CN110659355A - Conversation control method and system - Google Patents
Conversation control method and system Download PDFInfo
- Publication number
- CN110659355A CN110659355A CN201810713598.7A CN201810713598A CN110659355A CN 110659355 A CN110659355 A CN 110659355A CN 201810713598 A CN201810713598 A CN 201810713598A CN 110659355 A CN110659355 A CN 110659355A
- Authority
- CN
- China
- Prior art keywords
- user
- topic
- script
- information
- script information
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
- 238000000034 method Methods 0.000 title claims abstract description 34
- 238000013515 script Methods 0.000 claims abstract description 137
- 230000003993 interaction Effects 0.000 claims abstract description 45
- 238000004458 analytical method Methods 0.000 claims abstract description 33
- 230000008451 emotion Effects 0.000 claims abstract description 7
- 230000002452 interceptive effect Effects 0.000 claims description 75
- 230000014509 gene expression Effects 0.000 claims description 9
- 230000008859 change Effects 0.000 claims description 8
- 238000013461 design Methods 0.000 claims description 3
- 230000002996 emotional effect Effects 0.000 claims description 3
- 238000005192 partition Methods 0.000 claims description 3
- 238000012216 screening Methods 0.000 claims description 3
- 238000007619 statistical method Methods 0.000 claims description 3
- 238000005516 engineering process Methods 0.000 abstract description 3
- 238000012545 processing Methods 0.000 description 6
- 238000010586 diagram Methods 0.000 description 3
- 238000013473 artificial intelligence Methods 0.000 description 2
- 230000008569 process Effects 0.000 description 2
- 230000006978 adaptation Effects 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000004422 calculation algorithm Methods 0.000 description 1
- 230000010485 coping Effects 0.000 description 1
- 238000007405 data analysis Methods 0.000 description 1
- 230000006870 function Effects 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 230000036651 mood Effects 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
- 230000004044 response Effects 0.000 description 1
Images
Abstract
The embodiment of the invention provides a conversation control method and a conversation control system, which utilize a natural language understanding technology to provide a conversation control for a voice interaction system, select a single-turn conversation script or a plurality of turns of topic scripts according to intention understanding and emotion analysis input by a user, and judge topic changing, question and preference and the like according to interaction occurrence conditions. Meanwhile, the system models the obtained data by utilizing big data, tracks the preference topics of the user and the real-time heat of each topic, and is used for intelligently recommending the topics.
Description
Technical Field
The present invention relates to the field of session control technologies, and in particular, to a session control method and system.
Background
At present, there are two common dialog methods: one is question-and-answer type single-turn dialogue, and a background search script gives a reply depending on the text input by a user; another is a multi-turn dialog by writing a multi-turn question-and-answer script by predicting the likely responses of the user. The application scenes of the user and the system are not completely the same, the user looks rigid and unnatural by using the same conversation method under all the situations, the user and the system can have more natural conversation interaction after the user and the system are reasonably combined and integrated.
The prior proposal has the following problems:
the single-round conversation method adopts a coping strategy, replies are given aiming at the input of the user, the conversation method does not combine the user context, depends on the input of the user, belongs to passive conversation, can not carry out system deep conversation around specific topics, is only suitable for the interactive occasions of tools such as inquiry, broadcast and the like, and limits the use in a plurality of scenes.
The method has the advantages that the possible input of a user is predicted by means of multiple rounds of conversations and multiple rounds of conversation scripts, reply information is organized by the system according to the possible input of the user, replies of the user cannot be exhausted, inappropriate replies are often obtained when the input of the user exceeds the multiple rounds of conversation scripts, and the conversation system is rigid and loses intelligence only by means of the multiple rounds of conversations.
The combination of single round and many rounds, mostly carry out single round and many rounds of matching simultaneously to user's input at present, select one of them to reply according to the degree of confidence, this kind of mode has only simply added single round and many rounds of dialogues, has not solved the problem of the interactive naturalness of dialogue more than, and both disadvantages also exist simultaneously.
