1. An IPTV terminal-based mediation consultation method is characterized by comprising the following steps:

acquiring text information;

obtaining a plurality of rounds of man-machine conversations according to the text information;

generating a customized mediation consultation result according to multiple rounds of man-machine conversations;

recommending corresponding mediation resources according to any one or more of the multi-round man-machine conversation content, the mediation result and the user requirements;

and the user determines the needed mediation resource and sends the contents of the multiple rounds of man-machine conversation and/or the mediation consultation result to the selected mediation resource.

2. The IPTV-based mediation consultation method of claim 1, wherein said recommending corresponding mediation resources specifically includes:

recommending a community mediation organization related to the mediation consultation result to the user, and/or;

recommending a moderator to the user, wherein the moderator is related to the mediation consultation result.

3. The IPTV-end-based mediation consultation method of claim 2, wherein said recommending a corresponding mediation resource further comprises:

the introduction, public praise, actual combat experience and case of the community mediation organization;

the professional field, the working time and the actual combat experience and case of the moderator.

4. The IPTV end-based mediation consultation method of claim 1, wherein said obtaining text information specifically includes:

collecting a user voice signal and/or obtaining information of a text, a picture, an interactive page and a chat record, and converting the voice information and/or the information of the text, the picture, the interactive page and the chat record into text information.

5. The IPTV-based mediation consultation method according to claim 1, wherein obtaining a plurality of rounds of man-machine conversations according to the text information specifically comprises:

determining a mediation category domain based on the deep learning intention recognition model;

and generating a man-machine conversation based on a multi-round conversation model according to the mediation category domain.

6. The IPTV-based mediation consultation method of claim 1, wherein said generating a customized mediation consultation result according to a plurality of rounds of man-machine conversation specifically comprises:

for each man-machine conversation, collecting and assigning a mediation characteristic variable;

and generating a customized mediation consultation result based on a reasoning decision strategy model according to the mediation characteristic variable.

7. An IPTV terminal-based mediation consultation system is characterized by comprising:

the intelligent interaction equipment is used for acquiring text information; and performing voice interaction with the user;

the cloud server is used for generating a customized mediation consultation result; recommending corresponding mediation resources to the user;

the television is used for playing and/or displaying the mediation consultation result and displaying the recommended mediation resource;

the IPTV set top box is used for communication between the intelligent interaction equipment and the cloud service; and communication between the cloud server and the television.

8. The IPTV-based mediation consultation system of claim 7, wherein said cloud server includes a mediation resource database;

the mediation resource database is used for storing information of community mediation organizations and mediation personnel.

9. The IPTV-based mediation consultation system of claim 7, wherein said cloud server further comprises:

an intention recognition unit for determining a mediation category domain based on a deep-learned intention recognition model;

the multi-round conversation unit is used for generating a man-machine conversation based on a multi-round conversation model library;

the assignment variable unit is used for collecting and assigning a mediation characteristic variable for each man-machine conversation;

and the reasoning decision unit is used for generating a customized mediation consultation result based on a reasoning decision strategy model according to the mediation characteristic variable.

10. The IPTV-based mediated consultation system of claim 7, wherein said intelligent interactive device further comprises:

the information acquisition unit is used for acquiring voice signals of a user and/or; obtaining information of texts, pictures, interactive pages and chat records;

and the generating unit is used for converting the information of the voice signal and/or the text, the picture, the interactive page and the chat record into text information.

Background

With the development of artificial intelligence technology, in order to meet the consultation requirements of numerous consultants, a plurality of intelligent robots for consultation services appear, the existing robots only arrange the problems proposed by users and then give consultation results, and no relevant prompt is given for the subsequent processes such as how to handle, especially for the intelligent robots used for mediation work, the existing robots are only used for guiding the processes of how to mediate the work of the users, but cannot convey the fact events that the users need to mediate to the mediators, and the users need to communicate with the mediators many times.

Disclosure of Invention

Objects of the invention

In view of the above problems, the present invention is directed to a mediation consultation method and system based on an IPTV end, which recommend a corresponding mediation resource to a user according to a mediation consultation result and a requirement of the user consultation, so that the mediation resource can quickly and accurately know the problem of the user consultation, and provide a corresponding service to the user, thereby satisfying the user that the user can enjoy an accurate, timely, and effective mediation service without going out of home.

