CN112307166B

CN112307166B - Intelligent question-answering method and device, storage medium and computer equipment

Info

Publication number: CN112307166B
Application number: CN202011181280.2A
Authority: CN
Inventors: 高波
Original assignee: Tencent Technology Shenzhen Co Ltd
Current assignee: Tencent Technology Shenzhen Co Ltd
Priority date: 2020-10-29
Filing date: 2020-10-29
Publication date: 2024-01-30
Anticipated expiration: 2040-10-29
Also published as: CN112307166A

Abstract

The application discloses an intelligent question-answering method, an intelligent question-answering device, a storage medium and computer equipment; the application is related to the fields of database query and artificial intelligence machine learning, and can display a robot question-answer page of a client, wherein the robot question-answer page comprises a chat content editing area; in response to a query editing operation for the chat content editing area, displaying a query message on the robot question-answer page, the query message including query content for at least two applications; identifying the query message, and displaying a reply message on a robot question-answer page, wherein the reply message comprises reply contents obtained by identifying at least two virtual robots aiming at the query message, and the at least two virtual robots are in one-to-one correspondence with at least two applications; the method and the device can remarkably improve the question and answer efficiency of the virtual robot.

Description

Intelligent question-answering method and device, storage medium and computer equipment

Technical Field

The application relates to the field of robots, in particular to an intelligent question-answering method, an intelligent question-answering device, a storage medium and computer equipment.

Background

With the development of technology, information interaction can be performed on a computer device through a virtual robot, and the virtual robot can perform a session with a user through text, voice and other modes, for example, in the prior art, the user can ask a question on a chat page to another virtual robot, if the same question is required, the chat page needs to be switched from one virtual robot to another virtual robot, and then the question is asked to the other virtual robot.

In the research and practice process of the prior art, the inventor of the application finds that when a prior art user asks for questions to at least two virtual robots, the chat pages need to be switched, and information input needs to be repeated, so that the question and answer efficiency is low.

Disclosure of Invention

The embodiment of the application provides an intelligent question-answering method, an intelligent question-answering device, a storage medium and computer equipment, which can obviously enhance the question-answering efficiency of a virtual robot.

The embodiment of the application provides an intelligent question-answering method, which comprises the following steps:

displaying a robot question-answer page of a client, wherein the robot question-answer page comprises a chat content editing area;

in response to a query editing operation for the chat content editing area, displaying a query message on the robot question-answer page, the query message including query content for at least two applications;

And identifying the inquiry message, and displaying a reply message on the robot inquiry and reply page, wherein the reply message comprises reply contents obtained by identifying at least two virtual robots aiming at the inquiry message, and the at least two virtual robots are in one-to-one correspondence with the at least two applications.

Correspondingly, the application provides an intelligent question answering device, which comprises:

the page display module is used for displaying a robot question-answer page of the client, and the robot question-answer page comprises a chat content editing area;

a query display module for displaying a query message on the robot question page in response to a query editing operation for the chat content editing area, the query message including query content for at least two applications;

the reply display module is used for identifying the inquiry message and displaying a reply message on the robot inquiry and reply page, wherein the reply message comprises reply contents obtained by identifying at least two virtual robots aiming at the inquiry message, and the at least two virtual robots are in one-to-one correspondence with the at least two applications.

In some embodiments, the reply display module includes an identification sub-module and a reply display sub-module, wherein,

The identification sub-module is used for identifying the query message;

and the reply display sub-module is used for displaying a reply message on the robot question and answer page, wherein the reply message comprises reply contents obtained by identifying at least two virtual robots aiming at the query message, and the at least two virtual robots are in one-to-one correspondence with the at least two applications.

In some embodiments, the at least two virtual robots include a first virtual robot and a second virtual robot, the recognition sub-module includes a recognition unit, a determination unit, wherein,

the identification unit is used for carrying out intention identification on the query message to obtain the query intention of the query message;

and the determining unit is used for determining first application reply content of the first virtual robot and second application reply content of the second virtual robot based on the query intention.

In some embodiments, the reply display sub-module is specifically configured to:

and displaying a reply message of the first virtual robot or the second virtual robot on the robot question and answer page, wherein the reply message is determined according to the first application reply content and the second application reply content.

and displaying a reply message on the robot question-answer page, wherein the reply message comprises a first reply message and a second reply message, the first reply message comprises first application reply content of the first virtual robot, and the second reply message comprises second application reply content of the second virtual robot.

In some embodiments, the intelligent question answering apparatus further comprises:

the interaction acquisition module is used for acquiring the interaction content of the second reply message;

the interactive display module is used for displaying an interactive message of a second reply message of the first virtual robot aiming at the second virtual robot, and the interactive message comprises the interactive content.

the wake-up display module is used for responding to the robot wake-up editing operation aiming at the chat content editing area and displaying a wake-up message on the robot question-answer page;

the wake-up identification module is used for identifying the wake-up message so as to determine a target wake-up object from at least two virtual robots;

and the wake-up object module is used for displaying the chat content editing area of the target wake-up object on the robot question-answer page.

the wake-up interaction module is used for displaying wake-up interaction information aiming at the target wake-up object;

in this embodiment, the wake object module is specifically configured to:

and displaying the wake-up response message and the chat content editing area of the target wake-up object on the robot question page.

In some embodiments, the identification unit is specifically configured to:

preprocessing the query content in the query message to obtain a query character string;

and identifying the query character string through the trained intention identification model to obtain the query intention of the query message.

the sample acquisition module is used for acquiring a plurality of sample texts and labels thereof;

the text recognition module is used for recognizing the sample text through the intention recognition model to obtain a recognition result of the sample text;

and the parameter adjustment module is used for adjusting network parameters of the intention recognition model based on the recognition result of the sample text and the label so as to obtain a trained intention recognition model.

In some embodiments, the identification unit is specifically configured to:

performing intention recognition on the query message through the first virtual robot to obtain the query intention of the query message;

In this embodiment, the determining unit is specifically configured to:

transmitting, by the first virtual robot, the query intent to the second virtual robot;

determining, by the first virtual robot, first application reply content from a first application knowledge base based on the query intent;

and determining, by the second virtual robot, second application reply content from a second application knowledge base based on the query intent.

Correspondingly, the embodiment of the application also provides a storage medium, and the storage medium stores a computer program, and the computer program is suitable for being loaded by a processor to execute any intelligent question-answering method provided by the embodiment of the application.

