CN113282730A - Provincial conversation flow design method and system - Google Patents
Provincial conversation flow design method and system Download PDFInfo
- Publication number
- CN113282730A CN113282730A CN202110636068.9A CN202110636068A CN113282730A CN 113282730 A CN113282730 A CN 113282730A CN 202110636068 A CN202110636068 A CN 202110636068A CN 113282730 A CN113282730 A CN 113282730A
- Authority
- CN
- China
- Prior art keywords
- user
- nodes
- conversation
- dialog
- flow
- 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 38
- 238000013461 design Methods 0.000 title claims description 20
- 230000006399 behavior Effects 0.000 claims description 36
- 230000006870 function Effects 0.000 claims description 16
- 238000012938 design process Methods 0.000 claims description 15
- 230000006872 improvement Effects 0.000 claims description 6
- 230000008569 process Effects 0.000 description 5
- 238000012986 modification Methods 0.000 description 3
- 230000004048 modification Effects 0.000 description 3
- 230000000694 effects Effects 0.000 description 2
- 230000009191 jumping Effects 0.000 description 2
- 238000013473 artificial intelligence Methods 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000004364 calculation method Methods 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- 238000012545 processing Methods 0.000 description 1
- 230000004044 response Effects 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/30—Information retrieval; Database structures therefor; File system structures therefor of unstructured textual data
- G06F16/33—Querying
- G06F16/332—Query formulation
- G06F16/3329—Natural language query formulation or dialogue systems
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Mathematical Physics (AREA)
- Theoretical Computer Science (AREA)
- Data Mining & Analysis (AREA)
- Computational Linguistics (AREA)
- Human Computer Interaction (AREA)
- Databases & Information Systems (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Artificial Intelligence (AREA)
- User Interface Of Digital Computer (AREA)
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
- Machine Translation (AREA)
Abstract
The invention provides a method and a system for designing introspection conversation flow. The method comprises the following steps: initializing a global variable, and initializing intention definition, a value set and a function handle; the user carries out conversation; extracting dialog flow logic according to user dialog contents, and determining whether the user's question can be answered or not by judging the dialog flow logic; and recording the node behavior data, and performing self-learning by utilizing the node behavior data. The system comprises modules corresponding to the steps of the method.
Description
Technical Field
The invention provides a method and a system for designing a self-provincial conversation flow, and belongs to the technical field of conversation flow design.
Background
With the continuous development of information technology and artificial intelligence technology, intelligent robots and intelligent customer service are increasingly popularized, so that the robot is replaced by the robot, and the overall working efficiency of the society is improved. The intelligent customer service system is an application system based on knowledge base management, and comprises: knowledge management technology, natural language understanding technology, customer service conversation process management technology and the like, and has certain industry universality. The conversation content of the intelligent customer service is combed based on a theme-oriented knowledge base, the intention of the customer conversation is judged by extracting keywords in the customer conversation or constructing an intention recognition model based on a natural language understanding technology, and therefore the follow-up flow trend and answer conversation of the customer service robot are further guided according to the intention of the customer and a configured customer service conversation flow. However, the existing dialog flow design process still has the problems of complex dialog flow design process, high resource consumption and low dialog performance.
Disclosure of Invention
The invention provides a method and a system for designing a self-saving conversation flow, which are used for solving the problems of complex design process, large resource consumption and lower conversation performance of the traditional conversation flow, and adopt the following technical scheme:
a introspection dialog flow design method, the method comprising:
initializing a global variable, and initializing intention definition, a value set and a function handle;
the user carries out conversation;
extracting dialog flow logic according to user dialog contents, and determining whether the user's question can be answered or not by judging the dialog flow logic;
and recording the node behavior data, and performing self-learning by utilizing the node behavior data.
Further, the extracting dialog flow logic according to the user dialog content and determining whether the user's question can be answered by judging the dialog flow logic comprises:
judging question questions in user conversation, and if the question questions can be answered, normally answering the user questions; if the question asked by the user cannot be answered, the QA system is switched to for alternative answering.
Further, the recording of the node behavior data and the self-learning by using the node behavior data include:
recording the flow direction of each user's dialog in the dialog flow into node behavior data;
and self-learning is carried out by utilizing the node behavior data through learning behaviors so as to feed back problems in the design process of the conversation flow and provide improvement suggestions aiming at the problems in the design process of the conversation flow.
Further, the node types include the following categories: the system comprises slot filling nodes, function nodes, assignment nodes, sub-processes, branch nodes, return nodes, exit nodes and reply nodes.
