CN106503917A - The method for building up and reponse system of product appearance and awareness character relational model - Google Patents
The method for building up and reponse system of product appearance and awareness character relational model Download PDFInfo
- Publication number
- CN106503917A CN106503917A CN201610971240.5A CN201610971240A CN106503917A CN 106503917 A CN106503917 A CN 106503917A CN 201610971240 A CN201610971240 A CN 201610971240A CN 106503917 A CN106503917 A CN 106503917A
- Authority
- CN
- China
- Prior art keywords
- product
- awareness character
- appearance
- awareness
- character
- 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
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q10/00—Administration; Management
- G06Q10/06—Resources, workflows, human or project management; Enterprise or organisation planning; Enterprise or organisation modelling
- G06Q10/063—Operations research, analysis or management
- G06Q10/0631—Resource planning, allocation, distributing or scheduling for enterprises or organisations
- G06Q10/06315—Needs-based resource requirements planning or analysis
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q30/00—Commerce
- G06Q30/02—Marketing; Price estimation or determination; Fundraising
- G06Q30/0201—Market modelling; Market analysis; Collecting market data
- G06Q30/0203—Market surveys; Market polls
Landscapes
- Business, Economics & Management (AREA)
- Engineering & Computer Science (AREA)
- Human Resources & Organizations (AREA)
- Strategic Management (AREA)
- Entrepreneurship & Innovation (AREA)
- Development Economics (AREA)
- Economics (AREA)
- Finance (AREA)
- Accounting & Taxation (AREA)
- General Physics & Mathematics (AREA)
- Marketing (AREA)
- Physics & Mathematics (AREA)
- General Business, Economics & Management (AREA)
- Game Theory and Decision Science (AREA)
- Theoretical Computer Science (AREA)
- Data Mining & Analysis (AREA)
- Educational Administration (AREA)
- Operations Research (AREA)
- Quality & Reliability (AREA)
- Tourism & Hospitality (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The invention discloses the method for building up of product appearance and awareness character relational model, comprises the following steps：(1) product appearance form is deconstructed, obtains the property value that each attribute after product appearance is decomposed is different from, digital coding is carried out to attribute and property value；(2) sample data of the awareness character of sample product outward appearance is obtained, awareness character is sorted out and is processed, extract typical awareness character and each score of the sample product outward appearance on typical awareness character of product appearance；(3) the typical awareness character data of the attribute of each sample product outward appearance and the digital coding combination product outward appearance of property value are set up the relational model of product appearance and compound awareness character；The invention also discloses including the reponse system of the product of the relational model；The present invention can just give the guidance of designer's design direction at the design initial stage, and can provide corresponding design inspiration according to product design demand, improve the design efficiency of designer.
Description
Technical field
The present invention relates to product design technology field, more particularly to the foundation side of product appearance and awareness character relational model
Method and reponse system.
Background technology
Research shows that different crowd is had differences to the perceptual recognition of product, in the case of without survey of demands, if
Often there is certain deviation to the understanding of product in meter teacher and user.Additionally, user is fuzzy, compound to the perceptual recognition of product
's.Therefore, the mapping between product-forming and user perceive is often inaccurate, causes product accurately attract target to use
Family.
Researchers often used the conversion and mapping between accomplished in many ways kansei image and elements of product design in the past,
Such as neutral net, fuzzy set, genetic algorithm etc., but which typically defines distribution of the product in single perceptual dimension respectively,
Rather than compound kansei image as space integrally setting up system.During perception prediction, and to product in list
Perceptual value in one perceptual dimension is predicted respectively.The perceptual semantic system that this mode is constructed have ignored user's perception
Globality, the monoid to product in sensible space do not carry out deep excavation and visual presentation yet.Additionally, in the past this
Class research lacks effective data processing system, and the process that system is set up expends the time, and be as a result difficult to convincing, so I
Need a system that can process insufficient, fuzzy initial data, reducing modeling load has system simultaneously again
Effect and accurate.