Correspondingly, the embodiment of the application also provides computer equipment, which comprises a memory, a processor and a computer program stored on the memory and capable of running on the processor, wherein the processor realizes any intelligent question-answering method provided by the embodiment of the application when executing the computer program.

The method and the device can display a robot question-answer page of the client, wherein the robot question-answer page comprises a chat content editing area; in response to a query editing operation for the chat content editing area, displaying a query message on the robot question-answer page, the query message including query content for at least two applications; and identifying the query message, and displaying a reply message on a robot question-answer page, wherein the reply message comprises reply contents obtained by identifying at least two virtual robots aiming at the query message, and the at least two virtual robots are in one-to-one correspondence with the at least two applications.

The method and the device can display the query message on the robot query and answer page, the query message can comprise query contents aiming at least two applications, after the query contents are identified, the reply message can be displayed on the robot query and answer page, and can be obtained based on reply contents respectively made on the query message by at least two virtual robots corresponding to the at least two applications.

Drawings

In order to more clearly illustrate the technical solutions of the embodiments of the present application, the drawings that are needed in the description of the embodiments will be briefly introduced below, it being obvious that the drawings in the following description are only some embodiments of the present application, and that other drawings may be obtained according to these drawings without inventive effort for a person skilled in the art.

Fig. 1 is a schematic view of a scenario of an intelligent question-answering system provided in an embodiment of the present application;

fig. 2 is a schematic flow chart of an intelligent question-answering method provided in an embodiment of the present application;

FIG. 3 is a schematic diagram of page interaction of the intelligent question-answering method provided in the embodiment of the present application;

FIG. 4 is another page interaction schematic diagram of the intelligent question-answering method provided in the embodiment of the present application;

FIG. 5 is another page interaction schematic diagram of the intelligent question-answering method provided in the embodiment of the present application;

FIG. 6 is another page interaction schematic diagram of the intelligent question-answering method provided in the embodiment of the present application;

FIG. 7 is another page interaction schematic diagram of the intelligent question-answering method provided in the embodiment of the present application;

FIG. 8 is another page interaction schematic diagram of the intelligent question-answering method provided in the embodiment of the present application;

FIG. 9 is another page interaction schematic diagram of the intelligent question-answering method provided in the embodiment of the present application;

fig. 10 is an intention recognition architecture diagram of the intelligent question-answering method provided in the embodiment of the present application;

FIG. 11 is an illustration of an intent delivery example of a method of intelligent question-answering provided by embodiments of the present application;

FIG. 12 is another illustration of an intent delivery example of the intelligent question-answering method provided by embodiments of the present application;

fig. 13 is another flow chart of the intelligent question-answering method provided in the embodiment of the present application;

FIG. 14 is another page interaction schematic diagram of the intelligent question-answering method provided in the embodiment of the present application;

FIG. 15 is another page interaction schematic diagram of the intelligent question-answering method provided in the embodiment of the present application;

FIG. 16 is another page interaction schematic diagram of the intelligent question-answering method provided in the embodiment of the present application;

FIG. 17 is an exemplary diagram of an implementation flow of the intelligent question-answering method provided in an embodiment of the present application;

fig. 18 is a schematic structural diagram of an intelligent question-answering device according to an embodiment of the present application;

fig. 19 is a schematic structural diagram of a computer device according to an embodiment of the present application.

Detailed Description

The following description of the embodiments of the present application will be made clearly and completely with reference to the accompanying drawings of the embodiments of the present application, and it is apparent that the embodiments described in the present application are only some embodiments of the present application, but not all embodiments. All other embodiments, which can be made by those skilled in the art based on the embodiments herein without making any inventive effort, are intended to be within the scope of the present application.

The embodiment of the application provides an intelligent question answering method, an intelligent question answering device, a storage medium and computer equipment. Specifically, the embodiment of the application can be integrated in an intelligent question-answering system.

Artificial intelligence (Artificial Intelligence, AI) is the theory, method, technique and application system that uses a digital computer or a machine controlled by a digital computer to simulate, extend and extend human intelligence, sense the environment, acquire knowledge and use the knowledge to obtain optimal results. In other words, artificial intelligence is an integrated technology of computer science that attempts to understand the essence of intelligence and to produce a new intelligent machine that can react in a similar way to human intelligence. Artificial intelligence, i.e. research on design principles and implementation methods of various intelligent machines, enables the machines to have functions of sensing, reasoning and decision.

Machine Learning (ML) is a multi-domain interdisciplinary, involving multiple disciplines such as probability theory, statistics, approximation theory, convex analysis, algorithm complexity theory, etc. It is specially studied how a computer simulates or implements learning behavior of a human to acquire new knowledge or skills, and reorganizes existing knowledge structures to continuously improve own performance. Machine learning is the core of artificial intelligence, a fundamental approach to letting computers have intelligence, which is applied throughout various areas of artificial intelligence. Machine learning and deep learning typically include techniques such as artificial neural networks, confidence networks, reinforcement learning, transfer learning, induction learning, teaching learning, and the like.

Natural language processing (Nature Language processing, NLP) is an important direction in the fields of computer science and artificial intelligence. It is studying various theories and methods that enable effective communication between a person and a computer in natural language. Natural language processing is a science that integrates linguistics, computer science, and mathematics. Thus, the research in this field will involve natural language, i.e. language that people use daily, so it has a close relationship with the research in linguistics. Natural language processing techniques typically include text processing, semantic understanding, machine translation, robotic questions and answers, knowledge graph techniques, and the like.

The reinforcement learning model in the embodiments of the present application relates to the fields of machine learning of artificial intelligence, natural language processing, such as intention recognition models, intention recognition by means of trained intention recognition models, and the like, and will be specifically described in detail by the following embodiments.

The intelligent question-answering system can be integrated in computer equipment, the computer equipment can comprise a terminal, a server and the like, and the server can be an independent physical server, a server cluster or a distributed system formed by a plurality of physical servers, and a cloud server for providing cloud computing service. The terminal may be, but is not limited to, a smart phone, a tablet computer, a notebook computer, a desktop computer, a smart speaker, a smart watch, a smart television, etc. The terminal and the server may be directly or indirectly connected through wired or wireless communication, which is not limited herein.

