Mail processing method, device and system, intelligent agent, equipment and medium

Abstract

The invention discloses a mail processing method, device and system, an intelligent agent, electronic equipment, a storage medium and a program product, and relates to the field of artificial intelligence, in particular to the technical fields of mail processing, deep learning, large models and the like. According to the specific implementation scheme, in response to a received service mail, intention information of the service mail is determined; under the condition that the intention information represents that the to-be-processed service comprises information processing, obtaining service information based on a service mail; and executing an information processing operation based on the business information and the intention information by calling the large model.

Description

Technical Field

[0001] The present disclosure relates to the field of artificial intelligence technology, particularly to technical fields such as mail processing, deep learning, and large models. Specifically, it relates to a mail processing method, apparatus, system, intelligent agent, electronic device, storage medium, and program product. Background Art

[0002] In the digital age, email has become an important tool for daily communication and work exchanges. Emails used for work exchanges or business processing can also be referred to as business emails. The general characteristic of business emails is that business processing is carried out based on the business information indicated in the business email. With the rapid increase of online businesses, the number of business emails has become larger and larger, and thus it has become more and more difficult to manage business emails. Summary of the Invention

[0003] The present disclosure provides a mail processing method, apparatus, system, intelligent agent, electronic device, storage medium, and program product.

[0004] According to one aspect of the present disclosure, there is provided a mail processing method, including: in response to receiving a business email, determining the intent information of the business email; in the case where the intent information indicates that the business to be processed includes information processing, obtaining business information based on the business email; and by invoking a large model, performing an information processing operation based on the business information and the intent information.

[0005] According to another aspect of the present disclosure, there is provided a mail processing apparatus, including: a response module, configured to determine the intent information of the business email in response to receiving the business email; an information determination module, configured to obtain business information based on the business email in the case where the intent information indicates that the business to be processed includes information processing; and an invocation module, configured to perform an information processing operation by invoking a large model based on the business information and the intent information.

[0006] According to another aspect of the present disclosure, there is provided a mail processing system, including: an intermediate server, configured to obtain candidate emails and determine a candidate email as a business email in the case where the mailing address of the candidate email belongs to an authorized address; and a business server, configured to determine the intent information of the business email in response to receiving the business email from the intermediate server; in the case where the intent information indicates that the business to be processed includes information processing, obtaining business information based on the business email; and by invoking a large model, performing an information processing operation based on the business information and the intent information.

[0007] According to another aspect of the present disclosure, there is provided an agent, including: an input module for receiving input information; a processing module for determining a target task based on the input information received by the input module, determining a large model based on the target task, and obtaining output information by calling the large model to execute the method as described above; and an output module for outputting the output information obtained by the processing module.

[0008] According to another aspect of the present disclosure, there is provided an electronic device, including: at least one processor; and a memory communicatively connected to the at least one processor; wherein the memory stores instructions executable by the at least one processor, and when the instructions are executed by the at least one processor, the at least one processor is enabled to execute the method as described above.

[0009] According to another aspect of the present disclosure, there is provided a non-transitory computer-readable storage medium storing computer instructions, wherein the computer instructions are used to cause the computer to execute the method as described above.

[0010] According to another aspect of the present disclosure, there is provided a computer program product including a computer program, where the computer program implements the method as described above when executed by a processor.

[0011] It should be understood that the content described in this part is not intended to identify the key or important features of the embodiments of the present disclosure, nor is it used to limit the scope of the present disclosure. Other features of the present disclosure will become easily understood through the following description. BRIEF DESCRIPTION OF THE DRAWINGS

[0012] The drawings are used to better understand the solution and do not constitute a limitation to the present disclosure. Among them:

[0013] Figure 1 Schematically shows an exemplary system architecture to which the mail processing method, apparatus, and system according to the embodiments of the present disclosure can be applied;

[0014] Figure 2 Schematically shows a flowchart of the mail processing method according to the embodiments of the present disclosure;

[0015] Figure 3 Schematically shows a schematic diagram of determining intent information according to the embodiments of the present disclosure;

[0016] Figure 4 Schematically shows a schematic diagram of performing information processing operations according to the embodiments of the present disclosure;

[0017] Figure 5 Schematically shows a schematic diagram of performing information processing operations according to another embodiment of the present disclosure;

[0018] Figure 6Schematically shows a block diagram of a mail processing device according to an embodiment of the present disclosure;

[0019] Figure 7 Schematically shows a block diagram of an agent according to an embodiment of the present disclosure; and

[0020] Figure 8 Schematically shows a block diagram of an electronic device suitable for implementing a mail processing method according to an embodiment of the present disclosure. Detailed implementation manners

[0021] The following describes exemplary embodiments of the present disclosure with reference to the accompanying drawings. Various details of the embodiments of the present disclosure are included to assist understanding, and they should be considered merely exemplary. Therefore, those of ordinary skill in the art should recognize that various changes and modifications can be made to the embodiments described herein without departing from the scope and spirit of the present disclosure. Similarly, for clarity and conciseness, descriptions of well-known functions and structures are omitted in the following description.

[0022] Currently, for business mailboxes, such as non-personal mailboxes for handling company business, dedicated business personnel are generally assigned to handle business based on mails. For example, when a business person finds that a new mail has been received in the business mailbox, they view and read the mail, download attachments, process data, and upload the processed data to the system, etc.

