CN108153737A - A kind of method of semantic classification, system and dialog process system - Google Patents
A kind of method of semantic classification, system and dialog process system Download PDFInfo
- Publication number
- CN108153737A CN108153737A CN201711486033.1A CN201711486033A CN108153737A CN 108153737 A CN108153737 A CN 108153737A CN 201711486033 A CN201711486033 A CN 201711486033A CN 108153737 A CN108153737 A CN 108153737A
- Authority
- CN
- China
- Prior art keywords
- sentence
- vector
- current statement
- background
- group
- 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
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F40/00—Handling natural language data
- G06F40/30—Semantic analysis
-
- 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
- G06F16/35—Clustering; Classification
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F40/00—Handling natural language data
- G06F40/20—Natural language analysis
- G06F40/205—Parsing
- G06F40/211—Syntactic parsing, e.g. based on context-free grammar [CFG] or unification grammars
Abstract
The invention discloses a kind of method of semantic classification, system and dialog process system, including：All sentences in dialogue are obtained, and vector coding is carried out to every sentence, obtain the sentence vector of every sentence；Sentence vector is divided into current statement group and background sentence group, wherein, the sentence that current statement group includes current statement is vectorial, and the sentence vector in sentence vector of the background sentence group including all sentences in addition to the sentence vector of current statement, current statement is the last item sentence in dialogue；Average treatment is weighted to the sentence vector in background sentence group by attention mechanism, obtains background vector；The meaning vector of current statement is obtained according to the sentence vector sum background vector of current statement.The present invention can automatically extract the important information in background sentence group, and the important information of extraction and current statement are combined, to be better understood from the meaning of current statement, so as to improve accuracy of the processing module of dialog process system rear class in processing business.
Description
Technical field
The present invention relates to artificial intelligence field, more particularly to a kind of method of semantic classification, system and dialog process system
System.
Background technology
Dialog process system is always one very important branch of artificial intelligence field, these years not only by academia
Research pay attention to, and be also widely used in industrial quarters.A large amount of natural language processing has been used in dialog process system
Technology, a very important component part is semantic classification in the technology.The purpose of semantic classification is by the language of user session
Justice, be intended to distinguish, by the abstract meaning of natural language be converted into the relevant classification of business, in this way so as to conversational system root
Corresponding task is answered or performs according to the semantic classification generation of user accordingly.
Semantic classification of the prior art is divided into two kinds, one kind be classify to single sentence namely to current statement into
Row classification, another kind is classified to whole section of dialogue.In practical applications, if only classified to current statement, then
The background information of current statement will be lacked, current statement is made ambiguity occur, so as to can not correct understanding current statement meaning；
And directly entire dialogue is classified, since whole section of dialogue is spliced by a plurality of sentence, indistinguishable whole section of dialogue
Emphasis, so as to can not extract with the relevant important information of current statement, in this way can give dialog process system rear class processing mould
Block brings very big uncertainty.
Therefore, how to provide a kind of scheme for solving above-mentioned technical problem is that those skilled in the art need to solve at present
Problem.
Invention content
The object of the present invention is to provide a kind of method of semantic classification, system and dialog process systems, can automatically extract
Important information in background sentence group, and the important information of extraction and current statement are combined, to be better understood from current language
The meaning of sentence, so as to improve accuracy of the processing module of dialog process system rear class in processing business.
In order to solve the above technical problems, the present invention provides a kind of method of semantic classification, applied to dialog process system,
Including：
All sentences in dialogue are obtained, and vector coding is carried out to sentence every described, obtain every sentence
Sentence vector；
The sentence vector is divided into current statement group and background sentence group, wherein, the current statement group includes current
The sentence vector of sentence, the background sentence group include removing the sentence of the current statement in the sentence vector of all sentences
Sentence vector outside vector, the current statement are the last item sentence in the dialogue；
Average treatment is weighted to the sentence vector in the background sentence group by attention mechanism, obtain background to
Amount；
The meaning vector of the current statement is obtained according to background vector described in the sentence vector sum of the current statement.
