CN107256210A - The Situation of Students ' English Writing artificial intelligence system analyzed based on deep semantic - Google Patents
The Situation of Students ' English Writing artificial intelligence system analyzed based on deep semantic Download PDFInfo
- Publication number
- CN107256210A CN107256210A CN201710432046.4A CN201710432046A CN107256210A CN 107256210 A CN107256210 A CN 107256210A CN 201710432046 A CN201710432046 A CN 201710432046A CN 107256210 A CN107256210 A CN 107256210A
- Authority
- CN
- China
- Prior art keywords
- word
- chapters
- sections
- title
- artificial intelligence
- 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
- G06F40/279—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
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N5/00—Computing arrangements using knowledge-based models
- G06N5/02—Knowledge representation; Symbolic representation
- G06N5/022—Knowledge engineering; Knowledge acquisition
- G06N5/025—Extracting rules from data
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Computational Linguistics (AREA)
- Physics & Mathematics (AREA)
- Artificial Intelligence (AREA)
- General Health & Medical Sciences (AREA)
- Audiology, Speech & Language Pathology (AREA)
- Health & Medical Sciences (AREA)
- Data Mining & Analysis (AREA)
- Evolutionary Computation (AREA)
- Computing Systems (AREA)
- Mathematical Physics (AREA)
- Software Systems (AREA)
- Machine Translation (AREA)
Abstract
The present invention relates to a kind of Situation of Students ' English Writing artificial intelligence system analyzed based on deep semantic, including：Word sort module, for carrying out meaning of a word statistics to word, divides part of speech, and classification is stored in cloud knowledge base, while recording each word frequency of use situation；Sentence pattern sort module, for the progress classification of sentence pattern structure to be stored in cloud knowledge base, while recording the frequency of each sentence pattern；Paragraph analysis module, the central idea for analyzing each chapters and sections of generation；Title analysis module, the central idea for analyzing obtained chapters and sections according to paragraph analysis module, and analyze the degree of bringing out the theme for obtaining chapters and sections and title.Compared with prior art, the present invention analyzes the central idea of article with reference to sentence pattern with the classification of the word part of speech meaning of a word, finally gives the evaluation to article there is provided a kind of intelligent English Writing analysis mode, automatic exactly semanteme can be analyzed.
Description
Technical field
The present invention relates to English Writing analysis field, write more particularly, to a kind of English analyzed based on deep semantic
Make artificial intelligence system.
Background technology
For the development of natural language processing technique, no matter the commercial market of home and abroad, scientific research field, and bigger model
Rapid, accurate, depth analysis requirement always all pole of the human society enclosed for extensive, large-scale real text content
Its is urgent.Technological means of the global scientific and technological circle at present in terms of extensive real text is handled is broadly divided into the nature of (1) early stage
What advanced technology was many before the ratio that language processing techniques (NLP) and (2) are gradually formed for nearest 30 years uses mixing side
The treatment technology for the deep semantic analysis module that method is built.
The natural language processing technique (NLP) of early stage is also unique most prominent in terms of Text extraction the characteristics of
Technical functionality is-computer for big section word semantics recognition.By the technology, computer can recognize a big section words substantially
General idea.Such as a certain big section of literal expression one extremely complex scene and behavior, but pass through the technology, calculate
Machine can recognize that the main information included in fact in this section of words is：In the substantially behavior of some scene and period.The technology is made
To realize identification technology of the computer to text, during the research and use of nearly half a century, mainly it is used in machine and turns over
Translate, information retrieval, and the field such as information extraction, and the increasing application achievements obtained in extensive scope.
For the angle realized from technology, the technology is concentrated mainly in the analysis of morphology and syntax, rule-based mutually to be tied with statistics
The language analysis of conjunction.More there are phrase structure grammar, the vocabulary function that head drives in the syntactic analysis model of maturation at present
The technologies such as grammer, dependency grammar.After integrated a variety of natural language processing techniques, semantics recognition is developing progressively as comparative maturity
Application.