[0023] Using business personnel to monitor business mailboxes has the problems of consuming labor costs and possibly increasing the business processing time due to untimely viewing.

[0024] In view of this, embodiments of the present disclosure provide a mail processing method, device, system, electronic device, storage medium, program product, and agent, in order to use a large model to automatically process the to-be-processed business in business mails in an automated manner, improve the processing efficiency, and reduce the labor cost at the same time.

[0025] Figure 1 Schematically shows an exemplary system architecture to which the mail processing method, device, and system according to an embodiment of the present disclosure can be applied.

[0026] It should be noted that Figure 1 The shown is only an example of the system architecture to which the embodiments of the present disclosure can be applied, to help those skilled in the art understand the technical content of the present disclosure, but it does not mean that the embodiments of the present disclosure cannot be used in other devices, systems, environments, or scenarios. For example, in another embodiment, the exemplary system architecture to which the mail processing method and device can be applied may include a terminal device, but the terminal device can implement the mail processing method and device provided by the embodiments of the present disclosure without interacting with the server.

[0027] As Figure 1As shown, the mail processing system architecture 100 according to this embodiment may include an intermediate server 101 and a service server 102. The terminal device 103, the intermediate server 101, and the service server 102 are connected via a network 104. The network 104 is a medium for providing a communication link between the terminal device 103, the intermediate server 101, and the service server 102. The network 104 may include various connection types, such as wired and / or wireless communication links, etc.

[0028] Users can use the terminal device 103 to interact with the intermediate server 101 via the network 104 to receive or send messages, etc. Various communication client applications may be installed on the terminal device 103, such as knowledge reading applications, web browser applications, search applications, instant messaging tools, email clients, and / or social platform software, etc. (for example only).

[0029] The terminal device 103 may be various electronic devices with a display screen and supporting web browsing, including but not limited to smart phones, tablets, laptop computers, and desktop computers, etc.

[0030] The intermediate server 101 can receive the emails sent by the terminal device 103, and when it is determined that the mailing address of the email belongs to an authorized address, determine that the email is a business email, and send the business email to the service server to utilize the service server to perform business processing operations based on the business email.

[0031] The service server 102, in response to receiving the business email from the intermediate server 101, determines the intent information of the business email. When the intent information indicates that the business to be processed includes information processing, based on the business email, obtains business information. By invoking a large model, performs information processing operations based on the business information and the intent information.

[0032] It should be noted that the mail processing method provided by the embodiments of the present disclosure can generally be executed by the service server 102. Correspondingly, the mail processing device provided by the embodiments of the present disclosure can also be set in the service server 102.

[0033] It should be understood that Figure 1 the numbers of the terminal device, the network, the intermediate server, and the service server in

[0034] In the technical solution of the present disclosure, the processing of collection, storage, use, processing, transmission, provision, disclosure, and application, etc. of the user's personal information involved all comply with the provisions of relevant laws and regulations, take necessary confidentiality measures, and do not violate public order and good customs.

[0035] In the technical solutions of the present disclosure, before obtaining or collecting the user's personal information, the user's authorization or consent has been obtained.

[0036] It should be noted that the sequence numbers of the respective operations in the following methods are only used as representations of the operations for description purposes and should not be regarded as indicating the execution order of the respective operations. Unless explicitly stated, the method does not need to be executed exactly in the order shown.

[0037] Figure 2 The flowchart of the email processing method according to an embodiment of the present disclosure is schematically shown.

[0038] As Figure 2 shown, the method includes operations S210 to S230.

[0039] In operation S210, in response to receiving a business email, determine the intent information of the business email.

[0040] In operation S220, in the case where the intent information indicates that the business to be processed includes information processing, based on the business email, obtain business information.

[0041] In operation S230, by invoking a large model, perform information processing operations based on the business information and the intent information.

[0042] A business email, also known as a non-personal email, may refer to an email used for business processing. For example, an email sent to a corporate email account, and the email information is related to business. The email information of a business email may include the email subject, the sender, and the recipient, and may also include the email body and attachments, etc.

[0043] In the technical solutions of the present disclosure, the processing of obtaining, storing, using, processing, transmitting, and applying the business emails involved all comply with the provisions of relevant laws and regulations, obtain the consent or authorization of the user, take necessary confidentiality measures, and do not violate public order and good customs.

[0044] The intent information may be determined based on the email information of the business email. For example, perform intent recognition on the email information to obtain the intent information of the business email. The intent information may represent a business intent. For example, the business type or business theme of the business to be processed. However, it is not limited to this. It may also include other detailed information of the business to be processed, such as the operation process for performing the business to be processed. For example, an operation process of first performing operation A, then performing operation B, and then performing operation C.

[0045] Business information can refer to the relevant information used to process the business to be processed. Taking information processing as an example, business information can refer to the information to be processed. Specifically, taking news extraction-style information processing as an example, corresponding news content can be obtained based on the link address attached in the email body, and the news content can be used as business information. However, it is not limited to this. Taking report statistics-style information processing as an example, the report attachment attached in the business email can be downloaded, and the report attachment can be used as business information.

[0046] An information processing operation can be performed by calling a large model based on business information and intent information. For example, taking the news content as business information and "extracting news summary" as intent information and inputting them into the large model to use the large model to perform the information processing operation of summary generation and output the news summary. Also, for example, taking the report attachment as business information and "statistical workload" as intent information and inputting them into the large model to use the large model for report statistics and obtain the statistical report.

