CN106897266A - For the text handling method and system of intelligent robot - Google Patents
For the text handling method and system of intelligent robot Download PDFInfo
- Publication number
- CN106897266A CN106897266A CN201710082574.1A CN201710082574A CN106897266A CN 106897266 A CN106897266 A CN 106897266A CN 201710082574 A CN201710082574 A CN 201710082574A CN 106897266 A CN106897266 A CN 106897266A
- Authority
- CN
- China
- Prior art keywords
- text
- normalization
- text data
- core word
- similarity
- 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
- G06F40/20—Natural language analysis
- G06F40/205—Parsing
- G06F40/211—Syntactic parsing, e.g. based on context-free grammar [CFG] or unification grammars
-
- 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
- G06F16/3332—Query translation
- G06F16/3334—Selection or weighting of terms from queries, including natural language queries
-
- 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
- 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
- G06F40/30—Semantic analysis
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- Computational Linguistics (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Artificial Intelligence (AREA)
- Audiology, Speech & Language Pathology (AREA)
- General Health & Medical Sciences (AREA)
- Health & Medical Sciences (AREA)
- Databases & Information Systems (AREA)
- Data Mining & Analysis (AREA)
- Mathematical Physics (AREA)
- Acoustics & Sound (AREA)
- Human Computer Interaction (AREA)
- Machine Translation (AREA)
Abstract
The invention discloses a kind of text handling method and text processing system for intelligent robot.The intelligent robot is provided with robot operating system, and the method includes:Obtain pending text data;Core word extraction is carried out to the pending text data by syntactic analysis, being compared based on core word carries out Text normalization treatment, wherein, the text data to comparing normalization failure based on core word carries out Text normalization based on Arithmetic of Semantic Similarity;Output and/or storage Text normalization result.The inventive method can improve intelligent robot and the interaction capabilities with user, preferable Text normalization treatment can be carried out to pending text data, it is easy to follow-up Language Processing, it is possible to increase the language interaction capabilities of intelligent robot, better meets user's request.
Description
Technical field
The present invention relates to field in intelligent robotics, more particularly to a kind of text handling method for intelligent robot and it is
System.
Background technology
With continuing to develop for science and technology, the introducing of information technology, computer technology and artificial intelligence technology, machine
Industrial circle is progressively walked out in the research of people, gradually extend to the neck such as medical treatment, health care, family, amusement and service industry
Domain.
And people are promoted to anthropomorphic question and answer, independence for the requirement of the robot multiple mechanical action of substance of also conforming to the principle of simplicity
And intelligent intelligent robot, the key factor that man-machine interaction also just develops as decision intelligent robot, therefore, improve intelligence
The man-machine interaction efficiency of energy robot, the major issue as current urgent need to resolve, then corresponding, to the developer of robot
Requirement higher is it is also proposed, to improve for handling machine people properties of product and robot data's treatment efficiency.
The content of the invention
One of technical problems to be solved by the invention are to need to provide a kind of raising for handling machine people's properties of product
With robot data's treatment efficiency, the intelligent solution of intelligent robot is lifted.
In order to solve the above-mentioned technical problem, embodiments herein provide firstly a kind of text for intelligent robot
Processing method, the intelligent robot is provided with robot operating system, and the method includes:Obtain pending text data;
Core word extraction is carried out to the pending text data by syntactic analysis, being compared based on core word carries out Text normalization
Treatment, wherein, the text data to comparing normalization failure based on core word carries out text normalizing based on Arithmetic of Semantic Similarity
Change;Output and/or storage Text normalization result.
Preferably, in the step of carrying out Text normalization based on Arithmetic of Semantic Similarity, based on shortest path length and
Depth capacity in classification carrys out computing semantic similarity.
Preferably, the Arithmetic of Semantic Similarity is to be mixed with to calculate similarity and the language according to concept according to semantic content
Justice distance calculates the algorithm of similarity.
Preferably, in carrying out the step of Text normalization is processed based on core word comparison, by the textual data after normalization
According to being placed in same normalization list, wherein, using the most short text data of number of words as the list normalization entry, it is other
Text data as the list list member.
