Business processing method, device, equipment, medium and product

Abstract

The invention discloses a service processing method and device, equipment, a medium and a product, and the method comprises the steps: obtaining to-be-processed information when a service processing condition is satisfied, and determining whether an action item exists in the to-be-processed information; if the to-be-processed information exists, extracting the to-be-processed information, and performing data processing on the to-be-processed information to form sending path information and to-be-sent information; and sending the to-be-sent information to a corresponding client based on the action item and the sending path information. According to the technical scheme, automatic information integration and notification are realized, and the office convenience and efficiency are improved.

Description

Technical Field

[0001] The present invention relates to the field of computer technology, and in particular to a business processing method, device, equipment, medium and product. Background Art

[0002] In order to improve office efficiency, existing office systems often integrate functions such as document management, project management, and operation log storage.

[0003] However, this type of office system simply categorizes and stores data generated by the business through a cloud storage platform. The convenience of office work is limited to cloud interconnection, and its convenience needs to be improved. Summary of the invention

[0004] The present invention provides a business processing method, device, equipment, medium and product to realize automatic information integration and notification, and improve office convenience and efficiency.

[0005] In a first aspect, an embodiment of the present disclosure provides a service processing method, including:

[0006] When the business processing conditions are met, obtaining the information to be processed, and determining whether there is an action item in the information to be processed;

[0007] If so, extract the information to be processed, perform data processing on the information to be processed, and form information to be sent;

[0008] The information to be sent is sent to a corresponding client based on the action item.

[0009] In a second aspect, an embodiment of the present disclosure provides a service processing device, including:

[0010] An action item confirmation module is used to obtain the information to be processed and determine whether there is an action item in the information to be processed when the business processing conditions are met;

[0011] An information determination module, used for extracting the information to be processed, if any, and performing data processing on the information to be processed to form information to be sent;

[0012] The information sending module is used to send the information to be sent to the corresponding client based on the action item.

[0013] In a third aspect, an embodiment of the present disclosure provides an electronic device, including:

[0014] at least one processor; and

[0015] a memory communicatively connected to at least one processor; wherein,

[0016] The memory stores a computer program that can be executed by at least one processor, and the computer program is executed by at least one processor so that the at least one processor can execute a business processing method provided by the above-mentioned first aspect embodiment.

[0017] In a fourth aspect, an embodiment of the present disclosure provides a computer-readable storage medium, which stores computer instructions, and the computer instructions are used to enable a processor to implement a business processing method provided by the embodiment of the first aspect above when executed.

[0018] In a fifth aspect, an embodiment of the present disclosure provides a computer program product, which includes a computer program. When the computer program is executed by a processor, it implements a business processing method provided by the embodiment of the first aspect above.

[0019] A business processing method, device, equipment, medium and product of an embodiment of the present invention includes obtaining information to be processed when business processing conditions are met, and determining whether there is an action item in the information to be processed; if so, extracting the information to be processed, performing data processing on the information to be processed, and forming information to be sent; and sending the information to be sent to the corresponding client based on the action item. The above technical solution automatically collects and analyzes text information, realizes automated information integration and notification, and automatically executes actions that require operation, reduces manual operations, reduces production and operation costs, realizes automated business processing, and improves work efficiency.

[0020] It should be understood that the contents described in this section are not intended to identify the key or important features of the embodiments of the present invention, nor are they intended to limit the scope of the present invention. Other features of the present invention will become easily understood through the following description. BRIEF DESCRIPTION OF THE DRAWINGS

[0021] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the following briefly introduces the drawings required for use in the description of the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0022] Figure 1 is a flow chart of a service processing method provided by Embodiment 1 of the present invention;

[0023] Figure 2 is a flow chart of a business processing method provided by Embodiment 2 of the present invention;

[0024] Figure 3 is a structural diagram of a service processing device provided by Embodiment 3 of the present invention;

[0025] Figure 4 It is a structural schematic diagram of an electronic device provided in Embodiment 4 of the present invention. DETAILED DESCRIPTION

[0026] In order to enable those skilled in the art to better understand the scheme of the present invention, the technical scheme in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work should fall within the scope of protection of the present invention.