(II) technical scheme

As a first aspect of the present invention, the present invention discloses a mediation consultation method based on an IPTV end, including:

acquiring text information;

obtaining a plurality of rounds of man-machine conversations according to the text information;

generating a customized mediation consultation result according to multiple rounds of man-machine conversations;

recommending corresponding mediation resources according to any one or more of the multi-round man-machine conversation content, the mediation consultation result and the user requirement;

and the user determines the needed mediation resource and sends the contents of the multiple rounds of man-machine conversation and/or the mediation consultation result to the selected mediation resource.

In a possible implementation manner, the recommending a corresponding mediation resource specifically includes:

recommending a community mediation organization related to the mediation consultation result to the user, and/or;

recommending a moderator to the user, wherein the moderator is related to the mediation consultation result.

In a possible implementation manner, the recommending a corresponding mediation resource specifically further includes:

the introduction, public praise, actual combat experience and case of the community mediation organization;

the professional field, the working time and the actual combat experience and case of the moderator.

In a possible implementation manner, the obtaining text information specifically includes:

collecting a user voice signal and/or obtaining information of a text, a picture, an interactive page and a chat record, and converting the voice information and/or the information of the text, the picture, the interactive page and the chat record into text information.

In a possible implementation manner, obtaining multiple rounds of man-machine conversations according to the text information specifically includes:

determining a mediation category domain based on the deep learning intention recognition model;

and generating a man-machine conversation based on a multi-round conversation model according to the mediation category domain.

In a possible implementation manner, the generating a customized mediation consultation result according to multiple rounds of man-machine conversations specifically includes:

for each man-machine conversation, collecting and assigning a mediation characteristic variable;

and generating a customized mediation consultation result based on a reasoning decision strategy model according to the mediation characteristic variable.

As a second aspect of the present invention, the present invention also discloses an IPTV-based mediation consultation system, which includes:

the intelligent interaction equipment is used for acquiring text information; and performing voice interaction with the user;

the cloud server is used for generating a customized mediation consultation result; recommending corresponding mediation resources to the user;

the television is used for playing and/or displaying the mediation consultation result and displaying the recommended mediation resource;

the IPTV set top box is used for communication between the intelligent interaction equipment and the cloud service; and communication between the cloud server and the television.

In one possible embodiment, the cloud server includes a mediation resources database;

the mediation resource database is used for storing information of community mediation organizations and mediation personnel.

In one possible embodiment, the cloud server further includes:

an intention recognition unit for determining a mediation category domain based on a deep-learned intention recognition model;

the multi-round conversation unit is used for generating a man-machine conversation based on a multi-round conversation model library;

the assignment variable unit is used for collecting and assigning a mediation characteristic variable for each man-machine conversation;

and the reasoning decision unit is used for generating a customized mediation consultation result based on a reasoning decision strategy model according to the mediation characteristic variable.

In one possible implementation, the intelligent interactive device further includes:

the information acquisition unit is used for acquiring voice signals of a user and/or; obtaining information of texts, pictures, interactive pages and chat records;

and the generating unit is used for converting the information of the voice signal and/or the text, the picture, the interactive page and the chat record into text information.

(III) advantageous effects

The invention discloses a mediation consultation method and a mediation consultation system based on an IPTV terminal, which are based on intelligent interaction equipment, an IPTV set top box and external equipment, provide family mediation consultation service for a user, recommend corresponding mediation resources to the user according to the mediation consultation result and the requirement of the user, enable the mediation resources to quickly and accurately know the problem of the user consultation, and provide corresponding service for the user, thereby meeting the requirement that the user can enjoy accurate, timely and effective mediation service without going out of home.

Drawings

The embodiments described below with reference to the drawings are exemplary and intended to be used for explaining and illustrating the present invention and should not be construed as limiting the scope of the present invention.

FIG. 1 is a flowchart of a consultation mediating method based on an IPTV terminal disclosed by the invention;

FIG. 2 is a flow chart of the present disclosure for obtaining multiple rounds of human-machine conversation;

FIG. 3 is a flow diagram of the present disclosure for generating customized mediation advisory results;

fig. 4 is a schematic diagram of the IPTV-based mediation consultation system disclosed in the present invention.

