CN114186041A - Answer output method - Google Patents
Answer output method Download PDFInfo
- Publication number
- CN114186041A CN114186041A CN202111434023.XA CN202111434023A CN114186041A CN 114186041 A CN114186041 A CN 114186041A CN 202111434023 A CN202111434023 A CN 202111434023A CN 114186041 A CN114186041 A CN 114186041A
- Authority
- CN
- China
- Prior art keywords
- unit
- management
- knowledge
- service engine
- analysis
- 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
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
- G06F16/3331—Query processing
- G06F16/334—Query execution
- G06F16/3343—Query execution using phonetics
-
- 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
-
- 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
- G06F16/3331—Query processing
- G06F16/334—Query execution
- G06F16/3344—Query execution using natural language analysis
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F40/00—Handling natural language data
- G06F40/20—Natural language analysis
- G06F40/232—Orthographic correction, e.g. spell checking or vowelisation
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F40/00—Handling natural language data
- G06F40/20—Natural language analysis
- G06F40/237—Lexical tools
- G06F40/247—Thesauruses; Synonyms
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F40/00—Handling natural language data
- G06F40/20—Natural language analysis
- G06F40/279—Recognition of textual entities
- G06F40/284—Lexical analysis, e.g. tokenisation or collocates
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F40/00—Handling natural language data
- G06F40/20—Natural language analysis
- G06F40/279—Recognition of textual entities
- G06F40/289—Phrasal analysis, e.g. finite state techniques or chunking
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F40/00—Handling natural language data
- G06F40/20—Natural language analysis
- G06F40/279—Recognition of textual entities
- G06F40/289—Phrasal analysis, e.g. finite state techniques or chunking
- G06F40/295—Named entity recognition
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F40/00—Handling natural language data
- G06F40/30—Semantic analysis
Abstract
The invention discloses an answer output method, which comprises the following steps: a service engine for processing the analysis data; the backstage management platform, backstage pipeline platform's function includes: service management, channel management, system management, and the like; the system comprises a knowledge base, a database and a database, wherein a plurality of knowledge items are arranged in the knowledge base; the emotion recognition module analyzes, processes, induces and infers the text through a service engine and marks the emotion type of the current text; the voice synthesis module is used for preprocessing a voice signal through a service engine, extracting features, matching an acoustic model with a mode and converting the voice into a text by matching a language model and language processing; the semantic association processing platform is used for processing and analyzing character information input by a user; the method and the device realize the purpose of outputting the corresponding answer to the user efficiently, and improve the use experience at the same time.
Description
Technical Field
The invention relates to the technical field of information processing, in particular to an answer output method.
Background
One of the most important features of an intelligent robot system is that the intelligent robot can interact with a user in the form of natural language dialog. Through complex semantic analysis, the intelligent robot can adapt to semantic changes, changes according to topic changes, and utilizes context to deduce answers. It can also provide additional information to the user, providing advice to the user or guiding the user.
The intelligent robot system in the existing service platform is not provided with functions of full-text contact measures, fuzzy question answering, sensitive filtering, identification for short and the like, or the forward answering rate of analysis is too low due to too simple functions, so that a corresponding conclusion cannot be analyzed. And the problems of wrongly written characters input by the user, input guidance by the user and the like also need to be solved.
The intelligent robot system needs to solve the user problem by a semantic association method, and needs to be equipped with management measures, an analysis system, a log system and the like in the corresponding field. And needs to learn and upgrade itself.
While a user carries certain emotion and attitude in the process of interacting with the intelligent robot, the existing system does not take the user's question answering according to the attitude and emotional state of the user into consideration when outputting the user answer.
Therefore, a solution to one or more of the above problems is needed.