[0027] It should be noted that the terms "first", "second" and "target" in the specification and claims of the present invention and the above-mentioned drawings are used to distinguish similar objects, and are not necessarily used to describe a specific order or sequence. It should be understood that the data used in this way can be interchanged where appropriate, so that the embodiments of the present invention described herein can be implemented in an order other than those illustrated or described herein. In addition, the terms "including" and "having" and any variations thereof are intended to cover non-exclusive inclusions, for example, a process, method, system, product or device that includes a series of steps or units is not necessarily limited to those steps or units that are clearly listed, but may include other steps or units that are not clearly listed or inherent to these processes, methods, products or devices.

[0028] Embodiment 1

[0029] Figure 1 It is a flowchart of a business processing method provided in the first embodiment of the present invention. This embodiment can be applied to the situation of automatic integration and processing of office messages. The method can be executed by a business processing device, and the business processing device can be implemented in the form of hardware and / or software.

[0030] like Figure 1 As shown, the method includes:

[0031] S101. When a business processing condition is met, obtain information to be processed, and determine whether there is an action item in the information to be processed.

[0032] In this embodiment, the business processing conditions can be understood as conditions for performing automated business processing, which require user authorization, such as the client subscribing to the automated business processing service, the client not subscribing but actively calling the automated business processing service, or the client triggering the automated business processing control (or robot), etc. The information to be processed can be understood as the original information to be processed in the business application, which can include application data of applications such as announcement applications, chat applications, email applications, conference applications, and calendar applications. These application data form information to be processed based on the agreed form, such as chat applications, whose information to be processed includes sender, recipient, sending time, and chat content list; for email applications, their information to be processed also includes sender, recipient, sending time, and a document group composed of subject and body text; for other types of applications, their information to be processed is extracted through similar content, and this embodiment does not limit this. Action items can be understood as actions that need to be performed by business applications, such as push or reminder of conference applications, forwarding of email applications, sending of chat applications, adding schedules of calendar applications, and publishing of announcements of announcement applications. For some texts, there may be no specific actions that need to be performed, and the action of the action item is empty.

[0033] Specifically, when the client subscribes to the automated business processing service, or the client actively calls the automated business processing service or the client triggers the automated business processing control (or robot), it is determined that the user authorization is obtained and the business processing conditions are met, and the automated business processing can be performed. Therefore, the information to be processed based on the agreed form of each business application is obtained in real time through network transmission, and the information to be processed is parsed, identified and semantically analyzed, and it is determined whether there is an action item in the information to be processed based on the results of the semantic analysis.

[0034] It is understandable that after the user cancels the subscription to the automated business processing service, the business processing will no longer be automatically performed and the client will no longer receive such push tasks.

[0035] S102: If it exists, extract the information to be processed, perform data processing on the information to be processed, and form sending path information and information to be sent.

[0036] In this embodiment, the sending path information can be understood as the path used to send the information to be sent, which at least includes the sender and the receiver, and can also include the sending time. The information to be sent can be understood as concise information that can be directly sent to the client after processing and can characterize the entire semantics of the log message in the information to be processed. The information to be sent corresponds to the business application. When the business application is a chat application, the information to be sent is the summary information formed based on the chat content list. When the business application is an email application, the information to be sent is the summary information of the document group composed of the subject and the body text.

[0037] Specifically, if there are action items in the information to be processed, the information to be processed is extracted, and semantic analysis is performed on the information to be processed to determine the sender and receiver. In some cases, the sending time can also be extracted as the sending path information, and the remaining information representing the specific event in the extracted information is determined as the information to be sent.

[0038] S103: Send the information to be sent to the corresponding client based on the action item and the sending path information.

[0039] In this embodiment, the action item is executed based on the sending path information, and the information to be sent is sent to the associated client (terminal) through the corresponding business application.

[0040] For example, if the action item is sending an email through an email application, and the sending path information includes the sender, the recipient, and the sending time, the email application will send the information to be sent to the recipient's mailbox through the sender's mailbox at the sending time. For example, if the action item is pushing a message through a conference application, the conference application will push the message to be sent to the recipient's account through the sender's account at the sending time. The sender and the recipient of the conference application may be the same person, and the information to be sent will be pushed to each terminal (computer and mobile phone, etc.) synchronously through the conference application.