As shown in fig. 1, fig. 1 is a schematic view of a scenario of an intelligent question-answering system, which may be integrated on a terminal and a server, specifically:

the terminal may display a robot question-and-answer page of the client, the robot question-and-answer page including a chat content editing area, and then display a query message including query contents for at least two applications on the robot question-and-answer page in response to a query editing operation for the chat content editing area, the terminal may transmit the query message to the server, the server may identify the query message through at least two virtual robots, and transmit reply contents for the query message for each virtual robot to the terminal, and the terminal may display the reply message on the robot question-and-answer page based on the received at least two reply contents.

It should be noted that, the schematic view of the scenario of the intelligent question-answering system shown in fig. 1 is only an example, and the intelligent question-answering system and scenario described in the embodiments of the present application are for more clearly describing the technical solutions of the embodiments of the present application, and do not constitute a limitation on the technical solutions provided in the embodiments of the present application, and those skilled in the art can know that, with the evolution of the intelligent question-answering device and the appearance of a new service scenario, the technical solutions provided in the embodiments of the present application are equally applicable to similar technical problems.

The following will describe in detail. In this embodiment, a detailed description will be given of an intelligent question-answering method, which may be integrated on a computer device, as shown in fig. 2, and may include:

101. and displaying a robot question-answer page of the client, wherein the robot question-answer page comprises a chat content editing area.

The client may include a local program for providing services to the object (e.g., a user, a test script), the local program may be a program on a computer device used by the object, and the client may perform information transfer with the object through an input/output device connected to the computer device, so as to provide services to the object, for example, the client may display a robot question-answer page through a display device (e.g., a display), and for example, the client may also perform information transfer with the object through an audio input/output device (e.g., a speaker, a microphone).

The client in the application may be a plurality of types of clients, for example, the robot question-answering page may include a group chat page in an instant messaging client, for example, the client may be a functional client mainly including factors such as games, shopping, live broadcasting, music, video, finance, news, communities and the like, and the robot question-answering page may include a customer service page, an information feedback page and the like in the functional client.

The robot question-answering page comprises a page capable of editing and sending messages and displaying the messages, and the robot question-answering page comprises a message display area capable of displaying the messages sent by objects in the page. The object in the robot question page may include at least two virtual robots, each virtual robot corresponding to an application.

The robot question and answer page may further include a chat content editing area through which the object may edit and send messages, the chat content editing area may include a content input control, a content selection control, etc. for editing the message content, and a transmission control, etc. for determining to send messages, etc., and the form of the message in the chat question and answer page may include text, voice, image, video, etc., so that the object may include various types, such as a text input control, a drawing control, a video selection control, an expression package selection control, etc., through the content input control, the content selection control, etc.

For example, referring to fig. 3, a robot question and answer page 01 in the client is displayed, the robot question and answer page 01 may be a group chat page of a group "angel skirt" in the instant messaging client, and the robot question and answer page 01 includes a chat content editing area 011.

102. In response to a query editing operation for the chat content editing area, a query message is displayed on the robot question-answer page, the query message including query content for at least two applications.

In this application, in response to a condition or state that may be used to represent a dependency of an operation performed, one or more operations performed may be real-time or may have a set delay when the dependency is satisfied; without being specifically described, there is no limitation in the execution sequence of the plurality of operations performed.

Specifically, the query editing operation may include an input operation (such as typing text, drawing an image, etc.), a selection operation (such as selecting from several existing images, text), an upload operation (such as uploading a locally stored image, audio-video, etc.), etc., by which the object may send the content to be expressed to the client, and the virtual robot in the client may answer.

The computer device may obtain the query content input by the user through the query editing operation according to the query editing operation for the chat content editing area, and display a query message on the robot question-answering page, where the query message may include the query content, the identity of the object, and the like.

Wherein the query content may include questions entered by the object, the query content may be in the form of text, images, audio-video, etc., for at least two applications. The query message may include query content, and accordingly, the query message may include a text message, an image message, an audio message, a video message, and the like.

In this application, the query content may be for at least two applications, for example, the query content may be: in the prior art, only the inquiry message aiming at a single application can be answered on the robot inquiry page, if the inquiry is to be carried out aiming at the related information of two or more applications, the inquiry content aiming at each application, such as the 'month expense of the A software' and the 'month expense of the B software', needs to be respectively output before and after the robot inquiry page is switched, so that the complex operation degree can be obviously reduced, and the virtual robot inquiry and answer efficiency is improved.

For example, referring to fig. 4, in response to a query edit operation of the user's tab for the chat content edit area 011 in the robot question and answer page 01, a query message 021 is displayed in the robot question and answer page 02, the query message 021 including the identification of the user (nicknames "tab" and the query content "how much memory is occupied by each software" for the application a and the application B).

103. And identifying the query message, and displaying a reply message on a robot question-answer page, wherein the reply message comprises reply contents obtained by identifying at least two virtual robots aiming at the query message, and the at least two virtual robots are in one-to-one correspondence with the at least two applications.

The reply content may correspond to the query content, that is, the reply content may include content responding to the query content, and the reply content may be in the form of text, image, audio and video, and the like. The reply message may include reply content, and the number of reply messages may be at least one. The reply message may also include identification information for the virtual robot, which may include a nickname, avatar, etc.

In the application, the reply message can be determined based on reply content obtained by identifying the reply message by at least two virtual robots, namely, the application can realize that an object outputs the reply content for at least two applications on a robot reply page, the client displays the reply message containing the reply content on the robot reply page, the reply message is identified by at least two virtual robots which are in one-to-one correspondence with the at least two applications, and the reply message determined based on the reply content obtained by identification is displayed on the robot reply page.

One virtual robot may correspond to one application, the virtual robot corresponding to the application may solve related problems of the application, for example, the related problems may be activity information, usage information, after-sales information, etc. of the application, the application may be a product disclosed on the computer device for a specific or whole group, and the types of the application may include many, for example, games, shopping, live broadcasting, video, communication, etc.

Specifically, the query message can be identified in various manners, the query message is mainly identified as query content contained in the query message, when the query content is in the form of voice, image, video and the like, the query content can be identified in different manners, the query content in the form of voice can be converted into the text form through voice identification (Speech Technology), and then the query content in the form of text can be identified through semantic understanding (Semantic Understanding); the query content in the form of images, videos, etc. may be identified by image semantic understanding (ISU, image Semantic Understanding), video semantic understanding (VSU, video Semantic Understanding), etc.

For example, when the text form query content is identified, the identification can be performed based on a preset query template and reply content corresponding to the query template, and the target query template corresponding to the query content is determined by calculating the similarity between the query content and a plurality of preset query templates, so as to determine the target reply content corresponding to the target query template.

