CN113362045A

CN113362045A - Conference schedule generation method and device, electronic equipment and readable storage medium

Info

Publication number: CN113362045A
Application number: CN202110911260.4A
Authority: CN
Inventors: 廖学民
Original assignee: Beijing Mininglamp Software System Co ltd
Current assignee: Beijing Mininglamp Software System Co ltd
Priority date: 2021-08-10
Filing date: 2021-08-10
Publication date: 2021-09-07

Abstract

The present application provides a method, device, electronic device, and readable storage medium for generating a conference schedule, including: in response to a conference initiator's touch operation on a text chat interface, acquiring text chats conducted within a preset time period at least one piece of chat information generated; at least one piece of text information extracted from at least one piece of chat information is input into a pre-trained keyword extraction model to obtain at least one conference keyword and a conference label corresponding to each conference keyword, and Input into the pre-trained meeting content sorting model to obtain the initial meeting content to be initiated initiated by the meeting initiator; based on the initial content of the meeting to be initiated, the target meeting schedule is generated. In the present application, the corresponding target meeting schedule is automatically generated according to the text information extracted from the text chat, which reduces the data processing steps for generating the meeting schedule, and reduces the time for generating the meeting schedule, which helps to improve the generating efficiency of the meeting schedule. .

Description

Conference schedule generation method and device, electronic equipment and readable storage medium

Technical Field

The present application relates to the field of data processing technologies, and in particular, to a method and an apparatus for generating a meeting schedule, an electronic device, and a readable storage medium.

Background

In the daily work arrangement, in order to solve a common problem or to aggregate meeting and communication for different purposes, at this time, a conference needs to be arranged, people are aggregated to discuss a solution for a problem, at the present stage, for aggregation of a conference, a conference initiator needs to search the topic, time, conference place, and the like of the conference to edit and generate a conference schedule after meeting requirements are provided, which needs the conference initiator to search a large amount of data information, spend a large amount of time to screen out conference information related to the conference from the data information, and edit the conference schedule, which needs a large amount of time to generate a conference schedule, how to improve the generation efficiency of the conference schedule, and becomes a problem to be solved urgently.

Disclosure of Invention

In view of the above, an object of the present application is to provide a method and an apparatus for generating a meeting schedule, an electronic device, and a readable storage medium, in which when a meeting initiator performs a touch operation on a text chat interface and initiates a text chat, a corresponding target meeting schedule is generated automatically according to text information extracted from the text chat through a pre-trained keyword extraction model and a pre-trained conference content ranking model, so as to reduce data processing steps for generating the meeting schedule, reduce time for generating the meeting schedule, and help to improve efficiency of generating the meeting schedule.

In a first aspect, an embodiment of the present application provides a method for generating a meeting schedule, where the method includes:

responding to touch operation of a conference initiator on a text chat interface, and acquiring at least one piece of chat information generated by text chat performed by the conference initiator in a preset time period;

inputting at least one piece of text information extracted from the at least one piece of chat information into a pre-trained keyword extraction model to obtain at least one conference keyword and a conference label corresponding to each conference keyword;

inputting the at least one conference keyword and a conference label corresponding to each conference keyword into a pre-trained conference content sequencing model to obtain initial to-be-initiated conference content initiated by the conference initiator;

and generating a target meeting schedule based on the initial meeting content to be initiated.

In a possible implementation manner, the inputting at least one text message extracted from the at least one chat message into a pre-trained keyword extraction model to obtain at least one conference keyword corresponding to a conference initiator and a conference label corresponding to each conference keyword includes:

inputting the at least one piece of text information into a pre-trained keyword extraction model, and extracting at least one text keyword included in each piece of text information;

aiming at each piece of text information, based on a preset semantic judgment rule and a plurality of preset labeling keywords, at least one conference keyword and a conference label corresponding to each conference keyword included in the piece of text information are extracted from the at least one text keyword.

In a possible implementation manner, for each piece of text information, based on a preset semantic judgment rule and a plurality of preset labeling keywords, extracting at least one conference keyword included in the piece of text information and a conference tag corresponding to each conference keyword from the at least one text keyword includes:

and determining the semantic intention of the user outputting the text information based on the preset semantic judgment rule, and if the semantic intention is the target semantic intention, determining at least one conference keyword included in the text information and a conference label corresponding to each conference keyword based on at least one character keyword included in the text information.

In one possible embodiment, for each text keyword, a meeting keyword and a corresponding meeting tag are determined by:

and comparing the character key word with each labeling key word included in the preset plurality of labeling key words, if a labeling key word matched with the character key word exists in the plurality of labeling key words, determining the character key word as a conference key word, and determining a character label corresponding to the matched labeling key word as a conference label corresponding to the character key word.

In a possible implementation manner, inputting the at least one conference keyword and a conference label corresponding to each conference keyword into a pre-trained conference content ordering model to obtain an initial to-be-initiated conference content initiated by the conference initiator, where the method includes:

inputting the at least one conference keyword and a conference label corresponding to each conference keyword into a pre-trained conference content ordering model, and determining a score corresponding to each conference keyword according to a preset score corresponding to each conference label; the preset score represents the relevancy between the corresponding conference label and the conference, and the higher the relevancy between the corresponding conference label and the conference is, the higher the corresponding preset score is;

screening at least one target keyword with the score larger than a preset score threshold value from the plurality of conference keywords;

and determining the initial conference content to be initiated based on the at least one target keyword.

In a possible embodiment, the generating a target meeting schedule based on the initial to-be-initiated meeting content includes:

and responding to the editing operation of the conference initiator aiming at the initial conference content to be initiated, and generating a target conference schedule.