[0041] A business processing method provided by an embodiment of the present invention includes obtaining information to be processed when business processing conditions are met, and determining whether there is an action item in the information to be processed; if so, extracting the information to be processed, performing data processing on the information to be processed, forming sending path information and information to be sent; and sending the information to be sent to the corresponding client based on the action item and the sending path information. The above technical solution automatically collects and analyzes text information, realizes automated information integration and notification, and automatically executes actions that require operation, reduces manual operations, reduces production and operation costs, realizes automated business processing, and improves work efficiency.

[0042] Embodiment 2

[0043] Figure 2 It is a flowchart of a business processing method provided in the second embodiment of the present invention. This embodiment is a further optimization of any of the above embodiments and can be applied to the situation of automatic integration and processing of office messages. The method can be executed by a business processing device, and the business processing device can be implemented in the form of hardware and / or software.

[0044] like Figure 2 As shown, the method includes:

[0045] S201. When a business processing condition is met, obtain information to be processed and extract current keywords in the information to be processed.

[0046] In this embodiment, the current keyword can understand the keyword in the information to be processed identified at the current moment. The keyword is a word with specific semantic information, such as meeting, held a meeting, there is a meeting, conference, hold a meeting, etc. These are all used as keywords for conference applications. For example, send an email, sent an email, log in to the mailbox, receive emails, etc., these are all used as keywords for email applications.

[0047] Specifically, when the business processing conditions are met, the information to be processed based on the agreed form of each business application is obtained in real time through network transmission, and it is identified whether the information to be processed contains keywords. If so, the keywords are located and extracted.

[0048] Exemplarily, the information to be processed is information obtained from an email application, including: "Sender: a, Recipient: b, Sent time: c, Email subject is "Meeting at address e at time d", Email content is "Dear xx, ..., a meeting is scheduled to be held at address e at time d, the meeting topic is xxx, participants include xxxx, etc., please be advised"", thus, it can be identified that the current keyword is "meeting".

[0049] S202: Match the current keyword with a keyword matching library to determine whether there is a target keyword corresponding to the current keyword in the keyword matching library.

[0050] In this embodiment, the keyword matching library can be understood as a pre-built word library for matching business software. The keyword matching library is a key-value pair type keyword word library, with the current keyword-target keyword as the key-value pair, such as meeting-conference, meeting-conference, meeting-conference, etc., where the keyword in the first part of the key-value pair is the extracted current keyword, and the keyword in the second part of the key-value pair is the target keyword. The target keyword can be understood as the name of the business application that needs to perform an action, such as a conference application, an email application, and an announcement application.

[0051] Specifically, a keyword matching library is pre-built to define keyword key-value pairs, the extracted current keyword is identified through the keyword matching library, and it is determined whether there is a target keyword corresponding to the current keyword in the keyword matching library.

[0052] Continuing with the above example, the current keyword is "meeting". The key-value pair corresponding to "meeting" is searched from the keyword matching library, and "meeting-conference" is found. From this, it can be determined that the target keyword "meeting" corresponding to the current keyword exists in the keyword matching library.

[0053] For example, if the current keyword extracted from the chat application is "eating", and there is no key-value pair corresponding to "eating" in the keyword matching library, it is determined that there is no target keyword corresponding to the current keyword in the keyword matching library. At this time, there is no need to execute the next step and exit the automated business processing procedure.

[0054] S203: If it exists, locate the key sentence where the current keyword is located, perform semantic analysis on the key sentence through the action item confirmation model, and determine whether there is an action item in the information to be processed.

[0055] In this embodiment, the keyword can be understood as the sentence where the current keyword is located in the information to be processed. The action item confirmation model can be understood as a pre-trained model for identifying whether there is an action item in a sentence, which is implemented based on natural language processing (NLP) technology. Further, the action item confirmation model is a Transformer-based bidirectional encoder representation (Bidirectional Encoder Representation from Transformers, Bert) model. The Bert model is different from the traditional unidirectional language model or the shallow splicing method of two unidirectional language models for pre-training. Instead, it adopts a masked language model (masked language model, MLM) based on self-supervised learning, so that it can generate deep bidirectional language representation.