Disclosure of Invention
In view of the foregoing technical problems, embodiments of the present invention provide a dialog control method and system, which can provide a dialog control scheme for a voice/text dialog interaction system, so as to provide a more intelligent and efficient scheme for intelligent dialog.
An embodiment of the present invention provides a dialog control method, including the following steps:
acquiring interactive information input by a user, and acquiring script information corresponding to the interactive information in a database and pushing the script information to the user when the interactive information is determined to be non-topic ongoing interaction;
when the interactive information is determined to be the interaction carried out in the topic, further determining whether the user has the tendency of changing the topic;
when the user is determined to have a tendency of changing topics, selecting script information corresponding to a new topic according to the historical data of the user and pushing the script information to the user;
and when the user is determined not to have the tendency of changing topics, acquiring script information corresponding to the interaction information from a database and pushing the script information to the user.
Further comprising the steps of:
when script information corresponding to the interaction information cannot be acquired in a database, screening according to the historical data of the user: and pushing script information corresponding to the topics of the preference types of the users with historical data, and pushing script information corresponding to the hot topics in which the users participate in the current system for the new users without the historical data.
Further comprising the steps of:
when determining that the user has a tendency to change topics, further judging the tendency of the user: when the user is determined to be disliked of the current topic, selecting script information corresponding to the new topic according to historical data of the user and pushing the script information to the user; and when the user is a new user without historical data, pushing script information corresponding to the hot topics in which the user participates in the current system.
Further comprising the steps of:
when the user is determined to have the tendency of changing to a specific topic, searching the interactive information input by the user in a database, and selecting script information corresponding to the corresponding topic for pushing when the corresponding topic is searched; otherwise, the script information corresponding to the hot topics participated in by the user in the current system is pushed.
Further comprising the steps of:
when it is determined that the user does not have a tendency to change topics, continuing to select branch nodes of multiple rounds of scripts for the user: and when the corresponding branch node is determined to exist in the database, outputting script information corresponding to the corresponding node information and pushing the script information to a user.
Further comprising the steps of:
when determining that no corresponding branch node exists in the database, judging the relevance between the interactive information input by the user and the current topic:
and when the interactive information is determined to be related to the current topic, continuing to select script information corresponding to the current topic and pushing the script information to the user.
Further comprising the steps of:
when the interactive information is determined to be related to the current topic, judging whether the interactive information is a question intention, if so, selecting single-round script information corresponding to the interactive information to output; otherwise, continuing to output the script information corresponding to the next branch node.
Further comprising the steps of:
when the interactive information is determined to be irrelevant to the current topic, outputting a corresponding single-turn script; a retrieval database, which directly outputs script information corresponding to the topic when retrieving the topic matched with the interactive information; and when the topics matched with the interactive information are not searched, acquiring historical data of the user, pushing script information corresponding to the topics with preference types for the user with the historical data, and pushing script information corresponding to hot topics in which the user participates in the current system for a new user without the historical data.
The embodiment of the invention also provides a conversation control system, which comprises an intelligent analysis module, an intelligent selection module, an intelligent recommendation module, an intelligent statistic module and a bottom database module, and the conversation control system comprises the following modules:
the intelligent analysis module is used for analyzing the interactive information input by the user, understanding the semantic meaning of the interactive information, analyzing the user intention and analyzing the user emotion; transmitting the corresponding analysis result to the intelligent selection module and the intelligent recommendation module, and selecting the dialogue script information by the intelligent selection module or the intelligent recommendation module;
the intelligent selection module is used for receiving the analysis result of the intelligent analysis module, selecting the dialogue script information according to the analysis result and outputting the analyzed script information;
the intelligent recommendation module is used for recommending topic script information by using the result obtained by the intelligent analysis module and the historical data of the user and outputting the analyzed script information;
the intelligent statistical module is used for acquiring all data of the system, performing statistical analysis on the data, performing model establishment for each user, tracking the preference topics of the users, modeling the topics, recording hot topics in different periods, and storing statistical results in the bottom database module;
the bottom database module is used for storing the use data of the user and the single-round and multi-round dialogue scripts; the single-round dialogue script is subjected to partition management according to the categories of human design, encyclopedias, tools, call making, daily expressions, emotional expressions, fixed expressions and super class selling; the multi-turn dialog script is managed in a partitioned manner according to entertainment, education, adults, and children.