In one possible embodiment, the editing operation includes at least one of:

deleting operation, creating operation and updating operation;

the generating a target meeting schedule in response to the editing operation of the meeting initiator for the initial meeting content to be initiated comprises:

in response to the deletion operation of the conference initiator on the initial to-be-initiated conference content for the conference initiator, determining the conference content, which is obtained by deleting the content selected by the conference initiator from the initial to-be-initiated conference content, as the target conference schedule; or

In response to a newly-built operation of the conference initiator on the conference initiator aiming at the initial conference content to be initiated, determining the conference content which is edited by the conference initiator and added in the initial conference content to be initiated as the target conference schedule; or

And in response to the update operation of the conference initiator on the initial to-be-initiated conference content for the conference initiator, determining the conference content, which is replaced by the content edited by the conference initiator at the corresponding position in the initial to-be-initiated conference content, as the target conference schedule.

In a possible implementation manner, before the generating a target meeting schedule based on the initial to-be-initiated meeting content, the generating method further includes:

and canceling the display of the initial conference content to be initiated in response to the cancellation operation of the conference initiator on the initial conference content to be initiated.

In a possible implementation, the generating method further includes:

responding to the determination operation of the conference initiator aiming at the target conference schedule, and publishing the target conference schedule to at least one user; wherein the user is a participant of the target conference.

In a possible implementation, the generating method further includes:

acquiring the cancellation times of the cancellation operation of the conference initiator aiming at the initial conference content to be initiated in a preset test time period;

if the cancellation times are larger than a preset cancellation time threshold value, collecting at least one piece of character information generated when the conference initiator performs character chatting and at least one piece of updating information generated when the conference initiator performs character chatting after cancellation operation;

updating the keyword extraction model and the conference content ordering model based on the at least one piece of text information and the at least one piece of updating information to obtain an updated keyword extraction model and an updated conference content ordering model;

and analyzing at least one piece of text information generated by text chatting in a preset time period again based on the updated keyword extraction model and the updated conference content sequencing model to obtain a corresponding target conference schedule.

In a possible implementation manner, the keyword extraction model and the conference content ordering model are obtained by training based on a plurality of pieces of labeled data labeled in advance;

the labeled data is labeled based on a preset semantic judgment rule and a preset target keyword library and comprises data of keywords related to the conference initiation.

In a second aspect, an embodiment of the present application further provides a device for generating a meeting schedule, where the device includes:

the system comprises an information acquisition module, a text chat interface and a text chat interface, wherein the information acquisition module is used for responding to touch operation of a conference initiator on the text chat interface and acquiring at least one piece of chat information generated by text chat performed by the conference initiator in a preset time period;

the keyword extraction module is used for inputting at least one piece of text information extracted from the at least one piece of chat information into a pre-trained keyword extraction model to obtain at least one conference keyword and a conference label corresponding to each conference keyword;

the conference content generation module is used for inputting the at least one conference keyword and a conference label corresponding to each conference keyword into a pre-trained conference content sequencing model to obtain initial to-be-initiated conference content initiated by the conference initiator;

and the conference schedule generating module is used for generating a target conference schedule based on the initial to-be-initiated conference content.

In a third aspect, an embodiment of the present application further provides an electronic device, including: a processor, a storage medium and a bus, wherein the storage medium stores machine-readable instructions executable by the processor, when an electronic device runs, the processor and the storage medium communicate with each other through the bus, and the processor executes the machine-readable instructions to execute the steps of the method for generating a meeting schedule according to any one of the first aspect.

In a fourth aspect, the present application further provides a computer-readable storage medium, on which a computer program is stored, where the computer program is executed by a processor to perform the steps of the method for generating a meeting schedule according to any one of the first aspect.

The method, the device, the electronic device and the readable storage medium for generating the conference schedule respond to a touch operation of a conference initiator on a text chat interface, obtain at least one piece of chat information generated when a text chat is performed within a preset time period, correspondingly extract at least one piece of text information from the at least one piece of chat information, input the at least one piece of text information into a pre-trained keyword extraction model, and obtain at least one conference keyword and a conference label corresponding to each conference keyword; and inputting the obtained at least one conference keyword and a conference label corresponding to each conference keyword into a pre-trained conference content sequencing model to obtain initial to-be-initiated conference content initiated by a conference initiator, and obtaining a target conference schedule according to the initial to-be-initiated conference content. According to the method and the device, the corresponding target meeting schedule is generated automatically according to the text information extracted from the text chatting, the data processing steps for generating the meeting schedule are simplified, the time for generating the meeting schedule is reduced, and the generation efficiency of the meeting schedule is improved.

In order to make the aforementioned objects, features and advantages of the present application more comprehensible, preferred embodiments accompanied with figures are described in detail below.

Drawings

In order to more clearly illustrate the technical solutions of the embodiments of the present application, the drawings that are required to be used in the embodiments will be briefly described below, it should be understood that the following drawings only illustrate some embodiments of the present application and therefore should not be considered as limiting the scope, and for those skilled in the art, other related drawings can be obtained from the drawings without inventive effort.

Fig. 1 is a flowchart of a method for generating a meeting schedule according to an embodiment of the present disclosure;

fig. 2 is a flowchart of another method for generating a meeting schedule according to an embodiment of the present disclosure;

FIG. 3 is a schematic diagram of data flow provided by an embodiment of the present application;

FIG. 4 is a flow chart of conference generation and model update provided by an embodiment of the present application;

fig. 5 is a schematic structural diagram of a device for generating a meeting schedule according to an embodiment of the present disclosure;

fig. 6 is a second schematic structural diagram of a conference schedule generating apparatus according to an embodiment of the present application;

fig. 7 is a schematic structural diagram of an electronic device according to an embodiment of the present application.

Detailed Description

In order to make the objects, technical solutions and advantages of the embodiments of the present application clearer, the technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the drawings in the embodiments of the present application, and it is obvious that the described embodiments are only a part of the embodiments of the present application, and not all the embodiments. The components of the embodiments of the present application, generally described and illustrated in the figures herein, can be arranged and designed in a wide variety of different configurations. Thus, the following detailed description of the embodiments of the present application, presented in the accompanying drawings, is not intended to limit the scope of the claimed application, but is merely representative of selected embodiments of the application. Every other embodiment that can be obtained by a person skilled in the art without making creative efforts based on the embodiments of the present application falls within the protection scope of the present application.