[0056] A branch of natural language processing (NLP) is chapter-level document analysis, which identifies the theme and structure of documents, classifies documents, identifies the semantics of documents, and clusters multiple documents. At present, Chinese word segmentation technology is relatively mature, such as Python's Chinese word segmentation library Jieba, whose precise mode can accurately separate sentences and is suitable for text analysis; clustering algorithm, clustering algorithm attempts to classify data close to each other according to their distance, ensuring that the distances between each data in the same class are closest and most similar, and the distances between each data in different classes are farther; the goal of the classification problem is to determine which known sample a new sample belongs to based on certain features of known samples. Classification can be divided into binary classification and multivariate classification. In the present invention, it belongs to multivariate classification when the k value is greater than 2. Machine learning classification algorithms mainly include linear classifiers, support vector machines, kernel estimation, Boosting (enhancement) algorithms, decision trees and neural networks, bagging (methods). Generally speaking, the classification effect of boosting algorithms is better than that of bagging algorithms, and bagging algorithms are better than single classifier algorithms.

[0057] Specifically, if there is a target keyword corresponding to the current keyword in the keyword matching library, the key sentence containing the target keyword is located, and the key sentence is input into the pre-trained action item confirmation model (Bert model). The key sentence is identified through the action item confirmation model to determine the probability of a binary classification problem of whether there is an action item in the key sentence. It is understandable that the output results of the model may be different based on different training samples of the model and different key sentences input in actual applications. For example, the output results may include "whether xx action item exists: yes / no" and "the probability of p action item existing is m, the probability of q action item existing is n,..."

[0058] Exemplarily, the information to be processed is information obtained from an email application, including: "a sent an email to c at time b, and the subject of the email is "Meeting at address e at time d", and the content of the email is "Dear xx, ..., a meeting is planned to be held at address e at time d, the meeting theme is xxx, and the participants include xxxx, etc., please be informed"". The current keywords in the information to be processed are "sent an email" and "meeting", which can match the keywords "email" and "meeting", and locate the key sentence "a sent an email to c at time b" and the key sentence "meeting at address e at time d" where the current keywords "sent an email" are located, or locate the key sentence "a sent an email to c at time b, and the subject of the email is "Meeting at address e at time d" where the current keywords "sent an email" and "meeting" are located.

[0059] The key sentence "a sent an email to c at time b" is input into the action item confirmation model, and the output result is: "Is there an action item for sending email: No"; the key sentence "meeting at address e at time d" is input into the action item confirmation model, and the output result is: "Is there an action item for holding a meeting: Yes"; or, combining the two key sentences, "a sent an email to c at time b, and the email subject is "meeting at address e at time d" is input into the action item confirmation model, and the output result is: "The probability that there is an action item for sending email is 10%, and the probability that there is an action item for holding a meeting is 90%", and then the action item is determined to be a meeting.

[0060] S204: If it exists, extract the information to be processed and determine the sending path information.

[0061] In this embodiment, if it is determined that there is an action item in the information to be processed, the path information to be sent is determined from the information to be processed. For example, if the action item is to send an email, the sending path information corresponds to the sender and the receiver, and may also include the scheduled sending time; if the action item is to hold a meeting, the receiver, that is, the meeting account, may also be the receiving end, such as a mobile phone, and may also include the scheduled push time. The extraction of the sending path information here may also be implemented based on natural language processing technology.

[0062] S205: Determine whether the information to be processed meets the word count threshold.

[0063] In this embodiment, the word count threshold can be understood as a critical value for determining whether the information to be processed needs to be extracted from a summary. The word count threshold can be a fixed value (e.g., 100 words), or can be adjusted based on the action item. For example, the word count threshold corresponding to the meeting action item is 50 words, the word count threshold corresponding to the email action item is 200 words, and the word count threshold corresponding to the announcement action item is 100 words, etc. This embodiment does not set any limitation on this.

[0064] Specifically, based on the definition of the word count threshold, the word count of the information to be processed is compared with the word count threshold. If the word count of the information to be processed is greater than or equal to the word count threshold, it is determined that the information to be processed meets the word count threshold, otherwise, it is determined that it does not meet the word count threshold.

[0065] S206: If satisfied, semantic analysis and summary extraction are performed on the information to be processed through a summary extraction model to form information to be sent.

[0066] In this embodiment, the summary extraction model can be understood as a model for semantic feature extraction, which is a sequence-to-sequence (Seq2Seq) model. The encoder in the Seq2Seq model uses a recurrent neural network to convert the input sequence into a fixed-length context vector, while the decoder uses this vector and another recurrent neural network to gradually generate an output sequence. The working principle of Seq2Seq is the encoder Encoder, which is part of the Seq2Seq model and is responsible for converting the input sequence into a fixed-length context vector.