Disclosure of Invention
In order to solve one or more problems in the prior art, the invention provides an answer output method. The technical scheme adopted by the invention for solving the problems is as follows: an answer output method comprising: a service engine for processing the analysis data; the background management platform, the function of background pipeline platform includes at least: service management, channel management, system management, operation and maintenance management and knowledge management;
the knowledge base is provided with a plurality of knowledge items, and the knowledge base at least comprises: a general language knowledge base, a professional speech knowledge base and a professional business knowledge base;
an emotion recognition module, the emotion recognition module comprising: the emotion recognition module analyzes, processes, induces and infers the text through a service engine and marks the emotion type of the current text;
the voice synthesis module is used for preprocessing a voice signal through a service engine, extracting features, matching an acoustic model with a mode and converting the voice into a text by matching a language model and language processing;
a semantic association processing platform, the semantic association processing platform comprising:
s010, a text processing unit, wherein the text processing unit is used for Chinese word segmentation, part of speech tagging and named entity identification;
s020, a filtering unit, wherein the filtering unit is used for filtering illegal words, stop words and suffixes; s030, an error correction unit, wherein the error correction unit is used for pinyin error correction and English error correction;
s040, a lexical analysis unit, wherein the lexical analysis unit is used for determining relatively universal semantic information in the sentence and outputting a lexical meaning analysis result;
s050, a sentence matching unit is arranged in the sentence matching unit, a plurality of semantic matching templates are arranged in the sentence matching unit, a vector space similarity calculation equation is arranged in the sentence matching unit, and the word sense matching templates and the vector space similarity calculation equation are processed by using the word sense analysis result and one or more knowledge items are selected from the knowledge base.
Further, the knowledge management function of the background management platform provides: knowledge base management, ontology class management, part of speech management, advanced function management, knowledge point statistics management and knowledge instance management;
the content of knowledge instance management at least comprises: example name, attribute name, standard question, answer, dimension, effective time start and stop, information query, editing maintenance, information import and information export.
Further, the semantic association processing platform further comprises:
s041, a fuzzy question-answering unit is arranged in the fuzzy question-answering unit, and when the lexical analysis unit judges that the input question of the user is fuzzy or only single word, the guidance unit performs analysis processing through a service engine and outputs the most relevant knowledge items;
s042, a question-answering unit, wherein when the fuzzy question-answering unit outputs a plurality of knowledge items, the question-answering unit analyzes and processes the knowledge items through the service engine and provides questions to a user;
and S043, a sensitive filtering unit, wherein a sensitive vocabulary recognition module is arranged in the sensitive filtering unit, and the sensitive filtering unit recognizes illegal vocabularies input by a user through a service engine and outputs a preset answer caliber.
Further, the semantic association processing platform further comprises:
s060, a context association unit, in which a user context memory unit is arranged, and the context association unit analyzes an answer by using a service engine.
Further, the semantic association processing platform further comprises:
s070, a scene recognition unit, wherein the scene recognition unit recognizes the conversation scene through a service engine and really selects the knowledge base.
Further, the semantic association processing platform further comprises:
s080, a region identification unit, wherein the region identification unit is used for identifying the geographic position of the current user and determining the alternative knowledge base through a service engine.
Further, the semantic association processing platform further comprises:
s090, a semantic rule definition template, wherein the semantic rule definition module is used for being matched with the lexical analysis unit and the sentence matching unit to perform data processing through a service engine, and the semantic rule definition module is made, modified and added by a manager.
Further, the lexical analysis unit further includes: the system comprises a word class module, a text classification unit, a feature extraction unit and an automatic reasoning unit;
the word class module can at least perform synonym analysis, near synonym analysis, willingness word analysis, range word analysis, collective word analysis and full-simple-name recognition.
The beneficial effects obtained by the invention are as follows: according to the invention, through the cooperation and coordination among the service engine, the background management platform and the semantic association processing platform, the processing and analysis of the text and voice input of the user are realized, the user requirements and emotions are recognized and corresponding answers are made, the user is pacified or encouraged while the user problems are solved, and the user experience is further improved; the background management platform is used for docking and coordinating the answer output of the semantic association processing platform, and the semantic association processing platform processes and analyzes the data of each unit by using a service engine so as to select corresponding knowledge items from the knowledge base and realize the functions of recognizing characters and associating characters; the text processing unit, the filtering unit, the error correcting unit, the lexical analyzing unit, the sentence matching unit, other units and modules are skillfully arranged, and the high-accuracy recognition and analysis of the user's intention and the guidance of the user are realized in a matching way, so that the user can accurately and quickly find the knowledge items (contents) to be searched; the part is not a very clear user for the knowledge items which need to be searched, after the context association unit, the fuzzy question-answering unit and the question-answering unit are arranged, the user can be assisted to recall or find the knowledge items which need to be searched, and meanwhile, when a malicious user accesses through illegal words, the user can quickly recognize and react. The practical value of the invention is greatly improved.