[0047] Using the email processing method provided by the embodiments of the present disclosure, the corresponding business information is obtained according to the intent information, and an information processing operation is performed by calling a large model based on the business information and the intent information. Thus, different information extraction methods can be adopted for different business emails, which can make the matching degree between the extracted business information and the business expected information requirements of the business email higher, improve the accuracy of information extraction, and lay a foundation for subsequent information processing operations. In addition, different information processing operations can be adopted for different business emails, which can make the information processing operation meet the business expected task requirements, realize the flexibility of business processing and reduce the user operation complexity, and improve the convenience of business processing.

[0048] In related examples, based on multiple historical business emails, the general rules of the business to be processed can be determined, such as the acquisition method of business information and the operation process of information processing, and automated operation code can be generated to process business emails using the automated operation code.

[0049] Compared with the method of using running code, the method of using a large model can improve the flexibility of automated processing and avoid the problem of designing multiple different running programs for different processing processes.

[0050] The above provides a general description of the email processing method, and the following will describe each operation in detail.

[0051] According to the embodiments of the present disclosure, for the operation S210 as shown in Figure 2 , determining the intent information of the business email may include operations: taking a screenshot of the email display interface of the business email to obtain an email image. Performing intent recognition on the email image to obtain the intent information.

[0052] For example, a mail image can be input into an intent recognition model to obtain intent information. However, it is not limited to this. It is also possible to perform character recognition on the mail image to obtain the text content in the mail image. Based on the keywords in the text content, intent information can be obtained. Any method that can obtain intent information from the mail image is acceptable.

[0053] Optionally, the mail image may include information such as a mail title, mail body, etc. Based on the screenshot, the text and images displayed in the business mail can be quickly obtained. Optionally, the attachment is additional information. Although the specific content of the attachment cannot be analyzed in the mail image, the attachment title of the attachment can be carried in the mail image.

[0054] According to an embodiment of the present disclosure, by using the screenshot method to obtain a mail image and performing intent recognition on the mail image, it is possible to reduce the operation process of obtaining intent information, while reducing the redundant information volume and improving the recognition accuracy and effectiveness of the intent information.

[0055] The following will be through Figure 3 to further explain the determination of intent information.

[0056] Figure 3 Schematically shows a schematic diagram of determining intent information according to an embodiment of the present disclosure.

[0057] As Figure 3 shown, in response to receiving the business mail 310, open the business mail 310 to enter the mail display interface. Take a screenshot to obtain the mail image 320. The mail image 320 shows the mail subject 321: Statistics of workload, and fill in the workload statistics form. The names of the two attachment documents 322 are: Daily Workload Report and Statistics Form. The mail body 323: Please count the annual workload of each staff member in 2023 according to the Daily Workload Report and summarize it in the statistics form.

[0058] Perform intent recognition on the mail image 320 to obtain the intent information 330. The intent information may include information related to "workload statistics".

[0059] The above has described in detail how to obtain intent information. The following will describe in detail how to obtain business information.

[0060] According to an embodiment of the present disclosure, for the operation S220 as Figure 2 shown, based on the business mail, obtaining business information may include: determining an information acquisition strategy that matches the intent information. According to the method that matches the information acquisition strategy, based on the business mail, obtain business information.

[0061] For example, in the case where the intent information includes "processing link address information", the information acquisition strategy may include obtaining the content corresponding to the link address as the business information.

[0062] For example, in a manner matching the information acquisition strategy, based on business emails, business information can be obtained, which may include: obtaining the link address in the business email. Through the link address, business information is obtained.

[0063] The link address can be matched from the body or attachments of the business email according to the matching rule. Click on the link address to enter the web page corresponding to the link address, and business information is obtained based on the web page content.

[0064] By obtaining business information in the above manner, it can be ensured that in the case where the business information is not directly displayed in the business email, the link address is matched through the matching rule, and then the business information is obtained, thereby improving the convenience and richness of business information acquisition, and further meeting personalized needs.

[0065] According to another embodiment of the present disclosure, the information acquisition strategy may include an artificial intelligence generated content (AIGC) processing strategy.

[0066] In a manner matching the information acquisition strategy, based on business emails, business information can be obtained, which may include: extracting the information in the business email. Performing artificial intelligence generated content processing on the information in the business email to obtain business information.

[0067] For example, in the case where the intent information includes "rewriting or expanding", the simple email information in the business email can be extracted, and the large model is used to perform artificial intelligence generated content processing on the email information, such as rewriting or expanding, to obtain business information.

[0068] By obtaining business information in the above manner, it is possible to fully understand the relevant content of the business to be processed in the case of simple email information, and avoid hallucinations in subsequent information processing operations.

[0069] Optionally, the information acquisition strategy may further include "processing link address information in a generative manner".

[0070] In a manner matching the information acquisition strategy, based on business emails, business information can be obtained, which may include: obtaining the link address in the business email. Through the link address, initial business information is obtained. Performing artificial intelligence generated content processing on the initial business information to obtain business information.

[0071] Thereby, the business relevance of obtaining business information by performing multiple operations can be improved.

[0072] Optionally, information can also be directly extracted from one or more of the email body and attachments as business information, as long as it can support subsequent information processing operations of the large model.