The system comprises at least one client and a server, wherein the server comprises an intelligent analysis module, an intelligent selection module, an intelligent recommendation module, an intelligent statistic module and a bottom database module, wherein,
the client is used for interacting with the user, collecting interaction information input by the user and transmitting the interaction information to the server in real time; obtaining conversation script information returned by the server and displaying the conversation script information to a user;
and the server is used for interacting with the client, receiving the interactive information input by the user, analyzing the interactive information and returning the dialogue script information corresponding to the analyzed interactive information to the client.
The technical scheme has the following advantages or beneficial effects:
in various embodiments of the present invention, a method for dialog control is provided for a voice interactive system using natural language understanding techniques. And selecting a single-turn dialogue script or a plurality of turns of topic scripts according to the intention understanding and emotion analysis of the user input. And according to the conditions of interaction occurrence, the judgment of topic changing, sentence asking, hobby and the like is carried out, the complete flow is realized, the natural and smooth conversation is realized, and the psychological characteristics of people are met. Meanwhile, the system models the obtained data by utilizing big data. The method comprises the following steps of tracking preference topics of a user and the real-time heat degree of each topic for intelligent recommendation of the topics.
Drawings
Fig. 1 is a flowchart of a session control method according to an embodiment of the present invention.
Fig. 2 is a schematic structural diagram of a dialog control system according to a second embodiment of the present invention.
Fig. 3 is a flowchart illustrating a dialog control method according to a third embodiment of the present invention.
Detailed Description
Exemplary embodiments of the present invention will be described in more detail below with reference to the accompanying drawings. While exemplary embodiments of the invention are shown in the drawings, it should be understood that the invention can be embodied in various forms and should not be limited to the embodiments set forth herein. Rather, these embodiments are provided so that this disclosure will be thorough and complete, and will fully convey the scope of the invention to those skilled in the art.
In various embodiments of the present invention, a single-turn dialog script is used as one of the bases of the present system, and a multi-turn dialog script is used as one of the bases of the present system. A multi-turn dialogue script takes a topic as a unit, and can carry out multi-turn interaction, wherein each turn contains different branch node information. The scheme relies on intention analysis and emotion analysis input by a user, and provides technical support for naturalness and intelligence of conversation interaction by using an artificial intelligence algorithm. The method is carried out by combining with the historical data of the user, so that the dialog system has the personalized attribute of the user. The method considers different situations of user input in topics and out of topics, combines various condition judgment functions such as topic changing intention judgment, question judgment, topic relevance judgment and the like, and strives to make conversation more consistent with human psychology.
Fig. 1 is a flowchart of a session control method according to an embodiment of the present invention. As shown in fig. 1, the dialog control flow includes the following steps:
The present embodiment is based on a specific interactive system comprising a plurality of clients and at least one server. The client can be in various forms, for example, the client can be program software or a webpage of a PC side, and can also be an APP or a webpage of a mobile side. Of course, a specific device is also possible. For example, the client may be in the form of a story machine that is easy to communicate with children or other forms that children prefer.
The server of the system is located at the cloud end and is connected with each terminal in real time through a network. The server side is provided with a database, and the database comprises all data of interaction between each client side and the server, and is used for analyzing big data to obtain specific analysis data.
The client side is mainly used for collecting the interaction information of the user and transmitting the interaction information to the server in real time. The interaction information of the user is various data interacted between the user and the client. Including voice interaction, video capture, or other forms of interaction. For example, the simplest key way may be used for interaction.