Drawings
FIG. 1 is a diagram illustrating an answer output method according to the present invention;
FIG. 2 is a schematic block diagram I of a semantic association processing platform according to an answer output method of the present invention;
fig. 3 is a schematic block diagram II of a semantic association processing platform of an answer output method according to the present invention.
Detailed Description
In order to make the aforementioned objects, features and advantages of the present invention comprehensible, embodiments accompanied with figures are described in detail below. In the following description, numerous specific details are set forth in order to provide a thorough understanding of the present invention. This invention may, however, be embodied in many different forms and should not be construed as limited to the embodiments set forth herein.
As shown in fig. 1 to 3, the present invention discloses an answer output method, including: a service engine T04, the service engine T04 for processing the analysis data, the service engine T04 including an ASR speech engine for speech recognition; a background management platform T03, the functions of the background pipeline platform T03 at least comprising: service management, channel management, system management, operation and maintenance management and knowledge management;
the knowledge base is provided with a plurality of knowledge items, and the knowledge base at least comprises: a general language knowledge base, a professional speech knowledge base and a professional business knowledge base;
an emotion recognition module T05, the emotion recognition module T05 comprising: the emotion recognition module T05 analyzes, processes, induces and infers the text through the service engine T04 and marks the emotion type of the current text;
the speech synthesis module T06, the speech synthesis module T06 carries out preprocessing and feature extraction on speech signals through the service engine T04, carries out acoustic model and mode matching, and then converts the speech into text by matching with a language model and language processing;
a semantic association processing platform T02, the semantic association processing platform T02 comprising:
s010, a text processing unit, wherein the text processing unit is used for Chinese word segmentation, part of speech tagging and named entity identification;
s020, a filtering unit, wherein the filtering unit is used for filtering illegal words, stop words and suffixes; s030, an error correction unit, wherein the error correction unit is used for pinyin error correction and English error correction;
s040, a lexical analysis unit, wherein the lexical analysis unit is used for determining relatively universal semantic information in the sentence and outputting a lexical meaning analysis result;
s050, a sentence matching unit is arranged in the sentence matching unit, a plurality of semantic matching templates are arranged in the sentence matching unit, a vector space similarity calculation equation is arranged in the sentence matching unit, and the word sense matching templates and the vector space similarity calculation equation are processed by using the word sense analysis result and one or more knowledge items are selected from the knowledge base.
When the intelligent robot is applied, a client accesses the intelligent robot through a website or a telephone, the robot preferentially judges and extracts corresponding knowledge base classification according to client information (IP, number attribution and the like), guides the client to answer necessary query conditions, calculates through a service engine T04, replies a related commodity list for the first time, guides the client to input next screening conditions again, screens the related commodities replied for the first time again and then replies to the client, if the client has subsequent intention, the client pushes related information to a platform through an interface, and can also directly carry out ordering operation through the robot and push related order information to the platform. The robot can also provide relevant after-sale information inquiry robot to input relevant inquiry information through a client and push the inquiry information to the platform, the platform inquires immediately and returns relevant information, and the robot feeds back the inquiry information through a client input mode (online and telephone). And if the customer inputs the related product pictures, the robot receives corresponding information through related technologies such as image recognition and the like, and feeds back the commodities in the platform corresponding to the pictures in real time.
It is noted that a mass analysis tool is also included, which is used to help answer the following questions: what the correct rate of the intelligent robot to answer the user question is; what the error rate is; successfully solving the problem of the user; leave without resolution for the user; the most interesting questions of the user; the most hot questions asked by the user over a certain period of time.
To improve the intelligent robot, the quality analysis tool can help the platform answer the following answers: what the most frequent questions the user asks; which questions the user asks most frequently, but the robot cannot answer; which problems are easily misunderstood by the robot.
Specifically, the service management function of the background pipeline platform T03 provides: system service management, service parameter management, special welcome language configuration, conversation preprocessing management, signature activity management, survey and vote management, dynamic menu management and hotspot problem management. The knowledge management function of the background pipeline platform T03 provides: knowledge base management, ontology class management, part of speech management, advanced function management, knowledge point statistics management and knowledge instance management; the content of knowledge instance management at least comprises: example name, attribute name, standard question, answer, dimension, effective time start and stop, information query, editing maintenance, information import and information export.