[0067] Specifically, if the information to be processed meets the word count threshold, it indicates that the information to be processed itself has enough words or has enough words relative to the business application corresponding to the action item. Therefore, it is necessary to extract a summary of the information to be processed to simplify the content and shorten the word count. The information to be processed is input into the well-trained summary extraction model Seq2Seq to obtain the information to be sent output by Seq2Seq, and the word count of the information to be sent is less than the word count threshold.

[0068] S207: If not satisfied, the information to be processed is determined as information to be sent.

[0069] In this embodiment, if the information to be processed does not meet the word count threshold, it means that the number of words in the information to be processed itself is small enough or the number of words is small enough relative to the business application corresponding to the action item. There is no need to streamline the content of the information to be processed or shorten its word count. Therefore, the information to be processed can be directly determined as the information to be sent.

[0070] S208: Send the information to be sent to the corresponding client based on the action item and the sending path information.

[0071] A business processing method provided by an embodiment of the present invention includes obtaining information to be processed and extracting current keywords in the information to be processed when business processing conditions are met; matching the current keywords with a keyword matching library to determine whether there is a target keyword corresponding to the current keyword in the keyword matching library; if so, locating the key sentence where the current keyword is located, and performing semantic analysis on the key sentence through an action item confirmation model to determine whether there is an action item in the information to be processed; if so, extracting the information to be processed and determining the sending path information; judging whether the information to be processed meets the word count threshold; if so, performing semantic analysis and abstract extraction on the information to be processed through an abstract extraction model to form information to be sent; if not, determining the information to be processed as information to be sent; and sending the information to be sent to the corresponding client based on the action item and the sending path information. The above technical solution firstly builds a keyword word library in the modeling process, judges whether there is an action item for the sentence through keyword matching and the Bert model, finds the application category of the action to be executed in the sentence, and uses an algorithm to determine the application type, without manual selection, thereby reducing user operations. Secondly, in order to extract document information, the document summary is extracted in Seq2Seq mode, and the recipients obtained in the data preprocessing part are used to call the application obtained in the keyword matching to complete the action execution. Using deep learning methods, there is no need for manual intervention from analysis to automatic execution, and multiple office business applications in the office system are coordinated and scheduled to automatically determine whether there is action item information that needs to be executed in the office information, and automatically execute it, which greatly reduces the manual operation steps of users in such scenarios, reduces the action execution in the action item scenario in the document, improves the work efficiency of users, reduces production and operation costs, and submission efficiency.

[0072] Optionally, as a first optional embodiment of this embodiment, the method further includes:

[0073] a1. Obtain first sample data and second sample data, wherein the first sample data is a key sentence containing an action item, and the second sample data is a key sentence not containing an action item.

[0074] In this embodiment, a large amount of first sample data containing action items and second sample data not containing action items are obtained as training sample data and verification sample data to implement training of the action item confirmation model.

[0075] b1. Train the initial model based on the first sample data and the second sample data to obtain a fully trained action item confirmation model.

[0076] In this embodiment, the initial model can be understood as an untrained natural language processing model framework, which can be a model based on a neural network or a model constructed in other ways, and this embodiment does not limit this.

[0077] Specifically, the first sample data and the second sample data are mixed and shuffled, and a part is taken as a training sample and a part is taken as a verification sample. The training sample is input into the initial model to obtain an output result. The output result represents whether there is an action item and the training sample actually has an action item. The error of the model is determined, and the model parameters such as the loss function are adjusted. Based on the adjusted model parameters, the second round of iterative training is repeated, and the iterative cycle is repeated until the loss function converges or other parameters meet specific conditions, and the model training is determined to be complete. At this time, the action item confirmation model is verified based on the verification sample, and whether the model can be put into practical application is determined based on the verification result. If not, the action item confirmation model is retrained until the model can be put into practical application.

[0078] It can be understood that the above technical content is a training method for the action item confirmation model. In addition to the action item confirmation model which is pre-trained, the summary extraction model also needs to be pre-trained. Its specific training method is similar to that of the action item confirmation model, except that the training samples are different from the verification samples. The summary extraction model Seq2Seq uses the public data set CNN / Daily Mail data set as the training data required, and does not need to use the collected data.