Reference numerals:

700. an intelligent interaction device; 710. collecting an information unit; 720. a generating unit; 800. a cloud server; 810. mediating a resource database; 830. an intention identifying unit; 840. a multi-turn dialogue unit; 850. assigning a variable unit; 860. a reasoning decision unit; 900. an IPTV set-top box; 910. a television set.

Detailed Description

In order to make the implementation objects, technical solutions and advantages of the present invention clearer, the technical solutions in the embodiments of the present invention will be described in more detail below with reference to the accompanying drawings in the embodiments of the present invention.

It should be noted that: in the drawings, the same or similar reference numerals denote the same or similar elements or elements having the same or similar functions throughout. The embodiments described are some embodiments of the present invention, not all embodiments, and features in embodiments and embodiments in the present application may be combined with each other without conflict. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention.

A first embodiment of the IPTV-side based mediation consultation method of the present disclosure is described in detail below with reference to fig. 1 to 3. The embodiment is mainly applied to mediation consultation, and corresponding mediation resources are recommended to the user according to the mediation consultation result and the requirement of the user consultation, so that the mediation resources can quickly and accurately know the problem of the user consultation, and corresponding services are provided for the user, and therefore the user can enjoy accurate, timely and effective mediation services without going out of home.

As shown in fig. 1, the present embodiment mainly includes the following steps:

and step 100, acquiring text information.

In step 100, acquiring text information specifically includes the following two ways:

the first method is that voice signals of a user are collected and converted into text information, namely, a section of audio is converted into corresponding characters;

the second is to obtain information in the form of text, picture, interactive page, or even chat record, and extract text information related to mediation consultation from the information, and also to use the information as evidence information of mediation consultation, wherein the evidence information is used to prove the fact of the case, such as contract, invoice, recording, and the like.

The text, the picture, the interactive page and the chat record can be obtained from a mobile phone, or can be obtained from a computer or a paper piece.

In the embodiment of the application, the ASR technology, the MIC display and the voice board technology are adopted to convert the voice signal into text information.

And 200, obtaining multiple rounds of man-machine conversations according to the text information.

As shown in fig. 2, in step 200, according to the text information, multiple rounds of man-machine conversations are obtained, which specifically includes the following steps:

step 210, determining a mediation category domain based on an intention recognition model of deep learning;

and step 220, generating a man-machine conversation based on the multi-round conversation model according to the mediation type domain.

In the method, a cloud server obtains a Toronto man-machine conversation according to text information, wherein after the text information is received in the cloud server, a category consulted by a user is determined based on an intention recognition model of deep learning, a corresponding category domain is searched from a category domain library, after a mediation category domain consulted by the user is determined, a corresponding multi-turn conversation model is called from the Toronto conversation model library, a conversation mechanism between the user and equipment is generated based on the multi-turn conversation model, and mediation fact information is finally generated.

For example, the mediation issues consulted by current users are: debt dispute, when the user says: i borrow 5 ten thousand yuan cash to king on 3/4/2020, and then the king is returned after committing for one year, but the cash is not returned beyond half a year. The method comprises the steps that after receiving voice information of a user, the equipment converts the voice information into text information and sends the text information to a cloud server, after receiving the text information, the cloud server recognizes that the intention of the user is 'owing and overdue', after determining the intention of the user, finds a category domain matched with the 'owing and overdue' from a category domain library, and after determining the category domain, calls corresponding multi-turn conversation models from a multi-turn conversation model library to have a conversation with the user, for example: whether a debit exists or not and whether the money is returned by multiple parties or not are prompted are inquired, so that the user can more accurately and specifically describe the passing of things, and after multiple rounds of conversation, the mediation fact information consulted by the user is generated.

And 300, generating a customized mediation consultation result according to the multiple rounds of man-machine conversation.

As shown in fig. 3, in step 300, a customized mediation consultation result is generated according to multiple rounds of man-machine conversations, which specifically includes the following steps:

step 310, collecting and assigning a mediation characteristic variable for each man-machine conversation;

after the corresponding multi-round conversation model is called, each conversation between the equipment and the user receives and converts the voice signal of the user into the text information, the text information is collected and the mediation characteristic variable is assigned, so that the mediation characteristic content of each interaction between the user and the equipment is accurately determined.