First, an application scenario to which the present application is applicable will be described. The method and the device can be applied to the technical field of data processing. It is found through research that in a daily work schedule, in order to solve a common problem or for different purposes, a conference needs to be arranged and discussed and exchanged together, and a way of solving a problem needs to be discussed together.

Based on this, the embodiment of the application provides a method for generating a meeting schedule, so as to improve the generation efficiency of the meeting schedule.

Referring to fig. 1, fig. 1 is a flowchart of a method for generating a meeting schedule according to an embodiment of the present disclosure. As shown in fig. 1, a method for generating a meeting schedule provided in an embodiment of the present application includes:

s101, responding to touch operation of a conference initiator on a text chat interface, and acquiring at least one piece of chat information generated by text chat performed by the conference initiator in a preset time period.

S102, inputting at least one piece of text information extracted from the at least one piece of chat information into a pre-trained keyword extraction model to obtain at least one conference keyword and a conference label corresponding to each conference keyword.

S103, inputting the at least one conference keyword and the conference label corresponding to each conference keyword into a pre-trained conference content sequencing model to obtain the initial to-be-initiated conference content initiated by the conference initiator.

And S104, generating a target conference schedule based on the initial to-be-initiated conference content.

According to the method for generating the conference schedule, the corresponding target conference schedule is generated automatically according to the character information extracted from the character chatting, the data processing steps for generating the conference schedule are simplified, the time for generating the conference schedule is reduced, and the generation efficiency of the conference schedule is improved.

The following describes exemplary steps in an embodiment of the present application:

In the embodiment of the application, in response to a touch operation of a conference initiator on a text chat interface, at least one piece of chat information generated when the conference initiator initiates a text chat within a preset time period is generated.

Here, the scenario addressed in the embodiment of the present application may be a scenario in which a user initiates a text chat, for example, a session initiated in a group chat through enterprise WeChat or other communication software, or a one-to-one session with a specific user.

Here, the touch operation on the text chat interface may be a touch operation performed by the conference initiator through a text sending control included in the text chat interface, or a touch operation performed by the conference initiator on the text chat interface is a touch operation directed to the conference invitation control at a specific position on the text chat interface.

The touch operation may be a click operation, a long-time press operation, or a slide operation.

Here, the preset time period may be a time period that is set to be a time period from when the text chat interface is opened by the conference initiator to when the chat is ended by the conference initiator, and the preset time period is set to ensure timeliness of obtaining the chat information so as to ensure that a subsequently generated conference schedule is the latest conference schedule.

In one possible implementation, the chat message may be a text message in the chat process, or may be a picture message, a link message, or the like in the chat process.

In this embodiment of the application, at least one piece of corresponding text information is extracted from at least one piece of chat information acquired in step S101, and the extracted at least one piece of text information is input into a pre-trained keyword extraction model, so as to extract at least one conference keyword included in the text information and a conference label corresponding to each conference keyword.

Here, the conference label corresponds to a corresponding score to represent the degree of relevance of the corresponding keyword to the conference, and the higher the score is, the higher the degree of relevance of the corresponding keyword to the conference is.

For example, the score of the meeting tag corresponding to the meeting keyword "start" must be higher than the score of the meeting tag of the meeting keyword "start".

Here, for the chat information acquired in step S101, it is necessary to extract the text information to perform subsequent keyword calculation, and for the case that the chat information is the text information, the text information is directly input into a keyword extraction model trained in advance to be extracted; when the chat information is picture information, image recognition and character recognition technologies may be adopted to extract corresponding character information from the picture.

In a possible implementation manner, please refer to fig. 2, and fig. 2 is a flowchart of another conference schedule generation method provided in an embodiment of the present application. As shown in fig. 2, a method for generating a meeting schedule provided in an embodiment of the present application includes:

s201, inputting the at least one piece of text information into a pre-trained keyword extraction model, and extracting at least one text keyword included in each piece of text information.

In the embodiment of the application, the extracted at least one piece of text information is input into a keyword extraction model trained in advance, and at least one text keyword included in each piece of text information is firstly extracted.

Here, for each piece of text information, the text keywords in the piece of text information may be extracted according to semantic and grammar composition rules.

The extracted character keywords can be character keywords directly extracted through semantic and grammar composition rules, and can also be character keywords obviously not meeting keywords which are screened after one round of screening.

For example, in the case of word screening, after splitting a text message "i' e take a meeting at eight points" through semantic and grammar composition rules, keywords "i", "eight points", and "take a meeting" are obtained, where the keyword "i" is obvious and has little relationship with the meeting, and can be directly screened out.

S202, aiming at each piece of text information, extracting at least one conference keyword and a conference label corresponding to each conference keyword from the text information based on a preset semantic judgment rule and a plurality of preset labeling keywords.

In the embodiment of the application, for each piece of extracted text information, according to a preset semantic judgment rule and a plurality of preset labeling keywords, at least one conference keyword included in the piece of text information and a conference label corresponding to each conference keyword are extracted from the at least one text keyword corresponding to the piece of text information.

Here, the preset semantic judgment rule is mainly used to judge the semantic intention to be expressed by the piece of text information, for example, "i want XX", which expresses the requirement intention of the user who inputs the piece of text information, and at this time, the meeting requirement of the user may be extracted, so that a text keyword which really expresses the requirement intention of the user after "i want" is needed.

Here, the conference label is a label when the keyword extraction model is trained in advance, and similarly, the conference label corresponds to a corresponding score to represent the degree of correlation between the corresponding keyword and the conference, and the higher the score is, the higher the degree of correlation between the corresponding keyword and the conference is.

In a possible implementation manner, the step "extracting, for each piece of text information, at least one conference keyword and a conference tag corresponding to each conference keyword included in the piece of text information from the at least one text keyword based on a preset semantic judgment rule and a plurality of preset labeling keywords" includes:

a 1: and determining the semantic intention of the user outputting the text information based on the preset semantic judgment rule, and if the semantic intention is the target semantic intention, determining at least one conference keyword included in the text information and a conference label corresponding to each conference keyword based on at least one character keyword included in the text information.

