic diagram illustrating an example of determining whether to store an event log based on a water level line according to some embodiments of the present disclosure is shown;

[0013] Figure 5 A flowchart illustrating an information processing procedure according to some embodiments of the present disclosure is shown;

[0014] Figure 6 A block diagram of an apparatus for information processing according to some embodiments of the present disclosure is shown; and

[0015] Figure 7 A block diagram of an apparatus capable of implementing several embodiments of the present disclosure is shown. Detailed Implementation

[0016] It is understood that before using the technical solutions disclosed in the various embodiments of this disclosure, users should be informed of the type, scope of use, and usage scenarios of the user information involved in this disclosure in an appropriate manner in accordance with relevant laws and regulations, and user authorization should be obtained.

[0017] For example, upon receiving a user's active request, a prompt message is sent to the user to explicitly inform them that the requested operation will require the acquisition and use of their information. This allows the user to autonomously choose, based on the prompt message, whether to provide user information to the software or hardware such as the electronic device, application program, server, or storage medium performing the operations of this disclosed technical solution.

[0018] As an optional but non-limiting implementation, in response to a user's active request, sending a prompt message to the user can be done via a pop-up window, where the prompt message can be presented in text format. Furthermore, the pop-up window can also include a selection control allowing the user to choose "agree" or "disagree" to provide user information to the electronic device.

[0019] The activation of related functions, the acquisition of data, the processing and storage of data, etc., in the embodiments disclosed herein shall all require prior authorization from the user and other rights holders associated with the user, and shall comply with the relevant laws and regulations and the agreements between rights holders.

[0020] It is understood that before using the technical solutions disclosed in the various embodiments of this disclosure, relevant users should be informed of the type, scope of use, and usage scenarios of the information involved in this disclosure through appropriate means in accordance with relevant laws and regulations, and authorization should be obtained from the relevant users. Among them, relevant users may include any type of rights holder, such as individuals, enterprises, and groups.

[0021] For example, in response to receiving an active request from a user, a prompt message is sent to the relevant user to clearly inform the user that the requested operation will require obtaining and using the user's information, thereby enabling the relevant user to choose whether to provide information to the software or hardware such as the electronic device, application, server, or storage medium that performs the operation of the technical solution disclosed herein based on the prompt message.

[0022] As an optional but non-restrictive implementation, in response to a user's active request, a prompt message can be sent to the user, such as a pop-up window, where the prompt message can be presented in text format. Furthermore, the pop-up window can also include a selection control allowing the user to choose "agree" or "disagree" to provide information to the electronic device.

[0023] It is understood that the above notification and user authorization process are merely illustrative and do not constitute a limitation on the implementation of this disclosure. Other methods that comply with relevant laws and regulations may also be applied to the implementation of this disclosure.

[0024] It is understood that when adopting this technical solution, the data involved (including but not limited to the data itself, the acquisition, use, storage and transmission of the data) shall comply with the requirements of relevant laws, regulations and related provisions.

[0025] It is understood that the above notification and user authorization process are merely illustrative and do not constitute a limitation on the implementation of this disclosure. Other methods that comply with relevant laws and regulations may also be applied to the implementation of this disclosure.

[0026] Embodiments of this disclosure will now be described in more detail with reference to the accompanying drawings. While some embodiments of this disclosure are shown in the drawings, it should be understood that this disclosure can be implemented in various forms and should not be construed as limited to the embodiments set forth herein. Rather, these embodiments are provided to provide a more thorough and complete understanding of this disclosure. It should be understood that the accompanying drawings and embodiments of this disclosure are for illustrative purposes only and are not intended to limit the scope of protection of this disclosure.

[0027] It should be noted that the headings of any section / subsection provided herein are not limiting. Various embodiments are described throughout this document, and embodiments of any type may be included under any section / subsection. Furthermore, embodiments described in any section / subsection may be combined in any way with any other embodiments described in the same section / subsection and / or different sections / subsections.

[0028] In this document, unless explicitly stated otherwise, performing a step in response to A does not mean that the step is performed immediately after A, but may include one or more intermediate steps.

[0029] In the description of embodiments of this disclosure, the term "comprising" and similar terms should be understood as open-ended inclusion, i.e., "including but not limited to". The term "based on" should be understood as "at least partially based on". The term "one embodiment" or "the embodiment" should be understood as "at least one embodiment". The term "some embodiments" should be understood as "at least some embodiments". Other explicit and implicit definitions may also be included below. The terms "first", "second", etc., may refer to different or the same objects. Other explicit and implicit definitions may also be included below.

[0030] As used in this paper, the term "model" refers to a system that learns the relationship between inputs and outputs from training data, enabling it to generate corresponding outputs for a given input after training. Model generation can be based on machine learning techniques. Deep learning is a machine learning algorithm that uses multiple layers of processing units to process inputs and provide corresponding outputs. In this paper, "model" may also be referred to as a "machine learning model," a "machine learning network," or simply a "network," and these terms are used interchangeably. A model can also include different types of processing units or networks.

[0031] In some scenarios, such as automated workflows based on instant messaging (IM) messages, certain content is difficult to extract using simple keywords. Understandably, in an automated IM workflow, a simple workflow can be abstracted as follows: a real-time or near-real-time message acts as a trigger, initiating multiple actions. Typically, the message needs to be provided to subsequent nodes (e.g., machine learning models) to determine the next triggered action. The introduction of machine learning models can significantly broaden the scope of scenarios. Users can input very simple message text or perform only simple configurations to obtain content they are interested in. For example, configuring action trigger conditions, which are used to trigger actions in the automated workflow when certain conditions are met, means that when the information related to the action trigger conditions is content that the user is interested in, the action previously set by the user can be triggered. The content the user is interested in can be determined based on the user-input message text, such as the topic, category, examples, sentences, paragraphs, etc., expressed in natural language, or it can be multimedia files such as images and videos.

[0032] Subsequent nodes (e.g., machine learning models) rely on contextual information for categorization / summarization. In this case, providing user relationship content (e.g., single messages, text snippets, audio / video clips, etc.) to subsequent nodes will be ineffective. Furthermore, if subsequent nodes utilize machine learning models, considering performance limitations such as the amount of data transmitted per transmission by the current machine learning model, tokens provided to the machine learning model are wasted on prompt word transmission when single messages are short, easily leading to additional costs. Similar to messages, machine learning models perform poorly and waste transmission resources when text snippets or audio / video clips are too small.