Preferably, the process to sentence progress vector coding every described is：
Recognition with Recurrent Neural Network RNN vector codings are carried out to sentence every described.
Preferably, the process to sentence progress vector coding every described is：
Convolutional neural networks CNN vector codings are carried out to sentence every described.
Preferably, background vector obtains the current statement described in the sentence vector sum according to the current statement
The process of meaning vector is specially：
Background vector described in the sentence vector sum of the current statement by multilayer neural network is handled, obtains described work as
The meaning vector of preceding sentence.
In order to solve the above technical problems, the present invention also provides a kind of system of semantic classification, applied to dialog process system
System, including：
Coding module for obtaining all sentences in talking with, and carries out vector coding to sentence every described, obtains every
The sentence vector of sentence described in item；
Grouping module, for the sentence vector to be divided into current statement group and background sentence group, wherein, the current language
Sentence group includes the sentence vector of current statement, and the background sentence group includes except described working as in the sentence vector of all sentences
Sentence vector outside the sentence vector of preceding sentence, the current statement are the last item sentence in the dialogue；
Processing module is weighted average place for passing through attention mechanism to the sentence vector in the background sentence group
Reason obtains background vector；
Acquisition module obtains the current statement for background vector described in the sentence vector sum according to the current statement
Meaning vector.
Preferably, the coding module is specifically used for carrying out Recognition with Recurrent Neural Network RNN vector codings to sentence every described.
Preferably, the coding module is specifically used for carrying out convolutional neural networks CNN vector codings to sentence every described.
Preferably, the acquisition module is specifically used for：
Background vector described in the sentence vector sum of the current statement by multilayer neural network is handled, obtains described work as
The meaning vector of preceding sentence.
In order to solve the above technical problems, the present invention also provides a kind of dialog process system, including such as above-mentioned any one
The semantic classification system.
The present invention provides a kind of method of semantic classification, applied to dialog process system, including：Obtain the institute in dialogue
There is sentence, and vector coding is carried out to every sentence, obtain the sentence vector of every sentence；Sentence vector is divided into current statement
Group and background sentence group, wherein, current statement group includes the sentence vector of current statement, and background sentence group includes all sentences
Sentence vector in sentence vector in addition to the sentence vector of current statement, current statement are the last item sentence in dialogue；It is logical
It crosses attention mechanism and average treatment is weighted to the sentence vector in background sentence group, obtain background vector；According to current language
The sentence vector sum background vector of sentence obtains the meaning vector of current statement.
As it can be seen that in practical applications, all sentences in dialogue are carried out vector coding by scheme using the present invention in advance
And be grouped, influence of the sentence in background sentence group to current statement is integrally considered, and utilize attention by sentence vector
Sentence vector in background sentence group is weighted averagely, automatically extracts the important information in background sentence group by mechanism, will
The important information and current statement of extraction combine, to be better understood from the meaning of current statement, so as to improve dialog process
Accuracy of the processing module of system rear class in processing business.
The present invention also provides a kind of system of semantic classification and dialog process systems, have and above-mentioned semantic classification method
Identical advantageous effect.
Description of the drawings
It to describe the technical solutions in the embodiments of the present invention more clearly, below will be to institute in the prior art and embodiment
Attached drawing to be used is needed to be briefly described, it should be apparent that, the accompanying drawings in the following description is only some implementations of the present invention
Example, for those of ordinary skill in the art, without creative efforts, can also obtain according to these attached drawings
Obtain other attached drawings.
Fig. 1 is a kind of step flow chart of semantic classification method provided by the present invention；
Fig. 2 is a kind of structure diagram of semantic classification system provided by the present invention.
Specific embodiment
The core of the present invention is to provide a kind of method of semantic classification, system and dialog process system, can automatically extract
Important information in background sentence group, and the important information of extraction and current statement are combined, to be better understood from current language
The meaning of sentence, so as to improve accuracy of the processing module of dialog process system rear class in processing business.