It should be noted that, the dimensional functions in the pipeline of the knowledge embodiment are as follows: the knowledge can be organized and classified according to different logics and visual angles, such as business types, department organizations, time, regions and the like. One piece of knowledge can be input into a plurality of knowledge dimensions, and can be observed in the corresponding dimension, and when the knowledge is updated, the knowledge in all the dimensions is updated instantly. The multi-dimensional classification is supported, the downward extension can be realized in an infinite level, and the sequence and the parent-child relationship can be freely adjusted by knowledge classification. The enterprise business knowledge is displayed in full-channel targeted content through the dimensional management of the knowledge, and differentiated service experience is provided for all customers. From the perspective of business knowledge, the system can be divided into various business knowledge such as basic business, activity, data, information and the like, and can be maintained and loaded respectively by regions.
Specifically, the knowledge management function of the background pipeline platform T03 provides: knowledge base management, ontology class management, part of speech management, advanced function management, knowledge point statistics management and knowledge instance management; the content of knowledge instance management at least comprises: example name, attribute name, standard question, answer, dimension, effective time start and stop, information query, editing maintenance, information import and information export. The operation and maintenance management function of the background pipeline platform T03 provides: log management and statistical analysis management; the log management contents include: and managing a human-computer interaction log, an automatic question and answer list and an operation log.
It should be noted that the statistical analysis report is an important component of the background pipeline platform T03, and is an evaluation basis of the service effect and the online customer service function. All operations can be defined and processed in batches, sufficient and quantifiable visual data and statistical reports are provided for continuously improving the system and the decision support system, and the operations can be customized according to needs. The content managed by the system comprises: authorization pipelines (including user management, role management, authority management and resource management, which can increase, delete, modify and check platform users and roles thereof, and set resource paths of all functional modules); interface rights pipe (set type of interface service and rules to allow access to IP); system parameter management (parameters such as project ID, project name and system name can be set).
It should be noted that the service engine T04 has dense algorithm and dense data, may be one of the existing commonly used data analysis engines and can perform self-upgrade and learning through artificial intelligence, such as: TensorFlow, Hadoop, Apache Spark, and the like. The knowledge base is a plurality of business knowledge bases, can be the existing knowledge base, also can be the knowledge base that the administrator drafts, the content of the knowledge base is different professional nouns and its content generally, for example: the system comprises a general language knowledge base, a professional voice knowledge base and a professional business knowledge base, wherein the knowledge bases are used for providing services and assisting the system to perform character recognition, analysis and output. It is noted that the different units are typically different main functions or functional modules within the executing software. Generally, the service engine T04 includes a text-associative output module, and a basic consulting question-answering module for consulting, and the basic consulting question-answering module may include: chinese natural language, basic service information, service product information, activity information consultation and the like.
Specifically, as shown in fig. 2, the method further includes: s060, a context association unit, in which a user context memory unit is arranged, and the context association unit analyzes an answer by using a service engine T04. When the question of the user lacks some key information, the context correlation unit cooperates with the service engine T04 to analyze the question in combination with the user's context and give the most appropriate answer. The service engine T04 simulates human thinking and user interaction, some specific knowledge points have an incidence relation, and the engine has the capability of memorizing the user's context, can be combined with new questions for analysis and provides answers, and embodies the fluency of the robot question-answering process.
S070, a scene identification unit which identifies the conversation scene (topic type) and the actually alternative knowledge base through a service engine;
s080, a region identification unit, which is used for identifying the geographic position of the current user and determining the alternative knowledge base through a service engine T04;
s090, a semantic rule definition module is used for being matched with the lexical analysis unit and the sentence matching unit to perform data processing through a service engine T04, and the semantic rule definition module is made, modified and added by a manager. The future expansion maintenance is facilitated through the region identification unit and the semantic rule definition template according to the development condition of the service.
Specifically, as shown in fig. 3, the method further includes: and S041, a fuzzy question-answering unit is arranged in the fuzzy question-answering unit, and when the lexical analysis unit judges that the input question of the user is fuzzy or only a single word, the guidance unit performs analysis processing through a service engine T04 and outputs the most relevant knowledge item, so that the accuracy of system output is improved.
And S042, a question-returning unit, wherein when the fuzzy question-answering unit outputs a plurality of knowledge items, the question-returning unit analyzes the knowledge items through the service engine T04 and provides a question-returning for the user, and the auxiliary system acquires more user information so as to analyze the user' S intention.
And S043, a sensitive filtering unit, wherein a sensitive vocabulary recognition module is arranged in the sensitive filtering unit, and the sensitive filtering unit recognizes illegal vocabularies input by a user through a service engine T04 and outputs a preset answer caliber.