For example, referring to fig. 5, the inquiry content "how much memory each software occupies" in the inquiry message is identified, and a reply message 031 is displayed on the robot question-answer page 03, and the reply message 031 contains reply content "a application 566M B occupies 255M" and identification information (nickname "happy small a" and head portrait) of the virtual robot.

In some embodiments, the at least two virtual robots include a first virtual robot and a second virtual robot, and the step of "identifying the query message" may include:

performing intention recognition on the query message to obtain the query intention of the query message; based on the query intent, first application reply content of the first virtual robot and second application reply content of the second virtual robot are determined.

Where the query intent may include an intent to which the query content points, the query content is typically a human representation consistent with the expression habits of most users, the application may first understand the query content to give a reply content corresponding to the query content, and understand the query content may include determining the query intent of the query content, after which the reply content may be determined based on the query intent.

The scope of the query intent varies with the application for which the query content is intended, for example, the query intent of a game-type application may include a game duration, a game character, a game activity, a game version, a game rule, etc., and the query intent of a shopping-type application may include a consumption amount, a shopping activity, a new on commodity, etc. The query intent may be divided into multiple levels according to the degree of detail, for example, the consumption amount may be a primary query intent, the last three months consumption amount, the dress consumption amount, etc. may be a secondary query intent, etc.

Specifically, the intention recognition is performed on the query message, that is, the intention recognition is performed on the query content in the query message, for example, the intention recognition may be performed in various manners, for example, key information matching may be performed, each query intention may include a preset key information set, where the preset key information set may include a keyword, a key image, a key voice, a key picture, and the like, the query content is compared with a preset key information set, and if key information in the preset key information set exists in the query content, it is determined that the target query intention of the query content is the query content corresponding to the preset key information set.

The first application reply content may include reply content determined by the first virtual robot according to the query intention, and the second application reply content may include reply content determined by the second virtual robot according to the query intention. Specifically, the determining of the first application reply content or the second application reply content may include various manners, for example, a knowledge base of the first application or the second application may be obtained, the knowledge base may include a plurality of keywords in the first application or the second application and corresponding reply contents thereof, the identified query intention is matched with the knowledge base of the first application or the second application, and then the first application reply content or the second application reply content is determined according to the matched keywords.

In some embodiments, the step of displaying the reply message on the robot question and answer page may include:

and displaying a reply message of the first virtual robot or the second virtual robot on the robot question-answering page, wherein the reply message is determined according to the first application reply content and the second application reply content.

The number of the reply messages can be one, one reply message can be determined according to the reply content of the first application and the reply content of the second application, the reply contents of different applications can be integrated by the display mode, the reply contents are intensively displayed in one message, a more concise display effect can be obtained, and the differences of the reply contents of different applications can be displayed more intuitively.

One reply message corresponds to one virtual robot, and the virtual robot may be the first virtual robot or the second virtual robot, specifically, the virtual robot corresponding to the reply message may be determined randomly, may be determined based on the virtual robot corresponding to the historical reply message, or the like.

For example, referring to fig. 5, the reply message 031 corresponds to a randomly determined virtual robot of application a, and thus, the reply message 031 contains identification information (nickname "happy small a" and avatar a) of the first virtual robot of application a and reply content "a application 566M B application occupies 255M of memory.

In some embodiments, the number of reply messages may be at least two, where one reply message corresponds to reply content of one virtual robot, so that the display effect is more clear.

For example, referring to FIG. 6, the reply message includes a first reply message 0311 and a second reply message 0322, the first reply message 0311 including the first virtual robot identification information of application A (nickname "Happy little A" and avatar A) and the first application reply content "Strong hand", vendor big yield-! During the period from 10 months 2 to 10 months 10 days, the customer who logs in the A application to recharge the telephone charge can obtain the voucher giving the corresponding face value, the purchasing machine withstands cash-! The second reply message 0322 includes the second virtual robot identification information of the application B (nickname "BB households" and avatar B) and the second application reply content "buy mobile phone to BB, BB is a long push of multiple special price mobile phones, and you come.

In some embodiments, the intelligent question answering method may further include:

acquiring the interactive content of the second reply message; and displaying an interaction message of a second reply message of the first virtual robot aiming at the second virtual robot, wherein the interaction message comprises interaction content.

In order to enhance interaction feeling and context feeling of conversation with the virtual robots, in the application, interaction can also be performed between different virtual robots, wherein the interaction content can be the content of the virtual robot responding to a message of another virtual robot, and the interaction content can comprise characters, images, audios and videos and the like. The interactive contents may correspond to specific objects, for example, the second virtual robot may determine the same or similar interactive contents for the interactive contents of the first virtual robot, i.e., for any message sent by the first virtual robot; the interactive content may also correspond to specific message content, e.g., the same or similar interactive content may be determined for any virtual robot-sent activity introductory messages, etc.

Therefore, the first interactive content of the first reply message may also be acquired, and the interactive message of the first reply message of the second virtual robot for the first virtual robot may be displayed on the robot question page, where the interactive message may include the first interactive content.

The interactive content can be obtained from an interactive content database, the interactive content database can contain a plurality of preset interactive contents, different virtual robots can correspond to different interactive content databases, and the same interactive content database can be shared; the interactive content can also be obtained through an interactive content generation model, and the interactive content generation model can be an artificial intelligent network model which is built and trained according to requirements, and the like.

For example, referring to fig. 7, the robot question and answer page 03 may further include an interaction message 0313, where the interaction message 0313 includes an interaction content "@ BB caretaker of the second reply message of the first virtual robot to the second virtual robot has a taste in the old marketing number.

responding to the robot wakeup editing operation aiming at the chat content editing area, and displaying a wakeup message on a robot question-answering page; identifying the wake-up message to determine a target wake-up object from at least two virtual robots; and displaying a chat content editing area of the target wake-up object on the robot question-answering page.

The robot question-answering page can be a question-answering page of a virtual robot of a specific application, the robot question-answering page can comprise a quick question-answering channel of the virtual robot, for example, a quick question-answering control containing common query contents is triggered by an object, and the virtual robot can send a reply message aiming at the query contents contained on the quick question-answering control, so that the object can query more conveniently and quickly.