The embodiment of the present invention additionally provides a kind of text processing system for intelligent robot, the intelligent robot peace
Equipped with robot operating system, text processing system includes:Text data acquisition module, it obtains pending textual data
According to;Text normalization module, it passes through syntactic analysis carries out core word extraction to the pending text data, based on core
Word is compared and carries out Text normalization treatment, wherein, the text data to comparing normalization failure based on core word, based on semantic phase
Text normalization is carried out like degree algorithm;Result treatment module, its output and/or storage Text normalization result.
Preferably, the Text normalization module, it is further carrying out Text normalization based on Arithmetic of Semantic Similarity
When, based on the depth capacity in shortest path length and classification come computing semantic similarity.
Preferably, the Arithmetic of Semantic Similarity is to be mixed with to calculate similarity and the language according to concept according to semantic content
Justice distance calculates the algorithm of similarity.
Preferably, the Text normalization module, it is further carrying out Text normalization treatment based on core word comparison
When, the text data after normalization is placed in same normalization list, wherein, using the most short text data of number of words as the row
The normalization entry of table, other text datas as the list list member.
Compared with prior art, one or more embodiments in such scheme can have the following advantages that or beneficial effect
Really:
The embodiment of the present invention carries out syntactic analysis and obtains each textual data by the pending text data to user input
Core word in, being then based on core word comparison carries out Text normalization treatment to obtain Text normalization result.And, it is right
The text data of normalization failure is compared based on core word, Text normalization is carried out based on Arithmetic of Semantic Similarity.The above method
Intelligent robot and the interaction capabilities with user can be improved, preferable text can be carried out to pending text data
Normalized, is easy to follow-up Language Processing, it is possible to increase the language interaction capabilities of intelligent robot, better meets user
Demand.
Other features and advantages of the present invention will be illustrated in the following description, also, the partly change from specification
Obtain it is clear that or being understood by implementing technical scheme.The purpose of the present invention and other advantages can by
Specifically noted structure and/or flow are realized and obtained in specification, claims and accompanying drawing.
Brief description of the drawings
Accompanying drawing is used for providing to the technical scheme of the application or further understanding for prior art, and constitutes specification
A part.Wherein, the accompanying drawing of expression the embodiment of the present application is used to explain the technical side of the application together with embodiments herein
Case, but do not constitute the limitation to technical scheme.
Fig. 1 is the schematic flow sheet of the text handling method for intelligent robot of the application first embodiment.
The schematic diagram that Fig. 2 is processed for the Text normalization of the embodiment of the present application.
Fig. 3 is the structural representation of the text processing system for intelligent robot of the application second embodiment.
Specific embodiment
Describe embodiments of the present invention in detail below with reference to drawings and Examples, how the present invention is applied whereby
Technological means solves technical problem, and reaches the implementation process of relevant art effect and can fully understand and implement according to this.This Shen
Each feature that please be in embodiment and embodiment, can be combined with each other under the premise of not colliding, the technical scheme for being formed
Within protection scope of the present invention.
In addition, the flow of accompanying drawing can be in the such as one group computer system of computer executable instructions the step of illustrating
Middle execution.And, although show logical order in flow charts, but in some cases, can be with different from herein
Order performs shown or described step.
In existing field in intelligent robotics, most of robots can carry out single interactive voice with user, complete
Simple question and answer behavior is carried out into the task of user's imparting or with user.Existing this interactive mode be usually by with
The dynamic question of householder, wakes up what robot was interacted therewith, and robot is looked into according to the problem of user from Q & A database
Corresponding response message is looked for, and voice messaging is exported according to response message.But, what existing Q & A database requirement was inquired about asks
Topic sentence possesses certain (aspect such as grammer, form) normalization, and user with robot when interactive voice is carried out, it is necessary to export
Specification sex chromosome mosaicism, robot can just inquire corresponding result, and the influence of the factor such as individual character or age according to user, they
The problem of proposition often has uncertainty, causes robot to inquire about corresponding answer from database according to problem.And
And, it is this require user send specification sex chromosome mosaicism pattern also reduce user use robot interest, bring bad use
Experience at family.The embodiment of the present invention proposes a kind of text handling method for intelligent robot, and the method can be asked similar
Answer database to be updated, pending text data (such as substantial amounts of problem sentence) is carried out into Text normalization treatment, be
Follow-up Language Processing (such as robot answers a question) lays the foundation.Specifically, this method passes through syntactic analysis to pending
Text data carries out core word extraction, and being compared based on core word carries out Text normalization treatment, wherein, to being compared based on core word
The text data of failure is normalized, Text normalization is carried out based on Arithmetic of Semantic Similarity, be finally completed normalized process,
To improve for handling machine people properties of product and robot data's treatment efficiency.