Specifically, the lexical analysis unit further includes: a part of speech module; the word class module can at least perform synonym analysis, near synonym analysis, willingness word analysis, range word analysis, collective word analysis and full simple name recognition; the lexical analysis unit further includes at least: the system comprises a text classification unit, a feature extraction unit and an automatic reasoning unit; thereby realizing vocabulary understanding and association.
Through the cooperation and coordination among the service engine T04, the background management platform T03 and the semantic association processing platform T0, the text and voice input of the user is processed and analyzed, the user requirements and emotions are recognized and corresponding answers are made, the user is pacified or encouraged while the user problems are solved, and the user experience is improved; the background management platform is used for docking and coordinating the answer output of the semantic association processing platform, and the semantic association processing platform processes and analyzes the data of each unit by using a service engine so as to select corresponding knowledge items from the knowledge base and realize the functions of recognizing characters and associating characters; the text processing unit, the filtering unit, the error correcting unit, the lexical analyzing unit, the sentence matching unit, other units and modules are skillfully arranged, and the high-accuracy recognition and analysis of the user's intention and the guidance of the user are realized in a matching way, so that the user can accurately and quickly find the knowledge items (contents) to be searched; the part is not a very clear user for the knowledge items which need to be searched, after the context association unit, the fuzzy question-answering unit and the question-answering unit are arranged, the user can be assisted to recall or find the knowledge items which need to be searched, and meanwhile, when a malicious user accesses through illegal words, the user can quickly recognize and react. The practical value of the invention is greatly improved.
The above-described examples merely represent one or more embodiments of the present invention, which are described in greater detail and detail, but are not to be construed as limiting the invention. It should be noted that, for a person skilled in the art, several variations and modifications can be made without departing from the spirit of the invention, which falls within the scope of the invention. Therefore, the protection scope of the present invention should be subject to the appended claims.
Claims (8)
1. An answer output method comprising: the service engine is used for processing analysis data, and is characterized in that the background management platform at least comprises the following functions: service management, channel management, system management, operation and maintenance management and knowledge management;
the knowledge base is provided with a plurality of knowledge items, and the knowledge base at least comprises: a general language knowledge base, a professional speech knowledge base and a professional business knowledge base;
an emotion recognition module, the emotion recognition module comprising: the emotion recognition module analyzes, processes, induces and infers the text through a service engine and marks the emotion type of the current text;
the voice synthesis module is used for preprocessing a voice signal through a service engine, extracting features, matching an acoustic model with a mode and converting the voice into a text by matching a language model and language processing;
a semantic association processing platform, the semantic association processing platform comprising:
s010, a text processing unit, wherein the text processing unit is used for Chinese word segmentation, part of speech tagging and named entity identification;
s020, a filtering unit, wherein the filtering unit is used for filtering illegal words, stop words and suffixes;
s030, an error correction unit, wherein the error correction unit is used for pinyin error correction and English error correction;
s040, a lexical analysis unit, wherein the lexical analysis unit is used for determining relatively universal semantic information in the sentence and outputting a lexical meaning analysis result;
s050, a sentence matching unit is arranged in the sentence matching unit, a plurality of semantic matching templates are arranged in the sentence matching unit, a vector space similarity calculation equation is arranged in the sentence matching unit, and the word sense matching templates and the vector space similarity calculation equation are processed by using the word sense analysis result and one or more knowledge items are selected from the knowledge base.
2. The answer output method of claim 1, wherein the knowledge management function of the background management platform provides: knowledge base management, ontology class management, part of speech management, advanced function management, knowledge point statistics management and knowledge instance management;
the content of knowledge instance management at least comprises: example name, attribute name, standard question, answer, dimension, effective time start and stop, information query, editing maintenance, information import and information export.
3. An answer output method as claimed in claim 1, wherein the semantic association processing platform further comprises:
s041, a fuzzy question-answering unit is arranged in the fuzzy question-answering unit, and when the lexical analysis unit judges that the input question of the user is fuzzy or only single word, the guidance unit performs analysis processing through a service engine and outputs the most relevant knowledge items;
s042, a question-answering unit, wherein when the fuzzy question-answering unit outputs a plurality of knowledge items, the question-answering unit analyzes and processes the knowledge items through the service engine and provides questions to a user;
and S043, a sensitive filtering unit, wherein a sensitive vocabulary recognition module is arranged in the sensitive filtering unit, and the sensitive filtering unit recognizes illegal vocabularies input by a user through a service engine and outputs a preset answer caliber.