Since the robot question-answering page includes at least two virtual robots, when the object wants to switch the robot question-answering page of the current virtual robot to the robot question-answering page of other virtual robots, the robot wakeup editing operation can be performed, the current virtual robot can identify the wakeup message and determine the target wakeup object to be waken, wherein the target wakeup object can be one of the at least two virtual robots included in the robot question-answering page, then, the robot question-answering page for displaying the target wakeup object can be switched, and the chat content editing area of the robot question-answering page of the target wakeup object can include: a shortcut question-answer control for the target wake-up object, etc.

For example, referring to fig. 8, the robot question and answer page 04a of the first virtual robot includes a shortcut question and answer area 0412, and the shortcut question and answer area 0412 includes a shortcut question and answer control: the welfare activity button, the latest notice button and the happy attack button respond to the robot wakeup editing operation aiming at the chat content editing area, after the wakeup message 0411 of the robot question and answer page display object LL is identified, the robot question and answer page 04b of the second virtual robot of the target wakeup object is switched to be displayed, the robot question and answer page 04b comprises a shortcut question and answer area 0422 in the chat content editing area of the second virtual robot, and the shortcut question and answer area 0422 comprises shortcut question and answer controls: a new button on BB, a daily life button and a cargo wash button.

displaying a wake-up interaction message aiming at a target wake-up object;

in this embodiment, the step of displaying the chat content editing area of the target wake object on the robot question and answer page may include:

and displaying a wake-up response message and a chat content editing area of the target wake-up object on the robot question page.

In order to enhance the interaction effect and the interestingness, a user is provided with a more definite switching prompt, after the target awakening object is determined, the robot question-answer page can display awakening interaction information aiming at the target awakening object, and when the robot question-answer page displaying the target awakening object is switched, the chat content editing area of the target awakening object and the awakening response information of the target awakening object can be displayed.

For example, referring to fig. 9, after identifying the target wake-up object, the robot question-answer page may display a wake-up interaction message 042a of the first virtual robot for the target wake-up object, where the wake-up interaction message 042a may include identification information (nickname "happy small a" and head portrait) and text "@ BB small households of the first virtual robot, the target wake-up object second virtual robot may display a wake-up response message 042b, where the wake-up response message 042b may include identification information (nickname" BB small households "and head portrait) and text" do you living small households "of the second virtual robot, i are BB small households-! ".

In some embodiments, the step of "intent recognition of the query message to obtain the query intent of the query message" may include:

preprocessing query content in a query message to obtain a query string;

The intention recognition of the query message can be performed through a model, and the query content in the query message can be preprocessed before being recognized through the model to obtain a query character string which can be rapidly recognized and processed by the model, particularly, for the query content in a text form, the format conversion of the query content can be performed to be matched with the requirements of the model used by the query content; the text may be subject to correction and adjustment of spelling, words, etc. of the text, e.g., correction of misplaced words, deletion of non-specifically defined stop words, etc.

acquiring a plurality of sample texts and labels thereof; identifying the sample text through the intention identification model to obtain an identification result of the sample text; based on the recognition result of the sample text and the label, carrying out network parameter adjustment on the intention recognition model to obtain a trained intention recognition model.

After receiving the message sent by the object, the trained intention recognition model can recognize the content in the message to determine that the message is an inquiry message or a wake-up message, and the like.

The trained intent recognition model in the application can also be used for recognizing the query intent of query content, the target application aimed by the query message, the target query object aimed by the wake-up message and the like.

Specifically, the training and application framework of the intent recognition model may refer to fig. 10, wherein the sample text may include a training prediction and verification corpus, the training corpus may be used for training of the intent recognition model, the verification corpus may be used for testing of the intent recognition model, the configuration file may store the intent recognition model, and relevant parameters therein, specifically, relevant parameters may include a learning rate (parameter adjustment when used for model training), word vector dimensions (used for determining feature vector dimensions in the model), number of elements of the language model (used for determining the number of associated word vectors of the target word vector), number of training (i.e., number of times the model is trained, flexibly adjustable), a loss function (used for quantifying recognition effect of the model), number of barrels (used for adjusting accuracy of model training), number of threads (used for determining the number of models for parallel training, flexibly adjustable according to a training machine), and so on.

The relevant data and files required by training can be stored in a specific path before training, the relevant data and files generated in the training process can be stored in the specific path during training, the path can be adjusted according to the requirement and not be unique, for example, the prior art\conf\conf properties can be a storage catalog, the subdirectories under the catalog comprise fasttet\train, fasttet.pred, pred_corpus, train_corp\init_corpus, wherein the fasttet\train can comprise relevant files and instructions for executing model training, the fasttet.pred can comprise relevant files and instructions for executing model testing, the pred_corp can comprise a test set file containing test samples, and the train_corp_corp can comprise a training set file containing training samples.

The dist may be a storage directory at training time, and the subdirectories under the directory may include: clear_dataset, conf\conf. Properties, models, pred_corps, pred_corps\init_corps, wherein clear_dataset may include a preprocessed training set file, conf\conf. Properties may include related files and instructions for performing model training and model testing, models may include files of models obtained from each training, pred_corps may include files containing all trained sample data, pred_corps\init_corps may include training set files containing training samples.

The stored sample data (including sample data for training and testing and sample data after preprocessing) can be stored by a file in a csv format.

Referring to fig. 10, the training and testing of the intent recognition model can be completed by loading configuration files and other related files in the adaptation layer, specifically, in the training stage, corpus merging, corpus deduplication, format conversion, spelling error correction and unification, word segmentation, data set division and other preprocessing operations can be performed on training samples, wherein corpus merging and corpus deduplication are performed on sample data according to sample labels, repeated corpus under one sample label is deleted, format conversion is performed on the sample data to be converted into a format conforming to the specification of the model, spelling error correction and unification are performed on grammar errors (such as spelling, misprinting words and the like) in the sample data, word segmentation is performed on the sample data in sentence form to be divided into a plurality of words, so as to obtain word sequences of the sample data, and the training sample set for each training and each test sample set can be determined for all sample data.

After preprocessing is completed and a training sample set for training is obtained, model parameters can be loaded, the model is trained, the intention recognition effect of the model can be evaluated through the test sample set after training (namely, model evaluation output in fig. 10), then the training sample set can be selected to be expanded, the model is trained again so as to further improve the intention recognition effect of the model (namely, model persistence), then the test is carried out through the test sample set again, the training effect is verified (namely, model verification), and finally the intention recognition can be carried out through the trained intention recognition model.