[0073] The above has introduced in detail how to obtain business information, and the following will explain how to perform information processing operations.

[0074] According to an embodiment of the present disclosure, for the operation S230 as Figure 2 shown, by invoking a large model and performing information processing operations based on business information and intent information, it may include: by invoking a large model, determining a target tool based on intent information. Obtaining associated information related to the business object indicated by the business email by invoking the target tool. Updating the associated information to the business information.

[0075] The intent information can be input into the large model, and based on the information processing operations indicated by the intent information and information such as the business type, a target tool is determined.

[0076] The type of the target tool is not limited. For example, it may include functions, deep learning models, knowledge bases, or retrieval engines, etc. As long as it is a target tool that can assist in performing information processing operations. Associated information related to the business object indicated by the business email can be obtained by invoking the target tool.

[0077] For example, the business object indicated by the business email may include object A. The relevant associated information of object A, such as advanced deeds, can be obtained by invoking a target tool such as a retrieval engine, and the advanced deeds are updated to the business information including, for example, the work experience and learning experience of object A, to obtain a complete personal report of person A.

[0078] Also for example, the business object indicated by the business email may include object B. The business workload of object B in the two years of 2023 and 2024 can be counted from the attached report A in the business email by invoking a target tool such as a function, as the associated information related to object B, and the business workload is updated to the corresponding fields of the attached report B to obtain a workload statistical table.

[0079] According to an embodiment of the present disclosure, a target tool matching the intent information is invoked through a large model, and the target tool is used to assist the large model in performing information processing operations. Thus, different types of operation processing are performed through the large model and the target tool, improving the convenience of automated processing and reducing the probability of hallucinations when the large model performs information processing operations.

[0080] Optionally, determining a target tool based on intent information may include: determining a target tool from multiple candidate tools based on the matching degrees between multiple tool description information and the intent information respectively.

[0081] The tool description information of the candidate tool characterizes the tool features of the candidate tool. The tool description information may include one or more of tool type, tool function, and tool attribute.

[0082] The matching degree between the tool description information and the intent information of each candidate tool can be calculated using a large model to obtain multiple matching degrees. The candidate tool with the highest matching degree is determined from multiple candidate tools as the target tool.

[0083] The target tool can be determined by matching the tool function with the intent information. Compared with this, the method of using the tool description information to determine the target tool can utilize the integrity of the tool description information to improve the accuracy of matching, and thus improve the accuracy of tool invocation.

[0084] The following will be through Figure 4 to further explain the execution of information processing operations.

[0085] Figure 4 A schematic diagram of the execution of information processing operations according to an embodiment of the present disclosure is schematically shown.

[0086] As Figure 4 shown, intent recognition is performed on the business email 410 to obtain the intent information 420 "generate a deed commendation report". Based on the intent information 420, the information in the business email 410 is extracted, such as the deed information 430 of object M. Based on the intent information 420, artificial intelligence generative processing is performed on the deed information 430 to obtain the "deed commendation" article 440.

[0087] Optionally, the deed information can be a video attachment attached to the business email. A multi-modal generative large model can be used to process the video attachment to generate a "deed commendation" article.

[0088] As Figure 4 shown, the target tool of calling the knowledge base M420 can be determined by calling the large model M410, and the association information related to object M is obtained from the knowledge base M420. The association information can include information such as the work experience and learning experience of person M. The association information can be combined with the "deed commendation" article 440 to obtain the "deed commendation" report 450.

[0089] The following will illustrate another embodiment of the information processing operation.

[0090] According to another embodiment of the present disclosure, by calling a large model, performing an information processing operation based on business information and intent information may include: generating a business instruction based on the business information and the intent information. Sending the business instruction to the address indicated in the business email.

[0091] In the case where the intention information includes processing by a third party, a business instruction may be generated based on the business information and the intention information, and the business instruction may be sent to the address corresponding to the third party indicated in the business email, so that the third party responds to the business instruction and performs the corresponding business operation.

[0092] For example, business information can be obtained in the following manner: initial business information is obtained through a link address, and the initial business information is desensitized using an artificial intelligence generative processing method to generate business information.

[0093] Based on the business information and the "render as a web page" included in the intent information, a business instruction is generated. The business instruction is sent to the corresponding address of the third party, so that the third party responds to the business instruction, renders the anonymized business information as a web page, and publishes it on the corresponding website.

[0094] By using the embodiments of the present disclosure, a business instruction is generated based on business information and intention information. Based on the business instruction, when the business server is difficult to implement the business function, a complete information processing operation is completed with the help of a third party, thereby improving the application scope of email processing.

[0095] Optionally, the information processing operation may also include the following embodiments.

[0096] For example, taking sensitive content risk monitoring as an example, the content to be published can be used as business information, and the big model can be used to perform content desensitization operations based on the business information of the content to be published and the intention information of content desensitization to obtain the desensitized content. Based on the desensitized content and the intention information including publishing the content, a business instruction is generated, and the business instruction is sent to a third party so that the third party responds to the business instruction, renders the desensitized content and generates the web page.

[0097] For example, taking the generation of training samples as an example, the content of the email, such as news, can be used as business information, based on the news as business information and the intention information of generating training samples, the training samples are obtained by calling the big model, so as to use the training samples to fine-tune the big model.

[0098] The above text has described in detail the execution of information processing operations. The following text will describe an example in which the intent information represents the to-be-processed business including information forwarding.