Purpose, technical scheme and advantage to make the embodiment of the present invention are clearer, below in conjunction with the embodiment of the present invention
In attached drawing, the technical solution in the embodiment of the present invention is clearly and completely described, it is clear that described embodiment is
Part of the embodiment of the present invention, instead of all the embodiments.Based on the embodiments of the present invention, those of ordinary skill in the art
All other embodiments obtained without making creative work shall fall within the protection scope of the present invention.
Fig. 1 is please referred to, Fig. 1 is a kind of step flow chart of semantic classification method provided by the present invention, applied to dialogue
Processing system, including：
Step 1：All sentences in dialogue are obtained, and vector coding is carried out to every sentence, obtain the language of every sentence
Sentence vector；
Specifically, generally comprising a plurality of sentence in one section of dialogue, vector coding is carried out in chronological order to a plurality of sentence, this
Each sentence in dialogue can be all converted into the relatively low sentence vector of a dimension by sample, if for example, one section right
Words include ten sentences, then the sentence vector of first statement can be 1, and the sentence vector of Article 2 sentence is 2, with this
Analogize, the sentence vector of Article 10 sentence namely current statement is 10.
Specifically, dialog process system can be man-machine interactive processing system, or everybody dialog process system is removed
Include voice dialogue, further include word dialog, the present invention is not specifically limited dialog process system.
Step 2：Sentence vector is divided into current statement group and background sentence group, wherein, current statement group includes current language
The sentence vector of sentence, sentence in the sentence vector of background sentence group including all sentences in addition to the sentence vector of current statement to
Amount, current statement are the last item sentence in dialogue；
Specifically, sentence vector is divided into current statement group and background sentence group, actually by the sentence in dialogue into
Row grouping, all sentences before the last item sentence are all background sentence, and the last item sentence is then current statement, it is assumed that
One section of dialogue includes ten sentences, then and the sentence that background sentence group then includes first statement to Article 9 sentence is vectorial,
Every sentence so before Article 10 sentence is properly termed as background sentence, and the sentence vector Dan Weiyi groups of Article 10 sentence are
For current statement group, the present invention has carried out vector coding to all sentences in dialogue, has integrally considered to carry on the back by sentence vector
Influence of the scape sentence to current statement, so as to improve the correctness to the semantic understanding of current statement.
Step 3：Average treatment is weighted to the sentence vector in background sentence group by attention mechanism, obtains background
Vector；
Specifically, the sentence vector in background sentence group is calculated respectively with current statement, background sentence group is obtained
For middle each sentence vector relative to the attention force parameter of current statement, attention force parameter here can be understood as background language
The weight of any one sentence in sentence group, it is understood that it is the associated degree of any one background sentence and current statement,
Then sentence vector all in background sentence group according to attention matrix is weighted averagely, obtains the back of the body of background sentence group
Scape vector, wherein, attention matrix column represents the weight of each sentence vector in background sentence group.The present invention is using attention
Sentence vector in background sentence group is weighted the important information that averagely, can automatically extract background sentence by power mechanism, into
One step improves the correctness to the semantic understanding of current statement.
Step 4：The meaning vector of current statement is obtained according to the sentence vector sum background vector of current statement.
As a kind of preferred embodiment, containing for current statement is obtained according to the sentence vector sum background vector of current statement
The process of adopted vector is specially：
The sentence vector sum background vector of current statement by multilayer neural network is handled, obtains the meaning of current statement
Vector
Specifically, the sentence vector sum background vector to current statement is calculated by multilayer neural network, you can
To the meaning vector of current statement, it is to be understood that corresponding information not only includes the letter of current statement in meaning vector
Breath further includes the key message of background sentence, is the combination of current statement and background sentence key message, the present invention can basis
The key message of background sentence group extraction carries out disambiguation to current statement, accurately to be classified to current statement.