[0033] In view of this, embodiments of the present disclosure propose an improved scheme for information processing. According to various embodiments of the present disclosure, for a workflow in at least one workflow, if at least one event related to the workflow is identified, a corresponding event record for the at least one event is stored. Further, in response to the storage of the corresponding event record, event record information for the workflow is updated, the event record information indicating event occurrence information or event record usage information of the workflow from at least one dimension. Then, based on the corresponding event record information for the at least one workflow, the event records stored for the at least one workflow are used to determine an action for a target workflow.

[0034] According to embodiments of this disclosure, upon identifying an event related to the workflow, event records are stored to aggregate event records generated over a period of time, rather than processing individual event records. The aggregated event records are then provided to subsequent nodes to determine the actions to be performed. For example, a prompt message is generated based on the information from the aggregated event records and provided to a machine learning model to determine subsequent actions. This ensures that the event information provided to subsequent nodes (e.g., machine learning models) is sufficient and useful, thereby avoiding waste of information transmission resources. In this way, the efficiency of intelligent workflow processing can be improved.

[0035] Example Environment

[0036] Figure 1 A schematic diagram of an example environment 100 in which embodiments of the present disclosure can be implemented is shown. In this example environment 100, a component runtime platform 110 is able to support the operation of a business component 125. A user 140 can interact with the business component 125 via a client of the component runtime platform 110.

[0037] In some embodiments, service component 125 can be downloaded and installed on user 140's terminal device. In some embodiments, service component 125 can also be accessed in other ways, such as via a web page. Figure 1 In environment 100, in response to the startup of business component 125, the client of component runtime platform 110 can display the interface 150 of business component 125.

[0038] Business components 125 include, but are not limited to, one or more of the following: chat business components (also known as instant messaging (IM) components), document business components, audio and video conferencing business components, email business components, task business components, calendar business components, goal and key results (OKR) business components, etc. It is understandable that, although... Figure 1The diagram shows a single business component, but in reality, multiple business components can be installed on the component runtime platform 110. Multiple business components can be integrated on the component runtime platform 110, making it a multi-functional collaboration platform. When multiple business components are installed on a terminal device, these components can be integrated onto one or more component runtime platforms 110. Within the component runtime platform 110, users can launch different business components as needed to perform corresponding information processing, sharing, communication, etc. Business component 125 can provide content entity 126. Content entity 126 can be a content instance created by user 140 or other users on business component 125. For example, depending on the type of business component 125, content entity 126 can be a document (e.g., a Word document, a PDF document, a presentation, a spreadsheet, etc.), email, message (e.g., a conversation message on an instant messaging business component), calendar, schedule, task, audio, video, image, etc.

[0039] The client interface can display interaction windows between user 140 and other users. In these interaction windows, user 140 can engage in conversations with other users by inputting natural language, images, audio files, video files, web page files, etc. The interaction window may include a session window, such as a session window within the instant messaging module of an instant messaging service component or a target service component. In the session window, interactions between user 140 and other users can be presented in the form of session messages. Alternatively or additionally, the interaction window may also include other types of windows, such as floating window windows, where user 140 can trigger corresponding operations on the client by inputting commands, selecting shortcuts, etc.

[0040] The component runtime platform 110 can be deployed locally on each user's (140's) terminal device and / or supported by a server device. For example, user 140's terminal device can run a client of the component runtime platform 110, which can support interaction between user 140 and the server-provided component runtime platform 110. When the component runtime platform 110 runs locally on the user's terminal device, user 145 can directly interact with the local component runtime platform 110 using the terminal device. When the component runtime platform 110 runs on a server device, the server device can provide services to the client running on the terminal device based on the communication connection with the terminal device. The component runtime platform 110 can present a corresponding interface 150 to user 140 based on user 140's operations, to output and / or receive information related to component usage from user 140.

[0041] In some embodiments, the implementation of at least some functions of the business component 125, and / or at least some functions of the digital assistant 120, may be based on target models. During the operation of the business component 125, one or more target models 155 may be invoked. Target models 155 may be used to understand user input and to provide services based on the output of target models 155, such as providing responses to the user.

[0042] Although shown as independent of component runtime platform 110, one or more target models 155 may run on component runtime platform 110 or other remote servers. In some embodiments, target model 155 may be a machine learning model, deep learning model, learning model, neural network, etc. In some embodiments, the model may be based on a language model (LM). A language model is capable of question answering by learning from a large corpus. Target model 155 may also be based on other suitable models.

[0043] The component runtime platform 110 can run on suitable electronic devices. These electronic devices can be any type of computing-capable device, including terminal devices or server devices. Terminal devices can be any type of mobile terminal, fixed terminal, or portable terminal, including mobile phones, desktop computers, laptop computers, notebook computers, netbook computers, tablet computers, media computers, multimedia tablets, personal communication system (PCS) devices, personal navigation devices, personal digital assistants (PDAs), audio / video players, digital cameras / camcorders, positioning devices, television receivers, radio receivers, e-book devices, gaming devices, or any combination of the foregoing, including accessories and peripherals of these devices or any combination thereof. Server devices can include, for example, computing systems / servers, such as mainframes, edge computing nodes, computing devices in cloud environments, and so on. In some embodiments, the component runtime platform 110 can be implemented based on cloud services.

[0044] It should be understood that the structure and function of environment 100 are described for illustrative purposes only and do not imply any limitation on the scope of this disclosure.

[0045] The following description continues with reference to the accompanying drawings, which will continue to illustrate some exemplary embodiments of this disclosure. It should be understood that the pages shown in the drawings are merely examples, and various page designs are possible in practice. The various graphic elements on the page may have different arrangements and different visual representations, one or more elements may be omitted or replaced, and one or more other elements may also be present. The embodiments of this disclosure are not limited in this respect. Furthermore, in the following description, exemplary embodiments will be primarily described with respect to component runtime platform 110. It should be understood that the actions described with respect to component runtime platform 110 may be performed by applications, components, or suites (e.g., business component 125) on component runtime platform 110, or may be performed by applications, components, or suites in conjunction with their server (e.g., a server).