[0079] Optionally, as a second optional embodiment of this embodiment, the method further includes:

[0080] a2. Obtain the deviation information fed back by the client and analyze the deviation information to determine the cause of the deviation.

[0081] In this embodiment, the deviation information can be understood as information fed back by the client indicating that there is an error in the automated business processing, and the deviation reason can be understood as the reason for the error in the automated business processing, including action items and information to be sent.

[0082] Specifically, after the business process is automatically executed, the client evaluates the automatically executed business process. If a control that indicates satisfaction is triggered, no deviation information is generated. If a control that indicates dissatisfaction is triggered, a comment box pops up for the client to enter deviation information, such as "the information to be sent extracted from the summary is inaccurate" or "the business application to be executed is an email application instead of a conference application", etc. The deviation information fed back by the client is obtained, and the deviation information is automatically identified and parsed through a certain natural language processing algorithm, and the cause of the deviation is determined to be an action item or information to be sent.

[0083] Exemplarily, if the received deviation information is "the summary extracted information to be sent is inaccurate", the deviation cause is determined to be the information to be sent; if the received deviation information is "the business application to be executed is an email application rather than a conference application", the deviation cause is determined to be an action item.

[0084] b2. If the cause of the deviation is an action item, iteratively update the action item confirmation model.

[0085] In this embodiment, if the cause of the deviation is an action item, it is determined that a problem occurs in the processing steps of the action item confirmation model, and the action item confirmation model is retrained to update the action item confirmation model.

[0086] c2. If the deviation is caused by information to be sent, the summary extraction model is iteratively updated.

[0087] In this embodiment, if the cause of the deviation is the information to be sent, it is determined that a problem occurs in the processing step of the summary extraction model, and the summary extraction model is retrained to update the summary extraction model.

[0088] Embodiment 3

[0089] Figure 3 Schematic diagram of the structure of a service processing device provided by Embodiment 3 of the present invention. Figure 3 As shown, the device comprises:

[0090] The action item confirmation module 31 is used to obtain the information to be processed and determine whether there is an action item in the information to be processed when the business processing conditions are met;

[0091] An information determination module 32 is used to extract the information to be processed if it exists, perform data processing on the information to be processed, and form information to be sent;

[0092] The information sending module 33 is used to send the information to be sent to the corresponding client based on the action item.

[0093] The business processing device adopted in this technical solution realizes automatic information integration and notification, and improves office convenience and efficiency.

[0094] Optionally, the action item confirmation module 31 is specifically used for:

[0095] Acquire information to be processed, and extract current keywords from the information to be processed;

[0096] Matching the current keyword with a keyword matching library to determine whether there is a target keyword corresponding to the current keyword in the keyword matching library;

[0097] If so, locate the key sentence where the current keyword is located, perform semantic analysis on the key sentence through an action item confirmation model, and determine whether there is an action item in the information to be processed.

[0098] Optionally, the information determination module 32 is specifically configured to:

[0099] Extracting the information to be processed and determining sending path information;

[0100] Determining whether the information to be processed meets a word count threshold;

[0101] If the conditions are met, semantic analysis and summary extraction are performed on the information to be processed through a summary extraction model to form information to be sent;

[0102] If not satisfied, the information to be processed is determined as information to be sent.

[0103] Optionally, the sending path information includes at least a sender and a receiver.

[0104] Optionally, the device further includes a model training module, which is used to:

[0105] Acquire first sample data and second sample data, wherein the first sample data is a key sentence containing an action item, and the second sample data is a key sentence not containing an action item;

[0106] The initial model is trained based on the first sample data and the second sample data to obtain a fully trained action item confirmation model.

[0107] Optionally, the device further comprises a model iteration module, which is used to:

[0108] Obtaining deviation information fed back by the client and analyzing the deviation information to determine the cause of the deviation;

[0109] If the cause of the deviation is an action item, iteratively update the action item confirmation model;

[0110] If the cause of the deviation is information to be sent, the summary extraction model is iteratively updated.

[0111] The service processing device provided in the embodiment of the present invention can execute the service processing method provided in any embodiment of the present invention, and has the corresponding functional modules and beneficial effects of the execution method.