An introspection dialog flow design system, the system comprising:
the initialization module is used for initializing the global variable and initializing the intention definition, the value set and the function handle;
the dialogue module is used for carrying out dialogue by a user;
the judging module is used for extracting conversation flow logic according to the conversation content of the user and determining whether the problem of the user can be solved or not through judging the conversation flow logic;
and the learning module is used for recording the node behavior data and utilizing the node behavior data to carry out self-learning.
Further, the judging module comprises:
the question judging module is used for judging question questions in user conversation;
the answer module is used for normally answering the user question if the question can be answered; if the question asked by the user cannot be answered, the QA system is switched to for alternative answering.
Further, the learning module includes:
the recording module is used for recording the flow direction of the dialog of each user in the dialog flow into node behavior data;
and the self-learning module is used for carrying out self-learning by utilizing the node behavior data through learning behaviors, feeding back problems in the design process of the conversation flow and providing improvement suggestions aiming at the problems in the design process of the conversation flow.
Further, the node types include the following categories: the system comprises slot filling nodes, function nodes, assignment nodes, sub-processes, branch nodes, return nodes, exit nodes and reply nodes.
The invention has the beneficial effects that:
according to the method and the system for designing the introspection dialog flow, the branch jumping logic is contained in each node, and each node is provided with an independent logic unit, so that the expression capability and the flexibility are greatly enhanced. Meanwhile, the node subband statistical function can automatically improve the design of the conversation flow by self-saving, and can also assist a user in improving the design of the conversation flow. On the other hand, the sub-processes of the provincial dialog flow design method and system provided by the invention can abstract and form a template library, and can be used for quickly constructing and forming a high-level API (application program interface) for a user to use.
Drawings
FIG. 1 is a first flow chart of the method of the present invention;
FIG. 2 is a second flow chart of the method of the present invention;
fig. 3 is a system block diagram of the system of the present invention.
Detailed Description
The preferred embodiments of the present invention will be described in conjunction with the accompanying drawings, and it will be understood that they are described herein for the purpose of illustration and explanation and not limitation.
An embodiment of the present invention provides a introspection dialog flow design method, as shown in fig. 1 and fig. 2, the method includes:
s1, initializing global variables, and initializing intention definition, value set and function handle;
s2, the user carries out a conversation;
s3, extracting dialog flow logic according to the user dialog content, and determining whether the user' S question can be answered or not through judging the dialog flow logic;
and S4, recording the node behavior data, and performing self-learning by using the node behavior data.
Wherein the dialog flow logic is extracted according to the user dialog content, and is determined to be yes through judgment of the dialog flow logic
The question of the user can be solved, comprising:
s301, judging question questions in user conversation;
s302, if the questions can be answered, the user questions are answered normally; if the question asked by the user cannot be answered, the QA system is switched to for alternative answering.
The node behavior data recording and self-learning by utilizing the node behavior data comprises the following steps:
s401, recording the flow direction of each user in the conversation flow into node behavior data;
s402, self-learning is carried out by utilizing the node behavior data through learning behaviors, the self-learning is used for feeding back problems in the design process of the conversation flow, and improvement suggestions are provided for the problems in the design process of the conversation flow.
Meanwhile, in this embodiment, the node types include the following categories: the system comprises slot filling nodes, function nodes, assignment nodes, sub-processes, branch nodes, return nodes, exit nodes and reply nodes.
The effect of the above technical scheme is as follows: in the method for designing introspection dialog flow provided by this embodiment, the branch jump logic is contained in each node, and each node has an independent logic unit, so that the expression capability and flexibility are greatly enhanced. Meanwhile, the node subband statistical function can automatically improve the design of the conversation flow by self-saving, and can also assist a user in improving the design of the conversation flow. On the other hand, the sub-processes in the introspection dialog flow design method provided by the embodiment can be abstracted to form a template library, and can be used for quickly constructing and forming a high-level API, so that the method is convenient for users to use.
In the process of processing the dialog flow, any dialog flow is abstractly described as a directed graph, and the process of dialog by a user is a process of jumping among nodes of the dialog flow until the nodes are exited. The sub-process is defined as a sub-graph, the template is defined as the hyper-parameterization of the sub-graph, in the process of loading the sub-graph in the self-defined dialog flow design, the Zhan template and the hyper-parameters are added firstly, then the real sub-graph corresponding to the dialog flow is obtained through calculation, and then the dialog flow is executed.
This embodiment proposes an introspection dialog flow design system, as shown in fig. 3, the system includes:
the initialization module is used for initializing the global variable and initializing the intention definition, the value set and the function handle;
the dialogue module is used for carrying out dialogue by a user;
the judging module is used for extracting conversation flow logic according to the conversation content of the user and determining whether the problem of the user can be solved or not through judging the conversation flow logic;
and the learning module is used for recording the node behavior data and utilizing the node behavior data to carry out self-learning.