The application of main flow is the highly integrated product of front end speech recognition technology on current commercial market, as known to masses
The Siri of Apple Inc., and the well-known University of Science and Technology's news of domestic contrast fly the semantic identification of the related voice based on the release of this technology and answered
With.
In current semantic analysis spring tide, most of time and energy are all used in the place for natural language (NLP)
In reason, that is to say, that have very much limitation in actual life based on applying for semantics recognition technology.After all, the technology can only be known
The language message on not simple basis, however no matter business circles, educational circles, scientific circles for computerization semantics recognition real demand
The information on simple basis, but text include various dimensions, profound information-and this point are the technologies can not be real
Existing.Various circles of society's widespread demand and earlier technique based on sustainable growth can not meet the demand, the multi-dimensional semantic of higher level
Depth analysis just rises naturally from global scientific research academia.
Thus, according to the deduction of natural language and big data, a kind of pattern is gone into, and derive into one based on structuring
The module of autonomous learning, this process is undoubtedly inaccurate for artificial intelligence writing assistant.Reason has following several
Point, first, it is temporal to waste.This process is the accurate model of a kind of neither one itself, and this method is substantially by big
The fuzzy matching of data, is excavated for data are continuous, finds out similar model, and (either neural by Recognition with Recurrent Neural Network
Netspeak model, or LSTM models) continuous iterative cycles.Do not say first precision can how, light row is into a kind of model
Standard, may be accomplished by taking a period of time.Second, model accuracy.In the age that office software all carries English verification,
Engineering inveterate habit how can be allowed to carry the big data information bank of specific background demand, and check out a kind of new model automatically
It is a difficult thing.For example, how to allow machine to understand the judgment criteria of English composition, and student is drawn by this machine
Model goes to write whether English composition can really write out a complete articleAnswer is negative.Go into seriously reason, that is,
Machine does not have autonomous expert system, such as professional English composition, is to need to understand very much expert's note of this kind of judgment criteria
Enter information model.In this, as foundation stone, dynamic adjustment judgment criteria can be only achieved optimal most accurate judge based on artificial intelligence
As a result.
The content of the invention
It is an object of the present invention to overcome the above-mentioned drawbacks of the prior art and provide one kind is based on deep semantic
The Situation of Students ' English Writing artificial intelligence system of analysis.
The purpose of the present invention can be achieved through the following technical solutions：
A kind of Situation of Students ' English Writing artificial intelligence system analyzed based on deep semantic, including：
Word sort module, for carrying out meaning of a word statistics to word, divides part of speech, and classification is stored in cloud knowledge base,
Record each word frequency of use situation simultaneously；
Sentence pattern sort module, for the progress classification of sentence pattern structure to be stored in cloud knowledge base, while recording each sentence pattern
Frequency；
Paragraph analysis module, the central idea for analyzing each chapters and sections of generation；
Title analysis module, the central idea for analyzing obtained chapters and sections according to paragraph analysis module, and analysis are obtained
The degree of bringing out the theme of chapters and sections and title.
The title analysis module routine specifically includes step：
The central idea of obtained chapters and sections is analyzed according to paragraph analysis module, the keyword of chapters and sections is extracted；
Extract the keyword of title；
According to the meaning of a word of each word of cloud knowledge library storage, the pass between the keyword of chapters and sections and the keyword of title is generated
Connection degree, and using this degree of association as the chapters and sections degree of bringing out the theme.
The keyword of the chapters and sections is the immediate word of central idea of the meaning of a word and chapters and sections in chapters and sections, or the meaning of a word in dictionary
With the immediate word of central idea of chapters and sections,
The keyword of the title is the meaning of a word and the immediate word of title implication in title.
State the paragraph analysis module course of work and specifically include step：
According to the sentence pattern of sentence, and in sentence each word the meaning of a word and part of speech, determine the implication of each word in the sentence；
Extract the word meanings that part of speech in sentence is noun；
All word meanings according to being extracted in chapters and sections determine the thought of center chapters and sections.