[0046] It should be understood that the structure and function of the various elements in environment 100 are described for illustrative purposes only and do not imply any limitation on the scope of this disclosure.

[0047] The following description will continue with reference to the accompanying drawings, which will be used to describe some exemplary embodiments of this disclosure. Hereinafter, the exemplary embodiments will be described primarily with respect to component runtime platform 110. It should be understood that the actions described with respect to component runtime platform 110 can be performed by an application on component runtime platform 110, or can be performed by an application in conjunction with its server (e.g., server 130).

[0048] The following is for reference Figure 2 and Figure 3 This disclosure describes a scheme for information processing. Figure 2 A schematic diagram of an example framework 200 for information processing according to some embodiments of the present disclosure is shown. Figure 3 A flowchart of an example process 300 for information processing according to some embodiments of the present disclosure is shown.

[0049] In example architecture 200, in a configuration-mode scenario, user 140 (sometimes referred to as the workflow configurer) can configure a custom workflow through a front-end page. For example, a workflow can be defined by a work recipe. User 140 can set one or more work recipes. Each work recipe defines a workflow. In some embodiments, each workflow has its corresponding trigger, which can include one or more triggering conditions. For example, a message received by the component runtime platform 110. If the message is related to the workflow, the accumulated triggering conditions are met, and the event record related to the message is stored. Then, for the message, it can be determined whether the action triggering conditions are met. If the action triggering conditions are met, the action corresponding to the message is executed, such as a machine learning model analyzing the message, writing the message into a table, and so on.

[0050] After user 140 confirms the configuration state, the system can transition to the runtime state. The following continues to describe the message processing scheme performed by the component runtime platform 110 in the runtime state. In some embodiments, for a workflow in at least one workflow, if the component runtime platform 110 identifies at least one event related to that workflow, it stores the corresponding event record for that at least one event.

[0051] When in runtime, the component runtime platform 110 receives and detects event sources 210 from clients (e.g., user 140's client). For example, the component runtime platform 110 can detect user 140's message event 210-1, user 140's schedule event 210-2, user 140's task event, or other data source events 210-3, etc. In some embodiments, events may include, but are not limited to, received messages, schedules, emails, text snippets, audio / video snippets, etc. The corresponding event record for a message may include the message's sending time, sender, recipient, and the session in which it occurred. The corresponding event record for a schedule may include the schedule's start time, end time, participating members, etc. The corresponding event record for an email may include the email's start time, sender, recipient, etc. For ease of discussion, the embodiments of this disclosure will primarily use messages as examples in the following description.

[0052] In some embodiments, the component runtime platform 110 can use the triggering module 211 to identify at least one event related to the workflow. The triggering module 211 can perform basic checks using some parameters (e.g., rate limiting, parameter validation, request deduplication, etc.) to verify whether the detected event meets the preset triggering conditions. For example, in process 300, the component runtime platform 110 delivers (311) the received message event 210-1 to the filter 212. The component runtime platform 110 uses the filter 212 to perform primary and / or secondary filtering and classification (312) of the message event 210-1 according to the user-set rules. That is, the filter 212 can identify at least one message event 210-1 related to the workflow. Other types of events and message events have similar processing methods, so they will not be described in detail.

[0053] In some embodiments, the component runtime platform 110 stores corresponding event records for at least one event. Continuing the example above, in process 300, for at least one filtered and categorized message event 210-1, the component runtime platform 110 stores (313) the corresponding event records corresponding to at least one message event 210-1 to the accumulator module 213 according to the configuration. In some embodiments, the accumulator module 213 may store event record information in at least one dimension. In some embodiments, at least one dimension may include a first dimension and a second dimension. The first dimension may indicate event record usage time information, usage frequency information, or the time information that triggers the output of prompt information to the target model based on the event record. The second dimension may indicate event occurrence time information, occurrence frequency information, or event storage time information, or event storage number information. In some examples, the accumulator module 213 may include a storage device for storing corresponding event records for at least one event.

[0054] In some embodiments, the component runtime platform 110 updates the event log information for the workflow based on the storage of corresponding event logs. In some embodiments, the event log information indicates event occurrence information or event log usage information for any workflow from at least one dimension. In some examples, the component runtime platform 110 updates the event log information for the workflow based on its stored corresponding event logs for at least one event related to the workflow in at least one workflow. For example, updating the event occurrence information or event log usage information of the workflow, etc.

[0055] In some embodiments, the component runtime platform 110 determines a target workflow from at least one workflow based on event log information corresponding to at least one workflow. In some examples, the component runtime platform 110 can determine the target workflow from at least one workflow using the accumulator module 213 based on the event log information corresponding to at least one workflow. In some examples, the component runtime platform 110 can also determine multiple workflows, such as X workflows, from at least one workflow using the accumulator module 213 based on event logs stored for at least one workflow.

[0056] The accumulator module 213 is configured to aggregate events (i.e., store multiple event records) and, after slicing the events, invoke subsequent nodes (e.g., machine learning models) for processing. For example, it can slice the data corresponding to a streaming message into n slices, each slice potentially including 10 or 20 messages. Similarly, it can slice the received workflow-related email data into n slices, each slice potentially including 10 or 20 emails. This improves the processing efficiency for events (e.g., a single message) and reduces additional costs caused by performance limitations such as the amount of data transmitted per transmission by the machine learning model. The component runtime platform 110 will be described in detail below as determining the target workflow from at least one workflow.

[0057] In some embodiments, the component runtime platform 110 uses event records stored for the target workflow to determine actions for the target workflow. In example architecture 200, after execution by the accumulator module 213, the component runtime platform 110 uses the process scheduler 214 to schedule the sliced ​​events. Accordingly, the component runtime platform 110 calls the action module 215 to process the sliced ​​events to determine actions for the target workflow. For example, if the event record stored in the workflow is "Today's sales revenue for product XX is XXX", then the action corresponding to the workflow could be to add this record to the "Sales Revenue" table.