After receiving the interactive information, the server needs to identify the interactive information, for example, natural language understanding processing or image processing needs to be performed. Intent recognition and the resulting sentiment analysis are also required. The server carries out structured output on the dialogue information by utilizing artificial intelligence technologies such as voice recognition, age recognition, natural language understanding and the like, carries out semantic understanding and interaction based on voice input, and obtains voice interaction data of the children in real time to serve as a data basis of subsequent data analysis.
The server finds corresponding dialogue script information in the database according to the processing result, and then sends the content script information to the client for display to the user, so that interaction with the user is realized. Meanwhile, further interactive contents of the user are collected, and subsequent analysis processing is carried out.
After the interactive information input by the user is obtained, the content of the interactive information is obtained through analysis. And then, judging according to the content of the interactive information, determining whether the content of the interactive information is related to the currently ongoing topic, namely whether the interactive information is the currently ongoing interaction of the topic, and if so, continuing to perform subsequent interactive feedback operation on the topic. And if the interactive information is determined to be the interaction in the non-topic process, acquiring script information corresponding to the interactive information in a database and pushing the script information to the user.
The interaction here is a single round of dialogue, i.e. a one-to-one dialogue of explicit content. And when the interactive information input by the user is irrelevant to the current topic, searching and selecting the dialogue script information corresponding to the interactive information in the database according to the interactive information input by the user, and then feeding back and pushing the dialogue script information to the user. A simple interaction procedure is accomplished here.
When script information corresponding to the interaction information cannot be acquired in a database, screening according to the historical data of the user: and pushing script information corresponding to the topics of the preference types of the users with historical data, and pushing script information corresponding to the hot topics in which the users participate in the current system for the new users without the historical data.
And 102, when the interactive information is determined to be the interaction carried out in the topic, further determining whether the user has the tendency of changing the topic.
When the interactive information input by the user is still the interaction performed in the topic, multiple rounds of selection of the conversation nodes are required. It is further determined whether the user is inclined to change topics. This can be analyzed according to the mood, content, words, etc. contained in the interactive information input by the user.
And 103, when the user is determined to have a tendency of changing topics, selecting script information corresponding to a new topic according to the historical data of the user and pushing the script information to the user.
When determining that the user has a tendency to change topics, further analyzing the topic preference of the user, and pushing corresponding topic script information according to the preference. In general, the topic preferences of a user may be selected based on their historical data. And if the user does not have historical data, recommending according to the hot data.
When determining that the user has a tendency to change topics, further judging the tendency of the user: when the user is determined to be disliked of the current topic, selecting script information corresponding to the new topic according to historical data of the user and pushing the script information to the user; and when the user is a new user without historical data, pushing script information corresponding to the hot topics in which the user participates in the current system.
When the user is determined to have the tendency of changing to a specific topic, searching the interactive information input by the user in a database, and selecting script information corresponding to the corresponding topic for pushing when the corresponding topic is searched; otherwise, the script information corresponding to the hot topics participated in by the user in the current system is pushed.
And 104, when the user is determined not to have the tendency of changing topics, acquiring script information corresponding to the interaction information from a database and pushing the script information to the user.
When the user is determined not to have the intention of changing topics, corresponding processing is needed.
When it is determined that the user does not have a tendency to change topics, continuing to select branch nodes of multiple rounds of scripts for the user: and when the corresponding branch node is determined to exist in the database, outputting script information corresponding to the corresponding node information and pushing the script information to a user.
When determining that no corresponding branch node exists in the database, judging the relevance between the interactive information input by the user and the current topic:
and when the interactive information is determined to be related to the current topic, continuing to select script information corresponding to the current topic and pushing the script information to the user.
When the interactive information is determined to be related to the current topic, judging whether the interactive information is a question intention, if so, selecting single-round script information corresponding to the interactive information to output; otherwise, continuing to output the script information corresponding to the next branch node.