The analysis report that the artificial intelligence system is generated to english composition, report content includes：Error rate, tricky question rate,
Write type and composition specification scoring.
Compared with prior art, the present invention just has the advantage that：
1) with the classification of the word part of speech meaning of a word, the central idea of article is analyzed with reference to sentence pattern, finally given to article
Evaluate there is provided a kind of intelligent English Writing analysis mode, automatic exactly semanteme can be analyzed.
2) and as the substantial amounts of precipitation of this data is accumulated, the high level model algorithm of big data and earlier set is mutually printed
Card is with correcting, so that the improvement of implementation model and the modification of original character text message are improved.
3) operation of this process can light implementation model algorithm automatic evolution.
Brief description of the drawings
Fig. 1 is structural representation of the invention；
Fig. 2 is the word relationship map in the present invention for machine learning；
The word list according to parts of speech classification that Fig. 3 draws for machine learning in the present invention；
The part of speech relation list that Fig. 4 generates for machine learning in the present invention；
Wherein：1st, word sort module, 2, sentence pattern sort module, 3, paragraph analysis module, 4, title analysis module.
Embodiment
The present invention is described in detail with specific embodiment below in conjunction with the accompanying drawings.The present embodiment is with technical solution of the present invention
Premised on implemented, give detailed embodiment and specific operating process, but protection scope of the present invention is not limited to
Following embodiments.
A kind of Situation of Students ' English Writing artificial intelligence system analyzed based on deep semantic, including：
Word sort module 1, for carrying out meaning of a word statistics to word, divides part of speech, and classification is stored in cloud knowledge base,
Record each word frequency of use situation simultaneously；
Sentence pattern sort module 2, for the progress classification of sentence pattern structure to be stored in cloud knowledge base, while recording each sentence pattern
Frequency；
Paragraph analysis module 3, the central idea for analyzing each chapters and sections of generation；
Title analysis module 4, the central idea for analyzing obtained chapters and sections according to paragraph analysis module, and analysis are obtained
The degree of bringing out the theme of chapters and sections and title.
The course of work of title analysis module 4 specifically includes step：
The central idea of obtained chapters and sections is analyzed according to paragraph analysis module 3, the keyword of chapters and sections is extracted；
Extract the keyword of title；
According to the meaning of a word of each word of cloud knowledge library storage, the pass between the keyword of chapters and sections and the keyword of title is generated
Connection degree, and using this degree of association as the chapters and sections degree of bringing out the theme.
The keyword of chapters and sections is the immediate word of central idea of the meaning of a word and chapters and sections in chapters and sections, or the meaning of a word and chapter in dictionary
The immediate word of central idea of section, the general part of speech of keyword here is noun,
The keyword of title is the meaning of a word and the immediate word of title implication, the preferably noun in title in title.
The course of work of paragraph analysis module 3 specifically includes step：
According to the sentence pattern of sentence, and in sentence each word the meaning of a word and part of speech, determine the implication of each word in the sentence；
Extract the word meanings that part of speech in sentence is noun；
All word meanings according to being extracted in chapters and sections determine the thought of center chapters and sections.
The analysis report that artificial intelligence system is generated to english composition, report content includes：Error rate, tricky question rate, writing
Type and composition specification scoring.
The system is intended to make a multimodal platform, can give lessons and comment according to English for academic purpose expert, master of instruction
Sentence the experience of English Writing, extract substantial amounts of pattern, after pattern extraction is finished, system can make one automatically in this mode automatically
Sets of data stores specification.Have after this storage specification, system is entered row mode compiling, pattern quantization and mode iterative and updated,
A set of targetedly data model is intelligently built, and says the storage of this data model into our cloud knowledge base, and is based on
The regular data that regulation engine extracts correlation is stored in cloud rule base in the lump.
When system everything in readiness, student can upload article, into human-computer interaction interface.System is write according to student's
Make, scored.And the content push of corresponding training contents and correlation, and examination focus are searched out from cloud rule base
Analysis, situation change, and using a series of analysis modeling and the technology of excavation, to help student preferably to improve writing.