[0099] According to an embodiment of the present disclosure, when executing Figure 2 After the operation S210 shown, the email processing method may further include the following operation: when the intent information indicates that the business to be processed includes information forwarding, forwarding the information to be forwarded in the business email to the business object indicated by the business email using a communication method matching the intent information.

[0100] The communication method may include one or more of an email communication method, an instant messaging (IM) method, and a telephone communication method.

[0101] The business object indicated by the business email may be determined by the recipient address, but is not limited thereto, and may also be the business object described in the email body. As long as it can represent the object to be forwarded.

[0102] The information type of the information to be forwarded is not limited. For example, it may include one or more of text, video, image, QR code, and audio. As long as the information can be obtained through the business email.

[0103] Using the communication method that matches the intent information, the address book of this communication method can be obtained from the local storage space, such as contact phone numbers, instant messaging accounts, or mailing addresses, etc. Obtain the relevant communication information of the business object from the address book, and forward the information to be forwarded in the business email to the business object according to the communication information of the business object.

[0104] For example, in the software registration application scenario, after the business object registers using the enterprise email, the software provider will send the authentication email to the enterprise email. This enterprise email is not the user's email but the email address of the business email, so the business object cannot directly obtain the relevant information of the authentication email. The information can be used as the information to be forwarded, and the information to be forwarded can be notified to the business object through the IM instant messaging method, so that the business object can register the software according to the received information.

[0105] Using the embodiments of the present disclosure, the communication method can be determined based on the intent information, and the information to be forwarded is forwarded to the business object indicated by the business email according to the corresponding communication method, thereby improving the ability and efficiency of information sharing.

[0106] The following will be through Figure 5 To further explain the execution of information processing operations.

[0107] Figure 5 Schematically shows a schematic diagram of performing an information processing operation according to another embodiment of the present disclosure.

[0108] As Figure 5 shown, determine the intent information 520 of the business email 510. In the case where the intent information 520 represents that the business to be processed includes information processing and information forwarding, the information processing method, communication method, and communication information corresponding to the business object of the information to be forwarded can be determined based on the intent information.

[0109] As Figure 5As shown, an information acquisition strategy is determined based on the information processing method of the information to be forwarded. The business email 510 is processed for information in a manner matching the information acquisition strategy to obtain business information 530. By invoking the large model M510, an information processing operation is performed based on the business information 530 and the intent information 520 to obtain the information to be forwarded 540. According to the communication information of the business object, the information to be forwarded 540 in the business email is forwarded to the business object 550.

[0110] The above further explains the execution of the information processing operation. The following will elaborate on the acquisition method of the email.

[0111] According to an embodiment of the present disclosure, the execution subject of the above email processing method may include a business server. The business server can communicate with an intermediate server. The business email is received by the intermediate server in the following way: Obtain candidate emails. When it is determined that the mailing address of the candidate email belongs to an authorized address, the candidate email is determined as a business email and forwarded. When it is determined that the mailing address of the candidate email belongs to an unauthorized address, the candidate email is sent to the email box corresponding to the mailing address.

[0112] Optionally, the intermediate server can receive candidate emails sent from different clients by setting a common port.

[0113] An authorized address table can be preset. After the intermediate server receives a candidate email, it queries whether the mailing address is an authorized address from the authorized address table. If it is an authorized address, the candidate email is determined as a business email and forwarded to the business server. If it is not an authorized address, the candidate email is determined as a non-business email and the forwarding to the business server is abandoned.

[0114] The intermediate server can use the Simple Mail Transfer Protocol (SMTP) to send it to the business server in the form of a mail flow through the delivery port exposed by the business server.

[0115] In response to receiving a business email, the business server executes the above-mentioned email processing method.

[0116] Using the email processing method provided by the embodiments of the present disclosure, through the intermediate server to perform intermediate screening on the email, it can reduce the difficulty of the built-in email screening module of the business server. In addition, it can improve the parallel operation ability of email screening and business processing, thereby improving the processing and response efficiency of business emails.

[0117] Optionally, the business server can directly receive candidate emails sent by the client. The candidate emails are screened by pulling to obtain business emails.

[0118] Compared with the method of directly pulling candidate emails from the business server, the method of obtaining candidate emails using an intermediate server can improve the response efficiency, avoid delays, and in addition, can reduce the resource occupancy.

[0119] The operations performed by the business server and the operations performed by the intermediate server have been described above. The following will describe the email processing system obtained by combining the business server and the intermediate server.

[0120] An embodiment of the present disclosure provides an email processing system, including: an intermediate server, configured to obtain candidate emails and determine that a candidate email is a business email when it is determined that the mailing address of the candidate email belongs to an authorized address. A business server, configured to execute the email processing method as described above.

[0121] The email processing method as described above may include, for example, in response to receiving a business email from the intermediate server, determining the intent information of the business email. When the intent information indicates that the business to be processed includes information processing, based on the business email, obtaining business information. By invoking a large model, performing an information processing operation based on the business information and the intent information.