When the interactive information is determined to be irrelevant to the current topic, outputting a corresponding single-turn script; a retrieval database, which directly outputs script information corresponding to the topic when retrieving the topic matched with the interactive information; and when the topics matched with the interactive information are not searched, acquiring historical data of the user, pushing script information corresponding to the topics with preference types for the user with the historical data, and pushing script information corresponding to hot topics in which the user participates in the current system for a new user without the historical data.
In order to implement the above flow, another embodiment of the technical solution of the present invention provides a dialog control system, and fig. 2 is a schematic structural diagram of the dialog control system in the second embodiment of the present invention.
As shown in fig. 2, the system includes an intelligent analysis module 21, an intelligent selection module 22, an intelligent recommendation module 23, an intelligent statistics module 24, and a bottom database module 25, which are as follows:
the intelligent analysis module 21 is configured to analyze interactive information input by a user, understand semantic meanings of the interactive information, analyze user intentions, and analyze user emotions; transmitting the corresponding analysis result to the intelligent selection module 22 and the intelligent recommendation module 23, and selecting the dialogue script information by the intelligent selection module 22 or the intelligent recommendation module 23;
the intelligent selection module 22 is configured to receive the analysis result of the intelligent analysis module 21, select the dialog script information according to the analysis result, and output the parsed script information;
the intelligent recommendation module 23 is configured to recommend the topic script information by using the result obtained by the intelligent analysis module 21 and the historical data of the user, and output the parsed script information;
the intelligent statistical module 24 is configured to obtain all data of the system, perform statistical analysis on the data, perform model building for each user, track preferred topics of the user, model the topics, record trending topics at different periods, and store statistical results in the bottom database module;
the bottom database module 25 is used for storing the use data of the user and the single-round and multi-round dialogue scripts; the single-round dialogue script is subjected to partition management according to the categories of human design, encyclopedias, tools, call making, daily expressions, emotional expressions, fixed expressions and super class selling; the multi-turn dialog script is managed in a partitioned manner according to entertainment, education, adults, and children.
Further, the system comprises at least one client and a server, the server comprises an intelligent analysis module 21, an intelligent selection module 22, an intelligent recommendation module 23, an intelligent statistics module 24 and an underlying database module 25, wherein,
the client is used for interacting with the user, collecting interaction information input by the user and transmitting the interaction information to the server in real time; obtaining conversation script information returned by the server and displaying the conversation script information to a user;
and the server is used for interacting with the client, receiving the interactive information input by the user, analyzing the interactive information and returning the dialogue script information corresponding to the analyzed interactive information to the client.
In fact, the specific implementation flow of the embodiment of the present invention can be seen in fig. 3, which is a process diagram of a specific dialog control scheme, wherein,
and (4) searching and outputting single-round corresponding script information when the user inputs the interaction which occurs outside the topic, namely the interaction which is currently carried out on the non-topic, and simultaneously retrieving and pushing the topic. If the topics associated with the user input exist, directly outputting corresponding script information; the users push the topics of the preference types, and for new users without historical data, the hot topics in which the users participate in the current system are pushed. And outputting corresponding multi-turn script information after selecting the topics.
And if the interaction of the user occurs in the topic, namely the interaction currently performed in the topic, judging whether the user has the intention of changing the topic. If the user has a clear intention of changing topics, the tendency of the user is further judged, a new topic is selected through the historical data of the user only when the user does not like the current topic, the user with the historical data pushes the preferred topic, and the new user without the historical data pushes the hot topic participated by the user in the current system. If the user has an intention to switch to a specific topic, searching is performed for the input, the corresponding topic is selected when the corresponding topic is searched, and the trending topic is selected when the corresponding topic is not searched. And outputting corresponding script information after selecting the topics.
And if the user does not have the intention of clearly changing the topics, continuously selecting the branch nodes of the multiple rounds of scripts. And outputting corresponding node information when corresponding branch nodes exist. And when no corresponding branch node exists, judging the relevance between the user input and the topic: if the query is related to the topic, further judging whether the input is a query intention, if so, selecting single-turn script information to output, and if not, continuously outputting next node information; when the input is not related to the topic, outputting a corresponding single-turn script, searching a topic library, searching the topic matched with the input and directly outputting corresponding script information, calling historical data of the user when the corresponding topic is not searched, pushing the preference topic of the user with the historical data, and pushing the hot topic of the user of the current system for a new user without the historical data. And outputting corresponding script information after selecting the topics.