Wherein, the judging module comprises:
the question judging module is used for judging question questions in user conversation;
the answer module is used for normally answering the user question if the question can be answered; if the question asked by the user cannot be answered, the QA system is switched to for alternative answering.
Wherein the learning module comprises:
the recording module is used for recording the flow direction of the dialog of each user in the dialog flow into node behavior data;
and the self-learning module is used for carrying out self-learning by utilizing the node behavior data through learning behaviors, feeding back problems in the design process of the conversation flow and providing improvement suggestions aiming at the problems in the design process of the conversation flow.
Wherein the node types include the following categories:
slot filling node (slot filling)
Function node (function)
Assignment node (assignment)
Sub-process (flow)
Branch node (branch)
Return node (return)
Exit node (exit)
Return node (response)
The effect of the above technical scheme is as follows: in the introspection dialog flow design system provided by this embodiment, the branch jump logic is included in each node, and each node has an independent logic unit, so that the expression capability and flexibility are greatly enhanced. Meanwhile, the node subband statistical function can automatically improve the design of the conversation flow by self-saving, and can also assist a user in improving the design of the conversation flow. On the other hand, the sub-processes in the introspection dialog flow design system provided by the embodiment can be abstracted to form a template library, and can be used for quickly constructing and forming a high-level API, so that the user can use the API conveniently.
It will be apparent to those skilled in the art that various changes and modifications may be made in the present invention without departing from the spirit and scope of the invention. Thus, if such modifications and variations of the present invention fall within the scope of the claims of the present invention and their equivalents, the present invention is also intended to include such modifications and variations.
Claims (8)
1. A introspection dialog flow design method, characterized in that the method comprises:
initializing a global variable, and initializing intention definition, a value set and a function handle;
the user carries out conversation;
extracting dialog flow logic according to user dialog contents, and determining whether the user's question can be answered or not by judging the dialog flow logic;
and recording the node behavior data, and performing self-learning by utilizing the node behavior data.
2. The method of claim 1, wherein the extracting dialog flow logic according to the user dialog content and determining whether the user's question can be answered by judging the dialog flow logic comprises:
judging question questions in user conversation, and if the question questions can be answered, normally answering the user questions; if the question asked by the user cannot be answered, the QA system is switched to for alternative answering.
3. The method of claim 1, wherein the recording of node behavior data and the self-learning using node behavior data comprises:
recording the flow direction of each user's dialog in the dialog flow into node behavior data;
and self-learning is carried out by utilizing the node behavior data through learning behaviors so as to feed back problems in the design process of the conversation flow and provide improvement suggestions aiming at the problems in the design process of the conversation flow.
4. The method of claim 1, wherein the node types include the following categories: the system comprises slot filling nodes, function nodes, assignment nodes, sub-processes, branch nodes, return nodes, exit nodes and reply nodes.
5. An introspection dialog flow design system, the system comprising:
the initialization module is used for initializing the global variable and initializing the intention definition, the value set and the function handle;
the dialogue module is used for carrying out dialogue by a user;
the judging module is used for extracting conversation flow logic according to the conversation content of the user and determining whether the problem of the user can be solved or not through judging the conversation flow logic;
and the learning module is used for recording the node behavior data and utilizing the node behavior data to carry out self-learning.
6. The system of claim 5, wherein the determining module comprises:
the question judging module is used for judging question questions in user conversation;
the answer module is used for normally answering the user question if the question can be answered; if the question asked by the user cannot be answered, the QA system is switched to for alternative answering.
7. The system of claim 5, wherein the learning module comprises:
the recording module is used for recording the flow direction of the dialog of each user in the dialog flow into node behavior data;
and the self-learning module is used for carrying out self-learning by utilizing the node behavior data through learning behaviors, feeding back problems in the design process of the conversation flow and providing improvement suggestions aiming at the problems in the design process of the conversation flow.