First embodiment
Fig. 1 is the schematic flow sheet of the example one for being related to the text handling method for intelligent robot of the invention, should
Intelligent robot is preferably the robot for being provided with robot operating system, however, other have the tables such as voice, expression, action
Danone power, do not use the intelligent robot of the robot operating system (or equipment) the present embodiment can also be realized.Below
Each step involved by illustrating the method with reference to Fig. 1.
In step s 110, pending text data is obtained.
During text message is obtained, substantial amounts of text message can be captured or by input equipment by network
(such as keyboard, speech recognition apparatus etc.) directly enters pending text message.Pending text message can be a plurality of
The text data of sentence, multiple vocabulary mixing compositions, such as including vocabulary " lawyer " and sentence " lawyer of immigrant's case defense "
Text data.For follow-up Language Processing, user and the more preferable interactive voice of robot are realized, obtained in the present embodiment
Pending text data is mainly the problem data of the multi-form being applied in dialogue, and these problem sentences are mostly with sentence
Form statement, for example " Beethovan can be introduced ", the text such as " whom Beethovan is, I does not know, can tell me "
Notebook data.
In the step s 120, core word extraction is carried out to pending text data by syntactic analysis, based on core word
Comparison carries out Text normalization treatment.
Specifically, first the text message of different expression form in pending text data is sorted out, is obtained word
The text message of species, the text message of sentence species.Because the text of word species itself has been entry form, because
This need not again be processed to it.And for the text message of sentence species, in addition it is also necessary to will be directed to by syntactic analysis
Core word is extracted, and thinks that Text normalization below is prepared.
When core word extraction is carried out, can be realized by the way of syntactic analysis.Specifically, it is exactly according to given
Syntax rule collection, analyze the syntax rule employed in sentence forming process.The result of syntactic analysis is typically expressed as tree
Structure, the node of tree represents the title of the syntactic units of sentence.Specific algorithm can be the interdependent syntactic analysis side based on conversion
Method, in the analysis method based on conversion, dependency analysis are counted as performing some actions to input sentence, are built by these actions
Erect contacting between word and word in sentence.Each action is all by current analysis State Transferring to new state.Based on conversion
Analysis method do not search for the action sequence of global optimum, but use greedy strategy, select local according to current state
Optimal action a, action would not change again once performing, thus also known as deterministic parsing method.
When being analyzed to " lawyer of immigrant's case defense ", output format is as follows:
0 immigrant's attribute of noun 1
The word structure of 1 case noun 3
The 2 defense complements of noun 1
3 attribute of structural auxiliary word 4
The core word of 4 lawyer's noun -1
It is thus determined that the core word of above-mentioned sentence is " lawyer ".
In addition to taking algorithm above, can also be using being integrated with probability context-free grammar, based on nerve net
The interdependent syntactic analysis of network and the method for the interdependent syntactic analysis based on conversion.
After completing core word and extracting, being compared by core word carries out Text normalization treatment.In this example, by text
The level of this normalized mainly includes:Word and word, word and sentence, sentence and sentence.In the mistake that core word is compared
Cheng Zhong, if core word comparison result is consistent, the two is placed in same list, repeats this process, until by this
Normalized content traversal is needed to complete.For convenience the later stage use the normalization list, normalize list in, choose number of words
Minimum text data is used as normalization entry, other members as the normalization list.For example, " lawyer " and " immigrant
The lawyer of case defense ", core word is " lawyer ", the two is placed in same class list by core word contrast, due to " rule
The number of words of teacher " is minimum, then as normalization entry, and " lawyer of immigrant's case defense " used as the member of the list.
In step s 130, judge whether to compare the text data that normalization fails based on core word, if in the presence of,
Step S140 is performed, step S150 is otherwise performed.
In step S140, the text data to comparing normalization failure based on core word, based on Arithmetic of Semantic Similarity
Carry out Text normalization.