In various embodiments of the present invention, a method for dialog control is provided for a voice interactive system using natural language understanding techniques. And selecting a single-turn dialogue script or a plurality of turns of topic scripts according to the intention understanding and emotion analysis of the user input. And according to the conditions of interaction occurrence, the judgment of topic changing, sentence asking, hobby and the like is carried out, the complete flow is realized, the natural and smooth conversation is realized, and the psychological characteristics of people are met. Meanwhile, the system models the obtained data by utilizing big data. The method comprises the following steps of tracking preference topics of a user and the real-time heat degree of each topic for intelligent recommendation of the topics.
In addition, functional units in the embodiments of the present invention may be integrated into one processing unit, or each unit may be physically included alone, or two or more units may be integrated into one unit. The integrated unit can be realized in a form of hardware, or in a form of hardware plus a software functional unit.
The integrated unit implemented in the form of a software functional unit may be stored in a computer readable storage medium. The software functional unit is stored in a storage medium and includes several instructions to enable a computer device (which may be a personal computer, a server, or a network device) to execute some steps of the transceiving method according to various embodiments of the present invention. And the aforementioned storage medium includes: various media capable of storing program codes, such as a usb disk, a removable hard disk, a Read-Only Memory (ROM), a Random Access Memory (RAM), a magnetic disk, or an optical disk.
While the preferred embodiments of the present invention have been described, it should be understood that modifications and adaptations to those embodiments may occur to one skilled in the art without departing from the principles of the present invention and are within the scope of the present invention.
Claims (10)
1. A dialog control method, characterized by comprising the steps of:
acquiring interactive information input by a user, and acquiring script information corresponding to the interactive information in a database and pushing the script information to the user when the interactive information is determined to be non-topic ongoing interaction;
when the interactive information is determined to be the interaction carried out in the topic, further determining whether the user has the tendency of changing the topic;
when the user is determined to have a tendency of changing topics, selecting script information corresponding to a new topic according to the historical data of the user and pushing the script information to the user;
and when the user is determined not to have the tendency of changing topics, acquiring script information corresponding to the interaction information from a database and pushing the script information to the user.
2. The dialog control method of claim 1 further comprising the steps of:
when script information corresponding to the interaction information cannot be acquired in a database, screening according to the historical data of the user: and pushing script information corresponding to the topics of the preference types of the users with historical data, and pushing script information corresponding to the hot topics in which the users participate in the current system for the new users without the historical data.
3. The dialog control method of claim 1 further comprising the steps of:
when determining that the user has a tendency to change topics, further judging the tendency of the user: when the user is determined to be disliked of the current topic, selecting script information corresponding to the new topic according to historical data of the user and pushing the script information to the user; and when the user is a new user without historical data, pushing script information corresponding to the hot topics in which the user participates in the current system.
4. The dialog control method of claim 3 further comprising the steps of:
when the user is determined to have the tendency of changing to a specific topic, searching the interactive information input by the user in a database, and selecting script information corresponding to the corresponding topic for pushing when the corresponding topic is searched; otherwise, the script information corresponding to the hot topics participated in by the user in the current system is pushed.
5. The dialog control method of claim 1 further comprising the steps of:
when it is determined that the user does not have a tendency to change topics, continuing to select branch nodes of multiple rounds of scripts for the user: and when the corresponding branch node is determined to exist in the database, outputting script information corresponding to the corresponding node information and pushing the script information to a user.
6. The dialog control method of claim 5 further comprising the steps of:
when determining that no corresponding branch node exists in the database, judging the relevance between the interactive information input by the user and the current topic:
and when the interactive information is determined to be related to the current topic, continuing to select script information corresponding to the current topic and pushing the script information to the user.