In the embodiment of the application, the semantic intention of the user in the text information is determined according to a preset semantic judgment rule, if the semantic intention of the user is determined to be the target semantic intention, at least one conference key and a conference label corresponding to each conference key word are determined from at least one character key word included in the text information.

Here, the target semantic intent is an intent to characterize the user's needs, and the target semantic intent may be, for example, "i want," "i need," "reserve," and the like.

Here, after the text information expressing the user's desire intention of inputting the text information is determined, the user's desire may be extracted, and thus, a text keyword which truly expresses the user's desire intention behind the word representing the user's desire is required.

b 1: and comparing the character key word with each labeling key word included in the preset plurality of labeling key words, if a labeling key word matched with the character key word exists in the plurality of labeling key words, determining the character key word as a conference key word, and determining a character label corresponding to the matched labeling key word as a conference label corresponding to the character key word.

In the embodiment of the application, for each character keyword, the character keyword is compared with each labeling keyword included in a plurality of preset labeling keywords, if a keyword matched with the character keyword exists in the plurality of labeling keywords, the character keyword is determined as a conference keyword, and a character tag of the corresponding labeling keyword is determined as a conference tag of the character keyword.

When the keyword extraction model is trained, it is necessary to obtain a plurality of pieces of historical meeting information in different historical time periods, label the keywords in the historical meeting information according to labeling rules, dictionaries, etc., generate labeled keywords, and store the labeled keywords in a corresponding word database.

The labeling of the keywords can be performed in a labeling platform, and the labeled keywords can be words with high occurrence frequency in the historical conference information.

Here, for matching the labeled keyword and the text keyword, each character included in the text keyword may be matched and compared with a labeled character at the same position in the labeled keyword, and if each character included in the text keyword is identical with the labeled character at the same position in the labeled keyword, the text keyword and the labeled keyword may be considered to be identical.

In a possible implementation manner, the step of inputting the at least one conference keyword and a conference label corresponding to each conference keyword into a pre-trained conference content ordering model to obtain an initial conference content to be initiated by the conference initiator includes:

c 1: and inputting the at least one conference keyword and the conference label corresponding to each conference keyword into a pre-trained conference content sequencing model, and determining the score corresponding to each conference keyword according to the preset score corresponding to each conference label.

In the embodiment of the application, at least one conference keyword extracted from the trained keyword extraction model and a conference label corresponding to each conference keyword are input into the conference content sorting model, and a score corresponding to each conference keyword is determined according to a preset score indicated by the conference label corresponding to each conference keyword.

The preset score represents the relevance between the corresponding conference label and the conference, and the higher the relevance between the conference label and the conference, the higher the corresponding preset score.

Here, the method for determining the score corresponding to each conference keyword may be directly assigning the preset score of the corresponding conference label to the conference keyword, or assigning the preset score of the conference label to the conference keyword after multiplying the preset score of the conference label by a scene coefficient according to an actual application scene.

Here, the at least one conference keyword and the conference label corresponding to each conference keyword are input into a pre-trained conference content ranking model, each conference keyword may be sequentially input into the trained conference content ranking model, or all conference keywords may be simultaneously input into the trained conference content ranking model, and the setting may be performed according to a specific application scenario, which is not specifically limited herein.

c 2: and screening at least one target keyword with the score larger than a preset score threshold value from the plurality of conference keywords.

In this embodiment, at least one target keyword with a score larger than a preset score threshold is screened out from the plurality of conference keywords determined in step c1 according to the score corresponding to each conference keyword.

The preset score threshold is set to eliminate the interference items in the conference keywords, so that the defect of keyword screening in the keyword extraction model is overcome, and the accuracy of generating the conference schedule is improved.

In a possible implementation manner, for the screening of the target keywords, the keywords may be sorted in the descending order of the corresponding scores, and a plurality of conference keywords with the sorting positions located before the preset position are determined as the target keywords.

c 3: and determining the initial conference content to be initiated based on the at least one target keyword.

In the embodiment of the present application, according to at least one target keyword determined in step c2, the keywords are arranged and combined to determine the content of the initial conference to be initiated.

Here, an initial conference content template may be set, and the location of different conference contents is preliminarily defined, for example, the conference subject, the conference time, and the conference location are placed at the front position, and the information of the participants may be considered to be placed at the back position.

Here, the initial conference content template may adopt different UI designs according to different requirements, and is not particularly limited herein.

Referring to fig. 3, fig. 3 is a schematic diagram of data flow provided in the embodiment of the present application, including a natural semantic processing system and other systems in the industry, where the natural semantic processing system includes a labeling platform, a training platform, a deployment platform, and a knowledge base, and the labeling platform acquires labeling data, rules, dictionaries, etc. from the knowledge base and returns a labeling result to the knowledge base; the training platform acquires training data from the database and stores the model file back to the knowledge base after training; the deployment platform loads the model file from the knowledge base, and after the model file is processed by the deployment platform, the access log is stored back to the knowledge base; meanwhile, the knowledge base is connected with other systems in the industry to obtain data, and the processed data is returned to the other systems in the industry.

The model trained by the training platform comprises a keyword extraction model and a conference content ordering model, wherein the keyword extraction model and the conference content ordering model are obtained by training a plurality of pre-labeled labeling data, and the labeling data are labeled based on a preset semantic judgment rule and a preset target keyword library and comprise data of keywords related to initiating a conference.

Here, the other systems in the row extract information and label data with access to enterprise WeChat as the main data source of other systems, such as mailbox, Object and Key Result (OKR) system, Customer Relationship Management (CRM) system, etc.

In the embodiment of the application, after the initial content of the conference to be initiated is determined in step S103, updating and optimization may be performed again according to the initial content of the conference to be initiated, and finally, a corresponding target conference schedule is generated.

Here, the manner of sending the initial to-be-initiated conference content to the conference initiator may be to directly display the initial to-be-initiated conference content in a graphical user interface of the conference initiator in a pop-up message mode, or to send the initial to-be-initiated conference content to the conference initiator in the form of an email.