[0112] Embodiment 4

[0113] Figure 4: is a structural schematic diagram of an electronic device provided in Embodiment 4 of the present invention. The electronic device is intended to represent various forms of digital computers, such as laptop computers, desktop computers, workbenches, personal digital assistants, servers, blade servers, mainframe computers, and other suitable computers. The electronic device may also represent various forms of mobile devices, such as personal digital processing, cellular phones, smart phones, wearable devices (such as helmets, glasses, watches, etc.) and other similar computing devices. The components shown herein, their connections and relationships, and their functions are merely examples and are not intended to limit the implementation of the present invention described and / or required herein.

[0114] like Figure 4 As shown, the electronic device 40 includes at least one processor 41, and a memory connected to the at least one processor 41, such as a read-only memory (ROM) 42, a random access memory (RAM) 43, etc., wherein the memory stores a computer program that can be executed by at least one processor, and the processor 41 can perform various appropriate actions and processes according to the computer program stored in the read-only memory (ROM) 42 or the computer program loaded from the storage unit 48 to the random access memory (RAM) 43. In the RAM 43, various programs and data required for the operation of the electronic device 40 can also be stored. The processor 41, the ROM 42, and the RAM 43 are connected to each other through a bus 44. An input / output (I / O) interface 45 is also connected to the bus 44.

[0115] A number of components in the electronic device 40 are connected to the I / O interface 45, including: an input unit 46, such as a keyboard, a mouse, etc.; an output unit 47, such as various types of displays, speakers, etc.; a storage unit 48, such as a disk, an optical disk, etc.; and a communication unit 49, such as a network card, a modem, a wireless communication transceiver, etc. The communication unit 49 allows the electronic device 40 to exchange information / data with other devices through a computer network such as the Internet and / or various telecommunication networks.

[0116] The processor 41 may be a variety of general and / or special processing components with processing and computing capabilities. Some examples of the processor 41 include, but are not limited to, a central processing unit (CPU), a graphics processing unit (GPU), various special artificial intelligence (AI) computing chips, various processors running machine learning model algorithms, a digital signal processor (DSP), and any appropriate processor, controller, microcontroller, etc. The processor 41 executes the various methods and processes described above, such as a business processing method.

[0117] In some embodiments, the business processing method may be implemented as a computer program, which is tangibly contained in a computer-readable storage medium, such as a storage unit 48. In some embodiments, part or all of the computer program may be loaded and / or installed on the electronic device 40 via the ROM 42 and / or the communication unit 49. When the computer program is loaded into the RAM 43 and executed by the processor 41, one or more steps of the business processing method described above may be performed. Alternatively, in other embodiments, the processor 41 may be configured to execute the business processing method in any other appropriate manner (e.g., by means of firmware).

[0118] Various implementations of the systems and techniques described above herein can be implemented in digital electronic circuit systems, integrated circuit systems, field programmable gate arrays (FPGAs), application specific integrated circuits (ASICs), application specific standard products (ASSPs), systems on chips (SOCs), load programmable logic devices (CPLDs), computer hardware, firmware, software, and / or combinations thereof. These various implementations 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 special purpose or general purpose programmable processor that can receive data and instructions from a storage system, at least one input device, and at least one output device, and transmit data and instructions to the storage system, the at least one input device, and the at least one output device.

[0119] Computer programs for implementing the methods of the present invention may be written in any combination of one or more programming languages. These computer programs may be provided to a processor of a general-purpose computer, a special-purpose computer, or other programmable data processing device, so that when the computer program is executed by the processor, the functions / operations specified in the flow chart and / or block diagram are implemented. The computer program may be executed entirely on the machine, partially on the machine, partially on the machine and partially on a remote machine as a stand-alone software package, or entirely on a remote machine or server.

[0120] In the context of the present invention, a computer-readable storage medium may be a tangible medium that may contain or store a computer program for use by or in combination with an instruction execution system, device or equipment. A computer-readable storage medium may include, but is not limited to, electronic, magnetic, optical, electromagnetic, infrared, or semiconductor systems, devices or equipment, or any suitable combination of the foregoing. Alternatively, a computer-readable storage medium may be a machine-readable signal medium. A more specific example of a machine-readable storage medium may include an electrical connection based on one or more lines, a portable computer disk, 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 disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the foregoing.