7. The dialog control method of claim 6 further comprising the steps of:
when the interactive information is determined to be related to the current topic, judging whether the interactive information is a question intention, if so, selecting single-round script information corresponding to the interactive information to output; otherwise, continuing to output the script information corresponding to the next branch node.
8. The dialog control method according to claim 6 or 7, characterized by further comprising the steps of:
when the interactive information is determined to be irrelevant to the current topic, outputting a corresponding single-turn script; a retrieval database, which directly outputs script information corresponding to the topic when retrieving the topic matched with the interactive information; and when the topics matched with the interactive information are not searched, acquiring historical data of the user, pushing script information corresponding to the topics with preference types for the user with the historical data, and pushing script information corresponding to hot topics in which the user participates in the current system for a new user without the historical data.
9. The utility model provides a conversation control system, its characterized in that includes intelligent analysis module, intelligent selection module, intelligent recommendation module, intelligent statistics module and bottom database module, specifically as follows:
the intelligent analysis module is used for analyzing the interactive information input by the user, understanding the semantic meaning of the interactive information, analyzing the user intention and analyzing the user emotion; transmitting the corresponding analysis result to the intelligent selection module and the intelligent recommendation module, and selecting the dialogue script information by the intelligent selection module or the intelligent recommendation module;
the intelligent selection module is used for receiving the analysis result of the intelligent analysis module, selecting the dialogue script information according to the analysis result and outputting the analyzed script information;
the intelligent recommendation module is used for recommending topic script information by using the result obtained by the intelligent analysis module and the historical data of the user and outputting the analyzed script information;
the intelligent statistical module is used for acquiring all data of the system, performing statistical analysis on the data, performing model establishment for each user, tracking the preference topics of the users, modeling the topics, recording hot topics in different periods, and storing statistical results in the bottom database module;
the bottom database module is used for storing the use data of the user and the single-round and multi-round dialogue scripts; the single-round dialogue script is subjected to partition management according to the categories of human design, encyclopedias, tools, call making, daily expressions, emotional expressions, fixed expressions and super class selling; the multi-turn dialog script is managed in a partitioned manner according to entertainment, education, adults, and children.
10. The dialog control system of claim 9 wherein the system comprises at least one client and a server, the server comprising an intelligent analysis module, an intelligent selection module, an intelligent recommendation module, an intelligent statistics module, and an underlying database module, wherein,
the client is used for interacting with the user, collecting interaction information input by the user and transmitting the interaction information to the server in real time; obtaining conversation script information returned by the server and displaying the conversation script information to a user;
and the server is used for interacting with the client, receiving the interactive information input by the user, analyzing the interactive information and returning the dialogue script information corresponding to the analyzed interactive information to the client.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810713598.7A CN110659355A (en) | 2018-06-29 | 2018-06-29 | Conversation control method and system |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810713598.7A CN110659355A (en) | 2018-06-29 | 2018-06-29 | Conversation control method and system |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN110659355A true CN110659355A (en) | 2020-01-07 |
Family
ID=69027723
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201810713598.7A Pending CN110659355A (en) | 2018-06-29 | 2018-06-29 | Conversation control method and system |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN110659355A (en) |
Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105446491A (en) * | 2015-12-16 | 2016-03-30 | 北京光年无限科技有限公司 | Intelligent robot based interactive method and apparatus |
| CN106294854A (en) * | 2016-08-22 | 2017-01-04 | 北京光年无限科技有限公司 | A kind of man-machine interaction method for intelligent robot and device |
| CN106372195A (en) * | 2016-08-31 | 2017-02-01 | 北京光年无限科技有限公司 | Human-computer interaction method and device for intelligent robot |
| JP2017049427A (en) * | 2015-09-01 | 2017-03-09 | カシオ計算機株式会社 | Dialogue control apparatus, dialogue control method, and program |
| CN106599196A (en) * | 2016-12-14 | 2017-04-26 | 竹间智能科技（上海）有限公司 | Artificial intelligence conversation method and system |
| CN107193978A (en) * | 2017-05-26 | 2017-09-22 | 武汉泰迪智慧科技有限公司 | A kind of many wheel automatic chatting dialogue methods and system based on deep learning |
| CN108037905A (en) * | 2017-11-21 | 2018-05-15 | 北京光年无限科技有限公司 | A kind of interaction output method and intelligent robot for intelligent robot |
-
2018
- 2018-06-29 CN CN201810713598.7A patent/CN110659355A/en active Pending
Patent Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2017049427A (en) * | 2015-09-01 | 2017-03-09 | カシオ計算機株式会社 | Dialogue control apparatus, dialogue control method, and program |
| CN105446491A (en) * | 2015-12-16 | 2016-03-30 | 北京光年无限科技有限公司 | Intelligent robot based interactive method and apparatus |
| CN106294854A (en) * | 2016-08-22 | 2017-01-04 | 北京光年无限科技有限公司 | A kind of man-machine interaction method for intelligent robot and device |
| CN106372195A (en) * | 2016-08-31 | 2017-02-01 | 北京光年无限科技有限公司 | Human-computer interaction method and device for intelligent robot |
| CN106599196A (en) * | 2016-12-14 | 2017-04-26 | 竹间智能科技（上海）有限公司 | Artificial intelligence conversation method and system |
| CN107193978A (en) * | 2017-05-26 | 2017-09-22 | 武汉泰迪智慧科技有限公司 | A kind of many wheel automatic chatting dialogue methods and system based on deep learning |
| CN108037905A (en) * | 2017-11-21 | 2018-05-15 | 北京光年无限科技有限公司 | A kind of interaction output method and intelligent robot for intelligent robot |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN107943998B (en) | Man-machine conversation control system and method based on knowledge graph | |
| US10911840B2 (en) | Methods and systems for generating contextual data elements for effective consumption of multimedia | |
| US11645547B2 (en) | Human-machine interactive method and device based on artificial intelligence | |
| US20200301954A1 (en) | Reply information obtaining method and apparatus | |
| CN106548773B (en) | Child user searching method and device based on artificial intelligence | |
| US8285654B2 (en) | Method and system of providing a personalized performance | |
| US20150243279A1 (en) | Systems and methods for recommending responses | |
| KR102033388B1 (en) | Apparatus and method for question answering | |
| CN105068661A (en) | Man-machine interaction method and system based on artificial intelligence | |
| CN112364234A (en) | Automatic grouping system for online discussion | |
| KR20190046062A (en) | Method and apparatus of dialog scenario database constructing for dialog system | |
| CN106407287A (en) | Multimedia resource pushing method and system | |
| CN110019691A (en) | Conversation message treating method and apparatus | |
| KR102117287B1 (en) | Method and apparatus of dialog scenario database constructing for dialog system | |
| CN109313649A (en) | Voice-based knowledge sharing application for chat robots | |
| Chow et al. | A mixture-of-expert approach to rl-based dialogue management | |
| CN113420058B (en) | Conversational academic conference recommendation method based on combination of user historical behaviors | |
| Galety et al. | Personalized music recommendation model based on Machine Learning | |
| CN111427444B (en) | Control method and device of intelligent device | |
| Cervone et al. | Roving mind: a balancing act between open–domain and engaging dialogue systems | |
| KR20200143039A (en) | Method and apparatus of dialog scenario database constructing for dialog system | |
| CN110659355A (en) | Conversation control method and system | |
| CN110633410A (en) | Information processing method and device, storage medium, and electronic device | |
| Wu et al. | Interactive spoken content retrieval by deep reinforcement learning | |
| Hernández et al. | User-centric Recommendation Model for AAC based on Multi-criteria Planning |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| WD01 | Invention patent application deemed withdrawn after publication | ||
| WD01 | Invention patent application deemed withdrawn after publication |
Application publication date: 20200107 |