[0122] By the mutual cooperation of the intermediate server and the business server, the intermediate server can intercept emails sent to the enterprise email box, and when it is determined that a candidate email is a business email, the business server configured with a large model can execute the corresponding email processing method. By invoking the large model based on the business information and the intent information, performing an information processing operation, thereby being able to adopt different information extraction methods for different business emails, enabling the extracted business information to have a higher matching degree with the business expected information requirements of the business email, improving the accuracy of information extraction, and laying a foundation for subsequent information processing operations. In addition, different information processing operations can be adopted for different business emails, thereby making the information processing operation meet the business expected task requirements, realizing the flexibility of business processing and reducing the user operation complexity, and enhancing the convenience of business processing.

[0123] Figure 6 A block diagram of an email processing device according to an embodiment of the present disclosure is schematically shown.

[0124] As Figure 6 shown, the email processing device 600 includes: a response module 610, an information determination module 620, and an invocation module 630.

[0125] The response module 610 is configured to determine the intent information of a business email in response to receiving the business email.

[0126] An information determination module 620, configured to obtain service information based on a service email when the intent information represents that the service to be processed includes information processing.

[0127] An invocation module 630, configured to perform an information processing operation based on the service information and the intent information by invoking a large model.

[0128] According to an embodiment of the present disclosure, the invocation module includes: a tool invocation sub-module, an association sub-module, and an update sub-module.

[0129] The tool invocation sub-module is configured to determine a target tool based on the intent information by invoking a large model.

[0130] The association sub-module is configured to obtain association information related to the service object indicated in the service email by invoking the target tool.

[0131] The update sub-module is configured to update the association information to the service information.

[0132] According to an embodiment of the present disclosure, the tool invocation sub-module includes: a matching unit.

[0133] The matching unit is configured to determine a target tool from multiple candidate tools based on the matching degrees between multiple tool description information and the intent information, wherein the tool description information of the candidate tool represents the tool features of the candidate tool.

[0134] According to an embodiment of the present disclosure, the invocation module includes: an instruction generation sub-module and an instruction sending sub-module.

[0135] The instruction generation sub-module is configured to generate a service instruction based on the service information and the intent information by invoking a large model.

[0136] The instruction sending sub-module is configured to send the service instruction to the address indicated in the service email.

[0137] According to an embodiment of the present disclosure, the information determination module includes: an extraction sub-module and an artificial generation sub-module.

[0138] The extraction sub-module is configured to extract information from the service email.

[0139] The artificial generation sub-module is configured to perform artificial intelligence generative processing on the information in the service email to obtain service information.

[0140] According to an embodiment of the present disclosure, the information determination module includes: an address acquisition sub-module and an information acquisition sub-module.

[0141] The address acquisition sub-module is configured to acquire a link address in the service email.

[0142] An information acquisition sub-module, configured to acquire service information through a link address.

[0143] According to an embodiment of the present disclosure, the mail processing device further includes: an information forwarding module.

[0144] The information forwarding module is configured to, when the intent information indicates that the service to be processed includes information forwarding, use a communication method matching the intent information to forward the information to be forwarded in the service mail to the service object indicated by the service mail.

[0145] According to an embodiment of the present disclosure, the service mail is received by using an intermediate server in the following manner: acquiring candidate mails. When it is determined that the mailing address of the candidate mail belongs to an authorized address, determining the candidate mail as a service mail and forwarding it.

[0146] According to an embodiment of the present disclosure, the response module includes: a screenshot sub-module and an intent recognition sub-module.

[0147] The screenshot sub-module is configured to take a screenshot of the mail display interface of the service mail to obtain a mail image.

[0148] The intent recognition sub-module is configured to perform intent recognition on the mail image to obtain intent information.

[0149] Figure 7 A structural block diagram of an agent according to an embodiment of the present disclosure is schematically shown.

[0150] As Figure 7 shown, the AI-based agent 700 may include an input module 710, a processing module 720, and an output module 730.

[0151] The input module 710 is configured to receive input information.

[0152] The processing module 720 is configured to, based on the input information received by the input module, output information by executing the above-mentioned mail processing method.

[0153] The output module 730 is configured to output the output information obtained by the processing module.

[0154] According to an embodiment of the present disclosure, the input module 710 is responsible for receiving or perceiving information such as queries, requests, instructions, signals, or data from the outside world (such as users or the external environment), and converting it into a format that the agent 700 can understand and process. The input module 710 is the primary link for the agent 700 to interact with the outside world. It enables the agent 700 to efficiently and accurately obtain necessary "sensory" information from the outside world and respond to this information.

[0155] In the example, the input module 710 may input the service information and intent information described above.

[0156] In the example, the processing module 720 is the core support for the agent 700 to handle complex tasks. The processing module 720 can execute the email processing method described above.

[0157] In the example, the performance of the processing module 720 can be closely related to the large model on which the agent 700 is based. To give full play to the capabilities of the large model, the internal structure of the processing module 720 can be designed to be highly configurable and extensible to handle various different types of tasks and requirements in real scenarios.

[0158] In the example, after the agent 700 obtains the business information and intent information as input information, the processing module 720 can call the large model to perform information processing operations based on the business information and intent information, obtain output information for the input content, and pass the output information to the output module 730.

[0159] It can be understood that although the large model has excellent language understanding and generation capabilities, like humans, it can solve very limited tasks without any tools. When the agent 700 is given the ability to call tools, it can perform tasks such as completing mathematical operations with the help of a calculator, performing data analysis with the help of the Python language, and obtaining weather forecasts with the help of a search engine.

[0160] In the example, the output module 730 can output the business instructions or updated business information described above.