[0058] In some embodiments, the component runtime platform 110 uses event logs stored for at least one workflow to determine actions for at least one workflow. For ease of understanding, the following description uses one of the at least one workflows as an example. For a target workflow in the at least one workflow, the component runtime platform 110 generates prompt information for a target model based on the event logs stored for the target workflow. Then, the component runtime platform 110 determines a target action for the target workflow based on the output of the target model. In some embodiments, the component runtime platform 110 can determine the target action based on the output of the target model and the action configuration information of the target workflow. In some embodiments, the component runtime platform 110 can determine the relationship between the event logs stored for the target workflow and the target workflow based on the output of the target model. Then, the component runtime platform 110 determines the target action based on the relationship and the action configuration information of the target workflow.

[0059] In example architecture 200, component runtime platform 110 generates prompts for the target model based on the event records corresponding to the target workflow. That is, component runtime platform 110 can invoke a machine learning model to process event slices. Correspondingly, component runtime platform 110 can also tag each sliced ​​event to record and store the relationship between the target workflow and its corresponding event records. For example, tagging can include identifying whether an event belongs to a key node of the workflow and which key node it belongs to. Then, component runtime platform 110 determines the actions for the target workflow (e.g., sending a message, adding a table record, invoking a model, etc.) based on the relationship between the target workflow and its corresponding event records, as well as the action configuration information of the target workflow (e.g., rate limiting, parameter validation, request deduplication, etc.).

[0060] Continuing with process 300, the component runtime platform 110 stores (313) the corresponding event record corresponding to at least one filtered and categorized message event 210-1 to the accumulator module 213 according to the configuration. For each workflow, the accumulator module 213 can maintain trigger time dimension, event number dimension, event time dimension, etc. In some examples, the target workflow is selected based on the trigger time dimension, event number dimension, and event time dimension. The component runtime platform 110 can pop up the corresponding event record according to the identity identifier (ID) of the target workflow and deliver (315) it to the process scheduler 214.

[0061] Subsequently, the process scheduler 214 triggers (316) the consumption of message event 210-1, thereby initiating action module 215. In some examples, a trigger can be initiated at regular intervals, when the number of messages stored in the workflow accumulates to m, when the number of received emails reaches a certain number, or when a suitable slice is reached. Then, in response to the trigger event, action module 215 calls (317) model interface 216, which returns (318) the result to action module 215. Then, action module 215 executes the action for the target workflow.

[0062] The following describes how the component runtime platform 110 determines a target workflow from at least one workflow. In some embodiments, the component runtime platform 110 determines a corresponding score for each workflow in at least one dimension. Then, the component runtime platform 110 selects the target workflow from the at least one workflow based on the corresponding scores for each of the at least one dimensions determined for the at least one workflow.

[0063] In some examples, the component runtime platform 110 determines a score for each workflow in at least one dimension for any one of the at least one workflows. Then, the component runtime platform 110 selects a target workflow from the at least one workflow based on the corresponding scores of each workflow in at least one dimension.

[0064] As described above, the accumulator module 213 can store and maintain event log information in at least one dimension. In some embodiments, the first dimension may include a trigger time dimension, and the second dimension may include an event count dimension and an event time dimension. In some embodiments, the trigger time dimension indicates the next trigger time using the event log. For example, the trigger time dimension may indicate the start time of the window. In some examples, when each event log is temporarily inserted into the time window, the component runtime platform 110 may set the corresponding score of the time window to the current time, or the time of the next trigger and the time window size.

[0065] In some embodiments, the event count dimension indicates the number of event records that have been stored but not used. For example, the event count dimension may indicate the cumulative number of event records. In some examples, the component runtime platform 110 may perform an accumulation of the corresponding score on the event count dimension as each event record is inserted into a time window for temporary storage. In some embodiments, the event time dimension indicates the storage time or occurrence time of the most recent event record stored for the workflow. For example, the event time dimension may indicate the storage time of the previous event (e.g., the previous message).

[0066] In some examples, for each workflow in at least one workflow, the component runtime platform 110 can obtain a score for each workflow (e.g., workflow A) in each dimension of at least one dimension. Subsequently, the component runtime platform 110 weights the corresponding scores of workflow A in each dimension to determine a weighted score for workflow A, and then selects one or more workflows from the at least one workflow.

[0067] The above describes example processing for a single workflow. The following describes an example implementation of target workflow selection. In some embodiments, the component runtime platform 110 can also select a target workflow from at least one workflow in the following manner.

[0068] For at least one dimension, the component runtime platform 110 selects a predetermined number of workflows from at least one workflow based on the corresponding score of the at least one workflow in that dimension. In some examples, based on the corresponding score of the at least one workflow in the trigger time dimension, the top J workflows can be extracted from the at least one workflow. Based on the corresponding score of the at least one workflow in the event number dimension, the top K workflows can be extracted from the at least one workflow. Based on the corresponding score of the at least one workflow in the event time dimension, the top L workflows can be extracted from the at least one workflow.

[0069] Subsequently, the component runtime platform 110 determines one or more candidate workflows based on a predetermined number of workflows selected for at least one dimension. In some embodiments, the component runtime platform 110 performs a deduplication operation on the predetermined number of workflows selected for at least one dimension to remove duplicate workflows. Then, the component runtime platform 110 determines one or more candidate workflows based on the result of the deduplication operation. It should be understood that the deduplication operation is not mandatory, and in some embodiments, it may be omitted.

[0070] In some examples, the component runtime platform 110 performs deduplication on J workflows selected based on the trigger time dimension, K workflows selected based on the number of events dimension, and L workflows selected based on the event time dimension. Understandably, the component runtime platform 110 performs deduplication on J+K+L workflows. Then, based on the result of the deduplication operation, the component runtime platform 110 determines one or more candidate workflows.

[0071] Then, the component runtime platform 110 selects the target workflow from one or more candidate workflows. For example, a certain number of workflows, such as N workflows, can be selected from the candidate workflows. Alternatively, the candidate workflows can be divided into multiple batches (e.g., M batches) and used as target workflows in batches. In some embodiments, the component runtime platform 110 may select the target workflow from one or more candidate workflows in the following manner.

[0072] In some embodiments, a priority strategy may be employed to select a target workflow from candidate workflows. Specifically, for each candidate workflow in one or more candidate workflows, the component runtime platform 110 determines a weighted score for that candidate workflow based on the corresponding weights of at least one dimension and the corresponding score of that candidate workflow in at least one dimension. Then, the component runtime platform 110 selects a target workflow from the one or more candidate workflows based on the weighted scores determined separately for each of the one or more candidate workflows.