The present invention provides a kind of method of semantic classification, applied to dialog process system, including：Obtain the institute in dialogue
There is sentence, and vector coding is carried out to every sentence, obtain the sentence vector of every sentence；Sentence vector is divided into current statement
Group and background sentence group, wherein, current statement group includes the sentence vector of current statement, and background sentence group includes all sentences
Sentence vector in sentence vector in addition to the sentence vector of current statement, current statement are the last item sentence in dialogue；It is logical
It crosses attention mechanism and average treatment is weighted to the sentence vector in background sentence group, obtain background vector；According to current language
The sentence vector sum background vector of sentence obtains the meaning vector of current statement.
As it can be seen that in practical applications, all sentences in dialogue are carried out vector coding by scheme using the present invention in advance
And be grouped, influence of the sentence in background sentence group to current statement is integrally considered, and utilize attention by sentence vector
Sentence vector in background sentence group is weighted averagely, automatically extracts the important information in background sentence group by mechanism, will
The important information and current statement of extraction combine, to be better understood from the meaning of current statement, so as to improve dialog process
Accuracy of the processing module of system rear class in processing business.
On the basis of above-described embodiment：
As a kind of preferred embodiment, the process that vector coding is carried out to every sentence is：
Recognition with Recurrent Neural Network RNN vector codings are carried out to every sentence.
Specifically, RNN (Recurrent Neural Network, Recognition with Recurrent Neural Network) neural network is that a kind of node is determined
To the artificial neural network of connection cyclization, the internal state of this network can show that dynamic time sequence behavior remembers that its is pervious defeated
Enter, when being related to continuous and context-sensitive task (such as speech recognition), there is the advantage of bigger, the present invention passes through
LSTM (Long Short-Term Memory, shot and long term memory network) model of RNN encodes the sentence in dialogue, more
Accurately.
Certainly, vector coding is carried out to sentence in addition to RNN may be used, other modes can also be used, the present invention is herein
It does not limit.
As a kind of preferred embodiment, the process that vector coding is carried out to every sentence is：
Convolutional neural networks CNN vector codings are carried out to every sentence.
Specifically, CNN (Convolutional Neural Networks, convolutional neural networks) is a kind of special depth
Layer neural network model, in practical applications, the CNN model training times are short, can reduce the duration of statement coding, shorten semantic
The time of classification, and then meet the concurrent demand of dialog process system.
Certainly, vector coding is carried out to sentence in addition to CNN may be used, other modes can also be used, the present invention is herein
It does not limit.
Fig. 2 is please referred to, Fig. 2 is a kind of structure diagram of semantic classification system provided by the present invention, applied to dialogue
Processing system, including：
Coding module 1 for obtaining all sentences in talking with, and carries out vector coding to every sentence, obtains every
The sentence vector of sentence；
Grouping module 2, for sentence vector to be divided into current statement group and background sentence group, wherein, current statement group packet
The sentence for including current statement is vectorial, in sentence vector of the background sentence group including all sentences in addition to the sentence vector of current statement
Sentence vector, current statement be talk in the last item sentence；
Processing module 3 is weighted average treatment for passing through attention mechanism to the sentence vector in background sentence group,
Obtain background vector；
Acquisition module 4, for obtaining the meaning of current statement vector according to the sentence vector sum background vector of current statement.
As a kind of preferred embodiment, coding module 1 be specifically used for carrying out every sentence Recognition with Recurrent Neural Network RNN to
Amount coding.
As a kind of preferred embodiment, coding module 1 be specifically used for carrying out every sentence convolutional neural networks CNN to
Amount coding.
As a kind of preferred embodiment, acquisition module 4 is specifically used for：
The sentence vector sum background vector of current statement by multilayer neural network is handled, obtains the meaning of current statement
Vector.
The present invention also provides a kind of dialog process systems, include the semantic classification system of such as above-mentioned any one.
The present invention also provides a kind of system of semantic classification and dialog process systems, have and above-mentioned semantic classification method
Identical advantageous effect.
Above-mentioned reality is please referred to for the introduction of the system and dialog process system of a kind of semantic classification provided by the present invention
Example is applied, details are not described herein by the present invention.