[0161] The agent 700 according to the embodiments of the present disclosure can simply and effectively improve the degree of intelligence, and improve flexibility and versatility.

[0162] According to the embodiments of the present disclosure, the present disclosure also provides an electronic device, a readable storage medium, and a computer program product.

[0163] According to the embodiments of the present disclosure, an electronic device includes: at least one processor; and a memory communicatively connected to the at least one processor; wherein the memory stores instructions executable by the at least one processor, and the instructions are executed by the at least one processor to enable the at least one processor to execute the method as described above.

[0164] According to the embodiments of the present disclosure, a non-transitory computer-readable storage medium storing computer instructions, wherein the computer instructions are used to cause a computer to execute the method as described above.

[0165] According to the embodiments of the present disclosure, a computer program product includes a computer program that, when executed by a processor, implements the method as described above.

[0166] Figure 8 FIG. shows a schematic block diagram of an exemplary electronic device 800 that can be used to implement embodiments of the present disclosure. The electronic device is intended to represent various forms of digital computers, such as, laptop computers, desktop computers, workstations, personal digital assistants, servers, blade servers, mainframe computers, and other suitable computers. The electronic device can also represent various forms of mobile devices, such as, personal digital processors, cellular phones, smart phones, wearable devices, and other similar computing devices. The components shown herein, their connections and relationships, and their functions are merely exemplary and are not intended to limit the implementation of the present disclosure described and / or claimed herein.

[0167] As Figure 8 shown, the device 800 includes a computing unit 801 that can perform various appropriate actions and processes according to a computer program stored in a read-only memory (ROM) 802 or a computer program loaded from a storage unit 808 into a random access memory (RAM) 803. In the RAM 803, various programs and data required for the operation of the device 800 can also be stored. The computing unit 801, the ROM 802, and the RAM 803 are connected to each other via a bus 804. An input / output (I / O) interface 805 is also connected to the bus 804.

[0168] A plurality of components in the device 800 are connected to the input / output (I / O) interface 805, including: an input unit 806, such as a keyboard, a mouse, etc.; an output unit 807, such as various types of displays, speakers, etc.; a storage unit 808, such as a magnetic disk, an optical disk, etc.; and a communication unit 809, such as a network card, a modem, a wireless communication transceiver, etc. The communication unit 809 allows the device 800 to exchange information / data with other devices via a computer network such as the Internet and / or various telecommunication networks.

[0169] The computing unit 801 can be various general-purpose and / or special-purpose processing components with processing and computing capabilities. Some examples of the computing unit 801 include, but are not limited to, a central processing unit (CPU), a graphics processing unit (GPU), various dedicated artificial intelligence (AI) computing chips, various computing units running machine learning model algorithms, a digital signal processor (DSP), and any suitable processor, controller, microcontroller, etc. The computing unit 801 executes the various methods and processes described above, such as the mail processing method. For example, in some embodiments, the mail processing method can be implemented as a computer software program tangibly embodied in a machine-readable medium, such as the storage unit 808. In some embodiments, part or all of the computer program can be loaded and / or installed onto the device 800 via the ROM 802 and / or the communication unit 809. When the computer program is loaded into the RAM 803 and executed by the computing unit 801, one or more steps of the mail processing method described above can be executed. Alternatively, in other embodiments, the computing unit 801 can be configured to execute the mail processing method by any other suitable means (e.g., by means of firmware).

[0170] The various embodiments of the systems and techniques described above in this document can be implemented in digital electronic circuitry, integrated circuit systems, field programmable gate arrays (FPGA), application specific integrated circuits (ASIC), application specific standard products (ASSP), systems on a chip (SOC), complex programmable logic devices (CPLD), computer hardware, firmware, software, and / or combinations thereof. These various embodiments can include: being implemented in one or more computer programs that can be executed and / or interpreted on a programmable system including at least one programmable processor, which can be a dedicated or general-purpose programmable processor that receives data and instructions from a storage system, at least one input device, and at least one output device, and transmits the data and instructions to the storage system, the at least one input device, and the at least one output device.

[0171] The program code for implementing the methods of the present disclosure can be written in any combination of one or more programming languages. These program codes can be provided to the processor or controller of a general-purpose computer, a special-purpose computer, or other programmable data processing device, such that when the program codes are executed by the processor or controller, the functions / operations specified in the flowcharts and / or block diagrams are implemented. The program code can be executed entirely on the machine, partially on the machine, as a stand-alone software package partially on the machine and partially on a remote machine, or entirely on a remote machine or server.

[0172] In the context of this disclosure, a machine-readable medium can be a tangible medium that can contain or store a program for use by or in connection with an instruction execution system, apparatus, or device. A machine-readable medium can be a machine-readable signal medium or a machine-readable storage medium. A machine-readable medium can include, but is not limited to, electronic, magnetic, optical, electromagnetic, infrared, or semiconductor systems, apparatus, or devices, or any suitable combination of the foregoing. More specific examples of a machine-readable storage medium would include an electrical connection based on one or more wires, a portable computer diskette, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or Flash memory), an optical fiber, a portable compact disc read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the foregoing.