[0073] In some embodiments, the target workflow is determined from at least one workflow based on the corresponding weights of at least one dimension. In some examples, for each candidate workflow in one or more candidate workflows, the component runtime platform 110 assigns weights based on a priority strategy. In some embodiments, the weight of the dimension indicating event log usage information is greater than the weight of the dimension indicating event occurrence information. In some examples, the dimension indicating event log usage information, i.e., the trigger time dimension, has a high priority, thereby preventing certain content of interest to the user from having a proportion exceeding a threshold. For example, if a group chat session is active, and the score is calculated solely based on the number of events, the session will frequently trigger corresponding actions, leading to excessive resource consumption. To address this uneven resource consumption, the trigger time dimension can also be considered when the number of events processed reaches a predetermined number, preventing the active session from consuming too many resources and causing other events to be delayed. Therefore, to solve this problem, dimensions such as trigger time, number of events, and event time can be weighted according to their corresponding priorities to calculate a final weighted score, which is then used to select the target workflow from one or more candidate workflows.

[0074] In some embodiments, after each window is triggered, the component runtime platform 110 can reset the event records corresponding to the time, number, and session window, respectively, until the next event arrives. Thus, by adopting a priority strategy, resource preemption issues can be resolved, and the requirements for horizontal scaling can also be met. For example, as the number of users increases, the number of workflows and events also increases. The embodiments of this disclosure can flexibly cope with the increase in the number of workflows and events.

[0075] The above describes an example embodiment of selecting a target workflow from at least one workflow for event log slicing based on stored event records. Such at least one workflow can be considered as a group of workflows. In some embodiments, if the component runtime platform 110 needs to manage multiple workflows, these workflows can be divided into multiple groups of workflows. Each group of workflows can be managed and maintained separately as at least one workflow as described above.

[0076] In some embodiments, at least one workflow may be determined as follows: The component runtime platform 110 divides multiple workflows into multiple groups of workflows based on corresponding workflow attributes (e.g., identifier, name, occurrence time, etc.) of a plurality of configured workflows. In some embodiments, the component runtime platform 110 performs a hash operation on the workflow attributes of each of the multiple workflows to obtain an attribute hash value for each workflow. Accordingly, the component runtime platform 110 divides workflows with matching attribute hash values ​​into one group of workflows. Then, the component runtime platform 110 determines each of the multiple groups of workflows as at least one workflow.

[0077] In some examples, the component runtime platform 110 can hash each of multiple workflows using a hash function to obtain a hash value for each workflow. Subsequently, the component runtime platform 110 groups at least one workflow with the same hash value or whose hash value difference is less than a threshold into a single workflow group. Based on this approach, the component runtime platform 110 can divide multiple workflows into P groups, and each group can include an indefinite or predetermined number of workflows, such as 5000. Therefore, the component runtime platform 110 supports a machine learning model capable of executing 5000*P workflows. This approach avoids performance degradation caused by excessively long windows. Correspondingly, it also avoids server downtime (also known as "hot key") due to excessively concentrated traffic reaching the physical network card limit.

[0078] In some cases, because messages are not ordered and there is a delay in arrival, messages that should belong to the previous time window may be stored in the current window. Therefore, to address this, in some embodiments, a time-dimensional watermark for the workflow instance can be maintained based on a reference time. This watermark is used to compare with the message sending time, thereby determining the difference between the message sending time and the watermark. In other words, in this way, the watermark can be used to specify the allowable data delay to determine whether a message is stored.

[0079] Specifically, the component runtime platform 110 stores the corresponding event record for at least one event in the following manner: For any one of the at least one workflows, if the component runtime platform 110 identifies a first event related to that workflow, it determines whether a first time difference between the occurrence time of the first event and a reference time is less than a threshold. If the first time difference is less than the threshold, the component runtime platform 110 stores the event record for the first event for that workflow. In some embodiments, the reference time includes the current time.

[0080] In some examples, if the component runtime platform 110 identifies an event related to the workflow, it determines whether the time difference between the event's occurrence time and the waterline is less than a pre-set allowable data delay time configuration. If the time difference is less than the pre-set allowable data delay time configuration, the event record corresponding to the event is stored.

[0081] In some embodiments, for a workflow in at least one workflow, if the component runtime platform 110 identifies a second event related to that workflow, it determines whether a second time difference between the occurrence time of the second event and a reference time is less than a threshold. If the second time difference is greater than the threshold, the event record for storing the second event is discarded.

[0082] In some examples, if the component runtime platform 110 identifies an event related to the workflow, it determines whether the time difference between the event's occurrence time and the waterline is less than a pre-set allowable data delay time configuration. If the time difference is greater than the pre-set allowable data delay time configuration, the event record corresponding to the event is discarded.

[0083] Figure 4 A schematic diagram of an example architecture 400 for determining whether to store an event record based on a water level, according to some embodiments of the present disclosure, is shown. In example architecture 400, when a threshold is set to 1, the determination of whether to store an event record can be made in the following manner. Figure 4 In the example, when event 411, identified as 5 (e.g., time identifier 5), arrives, the water level 412 is 6, as known from example architecture 200. The difference between event 411 (time identifier 5) and water level 412 is 1. This difference is less than or equal to a pre-set threshold of 1. Accordingly, the event record corresponding to event 411 (time identifier 5) can be stored, such as... Figure 4 As shown.

[0084] If the threshold is set to 0.5, the method for determining whether to store an event record can be as follows: When event 411, identified as 5 (e.g., time identifier 5), arrives, as shown in example architecture 200, the waterline 412 is 6, and the difference between event 411 (time identifier 5) and waterline 412 is 1. This difference is greater than the preset difference of 0.5. Accordingly, the event record corresponding to event 411 (time identifier 5) is not stored. However, this is merely exemplary and is not intended to limit the scope of this disclosure. It is understood that events received late due to certain factors (e.g., network connectivity) may no longer be relevant to the current message. Therefore, the problem of delayed event reception can be solved, preventing delayed events (e.g., messages) from interfering with the judgment of the current context in the workflow.