[0121] To provide interaction with a user, the systems and techniques described herein may be implemented on an electronic device 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 trackball) through which the user can provide input to the electronic device. Other types of devices may also be used to provide interaction with the user; for example, the feedback provided to the user may be any form of sensory feedback (e.g., visual feedback, auditory feedback, or tactile feedback); and input from the user may be received in any form (including acoustic input, voice input, or tactile input).

[0122] The systems and techniques described herein may be implemented in a computing system that includes backend 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 frontend components (e.g., a user computer with a graphical user interface or a web browser through which a user can interact with implementations of the systems and techniques described herein), or a computing system that includes any combination of such backend components, middleware components, or frontend components. The components of the system may 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), a blockchain network, and the Internet.

[0123] A computing system may include a client and a server. The client and the server are generally remote from each other and usually interact through a communication network. The client and server relationship is generated by computer programs running on the corresponding computers and having a client-server relationship with each other. The server may be a cloud server, also known as a cloud computing server or cloud host, which is a host product in the cloud computing service system to solve the defects of difficult management and weak business scalability in traditional physical hosts and VPS services.

[0124] It should be understood that the various forms of processes shown above can be used to reorder, add or delete steps. For example, the steps described in the present invention can be executed in parallel, sequentially or in different orders, as long as the desired results of the technical solution of the present invention can be achieved, and this document does not limit this.

[0125] The above specific implementations do not constitute a limitation on the protection scope of the present invention. It should be understood by those skilled in the art that various modifications, combinations, sub-combinations and substitutions can be made according to design requirements and other factors. Any modification, equivalent substitution and improvement made within the spirit and principle of the present invention should be included in the protection scope of the present invention.

Claims

1. A business processing method, characterized in that: include: When the business processing conditions are met, obtaining the information to be processed, and determining whether there is an action item in the information to be processed; If so, extract the information to be processed, perform data processing on the information to be processed, and form sending path information and information to be sent; The information to be sent is sent to a corresponding client based on the action item and the sending path information.

2. The method according to claim 1, characterized in that The obtaining of the information to be processed and determining whether there is an action item in the information to be processed includes: Acquire information to be processed, and extract current keywords from the information to be processed; Matching the current keyword with a keyword matching library to determine whether there is a target keyword corresponding to the current keyword in the keyword matching library; If so, locate the key sentence where the current keyword is located, perform semantic analysis on the key sentence through an action item confirmation model, and determine whether there is an action item in the information to be processed.

3. The method according to claim 1, characterized in that The extracting the information to be processed, performing data processing on the information to be processed, and forming sending path information and information to be sent, includes: Extracting the information to be processed and determining sending path information; Determining whether the information to be processed meets a word count threshold; If the conditions are met, semantic analysis and summary extraction are performed on the information to be processed through a summary extraction model to form information to be sent; If not satisfied, the information to be processed is determined as information to be sent.

4. The method according to claim 1, characterized in that The sending path information at least includes a sender and a receiver.

5. The method according to claim 2, characterized in that: Also includes: Acquire first sample data and second sample data, wherein the first sample data is a key sentence containing an action item, and the second sample data is a key sentence not containing an action item; The initial model is trained based on the first sample data and the second sample data to obtain a fully trained action item confirmation model.

6. The method according to claim 2 or 3, characterized in that: Also includes: Obtaining deviation information fed back by the client and analyzing the deviation information to determine the cause of the deviation; If the cause of the deviation is an action item, iteratively update the action item confirmation model; If the cause of the deviation is information to be sent, the summary extraction model is iteratively updated.

7. A service processing device, characterized in that: include: An action item confirmation module is used to obtain the information to be processed and determine whether there is an action item in the information to be processed when the business processing conditions are met; An information determination module, used for extracting the information to be processed, if any, and performing data processing on the information to be processed to form information to be sent; The information sending module is used to send the information to be sent to the corresponding client based on the action item.

8. An electronic device, characterized in that: include: at least one processor; as well as a memory communicatively connected to the at least one processor; wherein, The memory stores a computer program executable by the at least one processor, and the computer program is executed by the at least one processor so that the at least one processor can execute a business processing method according to any one of claims 1 to 6.

9. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores computer instructions, and the computer instructions are used to enable a processor to implement a business processing method according to any one of claims 1 to 6 when executed.

10. A computer program product, characterized in that The computer program product comprises a computer program, and when the computer program is executed by a processor, the computer program implements a service processing method according to any one of claims 1 to 6.

Images