Here, the time for sending the initial to-be-initiated conference content to the conference initiator may be that after the chat of the conference initiator is finished, the conference initiator may not immediately confirm or modify the initial to-be-initiated conference content, and the initial to-be-initiated conference content may be stored in a fixed position and then processed.

Specifically, the initial conference content to be initiated may be sent to the conference initiator, and if the conference initiator does not modify the initial conference content to be initiated, the initial conference content to be initiated may be directly determined as the target conference schedule for publishing.

In a possible implementation manner, a conference initiator may perform an editing operation on an initial conference content to be initiated, and specifically, the step "generating a target conference schedule based on the initial conference content to be initiated" includes:

d 1: and responding to the editing operation of the conference initiator aiming at the initial conference content to be initiated, and generating a target conference schedule.

In one possible embodiment, the editing operation includes at least one of: deleting operation, creating operation and updating operation.

Specifically, the step of "generating a target meeting schedule in response to the editing operation of the meeting initiator on the initial meeting content to be initiated" includes:

(1) in response to the deletion operation of the conference initiator on the initial to-be-initiated conference content for the conference initiator, determining the conference content, which is obtained by deleting the content selected by the conference initiator from the initial to-be-initiated conference content, as the target conference schedule; or

(2) In response to a newly-built operation of the conference initiator on the conference initiator aiming at the initial conference content to be initiated, determining the conference content which is edited by the conference initiator and added in the initial conference content to be initiated as the target conference schedule; or

(3) And in response to the update operation of the conference initiator on the initial to-be-initiated conference content for the conference initiator, determining the conference content, which is replaced by the content edited by the conference initiator at the corresponding position in the initial to-be-initiated conference content, as the target conference schedule.

Here, after receiving the initial content of the conference to be initiated, the conference initiator may update inaccurate content therein, or delete unimportant content therein, or newly create missing content.

The specific operation may be: { new/delete/update } contact information, { new/delete/update } { business/customer } information, { new/delete/update } business management information, and so on.

In a possible embodiment, before the step "generating a target meeting schedule based on the initial to-be-initiated meeting content", the generating method further includes:

e 1: and canceling the display of the initial conference content to be initiated in response to the cancellation operation of the conference initiator on the initial conference content to be initiated.

In the embodiment of the application, when a cancel operation of a conference initiator for the initial to-be-initiated conference content is received, it is determined that the conference initiator does not want to initiate the conference, so that a conference schedule of the conference is cancelled, and the display of the initial to-be-initiated conference content is cancelled.

Here, when the content of the initial conference to be initiated is displayed to the conference initiator, the cancel control can also be displayed at the same time, and when the operation on the cancel control is determined to be received, the conference initiator is determined to cancel the schedule of the current conference.

In a possible real-time mode, the conference initiator may cancel the conference schedule only when there is a sudden time conflict, and therefore, after determining that the conference initiator cancels the conference schedule, the prompt of whether to reserve the conference schedule again may be displayed, and after receiving a confirmation message of the conference initiator, the corresponding conference date is obtained again, and the conference content after the conference date is updated is displayed to the conference initiator to wait for the re-confirmation of the conference initiator.

In a possible implementation, the generating method further includes:

f 1: and responding to the determination operation of the conference initiator aiming at the target conference schedule, and publishing the target conference schedule to at least one user.

In the embodiment of the application, the target meeting schedule is published to at least one user in response to the determination operation of the meeting initiator on the target meeting schedule.

Wherein the user is a participant of the target conference.

Here, similarly, when the content of the initial conference to be initiated is displayed to the conference initiator, the determination control is also displayed at the same time, and when it is determined that the operation on the determination control is received, it is determined that the conference initiator determines to issue the schedule of the current conference.

The determination of the participant of the target conference may be determined according to at least one piece of chat information generated by performing a text chat, and specifically may be a user mentioned by the conference initiator in the chat process, or a user responding to a conference offer made by the conference initiator.

Here, the method for issuing the target meeting schedule to at least one user may be to issue the target meeting schedule to each user by mail, after determining the participants of the target meeting, determine the mailbox of each user from the database, and then issue the target meeting schedule to each user by mailbox, so as to prompt each user to attend at the designated place on time.

In a possible implementation, the generating method further includes:

g 1: and acquiring the canceling times of canceling operation of the conference initiator aiming at the initial conference content to be initiated in a preset test time period.

In the embodiment of the application, the canceling times of canceling operation of the initial conference content to be initiated when a conference initiator receives the initial conference content to be initiated within a preset test time period are obtained.

Here, the keyword extraction model and the conference content ranking model are updated according to the number of times of canceling operation of the conference initiator on the pushed initial conference content to be initiated, so as to obtain a keyword extraction model which is relatively in accordance with the keyword extraction requirement, and generate a conference content ranking model with relatively accurate conference content.

Here, the preset test time period may be set according to a model update requirement, and may be flexibly set and adjusted according to an actual update requirement.

g 2: and if the cancellation times are larger than a preset cancellation time threshold value, collecting at least one piece of text information generated when the conference initiator performs text chatting and at least one piece of updating information generated when the conference initiator performs text chatting after cancellation operation.

In the embodiment of the application, if the fact that the number of times of cancellation is larger than the preset cancellation number threshold value in the preset test time period is determined, at least one piece of text information generated when a conference initiator conducts text chatting and at least one piece of updating information generated when the conference initiator conducts text chatting after cancellation operation are collected.

When the cancel time is determined to be greater than the preset cancel time threshold, it is determined that the generated initial conference content to be initiated cannot meet the requirements of the conference initiator well within a certain time period, even the determined initial conference content to be initiated is wrong, and then the keyword extraction model and the conference content ranking model need to be updated so as to obtain the keyword extraction model which is relatively in line with the keyword extraction requirements and generate the conference content ranking model with relatively accurate conference content.

In a possible implementation manner, it may also be determined that the generated initial content of the conference to be initiated does not meet the requirement of the conference initiator well within a certain time period according to a time interval between the conference initiator and the cancellation operation of the two adjacent initial content of the conference to be initiated, if the time interval is smaller than a preset time threshold, and each model needs to be updated according to the text information of the conference initiator.