[0085] In summary, this disclosure enables near real-time aggregation of streaming messages based on instant messaging. Accordingly, embodiments of this disclosure ensure no stability impact on downstream processes (e.g., messages, information streams, etc.). For example, since the capacity of a machine learning model is limited, this disclosure reduces additional costs by using a frequency control strategy, setting the number of requests processed per second (QPS) to be less than or equal to the capacity of the machine learning model.

[0086] Example process

[0087] Figure 5 A flowchart of a process 500 for information processing according to some embodiments of the present disclosure is shown. Process 500 can be implemented at the component runtime platform 110. Reference is made below. Figure 1 Describe the process 500.

[0088] In box 510, the component runtime platform 110, for a workflow in at least one workflow, stores a corresponding event record for the at least one event in response to the identification of at least one event associated with the workflow.

[0089] In box 520, the component runtime platform 110 updates the event log information or event log usage information for the workflow in response to the storage of the corresponding event logs.

[0090] In box 530, the component runtime platform 110 uses the event records stored for at least one workflow to determine the action for the target workflow based on the corresponding event record information for at least one workflow.

[0091] In some embodiments, at least one dimension includes at least one of the following: a first dimension indicating event log usage time information, usage frequency information, and event information that triggers the output of prompt information to the target model based on the event log; or a second dimension indicating event occurrence time information, occurrence frequency information, or event storage time information and event storage number information.

[0092] In some embodiments, the first dimension includes a trigger time dimension, which indicates the next trigger time for using the event record, and the second dimension includes at least one of the following: an event number dimension, which indicates the number of stored and unused event records, or an event time dimension, which indicates the storage time or occurrence time of the most recent event record stored for the workflow.

[0093] In some embodiments, using event logs stored separately for at least one workflow to determine actions for at least one workflow includes: determining a target workflow from at least one workflow based on corresponding event log information for at least one workflow; and determining actions for the target workflow using event logs stored for the target workflow.

[0094] In some embodiments, determining a target workflow from at least one workflow includes: determining a corresponding score for the workflow in at least one dimension for the workflow in at least one workflow; and selecting a target workflow from at least one workflow based on the corresponding scores for the at least one dimension determined for the at least one workflow.

[0095] In some embodiments, selecting a target workflow from at least one workflow includes: selecting a predetermined number of workflows from at least one workflow based on a corresponding score of at least one workflow in a dimension; determining one or more candidate workflows based on the predetermined number of workflows selected for each of the at least one dimension; and selecting a target workflow from the one or more candidate workflows.

[0096] In some embodiments, selecting a target workflow from one or more candidate workflows includes: for each candidate workflow in the one or more candidate workflows, determining a weighted score for the candidate workflow based on the corresponding weight of at least one dimension and the corresponding score of the candidate workflow in at least one dimension; and selecting a target workflow from the one or more candidate workflows based on the weighted scores determined for each of the one or more candidate workflows.

[0097] In some embodiments, the target workflow is determined from at least one workflow based on the corresponding weights of at least one dimension, and the weight of the dimension indicating the use of information in event recording is greater than the weight of the dimension indicating information about the occurrence of the event.

[0098] In some embodiments, determining one or more candidate workflows includes: performing a deduplication operation on a predetermined number of workflows selected for at least one dimension to remove duplicate workflows; and determining one or more candidate workflows based on the result of the deduplication operation.

[0099] In some embodiments, storing a corresponding event record for at least one event includes: for a workflow in at least one workflow, in response to identifying a first event associated with the workflow, determining whether a first time difference between the occurrence time of the first event and a reference time is less than a threshold; and in response to the first time difference being less than the threshold, storing an event record for the first event for the workflow.

[0100] In some embodiments, process 500 further includes: for a workflow in at least one workflow, in response to identifying a second event associated with the workflow, determining whether a second time difference between the occurrence time of the second event and a reference time is less than a threshold; and in response to the second time difference being greater than the threshold, discarding the event record of the second event.

[0101] In some embodiments, the reference time includes the current time.

[0102] In some embodiments, at least one workflow is determined by dividing the multiple workflows into multiple groups of workflows based on the corresponding workflow attributes of the configured multiple workflows; and identifying at least one workflow from each of the multiple groups of workflows.

[0103] In some embodiments, dividing multiple workflows into multiple groups of workflows includes: performing a hash operation on the workflow attributes of each of the multiple workflows to obtain an attribute hash value for each workflow; and dividing workflows with matching attribute hash values ​​into a group of workflows within the multiple groups of workflows.

[0104] In some embodiments, using event logs stored separately for at least one workflow to determine an action for at least one workflow includes: generating a prompt message for a target model based on event logs stored for the target workflow for a target workflow in at least one workflow; and determining a target action for the target workflow based on the output of the target model.

[0105] In some embodiments, determining the target action of the target workflow includes: determining the target action based on the output of the target model and the action configuration information of the target workflow.

[0106] In some embodiments, determining the target action for the target workflow includes: determining the relationship between the stored event records for the target workflow and the target workflow based on the output of the target model; and determining the target action based on the relationship and the action configuration information of the target workflow.

[0107] Example devices and equipment

[0108] Figure 6 A schematic structural block diagram of an apparatus 600 for information processing according to certain embodiments of the present disclosure is shown. The apparatus 600 may be implemented as or included in the component runtime platform 110. The various modules / components in the apparatus 600 may be implemented by hardware, software, firmware, or any combination thereof.

[0109] As shown in the figure, the device 600 includes an event log storage module 610, configured to store corresponding event logs for at least one event in response to identifying at least one event related to the workflow. The device 600 also includes a log information update module 620, configured to update event log information or event log usage information for the workflow in response to the storage of the corresponding event logs. The device 600 further includes an event log usage module 630, configured to use the event logs stored for at least one workflow based on the corresponding event log information for the at least one workflow to determine actions for a target workflow.

[0110] In some embodiments, at least one dimension includes at least one of the following: a first dimension indicating event log usage time information, usage frequency information, and time information that triggers the output of prompt information to the target model based on the event log; or a second dimension indicating event occurrence time information, occurrence frequency information, or event storage time information and event storage number information.