4. An answer output method as claimed in claim 1, wherein the semantic association processing platform further comprises:
s060, a context association unit, in which a user context memory unit is arranged, and the context association unit analyzes an answer by using a service engine.
5. An answer output method as claimed in claim 1, wherein the semantic association processing platform further comprises:
s070, a scene recognition unit, wherein the scene recognition unit recognizes the conversation scene through a service engine and really selects the knowledge base.
6. An answer output method as claimed in claim 1, wherein the semantic association processing platform further comprises:
s080, a region identification unit, wherein the region identification unit is used for identifying the geographic position of the current user and determining the alternative knowledge base through a service engine.
7. An answer output method as claimed in claim 1, wherein the semantic association processing platform further comprises:
s090, a semantic rule definition template, wherein the semantic rule definition module is used for being matched with the lexical analysis unit and the sentence matching unit to perform data processing through a service engine, and the semantic rule definition module is made, modified and added by a manager.
8. The answer output method according to claim 1, wherein the lexical analysis unit further includes: the system comprises a word class module, a text classification unit, a feature extraction unit and an automatic reasoning unit;
the word class module can at least perform synonym analysis, near synonym analysis, willingness word analysis, range word analysis, collective word analysis and full-simple-name recognition.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202111434023.XA CN114186041A (en) | 2021-11-29 | 2021-11-29 | Answer output method |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202111434023.XA CN114186041A (en) | 2021-11-29 | 2021-11-29 | Answer output method |
Publications (1)
Publication Number | Publication Date |
---|---|
CN114186041A true CN114186041A (en) | 2022-03-15 |
Family
ID=80541721
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN202111434023.XA Pending CN114186041A (en) | 2021-11-29 | 2021-11-29 | Answer output method |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN114186041A (en) |
Cited By (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN115080720A (en) * | 2022-06-29 | 2022-09-20 | 壹沓科技(上海)有限公司 | Text processing method, device, equipment and medium based on RPA and AI |
-
2021
- 2021-11-29 CN CN202111434023.XA patent/CN114186041A/en active Pending
Cited By (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN115080720A (en) * | 2022-06-29 | 2022-09-20 | 壹沓科技(上海)有限公司 | Text processing method, device, equipment and medium based on RPA and AI |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
US11334635B2 (en) | Domain specific natural language understanding of customer intent in self-help | |
Gu et al. | " what parts of your apps are loved by users?"(T) | |
US11170179B2 (en) | Systems and methods for natural language processing of structured documents | |
US9910886B2 (en) | Visual representation of question quality | |
CN110674271B (en) | Question and answer processing method and device | |
RU2704531C1 (en) | Method and apparatus for analyzing semantic information | |
US10861437B2 (en) | Method and device for extracting factoid associated words from natural language sentences | |
CN111179935B (en) | Voice quality inspection method and device | |
CN110597964A (en) | Double-record quality inspection semantic analysis method and device and double-record quality inspection system | |
CN111098312A (en) | Window government affairs service robot | |
CN112487140A (en) | Question-answer dialogue evaluating method, device, equipment and storage medium | |
WO2021010744A1 (en) | Method and device for analyzing sales conversation based on speech recognition | |
US20190377824A1 (en) | Schemaless systems and methods for automatically building and utilizing a chatbot knowledge base or the like | |
CN111767382A (en) | Method and device for generating feedback information and terminal equipment | |
CN114186040A (en) | Operation method of intelligent robot customer service | |
CN111782793A (en) | Intelligent customer service processing method, system and equipment | |
CN112765974A (en) | Service assisting method, electronic device and readable storage medium | |
CN110610003A (en) | Method and system for assisting text annotation | |
CN116955573B (en) | Question searching method, device, equipment and storage medium | |
CN114186041A (en) | Answer output method | |
CN113705207A (en) | Grammar error recognition method and device | |
CN116501960B (en) | Content retrieval method, device, equipment and medium | |
CN112288584A (en) | Insurance application processing method and device, computer readable medium and electronic equipment | |
CN116542676A (en) | Intelligent customer service system based on big data analysis and method thereof | |
CN116070599A (en) | Intelligent question bank generation and auxiliary management system |
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 |