By step S120, the normalized of the most contents of pending text data is had been completed, however,
Due to the difference between different sentences, it is likely that can there is the inconsistent situation of core word, in this case, lost for normalization
The text data for losing, is normalized based on Arithmetic of Semantic Similarity.It should be noted that the Arithmetic of Semantic Similarity is
It is mixed with the algorithm for similarity being calculated according to semantic content and similarity being calculated according to the semantic distance of concept, it is preferable that the language
Adopted similarity algorithm is come computing semantic similarity based on the depth capacity in shortest path length and classification.
The Arithmetic of Semantic Similarity model such as following formula:
Wherein, the maxdepth (c) in denominator represents the depth capacity in WordNet classification tree.The algorithm considers two
Concept c1, c2Between shortest path length length (c1,c2), for a fixed classification tree, between two concepts
Path is bigger, and semantic similarity is smaller.For example the semantic similarity of " post-doctor " and " administrative personnel " be 0.57, " personage " and
The semantic similarity of " post-doctor " is 0.4, then " post-doctor " and " administrative personnel " is divided into a class.
From from the perspective of information theory, the semantic similarity value that solution required by the algorithm is obtained is that path is provided between concept
Information content.The calculating of the semantic similarity is not limited solely to the semantic depth in same classification, while improving all semantemes
Connect the problem of equal weight.
Calculated by Arithmetic of Semantic Similarity above based on core word compare normalization failure text data in
Similarity between core word, further carries out Text normalization treatment.
In step S150, export and/or storage Text normalization result.
When follow-up speech processes, treatment is analyzed using the normalized result, completes robot and user's
Interactive voice.
In sum, the method for the present embodiment can improve intelligent robot and the interaction capabilities with user, treat
The text data for the treatment of can carry out preferable Text normalization treatment, be easy to follow-up Language Processing, improve for processor
Device people properties of product and robot data's treatment efficiency, so as to lift the language interaction capabilities of intelligent robot, better meet
User's request.
Second embodiment
Fig. 3 is the structured flowchart of the text processing system 300 for intelligent robot of the embodiment of the present invention.Intelligence therein
Energy machine is artificially provided with the robot of robot operating system.As shown in figure 3, the system 300 of the embodiment of the present application is mainly wrapped
Include:Text data acquisition module 310, Text normalization module 320 and result treatment module 330.
Text data acquisition module 310, it obtains pending text data.
Text normalization module 320, it passes through syntactic analysis carries out core word extraction to the pending text data,
Being compared based on core word carries out Text normalization treatment, wherein, the text data to comparing normalization failure based on core word, base
Text normalization is carried out in Arithmetic of Semantic Similarity.Text normalization module 320, it is further based on Arithmetic of Semantic Similarity
When carrying out Text normalization, based on the depth capacity in shortest path length and classification come computing semantic similarity.The semanteme
Similarity algorithm is to be mixed with the algorithm for calculating similarity according to semantic content and similarity being calculated according to the semantic distance of concept.
The Text normalization module 320, its further based on core word compare carry out Text normalization process when, will be pending
Used as normalization entry, other text datas are normalized the most short text data of number of words as needs in text data
List member.
Result treatment module 330, its output and/or storage Text normalization result.
The system 300 is configurable to the Text normalization model shown in Fig. 2.
By rationally setting, the system 300 of the present embodiment can perform each step of first embodiment, no longer go to live in the household of one's in-laws on getting married herein
State.
Because the method for the present invention describes what is realized in computer systems.The computer system can for example be set
In the control core processor of robot.For example, method described herein can be implemented as what can be performed with control logic
Software, it is performed by the CPU in robot operating system.Function as herein described can be implemented as storage to be had in non-transitory
Programmed instruction set in shape computer-readable medium.When implemented in this fashion, the computer program includes one group of instruction,
When group instruction is run by computer, it promotes computer to perform the method that can implement above-mentioned functions.FPGA can be temporary
When or be permanently mounted in non-transitory tangible computer computer-readable recording medium, for example ROM chip, computer storage,
Disk or other storage mediums.In addition to being realized with software, logic as herein described can utilize discrete parts, integrated electricity
What road and programmable logic device (such as, field programmable gate array (FPGA) or microprocessor) were used in combination programmable patrols
Volume, or embodied including any other equipment that they are combined.All such embodiments are intended to fall under model of the invention
Within enclosing.