Each embodiment is described by the way of progressive in this specification, the highlights of each of the examples are with other
The difference of embodiment, just to refer each other for identical similar portion between each embodiment.For device disclosed in embodiment
For, since it is corresponded to the methods disclosed in the examples, so description is fairly simple, related part is said referring to method part
It is bright.
The foregoing description of the disclosed embodiments enables professional and technical personnel in the field to realize or use the present invention.
A variety of modifications of these embodiments will be apparent for those skilled in the art, it is as defined herein
General Principle can be realized in other embodiments without departing from the spirit or scope of the present invention.Therefore, it is of the invention
The embodiments shown herein is not intended to be limited to, and is to fit to and the principles and novel features disclosed herein phase one
The most wide range caused.
Claims (9)
1. a kind of method of semantic classification, applied to dialog process system, which is characterized in that including：
All sentences in dialogue are obtained, and vector coding is carried out to sentence every described, obtain the sentence of every sentence
Vector；
The sentence vector is divided into current statement group and background sentence group, wherein, the current statement group includes current statement
Sentence vector, except the sentence vector of the current statement in the sentence vector of the background sentence group including all sentences
Outer sentence vector, the current statement are the last item sentence in the dialogue；
Average treatment is weighted to the sentence vector in the background sentence group by attention mechanism, obtains background vector；
The meaning vector of the current statement is obtained according to background vector described in the sentence vector sum of the current statement.
2. the according to the method described in claim 1, it is characterized in that, process that vector coding is carried out to sentence every described
For：
Recognition with Recurrent Neural Network RNN vector codings are carried out to sentence every described.
3. the according to the method described in claim 1, it is characterized in that, process that vector coding is carried out to sentence every described
For：
Convolutional neural networks CNN vector codings are carried out to sentence every described.
4. according to the method described in claim 1-3 any one, which is characterized in that the sentence according to the current statement
The process that background vector described in vector sum obtains the meaning vector of the current statement is specially：
Background vector described in the sentence vector sum of the current statement by multilayer neural network is handled, obtains the current language
The meaning vector of sentence.
5. a kind of system of semantic classification, applied to dialog process system, which is characterized in that including：
Coding module for obtaining all sentences in talking with, and carries out vector coding to sentence every described, obtains every institute
The sentence vector of predicate sentence；
Grouping module, for the sentence vector to be divided into current statement group and background sentence group, wherein, the current statement group
Sentence vector including current statement, the background sentence group include removing the current language in the sentence vector of all sentences
Sentence vector outside the sentence vector of sentence, the current statement are the last item sentence in the dialogue；
Processing module is weighted average treatment for passing through attention mechanism to the sentence vector in the background sentence group,
Obtain background vector；
Acquisition module obtains containing for the current statement for background vector described in the sentence vector sum according to the current statement
Adopted vector.
6. system according to claim 5, which is characterized in that the coding module be specifically used for sentence every described into
Row Recognition with Recurrent Neural Network RNN vector codings.
7. system according to claim 5, which is characterized in that the coding module be specifically used for sentence every described into
Row convolutional neural networks CNN vector codings.
8. according to the system described in claim 5-7 any one, which is characterized in that the acquisition module is specifically used for：
Background vector described in the sentence vector sum of the current statement by multilayer neural network is handled, obtains the current language
The meaning vector of sentence.