[0111] In some embodiments, the first dimension includes a trigger time dimension, which indicates the next trigger time for using the event record, and the second dimension includes at least one of the following: an event number dimension, which indicates the number of event records that have been stored and not used, or an event time dimension, which indicates the storage time or occurrence time of the most recent event record stored for the workflow.

[0112] In some embodiments, the event logging module 630 is further configured to determine a target workflow from at least one workflow based on corresponding event logging information for at least one workflow; and to determine actions for the target workflow using event logs stored for the target workflow.

[0113] In some embodiments, the event logging module 630 is further configured to determine a corresponding score for a workflow in at least one dimension for a workflow in at least one workflow; and to select a target workflow from at least one workflow based on the corresponding scores for the at least one dimension determined for at least one workflow.

[0114] In some embodiments, the event logging module 630 is further configured to, for at least one dimension, select a predetermined number of workflows from at least one workflow based on the corresponding score of at least one workflow in that dimension; determine one or more candidate workflows based on the predetermined number of workflows selected for each of the at least one dimension; and select a target workflow from the one or more candidate workflows.

[0115] In some embodiments, the event logging module 630 is further configured to determine a weighted score for each of the one or more candidate workflows based on the corresponding weights of at least one dimension and the corresponding score of the candidate workflow in at least one dimension; and to select a target workflow from the one or more candidate workflows based on the weighted scores determined for each of the one or more candidate workflows.

[0116] In some embodiments, the target workflow is determined from at least one workflow based on the corresponding weights of at least one dimension, and the weight of the dimension indicating the use of information in event recording is greater than the weight of the dimension indicating information about the occurrence of the event.

[0117] In some embodiments, the event logging module 630 is further configured to perform a deduplication operation on a predetermined number of workflows selected for at least one dimension to remove duplicate workflows; and to determine one or more candidate workflows based on the result of the deduplication operation.

[0118] In some embodiments, the apparatus 600 further includes a recording and storage module configured to, for a workflow in at least one workflow, determine whether a first time difference between the occurrence time of the first event and a reference time is less than a threshold in response to identifying a first event associated with the workflow; and store an event record of the first event for the workflow in response to the first time difference being less than the threshold.

[0119] In some embodiments, the apparatus 600 further includes a discard storage module configured to, for a workflow in at least one workflow, determine whether a second time difference between the occurrence time of the second event and a reference time is less than a threshold in response to identifying a second event associated with the workflow; and discard the storage of the event record of the second event in response to the second time difference being greater than the threshold.

[0120] In some embodiments, the reference time includes the current time.

[0121] In some embodiments, the event logging module 630 is further configured to divide the multiple workflows into multiple groups of workflows based on the corresponding workflow attributes of the configured multiple workflows; and to identify at least one workflow in each of the multiple groups of workflows.

[0122] In some embodiments, the apparatus 600 further includes a workflow partitioning module configured to perform a hash operation on the workflow attributes of each of a plurality of workflows to obtain an attribute hash value for each workflow; and to partition workflows with matching attribute hash values ​​into a group of workflows among a plurality of groups of workflows.

[0123] In some embodiments, the event logging module 630 is further configured to generate a prompt message for a target model based on event logs stored for the target workflow in at least one workflow; and to determine a target action for the target workflow based on the output of the target model.

[0124] In some embodiments, the recording module 630 is also configured to determine the target action based on the output of the target model and the action configuration information of the target workflow.

[0125] In some embodiments, the event logging module 630 is further configured to determine the relationship between the event logs stored for the target workflow and the target workflow based on the output of the target model; and to determine the target action based on the relationship and the action configuration information of the target workflow.

[0126] Figure 7 A block diagram is shown illustrating an electronic device 700 in which one or more embodiments of the present disclosure may be implemented. It should be understood that... Figure 7 The electronic device 700 shown is merely exemplary and should not be construed as limiting the functionality and scope of the embodiments described herein. Figure 7 The electronic device 700 shown can be used to achieve Figure 1 The component runtime platform 110.

[0127] like Figure 7 As shown, electronic device 700 is in the form of a general-purpose electronic device. Components of electronic device 700 may include, but are not limited to, one or more processors or processing units 710, memory 720, storage device 730, one or more communication units 740, one or more input devices 750, and one or more output devices 760. Processing unit 710 may be a physical or virtual processor and is capable of performing various processes according to programs stored in memory 720. In a multiprocessor system, multiple processing units execute computer-executable instructions in parallel to improve the parallel processing capability of electronic device 700.

[0128] Electronic device 700 typically includes multiple computer storage media. Such media can be any accessible media that is accessible to electronic device 700, including but not limited to volatile and non-volatile media, removable and non-removable media. Memory 720 can be volatile memory (e.g., registers, cache, random access memory (RAM)), non-volatile memory (e.g., read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), flash memory), or some combination thereof. Storage device 730 can be removable or non-removable media and can include machine-readable media, such as flash drives, disks, or any other media that can be used to store information and / or data and can be accessed within electronic device 700.

[0129] Electronic device 700 may further include additional removable / non-removable, volatile / non-volatile storage media. Although not explicitly stated... Figure 7 As shown, disk drives for reading from or writing to removable, non-volatile disks (e.g., "floppy disks") and optical disk drives for reading from or writing to removable, non-volatile optical disks can be provided. In these cases, each drive can be connected to a bus (not shown) via one or more data media interfaces. Memory 720 may include computer program product 725 having one or more program modules configured to perform various methods or actions of various embodiments of this disclosure.

[0130] The communication unit 740 enables communication with other electronic devices via a communication medium. Additionally, the functionality of the components of the electronic device 700 can be implemented using a single computing cluster or multiple computing machines capable of communicating via communication connections. Therefore, the electronic device 700 can operate in a networked environment using logical connections to one or more other servers, network personal computers (PCs), or another network node.

[0131] Input device 750 can be one or more input devices, such as a mouse, keyboard, trackball, etc. Output device 760 can be one or more output devices, such as a monitor, speaker, printer, etc. Electronic device 700 can also communicate with one or more external devices (not shown) via communication unit 740 as needed. These external devices include storage devices, display devices, etc., and can communicate with one or more devices that enable user interaction with electronic device 700, or with any device that enables electronic device 700 to communicate with one or more other electronic devices (e.g., network card, modem, etc.). Such communication can be performed via input / output (I / O) interface (not shown).