And step 320, generating a customized mediation consultation result based on the reasoning decision strategy model according to the mediation characteristic variable.

After the text information of the user is assigned with the mediation characteristic variable, the assigned mediation characteristic variable is input into the value inference decision strategy model, and then a customized mediation consultation result is generated, wherein the mediation consultation result comprises appeal, brief description of events, judicial materials and some relative suggestions of the user, and after the customized mediation consultation result is generated, the mediation consultation result is displayed and/or played to the user through an IPTV and a television, or is displayed in an interactive interface mode through a mobile phone, a computer, a projector and other equipment, and is accompanied with automatic voice generation and broadcast, wherein multimedia type interaction can be automatically generated and displayed, and the interaction comprises automatically generated animation, for example, a cartoon character is displayed on the television for broadcast and the like.

Step 400, recommending corresponding mediation resources according to any one or more of the multi-turn man-machine conversation content, mediation consultation results and user requirements.

In step 400, according to any one or more of the multi-turn human-computer conversation content, mediation consultation result and user requirement, recommending corresponding mediation resources, which specifically includes:

after the customized mediation consultation result is generated, the device can consult the requirement of the mediation resource needed by the user, and recommend the mediation resource meeting the user appeal to the user according to the requirement of the user, the mediation consultation result and the contents of the multi-turn man-machine conversation, so that diversified services are provided for the user.

In a possible embodiment, recommending a corresponding mediation resource specifically includes:

recommending community mediation organizations related to mediation consultation results to the users, and/or;

and recommending a moderator related to the mediation consultation result to the user.

In a possible embodiment, recommending a corresponding mediation resource specifically includes:

brief introduction, public praise, actual combat experience and case of community mediation organization;

the professional field, the practice time and the actual combat experience and case of the moderator.

When recommending corresponding mediation resources for a user, the IPVT can be connected with other equipment which can be watched and/or played by the user, such as a television, a mobile phone, a projector and the like, so as to present the mediation resources, the recommended mediation resources can be ordered to a certain extent according to the requirements of the user, and when recommending the mediation resources, the mediation resources can be known more deeply by watching actual combat cases of the mediation resources.

For example, the user is currently located in an Tianjin yellow river channel purple flower garden community, after the user finishes the mediation service consultation, the device consults the requirement of the user for mediation resources, if the user only needs the community mediation organization of the community where the user is located, the device recommends the community mediation organization of the purple flower garden community to the user according to the user requirement, the mediation result, debt dispute and the content of multiple rounds of man-machine conversation, such as debt and overdue, and the like, and processes the debt dispute, and simultaneously the device gives corresponding other suggestions according to the regulation consultation result of the user consultation, comprehensively sorts the suggestions, and sends the suggestions to a television, a mobile phone or other devices for the user to select.

Step 500, the user determines the needed mediation resource, and sends the contents of the multiple rounds of man-machine conversation and/or mediation consultation results to the selected mediation resource.

In step 500, the user determines the needed mediation resource, and sends the multiple rounds of human-computer conversation contents and/or mediation consultation results to the selected mediation resource, which specifically includes:

after the user selects the mediation resource, the device directly sends the multi-turn man-machine conversation content and/or the mediation consultation result to the mediation resource, the mediation resource can timely, accurately and effectively receive the user's appeal, and the user can ask for answers on line with the mediation resource through other devices such as a television and a mobile phone, so that more accurate and effective mediation service is provided for the user until the user solves the problem. For example, the user selects a moderator, the device sends the mediation consultation result of the user consultation to the moderator, the moderator can clearly and accurately know the event which the user needs to mediate, and the moderator can make a corresponding reply according to the requirement of the user.

The following describes in detail with reference to fig. 4, and based on the same inventive concept, the embodiment of the present invention further provides a first embodiment of an IPTV-side based consultation mediation system. Because the principle of the problem solved by the system is similar to the IPTV-end-based mediation and consultation method, the implementation of the system can refer to the implementation of the method, and repeated details are not repeated. The embodiment is mainly applied to the mediation consultation condition, and recommends the corresponding mediation resource to the user according to the mediation consultation result and the demand of the user consultation, so that the mediation resource can quickly and accurately know the problem of the user consultation, and provides the corresponding service for the user, thereby meeting the requirement that the user can enjoy the accurate, timely and effective mediation service without going out of home.