For our this set system, emphasis is to need to build a set of core schema platform, and defines the number of complete set
According to form.Based on this platform, intelligent classification, reassociate relevant information, by artificial intelligence and deep semantic technology, generates many
The pattern that even 1,100 kinds of kind, i.e., different students generate an optimal path (Path according to different Writing contents
Optimization)。
Each Executor is our a set of standard.Meanwhile, each Executor can be re-used, with the structure structure schemed
Into a series of Node, make it possible to easily to be iterated called, or even deduced.
The system also includes the analysis to all student's use habits.
The system also include for student written suggestion and further learn instruct knowledge base, with reach student improve
The purpose of writing.The module is presented in next chapters and sections.
The data source of the system framework is each English Writing content, error rate, the rate of bringing out the theme for including the student, and
Analyze writing strong point, the weakness of the student, like writing any class article, and the writing specified value difference how much, etc..These
Big data, can provide a series of study plans to improve writing ability for our follow-up help students in real time well.
Iterative learning framework comprises the following steps：
(1) a set of behind process is provided, the process can be based on similar Google's crawler technology, excavate the custom of each user,
And derived according to each user custom in cloud knowledge base.If network goes wrong, we can be accustomed in local cache user, etc.
When network leads to, then uploaded.Ensure that user's learning experience is not obstructed.
(2) cloud knowledge base is the detailed knowledge point of complete set.Each knowledge point is by discrete and there is provided SHA-1 marks
Label ensure uniqueness.Each label belongs to a ZONE.Each ZONE belongs to a PARATITION.Purpose is in order in millisecond
Level can be quickly found out corresponding knowledge point.
(3) cloud regulation engine, is the cloud rule base under a set of big data set.Substantial amounts of small rule are deposited in this storehouse
Then, this small rule such as IF XXX THEN YYY pattern storage.XXX and YYY are simply to judge very much.XXX and YYY
Can be another regular entrance either conclusion, afterwards the system will by calculation, derive, and calculate a conclusion,
This conclusion is exactly the study plan of user.
Being different from earlier technique natural language processing technique can only realize in the range of semantics recognition function, our times at first
The language semantic analytical technology entered is " complicated semanteme depth analysis technology ".It is mainly technically characterized by (fixed with mixed method
Property and quantitative approach) various dimensions modeling is carried out to the literal expression information of specific area, the model built up is referred to as that " multi-dimensional semantic is calculated
Method ".Such advanced algorithm model can carry out the rigorous analysis of depth to literal expression information, while text information each time
Analysis all formed a data precipitation；And as the substantial amounts of precipitation of this data is accumulated, big data and earlier set it is senior
Model algorithm carries out mutually confirmation with correcting, so that the improvement of implementation model and the modification of original character text message are improved.It is special
It is more not valuable to be, the operation of this process can light implementation model algorithm automatic evolution, or iteration upgrade-i.e.：
Realize that multi-dimensional semantic depth analysis artificial intelligence is independently evolved.And during this original character text message by automatic error-correcting with repairing
The part changed has become international academic community very forward position and commercially valuable part.Figure below is that multi-dimensional semantic is analyzed in production
Application in product design：
Word is embedded in (Word Embeddings)
Building deep learning model using neutral net can realize to single vocabulary near synonym, the database such as related term
In relation formation machine learning training pattern, so as to more accurately accomplish multidigit semantic analysis.
For example：W:Words → Rn is a parameterized function, and the word in some language is mapped to high dimension vector by it
(general 200 to 500 dimension).For example so：
" W (" cat ")=(0.2, -0.4,0.7 ...)
W (" mat ")=(0.0,0.6, -0.1 ...) "
After initialization, one random vector of each word correspondence in W.It can learn significant vector and appoint to perform
Business.
One network of training predicts whether 5 tuple (5-gram) (continuous 5 words) ' sets up ' by it.We can be with
At will select the tuple of a pile 5 (such as cat sat on the mat) and then one of word is at will changed into another word (ratio
Such as cat sat song the mat), then the 5 tuples estimation of half can all become absurd and null(NUL).