The intent recognition model in the present application may be built based on a text classification algorithm, such as a fast text (fastText) classification algorithm, a continuous word bag (CBOW, continues bag of words) algorithm, etc., for example, when fastText is used, sample data may be converted into a format conforming to the regulation of fastText when format conversion is performed, and a penalty value may be calculated based on hierarchical softmax (a penalty function).

In some embodiments, the step of "intent recognition of the query message to obtain a query intent of the query message" includes:

The step of determining the first application reply content of the first virtual robot and the second application reply content of the second virtual robot based on the query intention may include:

sending, by the first virtual robot, an inquiry intention to the second virtual robot; determining, by the first virtual robot, first application reply content from a first application knowledge base based on the query intent; determining, by the second virtual robot, second application reply content from the second application knowledge base based on the query intent.

In order to improve the efficiency of the virtual robots, all the virtual robots contained in the robot question-answering page can answer the query message provided by the object, one virtual robot can conduct intention recognition on the query message to obtain the query intention of the query message, and then the query intention is transmitted to each virtual robot in an intention transmission mode.

In this embodiment, the first virtual robot may perform intention recognition on the query message, obtain a query intention of the query message, and send the query intention to the second virtual robot, so that the first virtual robot may determine the first application reply content from the first application knowledge base based on the query intention, and the second virtual robot may determine the second application reply content from the second application knowledge base based on the query intention.

The robot performing the intention recognition may be determined randomly, may also be determined according to a virtual robot to which the current robot question page belongs, and so on.

For example, referring to fig. 11, the a robot may perform preliminary processing on the query message through the natural language processing module to obtain a query string, perform intent recognition according to the query string by the a robot to obtain an intent recognition result, and transmit the intent recognition result to the dialogue management module of the B robot and the dialogue management module of the a robot, so as to obtain reply content of the a robot and reply content of the B robot respectively.

For another example, referring to fig. 12, the robot question-answer page may include a plurality of users, demands for virtual robots among users may be displayed in a group by means of messages, the users may determine each other's intention by means of messages, and a plurality of robots may be included in the robot question-answer page, one robot may recognize the user's message and transmit the recognized intention among the robots, so that each robot may answer the user's message.

The method described in the above embodiments is described in further detail below by way of example.

The embodiment will be described in detail by taking an intelligent question-answering method integrated in an instant messaging client as an example. For example, the application may include applying XX glowing and applying DD, wherein the glowing small fox may be a virtual robot for XX glowing, the glowing small fox may be a first virtual robot, the small soy sauce may be a virtual robot for applying DD, and the small soy sauce may be a second virtual robot.

Fig. 13 is a schematic flow chart of an intelligent question-answering method according to an embodiment of the present application, as shown in fig. 13. The intelligent question-answering method can comprise the following steps:

201. the computer device displays a first robot question-answer page of the first virtual robot, the robot display page including a shortcut question-answer area and a chat content editing area.

For example, referring to fig. 14, the first robot question and answer page 09a of the glory small fox may include a shortcut question and answer area and a chat content editing area, and the shortcut question and answer area may include several shortcut buttons: canyon chat buttons, hero encyclopedia buttons, and welfare activity buttons.

202. The computer device displays a wake-up message on a first robot question-answer page of the first virtual robot in response to an operation for the chat question-answer editing area.

For example, referring to fig. 14, after the user LL performs the editing operation in the chat question-answer editing area, the first robot question-answer page 09a displays a wake-up message "@ glowing small fox help me wake-up small soy sauce" of the user LL.

203. The computer equipment determines that the target awakening object is the second virtual robot based on the awakening message, and switches to display a second robot question-answer page of the second virtual robot, wherein the second robot question-answer page comprises a chat content editing area and awakening response messages of the second virtual robot.

For example, referring to fig. 14, the glory small fox recognizes the wake-up message, determines that the target wake-up object is small soy sauce, switches to display the second robot question-answer page 09b, and displays the wake-up interaction message "@ small soy sauce fast out" of the glory small fox for the wake-up message on the second robot question-answer page 09b, the chat content editing area of the small soy sauce, and the chat content editing area of the small soy sauce may include a shortcut question-answer area, and the shortcut question-answer area may include a plurality of shortcut buttons: welfare activity buttons, latest bulletin buttons, and copy attack buttons.

The second robot question and answer page 09c may then display a wake-up response message "good for young soy sauce, i am all the year round with ai mei young soy sauce, what is wrong, although i am asking me, fast @ i + you want to say, try a bar, surprise etc. you come,

In addition, after the target wake-up object is determined to be the small soy sauce, a wake-up interaction message "@ the small soy sauce is displayed on the first robot question-answering page, and then the second robot question-answering page displaying the small soy sauce is switched.

204. The computer device responds to the query editing operation aiming at the chat content editing area, and displays a query message on a second robot question-answer page, wherein the query message comprises query content aiming at a first application and a second application, the first application corresponds to a first virtual robot, and the second application corresponds to a second virtual robot.

For example, referring to fig. 14, the user LL may issue an inquiry message "what is the most recent activity of each game? The query message is for all virtual robots in the group, i.e. glowing small foxes and small soy sauce.

205. The computer device performs intent recognition on the query message and determines the query intent of the query message.

For example, the computer device may perform intent recognition on the query message via an intent recognition model to determine that the query intent of the query message is "active".

In performing the intention recognition, a plurality of preset query intents may be included, and the query recognition may be to determine a query intention corresponding to the query message from among the preset query intents, for example, the preset query intents may include "small fox", "small soy sauce", and "wake up virtual robot". The intention recognition can be performed through an intention model, and a sample corpus used by the intention recognition model in training can comprise three labels, namely three preset query intention small foxes (fox), small soy sauce (soy_sace) and a wake-up virtual robot (summon), wherein the following table can be part of the sample corpus and the corresponding sample labels thereof:

Sample corpus	Sample label
		XX glows with hero	fox
Josepia what equipment	fox
		How many copies the DD has	soy_sauce
How to make DD poems	soy_sauce
		Help me wake up small fox	summon
Help me wake-up small soy sauce	summon

In some embodiments, the wake-up virtual robot may be a primary preset query intention, under which a secondary preset query intention may be included, such as a soy sauce wake-up (soy_task_query), and the following table may include a portion of the secondary preset query intention and its corresponding sample corpus:

206. the computer device displays a second reply message of the second virtual robot and a first reply message of the first virtual robot on a second robot question-answer page, wherein the second reply message comprises second reply content determined by the second virtual robot according to the query intention, and the first reply message comprises first reply content determined by the first virtual robot according to the query intention.