Here, after acquiring the text information and the update information of the conference initiator, the information items may be classified in advance, for example, information items that appear repeatedly in the same type, information items that appear continuously in different types, and the like.

g 3: and updating the keyword extraction model and the conference content sequencing model based on the at least one piece of character information and the at least one piece of updating information to obtain an updated keyword extraction model and an updated conference content sequencing model.

In this embodiment of the application, according to the at least one piece of text information determined in step g2 and the at least one piece of update information, the model parameters of the keyword extraction model are updated, so as to obtain an updated keyword extraction model and an updated conference content ranking model.

Here, in the embodiment of the present application, the word stock of the labeled keyword is mainly subdivided, including but not limited to: modifying labeled keywords, deleting labeled keywords and adding labeled keywords in a word stock, and after the word stock labeled with the labeled keywords is updated, performing update training on the keyword extraction model according to all the updated labeled keywords to obtain an updated keyword extraction model;

in a possible implementation manner, the conference content ordering model can be updated, scores corresponding to different conference labels need to be adjusted, and the conference content ordering model is updated according to the adjusted conference keywords and the adjusted conference labels, so that the updated conference content ordering model is obtained.

g 4: and analyzing at least one piece of text information generated by text chatting in a preset time period again based on the updated keyword extraction model and the updated conference content sequencing model to obtain a corresponding target conference schedule.

In this embodiment of the application, according to the updated keyword extraction model and the conference content ranking model obtained in step g3, at least one piece of text information generated by text chatting performed within the preset time period is analyzed again, and a corresponding target conference schedule is generated according to the updated keyword extraction model and the updated conference content ranking model.

After the updated keyword extraction model and the updated conference content ranking model are updated, the number of times of canceling operations of the conference initiator needs to be monitored, and the updated keyword extraction model and/or the updated conference content ranking model are adjusted again, so that the accuracy of the target conference schedule is ensured.

Taking a scene generated by a meeting schedule as an example, describing steps of meeting generation and model update in the technical scheme of the present application, please refer to fig. 4, where fig. 4 is a flowchart of meeting generation and model update provided in the embodiment of the present application, and as shown in fig. 4, the method includes the following steps:

the method comprises the following steps: an invitation generation step:

1. firstly, starting a system to access a data source, illustratively, accessing data by taking enterprise WeChat as a main access data source and taking other system data sources such as a mailbox, an OKR system, a CRM system and the like as auxiliary access data sources;

2. extracting key information by a natural language understanding model (equivalent to a keyword extraction model);

3. data is preprocessed and labeled in a natural language understanding model (equivalent to a keyword extraction model);

4. calculating and sequencing a recommendation model (equivalent to a conference content sequencing model);

5. recommending an invited user terminal product page;

6. storing the data in a database;

step two: user operation and model updating steps:

1. user behavior is obtained, including but not limited to: confirming the meeting, rejecting the meeting, canceling the meeting, modifying the meeting, supplementing contact information, forwarding the meeting, replying, and the like.

2. Collecting user behavior data;

3. judging whether optimization is needed or not from two dimensions of an algorithm effect index and a product operation index;

4. if not, ending the process, and if so, updating the natural language understanding model (equivalent to a keyword extraction model) and the recommendation model (equivalent to a conference content sorting model).

The method for generating the conference schedule, provided by the embodiment of the application, is used for responding to a touch operation of a conference initiator on a text chat interface, obtaining at least one piece of chat information generated when a text chat is performed within a preset time period, correspondingly extracting at least one piece of text information from the at least one piece of chat information, inputting the at least one piece of text information into a pre-trained keyword extraction model, and obtaining at least one conference keyword and a conference label corresponding to each conference keyword; and inputting the obtained at least one conference keyword and a conference label corresponding to each conference keyword into a pre-trained conference content sequencing model to obtain initial to-be-initiated conference content initiated by a conference initiator, and obtaining a target conference schedule according to the initial to-be-initiated conference content. According to the method and the device, the corresponding target meeting schedule is generated automatically according to the text information extracted from the text chatting, the data processing steps for generating the meeting schedule are simplified, the time for generating the meeting schedule is reduced, and the generation efficiency of the meeting schedule is improved.

Based on the same inventive concept, the embodiment of the present application further provides a device for generating a meeting schedule corresponding to the method for generating a meeting schedule, and because the principle of solving the problem of the device in the embodiment of the present application is similar to that of the method for generating a meeting schedule in the embodiment of the present application, the implementation of the device can refer to the implementation of the method, and repeated details are omitted.

Referring to fig. 5 and fig. 6, fig. 5 is a first schematic structural diagram of a device for generating a meeting schedule provided in an embodiment of the present application, and fig. 6 is a second schematic structural diagram of the device for generating a meeting schedule provided in the embodiment of the present application. As shown in fig. 5, the generating means 500 comprises:

an information obtaining module 510, configured to obtain, in response to a touch operation of a conference initiator on a text chat interface, at least one piece of chat information generated by text chat performed by the conference initiator within a preset time period;

a keyword extraction module 520, configured to input at least one piece of text information extracted from the at least one piece of chat information into a pre-trained keyword extraction model to obtain at least one conference keyword and a conference label corresponding to each conference keyword;

a conference content generating module 530, configured to input the at least one conference keyword and a conference label corresponding to each conference keyword into a pre-trained conference content ordering model, so as to obtain an initial to-be-initiated conference content initiated by the conference initiator;

and a meeting schedule generating module 540, configured to generate a target meeting schedule based on the initial content of the meeting to be initiated.