It should be understood that disclosed embodiment of this invention is not limited to ad hoc structure disclosed herein, process step
Or material, and the equivalent substitute of these features that those of ordinary skill in the related art are understood should be extended to.Should also manage
Solution, term as used herein is only used for describing the purpose of specific embodiment, and is not intended to limit.
" one embodiment " or " embodiment " mentioned in specification means special characteristic, the structure for describing in conjunction with the embodiments
Or characteristic is included at least one embodiment of the present invention.Therefore, the phrase " reality that specification various places throughout occurs
Apply example " or " embodiment " same embodiment might not be referred both to.
While it is disclosed that implementation method as above, but described content is only to facilitate understanding the present invention and adopting
Implementation method, is not limited to the present invention.Any those skilled in the art to which this invention pertains, are not departing from this
On the premise of the disclosed spirit and scope of invention, any modification and change can be made in the formal and details implemented,
But scope of patent protection of the invention, must be still defined by the scope of which is defined in the appended claims.
Claims (8)
1. a kind of text handling method for intelligent robot, the intelligent robot is provided with robot operating system, should
Method includes:
Obtain pending text data;
Core word extraction is carried out to the pending text data by syntactic analysis, carrying out text based on core word comparison returns
One change is processed, wherein, the text data to comparing normalization failure based on core word carries out text based on Arithmetic of Semantic Similarity
Normalization;
Output and/or storage Text normalization result.
2. text handling method according to claim 1, it is characterised in that text is being carried out based on Arithmetic of Semantic Similarity
In normalized step,
Based on the depth capacity in shortest path length and classification come computing semantic similarity.
3. text handling method according to claim 1 and 2, it is characterised in that
The Arithmetic of Semantic Similarity is to be mixed with calculate similarity according to semantic content and be calculated according to the semantic distance of concept
The algorithm of similarity.
4. the text handling method according to any one of claims 1 to 3, it is characterised in that compared based on core word
In carrying out the step of Text normalization is processed,
By the text data after normalization be placed on it is same normalization list in, wherein, using the most short text data of number of words as this
The normalization entry of list, other text datas as the list list member.
5. a kind of text processing system for intelligent robot, the intelligent robot is provided with robot operating system, should
Text processing system includes:
Text data acquisition module, it obtains pending text data;
Text normalization module, it passes through syntactic analysis carries out core word extraction to the pending text data, based on core
Heart word is compared and carries out Text normalization treatment, wherein, the text data to comparing normalization failure based on core word, based on semanteme
Similarity algorithm carries out Text normalization;
Result treatment module, its output and/or storage Text normalization result.
6. text processing system according to claim 5, it is characterised in that the Text normalization module, its is further
When Text normalization is carried out based on Arithmetic of Semantic Similarity, calculated based on the depth capacity in shortest path length and classification
Semantic similarity.
7. the text processing system according to claim 5 or 6, it is characterised in that
The Arithmetic of Semantic Similarity is to be mixed with calculate similarity according to semantic content and be calculated according to the semantic distance of concept
The algorithm of similarity.
8. the text processing system according to any one of claim 5~7, it is characterised in that the Text normalization mould
Block, it further when Text normalization treatment is carried out based on core word comparison, the text data after normalization is placed on same
In normalization list, wherein, using the most short text data of number of words as the list normalization entry, other text datas make
It is the list member of the list.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201710082574.1A CN106897266A (en) | 2017-02-16 | 2017-02-16 | For the text handling method and system of intelligent robot |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201710082574.1A CN106897266A (en) | 2017-02-16 | 2017-02-16 | For the text handling method and system of intelligent robot |
Publications (1)
Publication Number | Publication Date |
---|---|
CN106897266A true CN106897266A (en) | 2017-06-27 |
Family
ID=59198210