9. a kind of dialog process system, which is characterized in that including the semantic classification system as described in claim 5-8 any one
System.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201711486033.1A CN108153737A (en) | 2017-12-30 | 2017-12-30 | A kind of method of semantic classification, system and dialog process system |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201711486033.1A CN108153737A (en) | 2017-12-30 | 2017-12-30 | A kind of method of semantic classification, system and dialog process system |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN108153737A true CN108153737A (en) | 2018-06-12 |
Family
ID=62460421
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201711486033.1A Pending CN108153737A (en) | 2017-12-30 | 2017-12-30 | A kind of method of semantic classification, system and dialog process system |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN108153737A (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109597994A (en) * | 2018-12-04 | 2019-04-09 | 挖财网络技术有限公司 | Short text problem semantic matching method and system |
| CN111339256A (en) * | 2020-02-28 | 2020-06-26 | 支付宝(杭州)信息技术有限公司 | Method and device for text processing |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105389307A (en) * | 2015-12-02 | 2016-03-09 | 上海智臻智能网络科技股份有限公司 | Statement intention category identification method and apparatus |
| CN106776828A (en) * | 2016-11-24 | 2017-05-31 | 竹间智能科技（上海）有限公司 | For keeping conversational system to talk with the method and system of continuity |
| CN106815211A (en) * | 2016-12-22 | 2017-06-09 | 深圳爱拼信息科技有限公司 | A kind of method that document subject matter modeling is carried out based on circulation focus mechanism |
-
2017
- 2017-12-30 CN CN201711486033.1A patent/CN108153737A/en active Pending
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105389307A (en) * | 2015-12-02 | 2016-03-09 | 上海智臻智能网络科技股份有限公司 | Statement intention category identification method and apparatus |
| CN106776828A (en) * | 2016-11-24 | 2017-05-31 | 竹间智能科技（上海）有限公司 | For keeping conversational system to talk with the method and system of continuity |
| CN106815211A (en) * | 2016-12-22 | 2017-06-09 | 深圳爱拼信息科技有限公司 | A kind of method that document subject matter modeling is carried out based on circulation focus mechanism |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109597994A (en) * | 2018-12-04 | 2019-04-09 | 挖财网络技术有限公司 | Short text problem semantic matching method and system |
| CN109597994B (en) * | 2018-12-04 | 2023-06-06 | 挖财网络技术有限公司 | Short text problem semantic matching method and system |
| CN111339256A (en) * | 2020-02-28 | 2020-06-26 | 支付宝(杭州)信息技术有限公司 | Method and device for text processing |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN111477216B (en) | Training method and system for voice and meaning understanding model of conversation robot | |
| CN113505205A (en) | System and method for man-machine conversation | |
| CN108549637A (en) | Method for recognizing semantics, device based on phonetic and interactive system | |
| CN109885670A (en) | A kind of interaction attention coding sentiment analysis method towards topic text | |
| CN109388700A (en) | A kind of intension recognizing method and system | |
| CN110675859B (en) | Multi-emotion recognition method, system, medium, and apparatus combining speech and text | |
| CN106297773A (en) | A kind of neutral net acoustic training model method | |
| CN107657313B (en) | System and method for transfer learning of natural language processing task based on field adaptation | |
| CN107679225B (en) | Reply generation method based on keywords | |
| CN111179917B (en) | Speech recognition model training method, system, mobile terminal and storage medium | |
| GB2424502A (en) | Apparatus and method for model adaptation for spoken language understanding | |
| CN109344242A (en) | A kind of dialogue answering method, device, equipment and storage medium | |
| CN113987179A (en) | Knowledge enhancement and backtracking loss-based conversational emotion recognition network model, construction method, electronic device and storage medium | |
| CN106548777B (en) | Data processing method and device for intelligent robot | |
| CN105446955A (en) | Adaptive word segmentation method | |
| EP3816858A3 (en) | Character recognition method and apparatus, electronic device and computer readable storage medium | |
| CN115309877A (en) | Dialog generation method, dialog model training method and device | |
| CN114547293A (en) | Cross-platform false news detection method and system | |
| CN108153737A (en) | A kind of method of semantic classification, system and dialog process system | |
| CN114238373A (en) | Method and device for converting natural language question into structured query statement | |
| CN106708950B (en) | Data processing method and device for intelligent robot self-learning system | |
| CN109933773A (en) | A kind of multiple semantic sentence analysis system and method | |
| CN112801721A (en) | Information processing method, information processing device, electronic equipment and storage medium | |
| CN111680514B (en) | Information processing and model training method, device, equipment and storage medium | |
| CN106682642A (en) | Multi-language-oriented behavior identification method and multi-language-oriented behavior identification 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 | ||
| RJ01 | Rejection of invention patent application after publication | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20180612 |