8. The system of claim 5, wherein the node types include the following categories: the system comprises slot filling nodes, function nodes, assignment nodes, sub-processes, branch nodes, return nodes, exit nodes and reply nodes.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110636068.9A CN113282730A (en) | 2021-06-08 | 2021-06-08 | Provincial conversation flow design method and system |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110636068.9A CN113282730A (en) | 2021-06-08 | 2021-06-08 | Provincial conversation flow design method and system |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN113282730A true CN113282730A (en) | 2021-08-20 |
Family
ID=77283946
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202110636068.9A Pending CN113282730A (en) | 2021-06-08 | 2021-06-08 | Provincial conversation flow design method and system |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN113282730A (en) |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2000339314A (en) * | 1999-05-25 | 2000-12-08 | Nippon Telegr & Teleph Corp <Ntt> | Automatic answering method, dialog analyzing method, answer sentence generating method and their device and medium with their program recorded thereon |
| CN110377720A (en) * | 2019-07-26 | 2019-10-25 | 中国工商银行股份有限公司 | The more wheel exchange methods of intelligence and system |
| CN110472030A (en) * | 2019-08-08 | 2019-11-19 | 网易（杭州）网络有限公司 | Man-machine interaction method, device and electronic equipment |
| CN111368060A (en) * | 2020-05-27 | 2020-07-03 | 支付宝(杭州)信息技术有限公司 | Self-learning method, device and system for conversation robot, electronic equipment and medium |
| CN111597312A (en) * | 2020-04-07 | 2020-08-28 | 北京捷通华声科技股份有限公司 | Method and device for generating multi-turn dialogue script |
| CN112784024A (en) * | 2021-01-11 | 2021-05-11 | 软通动力信息技术(集团)股份有限公司 | Man-machine conversation method, device, equipment and storage medium |
-
2021
- 2021-06-08 CN CN202110636068.9A patent/CN113282730A/en active Pending
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2000339314A (en) * | 1999-05-25 | 2000-12-08 | Nippon Telegr & Teleph Corp <Ntt> | Automatic answering method, dialog analyzing method, answer sentence generating method and their device and medium with their program recorded thereon |
| CN110377720A (en) * | 2019-07-26 | 2019-10-25 | 中国工商银行股份有限公司 | The more wheel exchange methods of intelligence and system |
| CN110472030A (en) * | 2019-08-08 | 2019-11-19 | 网易（杭州）网络有限公司 | Man-machine interaction method, device and electronic equipment |
| CN111597312A (en) * | 2020-04-07 | 2020-08-28 | 北京捷通华声科技股份有限公司 | Method and device for generating multi-turn dialogue script |
| CN111368060A (en) * | 2020-05-27 | 2020-07-03 | 支付宝(杭州)信息技术有限公司 | Self-learning method, device and system for conversation robot, electronic equipment and medium |
| CN112784024A (en) * | 2021-01-11 | 2021-05-11 | 软通动力信息技术(集团)股份有限公司 | Man-machine conversation method, device, equipment and storage medium |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN108073600B (en) | Intelligent question-answer interaction method and device and electronic equipment | |
| CN108197167A (en) | Human-computer dialogue processing method, equipment and readable storage medium storing program for executing | |
| KR101369810B1 (en) | Empirical Context Aware Computing Method For Robot | |
| US6363301B1 (en) | System and method for automatically focusing the attention of a virtual robot interacting with users | |
| Azab et al. | Simulation methods for changeable manufacturing | |
| CN110750298B (en) | AI model compiling method, equipment and storage medium | |
| CN112199486A (en) | Task type multi-turn conversation method and system for office scene | |
| CN117573834B (en) | Multi-robot dialogue method and system for software-oriented instant service platform | |
| CN111930912A (en) | Dialogue management method, system, device and storage medium | |
| CN109693244A (en) | The method and device of optimization dialogue robot | |
| CN113806501A (en) | Method for training intention recognition model, intention recognition method and equipment | |
| CN113282730A (en) | Provincial conversation flow design method and system | |
| CN111309882A (en) | Method and device for realizing intelligent customer service question answering | |
| CN111680148B (en) | Method and device for intelligently responding to question of user | |
| CN111949776B (en) | User tag evaluation method and device and electronic equipment | |
| CN113868396A (en) | Task intelligent dialogue construction method and system based on knowledge graph | |
| Stoyanchev et al. | Knowledge-graph driven information state approach to dialog | |
| CN111881266B (en) | Response method and device | |
| CN115344673A (en) | Dialogue system dialogue method, device and storage medium | |
| David et al. | Interaction Templates: A Data-Driven Approach for Authoring Robot Programs | |
| CN110110057A (en) | A kind of dynamic words art group system and its application in Intelligent dialogue robot | |
| CN110222159A (en) | A kind of text based human-computer interaction training method and system | |
| CN117370638B (en) | Method and device for decomposing and scheduling basic model task with enhanced thought diagram prompt | |
| CN113255373B (en) | ARM side offline dialogue system, device and storage medium based on Rasa framework | |
| CN118364068A (en) | Intelligent question-answering method, device, equipment and medium |
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 |