In one possible real-time manner, as shown in fig. 6, the generating apparatus 500 further includes a conference canceling module 550, and the conference canceling module 550 is configured to:

In one possible real-time manner, as shown in fig. 6, the generating device 500 further includes a meeting notification module 560, and the meeting notification module 560 is configured to:

In one possible real-time approach, as shown in fig. 6, the generating apparatus 500 further includes a model updating module 570, and the model updating module 570 is configured to:

In a possible real-time manner, when the keyword extraction module 520 is configured to input at least one piece of text information extracted from the at least one piece of chat information into a pre-trained keyword extraction model to obtain at least one conference keyword corresponding to a conference initiator and a conference label corresponding to each conference keyword, the keyword extraction module 520 is configured to:

In a possible real-time manner, when the keyword extraction module 520 is configured to extract, for each piece of text information, at least one conference keyword and a conference tag corresponding to each conference keyword included in the piece of text information from the at least one text keyword based on a preset semantic judgment rule and a plurality of preset labeling keywords, the keyword extraction module 520 is configured to:

In one possible real-time approach, for each text keyword, the keyword extraction module 520 is configured to determine a meeting keyword and a corresponding meeting tag by:

In a possible real-time manner, when the conference content generating module 530 is configured to input the at least one conference keyword and the conference label corresponding to each conference keyword into a pre-trained conference content ordering model to obtain the initial to-be-initiated conference content initiated by the conference initiator, the conference content generating module 530 is configured to:

In a possible real-time manner, when the meeting schedule generating module 540 is configured to generate a target meeting schedule based on the initial content of the meeting to be initiated, the meeting schedule generating module 540 is configured to:

In one possible real-time approach, the editing operation includes at least one of:

deleting operation, creating operation and updating operation;

when the meeting schedule generating module 540 is configured to generate a target meeting schedule in response to the editing operation of the meeting initiator on the initial meeting content to be initiated, the meeting schedule generating module 540 is configured to:

In a possible real-time mode, the keyword extraction model and the conference content ordering model are obtained by training based on a plurality of pre-labeled data;

The device for generating the conference schedule, provided by the embodiment of the application, is used for responding to a touch operation of a conference initiator on a text chat interface, obtaining at least one piece of chat information generated when a text chat is performed within a preset time period, correspondingly extracting at least one piece of text information from the at least one piece of chat information, inputting the at least one piece of text information into a pre-trained keyword extraction model, and obtaining at least one conference keyword and a conference label corresponding to each conference keyword; and inputting the obtained at least one conference keyword and a conference label corresponding to each conference keyword into a pre-trained conference content sequencing model to obtain initial to-be-initiated conference content initiated by a conference initiator, and obtaining a target conference schedule according to the initial to-be-initiated conference content. According to the method and the device, the corresponding target meeting schedule is generated automatically according to the text information extracted from the text chatting, the data processing steps for generating the meeting schedule are simplified, the time for generating the meeting schedule is reduced, and the generation efficiency of the meeting schedule is improved.

Referring to fig. 7, fig. 7 is a schematic structural diagram of an electronic device according to an embodiment of the present disclosure. As shown in fig. 7, the electronic device 700 includes a processor 710, a memory 720, and a bus 730.

The memory 720 stores machine-readable instructions executable by the processor 710, when the electronic device 700 runs, the processor 710 communicates with the memory 720 through the bus 730, and when the machine-readable instructions are executed by the processor 710, the steps of the method for generating a meeting schedule in the method embodiments shown in fig. 1 and fig. 2 may be performed.

An embodiment of the present application further provides a computer-readable storage medium, where a computer program is stored on the computer-readable storage medium, and when the computer program is executed by a processor, the steps of the method for generating a meeting schedule in the method embodiments shown in fig. 1 and fig. 2 may be executed.

It is clear to those skilled in the art that, for convenience and brevity of description, the specific working processes of the above-described systems, apparatuses and units may refer to the corresponding processes in the foregoing method embodiments, and are not described herein again.

In the several embodiments provided in the present application, it should be understood that the disclosed system, apparatus and method may be implemented in other ways. The above-described embodiments of the apparatus are merely illustrative, and for example, the division of the units is only one logical division, and there may be other divisions when actually implemented, and for example, a plurality of units or components may be combined or integrated into another system, or some features may be omitted, or not executed. In addition, the shown or discussed mutual coupling or direct coupling or communication connection may be an indirect coupling or communication connection of devices or units through some communication interfaces, and may be in an electrical, mechanical or other form.

The units described as separate parts may or may not be physically separate, and parts displayed as units may or may not be physical units, may be located in one place, or may be distributed on a plurality of network units. Some or all of the units can be selected according to actual needs to achieve the purpose of the solution of the embodiment.

In addition, functional units in the embodiments of the present application may be integrated into one processing unit, or each unit may exist alone physically, or two or more units are integrated into one unit.

The functions, if implemented in the form of software functional units and sold or used as a stand-alone product, may be stored in a non-volatile computer-readable storage medium executable by a processor. Based on such understanding, the technical solution of the present application or portions thereof that substantially contribute to the prior art may be embodied in the form of a software product stored in a storage medium and including instructions for causing a computer device (which may be a personal computer, a server, or a network device) to execute all or part of the steps of the method according to the embodiments of the present application. And the aforementioned storage medium includes: various media capable of storing program codes, such as a usb disk, a removable hard disk, a Read-Only Memory (ROM), a Random Access Memory (RAM), a magnetic disk, or an optical disk.

Finally, it should be noted that: the above-mentioned embodiments are only specific embodiments of the present application, and are used for illustrating the technical solutions of the present application, but not limiting the same, and the scope of the present application is not limited thereto, and although the present application is described in detail with reference to the foregoing embodiments, those skilled in the art should understand that: any person skilled in the art can modify or easily conceive the technical solutions described in the foregoing embodiments or equivalent substitutes for some technical features within the technical scope disclosed in the present application; such modifications, changes or substitutions do not depart from the spirit and scope of the exemplary embodiments of the present application, and are intended to be covered by the scope of the present application. Therefore, the protection scope of the present application shall be subject to the protection scope of the claims.

Claims

1. a generation method of a meeting schedule, is characterized in that, described generation method comprises:

Acquiring at least one piece of chat information generated by the text chat performed by the conference initiator within a preset time period in response to the conference initiator's touch operation on the text chat interface;

Inputting the at least one conference keyword and the conference label corresponding to each conference keyword into the pre-trained conference content sorting model to obtain the initial conference content to be initiated initiated by the conference initiator;

Based on the initial meeting content to be initiated, a target meeting schedule is generated.