[0173] In order to provide interaction with a user, the systems and techniques described herein can be implemented on a computer having: a display device (e.g., a CRT (cathode ray tube) or LCD (liquid crystal display) monitor) for displaying information to the user; and a keyboard and a pointing device (e.g., a mouse or a trackball) by which the user can provide input to the computer. Other kinds of devices can also be used to provide interaction with the user; for example, the feedback provided to the user can be any form of sensory feedback (e.g., visual feedback, auditory feedback, or tactile feedback); and input from the user can be received in any form (including acoustic input, speech input, or tactile input).

[0174] The systems and techniques described herein can be implemented in a computing system that includes back-end components (e.g., as a data server), or a computing system that includes middleware components (e.g., an application server), or a computing system that includes front-end components (e.g., a user computer having a graphical user interface or a web browser through which the user can interact with an implementation of the systems and techniques described herein), or a computing system that includes any combination of such back-end components, middleware components, or front-end components. The components of the system can be interconnected by any form or medium of digital data communication (e.g., a communication network). Examples of communication networks include: a local area network (LAN), a wide area network (WAN), and the Internet.

[0175] A computer system can include a client and a server. The client and the server are generally remote from each other and typically interact through a communication network. The client-server relationship is generated by computer programs running on the respective computers and having a client-server relationship with each other. The server can be a cloud server, a server of a distributed system, or a server incorporating a blockchain.

[0176] It should be understood that the various forms of processes shown above can be used, with steps reordered, added or deleted. For example, the steps described in this disclosure can be executed in parallel, sequentially, or in different orders, as long as the desired results of the technical solutions disclosed in this disclosure can be achieved, and no limitations are imposed herein.

[0177] The above specific embodiments do not constitute a limitation on the protection scope of this disclosure. Those skilled in the art should understand that various modifications, combinations, sub - combinations, and substitutions can be made according to design requirements and other factors. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of this disclosure shall be included within the protection scope of this disclosure.

Claims

1. A mail processing method, comprising: In response to receiving a business email, determining intent information of the business email; In a case where the intention information indicates that the to-be-processed business includes information processing, obtaining business information based on the business email; as well as By calling the big model, an information processing operation is performed based on the business information and the intention information.

2. The method according to claim 1, wherein: The performing of information processing operations based on the business information and the intention information by calling the big model includes: Determine a target tool based on the intention information by calling the large model; Acquiring association information related to the business object indicated by the business email by calling the target tool; and The associated information is updated into the business information.

3. The method according to claim 2, wherein: The determining of the target tool based on the intention information includes: Based on the matching degree between a plurality of tool description information and the intention information respectively, the target tool is determined from a plurality of candidate tools, wherein the tool description information of the candidate tools represents the tool features of the candidate tools.

4. The method according to claim 1, wherein: The performing of information processing operations based on the business information and the intention information by calling the big model includes: Generate a business instruction based on the business information and the intention information by calling the big model; and The business instruction is sent to the address indicated by the business email.

5. The method according to any one of claims 1 to 4, wherein: The obtaining of business information based on the business email includes: extracting information from the business email; and The information in the business email is processed using artificial intelligence generation to obtain the business information.

6. The method according to any one of claims 1 to 4, wherein: The obtaining of business information based on the business email includes: Obtaining the link address in the business email; and The business information is obtained through the link address.

7. The method according to any one of claims 1 to 6, further comprising: In the case where the intent information indicates that the to-be-processed business includes information forwarding, the to-be-forwarded information in the business email is forwarded to the business object indicated by the business email using a communication method that matches the intent information.

8. The method according to any one of claims 1 to 7, wherein: The business email is received by using the intermediate server in the following manner: Get candidate emails; When it is determined that the mailing address of the candidate mail belongs to the authorized address, the candidate mail is determined to be a business mail and is forwarded.

9. The method according to any one of claims 1 to 8, wherein: The determining of the intention information of the business email includes: Taking a screenshot of the email display interface of the business email to obtain an email image; and Perform intent recognition on the mail image to obtain the intent information.

10. A mail processing device, comprising: A response module, configured to determine the intention information of the business email in response to receiving the business email; An information determination module, configured to obtain business information based on the business email when the intention information indicates that the business to be processed includes information processing; as well as A calling module is used to perform information processing operations based on the business information and the intention information by calling the big model.

11. A mail processing system, comprising: The intermediate server is used to obtain the candidate mail, and determine that the candidate mail is a business mail when it is determined that the mailing address of the candidate mail is an authorized address; as well as A business server, configured to determine the intent information of the business email in response to receiving the business email from the intermediate server; In a case where the intention information indicates that the to-be-processed business includes information processing, obtaining business information based on the business email; And by calling the big model, performing information processing operations based on the business information and the intention information.

12. An intelligent agent, comprising: An input module, used for receiving input information; a processing module, configured to determine a target task based on the input information received by the input module, determine a large model based on the target task, execute the method according to any one of claims 1 to 9 by calling the large model, and output information; An output module is used to output the output information obtained by the processing module.

13. An electronic device comprising: at least one processor; as well as a memory communicatively connected to the at least one processor; wherein, The memory stores instructions that can be executed by the at least one processor, and the instructions are executed by the at least one processor to enable the at least one processor to perform the method according to any one of claims 1 to 9.

14. A non-transitory computer-readable storage medium storing computer instructions, wherein: The computer instructions are used to cause the computer to execute the method according to any one of claims 1-9.

15. A computer program product, comprising a computer program, which, when executed by a processor, implements the method according to any one of claims 1 to 9.

Images