For example, referring to fig. 14, the second robot question page 09d may display a first answer message of a glowing small fox and a second answer message of a small soy sauce.

Further, referring to fig. 15, the query message in the robot question page 08a of the user LL may be "@ little fox DD and XX glowing which is rich in money", the query message at this time being for XX glowing and DD, at which time the reply message "the consumption by your last month in XX glowing is 1000 yuan in total, the consumption in DD may be 500 yuan in total" by only the glowing little fox in the robot question page 08 b. The desire to know more detailed running information can help you call the small soy sauce-! ".

207. The computer device displays an interactive message of a second reply message of the first virtual robot to the second virtual robot.

For example, referring to fig. 14, the second robot question and answer page 09e may display the interactive message "@ small soy sauce with old marketing number inner taste" of the glory small fox to the second answer message.

The interactive content in the interactive message may be preset through the interactive content generation page, referring to fig. 16, for the query content, "the underground city has recently been upgraded to send one hundred gold coins, the answer 1 (i.e. an interactive content)" has the taste in the marketing number "may be manually input. By the method, the interactive content can be flexibly and variously set for the query content.

For example, referring to fig. 17, a user may chat in a group and send a message, where the public corpus natural language understanding module in the present application may perform text preprocessing and intent recognition on the message, determine whether to wake up the a robot according to the intent recognition result, if so, send an inquiry message to the dialogue natural language processing module of the a robot, perform text preprocessing and intent recognition, determine whether to wake up the B robot, if so, wake up the B robot, and transmit an inquiry intent, and output a reply message of the inquiry message to the user according to the knowledge base one of the a robot and the knowledge base two of the B robot, and the dialogue management module and the language generating module.

In order to facilitate better implementation of the intelligent question-answering method provided by the embodiment of the application, the embodiment of the application also provides a device based on the intelligent question-answering method. The meaning of the nouns is the same as that of the intelligent question-answering method, and specific implementation details can be referred to the description of the method embodiment.

As shown in fig. 18, fig. 18 is a schematic structural diagram of an intelligent question and answer device according to an embodiment of the present application, where the intelligent question and answer device may include a page display module 301, an inquiry display module 302, and a answer display module 303, where,

the page display module 301 is configured to display a robot question-answer page of the client, where the robot question-answer page includes a chat content editing area;

A query display module 302 for displaying a query message on the robot question page in response to a query editing operation for the chat content editing area, the query message including query content for at least two applications;

the reply display module 303 is configured to identify the query message, and display a reply message on a robot question-and-answer page, where the reply message includes reply contents obtained by identifying at least two virtual robots for the query message, and the at least two virtual robots are in one-to-one correspondence with the at least two applications.

the identification sub-module is used for identifying the query message;

In some embodiments, the at least two virtual robots comprise a first virtual robot and a second virtual robot, the recognition submodule comprises a recognition unit, a determination unit, wherein,

and a determining unit configured to determine first application reply content of the first virtual robot and second application reply content of the second virtual robot based on the query intention.

the interactive display module is used for displaying an interactive message of a second reply message of the first virtual robot aiming at the second virtual robot, wherein the interactive message comprises interactive contents.

the wake-up display module is used for responding to robot wake-up editing operation aiming at the chat content editing area and displaying wake-up information on the robot question-answer page;

in this embodiment, the wake object module is specifically configured to:

In some embodiments, the identification unit is specifically configured to:

in some embodiments, the identification unit is specifically configured to:

preprocessing query content in a query message to obtain a query string;

and the parameter adjustment module is used for adjusting network parameters of the intention recognition model based on the recognition result of the sample text and the label so as to obtain the trained intention recognition model.

In some embodiments, the identification unit is specifically configured to:

in this embodiment, the determining unit is specifically configured to:

sending, by the first virtual robot, an inquiry intention to the second virtual robot;

determining, by the second virtual robot, second application reply content from the second application knowledge base based on the query intent.

In the application, the page display module 301 displays a robot question-answer page of the client, where the robot question-answer page includes a chat content editing area; the query display module 302 displays a query message on the robot question page in response to a query editing operation for the chat content editing area, the query message including query content for at least two applications; the reply display module 303 identifies the query message, and displays a reply message on the robot question-answer page, where the reply message includes reply contents obtained by identifying at least two virtual robots for the query message, and the at least two virtual robots are in one-to-one correspondence with the at least two applications.

In addition, the embodiment of the present application further provides a computer device, which may be a terminal or a server, as shown in fig. 19, which shows a schematic structural diagram of the computer device according to the embodiment of the present application, specifically:

the computer device may include one or more processors 401 of a processing core, memory 402 of one or more computer readable storage media, a power supply 403, and an input unit 404, among other components. Those skilled in the art will appreciate that the computer device structure shown in FIG. 19 is not limiting of the computer device and may include more or fewer components than shown, or may be combined with certain components, or a different arrangement of components. Wherein:

The processor 401 is a control center of the computer device, connects various parts of the entire computer device using various interfaces and lines, and performs various functions of the computer device and processes data by running or executing software programs and/or modules stored in the memory 402, and calling data stored in the memory 402, thereby performing overall monitoring of the computer device. Optionally, processor 401 may include one or more processing cores; preferably, the processor 401 may integrate an application processor and a modem processor, wherein the application processor mainly processes an operating system, a user page, an application program, etc., and the modem processor mainly processes wireless communication. It will be appreciated that the modem processor described above may not be integrated into the processor 401.

The memory 402 may be used to store software programs and modules, and the processor 401 executes various functional applications and data processing by executing the software programs and modules stored in the memory 402. The memory 402 may mainly include a storage program area and a storage data area, wherein the storage program area may store an operating system, an application program (such as a sound playing function, an image playing function, etc.) required for at least one function, and the like; the storage data area may store data created according to the use of the computer device, etc. In addition, memory 402 may include high-speed random access memory, and may also include non-volatile memory, such as at least one magnetic disk storage device, flash memory device, or other volatile solid-state storage device. Accordingly, the memory 402 may also include a memory controller to provide the processor 401 with access to the memory 402.

The computer device further comprises a power supply 403 for supplying power to the various components, preferably the power supply 403 may be logically connected to the processor 401 by a power management system, so that functions of charge, discharge, and power consumption management may be performed by the power management system. The power supply 403 may also include one or more of any of a direct current or alternating current power supply, a recharging system, a power failure detection circuit, a power converter or inverter, a power status indicator, and the like.