The model of training can take out the sign vector of each word in 5 tuples by W, input to another and cry R's
Module, module R can attempt to predict that this 5 tuple is ' establishment ' either ' broken '.Then it is desirable that seeing：
" R (W (" cat "), W (" sat "), W (" on "), W (" the "), W (" mat "))=1
R (W (" cat "), W (" sat "), W (" song "), W (" the "), W (" mat "))=0 "
As shown in Fig. 2 direct feel once word embedded space, we can be visualized with t-SNE to it.t-
SNE is a complicated high dimensional data visualization technique.
" map " that this word is constituted is more directly perceived for us.Similar word from closely.Another method is to see pair
For one given word, as shown in figure 3, which other word is nearest from it.We can see that these words are all again
It is much like.
After more complicated relation is encoded out, the database of multidigit semantic analysis can form relation as shown in Figure 4
Phrase.
Claims (5)
1. a kind of Situation of Students ' English Writing artificial intelligence system analyzed based on deep semantic, it is characterised in that including：
Word sort module, for carrying out meaning of a word statistics to word, divides part of speech, and classification is stored in cloud knowledge base, simultaneously
Record each word frequency of use situation；
Sentence pattern sort module, for the progress classification of sentence pattern structure to be stored in cloud knowledge base, while recording the frequency of each sentence pattern
Rate；
Paragraph analysis module, the central idea for analyzing each chapters and sections of generation；
Title analysis module, the central idea for analyzing obtained chapters and sections according to paragraph analysis module, and analysis obtains chapters and sections
With the degree of bringing out the theme of title.
2. a kind of Situation of Students ' English Writing artificial intelligence system analyzed based on deep semantic according to claim 1, it is special
Levy and be, the title analysis module routine specifically includes step：
The central idea of obtained chapters and sections is analyzed according to paragraph analysis module, the keyword of chapters and sections is extracted；
Extract the keyword of title；
According to the meaning of a word of each word of cloud knowledge library storage, the association between the keyword of chapters and sections and the keyword of title is generated
Degree, and using this degree of association as the chapters and sections degree of bringing out the theme.
3. a kind of Situation of Students ' English Writing artificial intelligence system analyzed based on deep semantic according to claim 2, it is special
Levy and be, the keywords of the chapters and sections is the immediate word of central idea of the meaning of a word and chapters and sections in chapters and sections, or the meaning of a word in dictionary
With the immediate word of central idea of chapters and sections,
The keyword of the title is the meaning of a word and the immediate word of title implication in title.
4. a kind of Situation of Students ' English Writing artificial intelligence system analyzed based on deep semantic according to claim 1, it is special
Levy and be, the paragraph analysis module course of work specifically includes step：
According to the sentence pattern of sentence, and in sentence each word the meaning of a word and part of speech, determine the implication of each word in the sentence；
Extract the word meanings that part of speech in sentence is noun；
All word meanings according to being extracted in chapters and sections determine the thought of center chapters and sections.