2. The generation method according to claim 1, wherein the at least one piece of text information extracted from the at least one piece of chatting information is input into a pre-trained keyword extraction model, and the conference initiator is obtained The corresponding at least one conference keyword and the conference label corresponding to each conference keyword, including:

For each piece of text information, based on a preset semantic judgment rule and a plurality of preset annotation keywords, at least one conference keyword included in the piece of text information and each The conference label corresponding to the conference keyword.

3. The generation method according to claim 2, wherein, for each piece of text information, based on a preset semantic judgment rule and a plurality of preset label keywords, extracting from the at least one text keyword At least one conference keyword included in the text message and the conference label corresponding to each conference keyword are displayed, including:

Based on the preset semantic judgment rule, determine the semantic intention of the user who outputs the piece of text information, and if the semantic intention is the target semantic intention, determine the piece of text information based on at least one text keyword included in the piece of text information At least one conference keyword and a conference tag corresponding to each conference keyword are included in the .

4. The generating method according to claim 3, wherein, for each text keyword, a conference keyword and a corresponding conference label are determined through the following steps:

Comparing the text keyword with each marked keyword included in the preset multiple marked keywords, if there is a marked keyword matching the text keyword in the multiple marked keywords, The text keyword is determined as the conference keyword, and the text tag corresponding to the matched marked keyword is determined as the conference tag corresponding to the text keyword.

5. The generation method according to claim 1, wherein the at least one conference keyword and the conference label corresponding to each conference keyword are input into a pre-trained conference content sorting model to obtain the conference The content of the initial meeting to be initiated initiated by the initiator, including:

Input the at least one conference keyword and the conference tag corresponding to each conference keyword into the pre-trained conference content sorting model, and determine the corresponding conference keyword according to the preset score corresponding to each conference tag. The score; wherein, the preset score represents the correlation between the corresponding conference label and the conference, and the higher the correlation with the conference, the higher the preset score corresponds to;

Screen out at least one target keyword with a score greater than a preset score threshold from a plurality of conference keywords;

Based on the at least one target keyword, the initial meeting content to be initiated is determined.

6. The generating method according to claim 1, wherein the generating a target meeting schedule based on the initial meeting content to be initiated comprises:

A target conference schedule is generated in response to an editing operation by the conference initiator with respect to the initially to-be-initiated conference content.

7. The generating method according to claim 6, wherein the editing operation comprises at least one of the following:

Delete operation, new operation, update operation;

The generating a target conference schedule in response to the conference initiator's editing operation on the initially to-be-initiated conference content, including:

In response to the conference initiator's deletion operation on the initially to-be-initiated conference content, the conference content after the content selected by the conference initiator is deleted from the initial to-be-initiated conference content is determined. the agenda for the said target meeting; or

In response to the conference initiator's new operation on the initially to-be-initiated conference content, the conference content after adding the content edited by the conference initiator to the initial to-be-initiated conference content is determined to be the desired conference content. the target meeting schedule; or

In response to the conference initiator's update operation on the initially to-be-initiated conference content, the conference in which the corresponding position in the initially to-be-initiated conference content is replaced with the content edited by the conference initiator The content is determined as the target meeting schedule.

8. The generating method according to claim 1, characterized in that, before generating a target meeting schedule based on the initial meeting content to be initiated, the generating method further comprises:

In response to the conference initiator's cancel operation on the initially to-be-initiated conference content, the display of the initially to-be-initiated conference content is canceled.

9. The generation method according to claim 1, wherein the generation method further comprises:

In response to the determination operation of the conference initiator on the target conference schedule, the target conference schedule is published to at least one user, wherein the user is a participant of the target conference.

10. The generating method according to claim 1, wherein the generating method further comprises:

Acquiring the number of times of cancellation of the conference initiator's cancellation operation with respect to the initially to-be-initiated conference content within a preset test time period;

If the number of cancellations is greater than the preset number of cancellations threshold, collect at least one piece of text information generated by the conference initiator during text chatting, and at least one piece of updated information generated by the conference initiator during text chatting after the cancellation operation ;

Based on the at least one piece of text information and the at least one piece of update information, the keyword extraction model and the conference content sorting model are updated to obtain an updated updated keyword extraction model and an updated updated conference content sorting model ;

Based on the updated keyword extraction model and the updated meeting content sorting model, at least one piece of text information generated by the text chat within a preset time period is analyzed again to obtain a corresponding target meeting schedule.

11. The generating method according to claim 1, wherein the keyword extraction model and the conference content sorting model are obtained based on a plurality of pre-labeled data training;

The annotated data is annotated based on a preset semantic judgment rule and a preset target keyword library, and includes data of keywords related to initiating a conference.

12. An apparatus for generating a meeting schedule, wherein the generating apparatus comprises:

an information acquisition module, configured to acquire at least one piece of chat information generated by the text chat performed by the conference initiator within a preset time period in response to the conference initiator's touch operation on the text chat interface;

A keyword extraction module, configured to input at least one piece of text information extracted from the at least one piece of chat information into a pre-trained keyword extraction model to obtain at least one conference keyword and a conference label corresponding to each conference keyword ;

A meeting content generation module, configured to input the at least one meeting keyword and the meeting label corresponding to each meeting keyword into the pre-trained meeting content sorting model, and obtain the initial meeting content to be initiated initiated by the meeting initiator ;

A meeting schedule generation module is configured to generate a target meeting schedule based on the initial meeting content to be initiated.

13. An electronic device, comprising: a processor, a storage medium, and a bus, wherein the storage medium stores machine-readable instructions executable by the processor, and when the electronic device runs, the processor is connected to a bus. The storage media communicate through a bus, and the processor executes the machine-readable instructions to execute the steps of the method for generating a meeting schedule according to any one of claims 1 to 11.

14. A computer-readable storage medium, characterized in that, a computer program is stored on the computer-readable storage medium, and when the computer program is run by a processor, the execution of the meeting schedule according to any one of claims 1 to 11 is executed. Generate the steps of the method.