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201710082574.1A Pending CN106897266A (en) | 2017-02-16 | 2017-02-16 | For the text handling method and system of intelligent robot |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN106897266A (en) |
Cited By (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN108920488A (en) * | 2018-05-14 | 2018-11-30 | 平安科技(深圳)有限公司 | The natural language processing method and device that multisystem combines |
CN110889285A (en) * | 2018-08-16 | 2020-03-17 | 阿里巴巴集团控股有限公司 | Method, apparatus, device and medium for determining core word |
Citations (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN102937901A (en) * | 2012-10-17 | 2013-02-20 | 武汉钢铁(集团)公司 | Multi-tenant architecture design method |
CN103020230A (en) * | 2012-12-14 | 2013-04-03 | 中国科学院声学研究所 | Semantic fuzzy matching method |
CN106202159A (en) * | 2016-06-23 | 2016-12-07 | 深圳追科技有限公司 | A kind of man-machine interaction method of customer service system |
-
2017
- 2017-02-16 CN CN201710082574.1A patent/CN106897266A/en active Pending
Patent Citations (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN102937901A (en) * | 2012-10-17 | 2013-02-20 | 武汉钢铁(集团)公司 | Multi-tenant architecture design method |
CN103020230A (en) * | 2012-12-14 | 2013-04-03 | 中国科学院声学研究所 | Semantic fuzzy matching method |
CN106202159A (en) * | 2016-06-23 | 2016-12-07 | 深圳追科技有限公司 | A kind of man-machine interaction method of customer service system |
Non-Patent Citations (2)
Title |
---|
刘小明 等: "一种结合本体和焦点的问题分类方法", 《北京理工大学学报》 * |
董才正 等: "面向问答社区的中文问题分类", 《计算机应用》 * |
Cited By (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN108920488A (en) * | 2018-05-14 | 2018-11-30 | 平安科技(深圳)有限公司 | The natural language processing method and device that multisystem combines |
CN110889285A (en) * | 2018-08-16 | 2020-03-17 | 阿里巴巴集团控股有限公司 | Method, apparatus, device and medium for determining core word |
CN110889285B (en) * | 2018-08-16 | 2023-06-16 | 阿里巴巴集团控股有限公司 | Method, device, equipment and medium for determining core word |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN109493166B (en) | Construction method for task type dialogue system aiming at e-commerce shopping guide scene | |
EP3696810B1 (en) | Training encoder model and/or using trained encoder model to determine responsive action(s) for natural language input | |
CN107563498B (en) | Image description method and system based on visual and semantic attention combined strategy | |
KR20190019962A (en) | Architectures and processes for computer learning and understanding | |
CN112818107B (en) | Conversation robot for daily life and chat method thereof | |
CN108536681A (en) | Intelligent answer method, apparatus, equipment and storage medium based on sentiment analysis | |
US20220092441A1 (en) | Training method and apparatus, dialogue processing method and system, and medium | |
CN102662931A (en) | Semantic role labeling method based on synergetic neural network | |
CN107273358A (en) | A kind of end-to-end English structure of an article automatic analysis method based on pipe modes | |
CN116561538A (en) | Question-answer scoring method, question-answer scoring device, electronic equipment and storage medium | |
KR102122081B1 (en) | Method for providing artificial intelligence based natural language corpus building service using interactive learning between artificial collector, analyzer and inspector, and building | |
Tyagi et al. | Demystifying the role of natural language processing (NLP) in smart city applications: background, motivation, recent advances, and future research directions | |
CN103049490B (en) | Between knowledge network node, attribute generates system and the method for generation | |
CN112052318A (en) | Semantic recognition method and device, computer equipment and storage medium | |
CN108735198B (en) | Phoneme synthesizing method, device and electronic equipment based on medical conditions data | |
KR20160058531A (en) | Method for establishing syntactic analysis model using deep learning and apparatus for perforing the method | |
CN114153994A (en) | Medical insurance information question-answering method and device | |
Veena et al. | A graph-based relation extraction method for question answering system | |
CN114647713A (en) | Knowledge graph question-answering method, device and storage medium based on virtual confrontation | |
CN106897266A (en) | For the text handling method and system of intelligent robot | |
CN111241843B (en) | Semantic relation inference system and method based on composite neural network | |
Xia et al. | Study of text emotion analysis based on deep learning | |
Jin et al. | Textual content prediction via fuzzy attention neural network model without predefined knowledge | |
CN115345177A (en) | Intention recognition model training method and dialogue method and device | |
CN117373591A (en) | Disease identification method and device for electronic medical record, electronic equipment and storage medium |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
PB01 | Publication | ||
PB01 | Publication | ||
SE01 | Entry into force of request for substantive examination | ||
RJ01 | Rejection of invention patent application after publication |
Application publication date: 20170627 |
|
RJ01 | Rejection of invention patent application after publication |