As shown in fig. 4, the present embodiment mainly includes an intelligent interaction device 700, a cloud server 800, an IPTV set-top box 900 and a television 910, where the intelligent interaction device 700 is connected to the IPTV set-top box, the IPTV set-top box 900 is connected to the cloud server 800 by using the internet, and the IPTV set-top box 900 is connected to the television by using the internet.

The intelligent interaction device 700 is used for acquiring text information; and performing voice interaction with the user;

the cloud server 800 is used for generating customized mediation consultation results and recommending relevant mediation resources to the user;

the television 910 is used for playing and/or displaying mediation consultation results and displaying recommended mediation resources;

the IPTV set top box 900 is used for communication between the intelligent interaction device and the cloud server; and communication between the cloud server and the television;

cloud server 800 includes a mediation resources database 810, mediation resources database 810 for storing information of community mediation organizations and mediators. A large number of moderator in different fields and different types are prestored in the mediation resource database 810 so as to provide different mediation resources for users to meet the requirements of different users, and meanwhile, the contents of the mediation resource database can be modified and increased in real time or regularly according to the actual requirements.

Specifically, a user can use the home intelligent interaction device 700 to perform mediation consultation, the intelligent interaction device 700 collects voice information, texts, pictures, interaction pages, chat records of the user and converts the voice information, the texts, the pictures, the interaction pages, the chat records and other information provided by the user into text information, the text information is sent to the cloud server 800 through the IPTV set-top box 900 by using the internet of things, the cloud server 800 generates a mediation consultation result by using the received text information, and the mediation consultation result is sent to the television by using the IPTV set-top box 900 to play and/or display the mediation consultation result for the user.

The cloud server 800 further recommends corresponding mediation resources to the user according to the mediation consultation result, the multi-turn man-machine conversation contents and the user requirements, and sends the recommended corresponding mediation resources to the IPTV set top box 900, the IPTV set top box 900 sends the mediation resources to the television to play and/or display mediation resource information for the user, the user can select mediation resources meeting the self requirements according to the recommended mediation resources, after the user selects, the cloud service sends the mediation consultation result and/or the multi-turn man-machine conversation contents to the mediation resources selected by the user, and therefore the user can enjoy accurate, timely and effective mediation service without going out.

In one possible implementation, the cloud server 800 includes: an intention recognition unit 830, a multi-turn dialogue unit 840, an assignment variable unit 850 and an inference decision unit 860, wherein the intention recognition unit 830 is configured to determine a mediation category domain based on a deep learning intention recognition model; the multi-turn dialogue unit 840 is used for generating a man-machine dialogue based on the multi-turn dialogue model library; the assignment variable unit 850 is used for collecting and assigning the mediation characteristic variable for each man-machine conversation; and the reasoning decision unit 860 is used for generating a customized mediation consultation result based on the reasoning decision strategy model according to the mediation characteristic variable.

The man-machine conversation generated by the multi-turn conversation unit 840 can interact with the user in a voice form, specifically, the man-machine conversation can be sent to the IPTV set-top box 900 by the sound box of the intelligent interaction device 700, and the IPTV set-top box 900 plays voice information by using the sound box of the television to interact with the user; or in the form of a menu interface, and is sent to the IPTV set-top box 900 to be displayed on the television, and the user uses the television remote controller to select the menu in an interactive manner; and an interactive mode for displaying the menu interface and selecting can be also utilized by a mobile phone and a computer.

In a possible implementation, the intelligent interaction device 700 further includes an information collecting unit 710 and a generating unit 720, where the information collecting unit 710 is configured to collect a voice signal of a user, or; the method comprises the steps of obtaining information of texts, pictures, interactive pages and chat records, converting the information of the texts, the pictures, the interactive pages and the chat records into text information by the generating unit 720 according to voice signals and/or the information of the texts, the pictures, the interactive pages and the chat records, and converting the voice signals into the text information by the generating unit 720 by adopting an ASR technology, an MIC display technology and a voice board technology.

The above description is only for the specific embodiments of the present invention, but the scope of the present invention is not limited thereto, and any changes or substitutions that can be easily conceived by those skilled in the art within the technical scope of the present invention are included in the scope of the present invention. Therefore, the protection scope of the present invention shall be subject to the protection scope of the claims.