[0132] According to an exemplary implementation of this disclosure, a computer-readable storage medium is provided that stores computer-executable instructions thereon, wherein the computer-executable instructions are executed by a processor to implement the methods described above. According to an exemplary implementation of this disclosure, a computer program product is also provided, which is tangibly stored on a non-transitory computer-readable medium and includes computer-executable instructions, which are executed by a processor to implement the methods described above.

[0133] Various aspects of this disclosure are described herein with reference to flowchart illustrations and / or block diagrams of methods, apparatuses, devices, and computer program products implemented according to this disclosure. It should be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer-readable program instructions.

[0134] These computer-readable program instructions can be provided to a processing unit of a general-purpose computer, a special-purpose computer, or other programmable data processing apparatus to produce a machine such that, when executed by the processing unit of the computer or other programmable data processing apparatus, they create means for implementing the functions / actions specified in one or more blocks of the flowchart and / or block diagram. These computer-readable program instructions can also be stored in a computer-readable storage medium that causes a computer, programmable data processing apparatus, and / or other device to operate in a particular manner. Thus, the computer-readable medium storing the instructions comprises an article of manufacture that includes instructions for implementing aspects of the functions / actions specified in one or more blocks of the flowchart and / or block diagram.

[0135] Computer-readable program instructions can be loaded onto a computer, other programmable data processing apparatus, or other device to cause a series of operational steps to be performed on the computer, other programmable data processing apparatus, or other device to produce a computer-implemented process, thereby causing the instructions that execute on the computer, other programmable data processing apparatus, or other device to perform the functions / actions specified in one or more boxes of a flowchart and / or block diagram.

[0136] The flowcharts and block diagrams in the accompanying drawings illustrate the architecture, functionality, and operation of possible implementations of systems, methods, and computer program products according to various embodiments of this disclosure. In this regard, each block in a flowchart or block diagram may represent a module, segment, or portion of an instruction, which contains one or more executable instructions for implementing the specified logical function. In some alternative implementations, the functions indicated in the blocks may occur in a different order than those indicated in the drawings. For example, two consecutive blocks may actually be executed substantially in parallel, and they may sometimes be executed in reverse order, depending on the functions involved. It should also be noted that each block in the block diagrams and / or flowcharts, and combinations of blocks in the block diagrams and / or flowcharts, may be implemented using a dedicated hardware-based system that performs the specified function or action, or using a combination of dedicated hardware and computer instructions.

[0137] Various implementations of this disclosure have been described above. These descriptions are exemplary and not exhaustive, nor are they limited to the disclosed implementations. Many modifications and variations will be apparent to those skilled in the art without departing from the scope and spirit of the described implementations. The terminology used herein is chosen to best explain the principles, practical applications, or improvements to technology in the market, or to enable others skilled in the art to understand the various implementations disclosed herein.

Claims

1. An information processing method, comprising: storing, for a workflow of at least one workflow, a respective event record for at least one event associated with the workflow in response to identifying the at least one event, to aggregate event records generated by the at least one workflow over a period of time; updating, in response to the storing of the respective event record, event record information for the workflow, the event record information indicating, from at least one dimension, event occurrence information or event record usage information of the workflow; determining, based on the respective event record information for the at least one workflow, a target workflow from the at least one workflow; and using event records stored for the target workflow over the period of time and a target model to determine an action for the target workflow. 2.The method of claim 1, wherein the at least one dimension comprises at least one of: a first dimension indicating event record usage time information, usage frequency information, or time information triggering output of prompt information based on the event record to the target model; or a second dimension indicating event occurrence time information, occurrence frequency information, or event storage time information, event storage number information. 3.The method of claim 1, wherein determining the target workflow from the at least one workflow comprises: determining, for a workflow of the at least one workflow, a respective score of the workflow in the at least one dimension; and selecting, based on the respective scores of the at least one dimension respectively determined for the at least one workflow, the target workflow from the at least one workflow. 4.The method of claim 1, wherein storing the respective event record for the at least one event comprises: determining, for a workflow of at least one workflow, in response to identifying a first event associated with the workflow, whether a first time difference between an occurrence time of the first event and a reference time is less than a threshold value; and storing, in response to the first time difference being less than the threshold value, an event record of the first event for the workflow. 5.The method of claim 1, wherein the at least one workflow is determined by: dividing a plurality of workflows, which have been configured, into a plurality of groups of workflows based on respective workflow attributes of the plurality of workflows; and determining the plurality of groups of workflows as the at least one workflow respectively. 6.The method of claim 5, wherein dividing the plurality of workflows into the plurality of groups of workflows comprises: performing a hash operation on the workflow attribute of each workflow of the plurality of workflows to obtain an attribute hash value of each workflow; and dividing the workflows with matching attribute hash values into a group of workflows of the plurality of groups of workflows. 7.The method of claim 1, wherein using the event records respectively stored for the target workflow and the target model to determine the action for the target workflow comprises: generating, for a target workflow of the at least one workflow, prompt information provided to the target model based on the event records stored for the target workflow; and ​ ​ ​ ​ ​ determine a target action for the target workflow based on an output of the target model.

8. An apparatus for information processing, comprising: an event record storing module configured to, for a workflow of at least one workflow, in response to identifying at least one event related to the workflow, store a respective event record for the at least one event to aggregate event records generated by the at least one workflow over a period of time; an record information updating module configured to, in response to the storing of the respective event record, update event record information for the workflow, the event record information indicating event occurrence information or event record usage information of the workflow from at least one dimension; a workflow determining module configured to determine a target workflow from the at least one workflow based on respective event record information for the at least one workflow; and an event record usage module configured to use event records stored for the target workflow over the period of time and a target model to determine an action for the target workflow.

9. An electronic device, comprising: at least one processing unit; and at least one memory coupled to the at least one processing unit and storing instructions for execution by the at least one processing unit, the instructions, when executed by the at least one processing unit, cause the electronic device to perform the method according to any one of claims 1 to 7.

10. A computer readable storage medium having stored thereon a computer program, the computer program being executable by a processor to implement the method according to any one of claims 1 to 7. ​

Images