The computer device may also include an input unit 404, which input unit 404 may be used to receive input numeric or character information and to generate keyboard, mouse, joystick, optical or trackball signal inputs related to user settings and function control.

Although not shown, the computer device may further include a display unit or the like, which is not described herein. In particular, in this embodiment, the processor 401 in the computer device loads executable files corresponding to the processes of one or more application programs into the memory 402 according to the following instructions, and the processor 401 executes the application programs stored in the memory 402, so as to implement various functions as follows:

Displaying a robot question-answer page of the client, wherein the robot question-answer page comprises a chat content editing area; in response to a query editing operation for the chat content editing area, displaying a query message on the robot question-answer page, the query message including query content for at least two applications; and identifying the query message, and displaying a reply message on a robot question-answer page, wherein the reply message comprises reply contents obtained by identifying at least two virtual robots aiming at the query message, and the at least two virtual robots are in one-to-one correspondence with the at least two applications.

The specific implementation of each operation above may be referred to the previous embodiments, and will not be described herein.

According to one aspect of the present application, there is provided a computer program product or computer program comprising computer instructions stored in a computer readable storage medium. The processor of the computer device reads the computer instructions from the computer-readable storage medium, and the processor executes the computer instructions to cause the computer device to perform the methods provided in the various alternative implementations of the above embodiments.

It will be appreciated by those of ordinary skill in the art that all or part of the steps of the various methods of the above embodiments may be performed by a computer program, or by computer program control related hardware, which may be stored in a computer readable storage medium and loaded and executed by a processor.

To this end, the embodiments of the present application also provide a storage medium having stored therein a computer program that can be loaded by a processor to perform the steps of any of the intelligent question-answering methods provided by the embodiments of the present application. For example, the computer program may perform the steps of:

Wherein the storage medium may include: read Only Memory (ROM), random access Memory (RAM, random Access Memory), magnetic or optical disk, and the like.

The steps in any of the intelligent question-answering methods provided in the embodiments of the present application may be executed by the computer program stored in the storage medium, so that the beneficial effects that any of the intelligent question-answering methods provided in the embodiments of the present application may be achieved, which are described in detail in the previous embodiments and are not repeated herein.

The foregoing has described in detail the methods, apparatuses, storage media and computer devices for intelligent question-answering provided by the embodiments of the present application, and specific examples have been applied to illustrate the principles and embodiments of the present application, where the foregoing description of the embodiments is only for aiding in understanding the methods and core ideas of the present application; meanwhile, as those skilled in the art will vary in the specific embodiments and application scope according to the ideas of the present application, the contents of the present specification should not be construed as limiting the present application in summary.

Claims

1. An intelligent question-answering method is characterized by comprising the following steps:

in response to an inquiry editing operation for the chat content editing area, displaying an inquiry message on the robot inquiry page, wherein the inquiry message comprises inquiry content for at least two applications, the at least two applications are in one-to-one correspondence with at least two virtual robots, and the at least two virtual robots comprise a first virtual robot and a second virtual robot;

identifying the query message, and displaying a reply message on the robot question-answer page, wherein the reply message comprises a reply message of the first virtual robot or the second virtual robot, the reply message is determined according to first application reply content of the first virtual robot and second application reply content of the second virtual robot, the first application reply content and the second application reply content are determined based on the query message, or the reply message comprises a first reply message and a second reply message, the first reply message comprises first application reply content of the first virtual robot, and the second reply message comprises second application reply content of the second virtual robot.

2. The method of claim 1, wherein the at least two virtual robots comprise a first virtual robot and a second virtual robot, the identifying the query message comprising:

performing intention recognition on the query message to obtain the query intention of the query message;

based on the query intent, first application reply content of the first virtual robot and second application reply content of the second virtual robot are determined.

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

acquiring the interactive content of the second reply message;

and displaying an interactive message of a second reply message of the first virtual robot for the second virtual robot, wherein the interactive message comprises the interactive content.

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

responding to a robot wakeup editing operation aiming at the chat content editing area, and displaying a wakeup message on the robot question-answering page;

identifying the wake-up message to determine a target wake-up object from at least two virtual robots;

and displaying the chat content editing area of the target wake-up object on the robot question-answering page.

5. The method of claim 4, wherein after identifying the wake-up message to determine a target wake-up object from at least two virtual robots, further comprising:

displaying a wake-up interaction message aiming at the target wake-up object;

the chat content editing area for displaying the target wake-up object on the robot question-answer page comprises the following steps:

6. The method of claim 2, wherein the performing the intent recognition on the query message to obtain the query intent of the query message comprises:

7. The method of claim 6, wherein the method further comprises:

acquiring a plurality of sample texts and labels thereof;

identifying the sample text through the intention identification model to obtain an identification result of the sample text;

and based on the recognition result of the sample text and the label, carrying out network parameter adjustment on the intention recognition model to obtain a trained intention recognition model.

8. The method of claim 2, wherein the performing the intent recognition on the query message to obtain the query intent of the query message comprises:

the determining, based on the query intent, first application reply content of the first virtual robot and second application reply content of the second virtual robot includes:

9. The utility model provides a virtual robot intelligence question answering device which characterized in that includes:

the query display module is used for responding to the query editing operation aiming at the chat content editing area and displaying a query message on the robot query page, wherein the query message comprises query content aiming at least two applications, the at least two applications are in one-to-one correspondence with at least two virtual robots, and the at least two virtual robots comprise a first virtual robot and a second virtual robot;

The reply display module is used for identifying the query message and displaying a reply message on the robot question and answer page, wherein the reply message comprises a reply message of the first virtual robot or the second virtual robot, the reply message is determined according to first application reply content of the first virtual robot and second application reply content of the second virtual robot, the first application reply content and the second application reply content are determined based on the query message, or the reply message comprises a first reply message and a second reply message, the first reply message comprises first application reply content of the first virtual robot, and the second reply message comprises second application reply content of the second virtual robot.

10. A storage medium storing a plurality of computer programs adapted to be loaded by a processor for performing the steps of the method according to any one of claims 1 to 8.

11. A computer device comprising a memory, a processor and a computer program stored on the memory and executable on the processor, wherein the processor implements the steps of the method according to any one of claims 1 to 8 when the computer program is executed.