5. a kind of Situation of Students ' English Writing artificial intelligence system analyzed based on deep semantic according to claim 1, it is special
Levy and be, the analysis report that the artificial intelligence system is generated to english composition, report content includes：Error rate, tricky question rate, write
Make type and composition specification scoring.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710432046.4A CN107256210A (en) | 2017-06-09 | 2017-06-09 | The Situation of Students ' English Writing artificial intelligence system analyzed based on deep semantic |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710432046.4A CN107256210A (en) | 2017-06-09 | 2017-06-09 | The Situation of Students ' English Writing artificial intelligence system analyzed based on deep semantic |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN107256210A true CN107256210A (en) | 2017-10-17 |
Family
ID=60023523
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201710432046.4A Pending CN107256210A (en) | 2017-06-09 | 2017-06-09 | The Situation of Students ' English Writing artificial intelligence system analyzed based on deep semantic |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN107256210A (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110264792A (en) * | 2019-06-17 | 2019-09-20 | 上海元趣信息技术有限公司 | One kind is for pupil's composition intelligent tutoring system |
| CN110647625A (en) * | 2018-06-27 | 2020-01-03 | 上海意仕腾教育科技有限公司 | Training method of English writing evaluation system |
Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20160034483A1 (en) * | 2014-07-31 | 2016-02-04 | Kobo Incorporated | Method and system for discovering related books based on book content |
| CN106126613A (en) * | 2016-06-22 | 2016-11-16 | 苏州大学 | One composition of digressing from the subject determines method and device |
-
2017
- 2017-06-09 CN CN201710432046.4A patent/CN107256210A/en active Pending
Patent Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20160034483A1 (en) * | 2014-07-31 | 2016-02-04 | Kobo Incorporated | Method and system for discovering related books based on book content |
| CN106126613A (en) * | 2016-06-22 | 2016-11-16 | 苏州大学 | One composition of digressing from the subject determines method and device |
Non-Patent Citations (1)
| Title |
|---|
| 巩捷甫: "面向语文作文自动评阅的修辞手法识别系统的设计与实现", 《中国优秀硕士学位论文全文数据库信息科技辑》 * |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110647625A (en) * | 2018-06-27 | 2020-01-03 | 上海意仕腾教育科技有限公司 | Training method of English writing evaluation system |
| CN110264792A (en) * | 2019-06-17 | 2019-09-20 | 上海元趣信息技术有限公司 | One kind is for pupil's composition intelligent tutoring system |
| CN110264792B (en) * | 2019-06-17 | 2021-11-09 | 上海元趣信息技术有限公司 | Intelligent tutoring system for composition of pupils |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| JP6309644B2 (en) | Method, system, and storage medium for realizing smart question answer | |
| CN108874878A (en) | A kind of building system and method for knowledge mapping | |
| CN108717423B (en) | Code segment recommendation method based on deep semantic mining | |
| CN111581474B (en) | Evaluation object extraction method of case-related microblog comments based on multi-head attention system | |
| Dušek et al. | Training a natural language generator from unaligned data | |
| CN109213861A (en) | In conjunction with the tourism evaluation sensibility classification method of At_GRU neural network and sentiment dictionary | |
| Ockeloen et al. | BiographyNet: Managing Provenance at Multiple Levels and from Different Perspectives. | |
| CN109213856A (en) | A kind of method for recognizing semantics and system | |
| CN108665141B (en) | Method for automatically extracting emergency response process model from emergency plan | |
| CN112102813B (en) | Speech recognition test data generation method based on context in user comment | |
| CN106257455A (en) | A kind of Bootstrapping algorithm based on dependence template extraction viewpoint evaluation object | |
| WO2024011813A1 (en) | Text expansion method and apparatus, device, and medium | |
| CN107656921A (en) | A kind of short text dependency analysis method based on deep learning | |
| CN112417161B (en) | Method and storage device for recognizing upper and lower relationships of knowledge graph based on mode expansion and BERT classification | |
| CN107688583A (en) | The method and apparatus for creating the training data for natural language processing device | |
| CN112364132A (en) | Similarity calculation model and system based on dependency syntax and method for building system | |
| CN108491459B (en) | Optimization method for software code abstract automatic generation model | |
| CN115858750A (en) | Power grid technical standard intelligent question-answering method and system based on natural language processing | |
| CN107256210A (en) | The Situation of Students ' English Writing artificial intelligence system analyzed based on deep semantic | |
| Volkova et al. | Lightly supervised learning of procedural dialog systems | |
| CN107818078B (en) | Semantic association and matching method for Chinese natural language dialogue | |
| CN106202033B (en) | A kind of adverbial word Word sense disambiguation method and device based on interdependent constraint and knowledge | |
| CN104750484A (en) | Code abstract generation method based on maximum entropy model | |
| CN113569560A (en) | Automatic scoring method for Chinese bilingual composition | |
| Wu | A Computational Neural Network Model for College English Grammar Correction |
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: 20171017 |