It is seen that, user is not only related to product-forming to the perceptual knowledge of product, also with product colour, material etc. closely
Related.Therefore, when product perception semantic system is set up, the modelling element of product will not only be paid close attention to, in addition it is also necessary to pay close attention to product color
Characteristic in color, material.Therefore, how the sensible space formed by the different dimensions such as product-forming, color, material to be carried
Conclusion is taken, setting up compound awareness character system becomes particularly significant.
User feeling demand is combined with product design, only product design does not provide data support, additionally it is possible to help produce
Product are just made prediction to the possible emotional experience of user in the design phase and are assessed, and increase successfully several after product enters into market
Rate, reduces the operating cost of product, improves the efficiency of design work.
Content of the invention
The invention provides the method for building up of product appearance and awareness character relational model, sets up product appearance and user's heart
Accurate mapping between reason, improves the design efficiency of designer and the emotion of design.
Product appearance and the method for building up of awareness character relational model, comprise the following steps：
(1) product appearance form is deconstructed, is obtained the property value that each attribute after product appearance is decomposed is different from,
Digital coding is carried out to attribute and property value；And obtain product appearance attribute and disassemble matrix；
(2) sample data of the awareness character of sample product outward appearance is obtained, awareness character is sorted out and is processed, extracted
The typical awareness character of product appearance and each score of the sample product outward appearance on typical awareness character；The score be by
The quantization of the tendency on the typical awareness character.
(3) the sample product outward appearance of step (2) is deconstructed according to step (1), is obtained each sample product outward appearance
Attribute and the digital coding of property value, the typical awareness character of the product appearance obtained in conjunction with step (2) and each sample are produced
Score of the product outward appearance on typical awareness character sets up the relational model of product appearance and compound awareness character.
Step (1) detailed process is as follows：
1-1, acquisition product appearance combination.Collection specific products outward appearance picture, collection source include network selling platform,
Product official website, product magazine, outdoor products advertisement etc., obtain product appearance data.
1-2, product is sorted out according to product appearance data, and the allusion quotation of various product is chosen using clustering method
Type outward appearance, obtains typical products outward appearance.
1-3, finally destructing appearance attribute.The method for taking form in morphological analysis to break up, product appearance form is entered
Row destructing, obtains the property value that each attribute after product appearance is decomposed is different from.It is simultaneously introduced the spies such as product color, material
Property, extract its property value in this dvielement.Destructing attribute and property value are carried out digital coding, product appearance attribute is obtained
Disassemble matrix.
In order to improve the accuracy of model, while simplified model, it is preferred that in step (2), awareness character is sorted out
With comprising the following steps that for processing：
2-1, the similarity degree according to sample data, sort out to product awareness character, and preliminary extraction is primary perceptual special
Levy；The gatherer process of sample data is：The awareness character of product is selected in collection, and collection source includes user's interview, network selling
Platform, product official website, product magazine, outdoor products advertisement etc..
2-2, to the compound semantic feature matrix of the primary awareness character construction of the sample product outward appearance of step 2-1, using language
Adopted difference scale assesses tendency of the product appearance in different awareness character dimensions, obtains product obtaining on primary awareness character
Point；
The score of 2-3, the sample product outward appearance obtained by step 2-2 on primary awareness character carries out factorial analysis, carries
Take typical awareness character and each score of the sample product outward appearance on typical awareness character.
Preferably, in step (3), opening relationships model adopts Multielement statistical analysis method, support vector regression method.
Opening relationships model can adopt multiple destructing methods, such as classical parametric statistics or neutral net etc., but this
Bright when setting up preferably using the method for support vector regression.Its main reason is that support vector regression small sample, non-linear,
There is advantage in the pattern-recognition of high dimension data, while the problem of local minimum can be overcome, generalization ability is preferable.It is preferred that
, in step (3), opening relationships model adopts support vector regression method, comprises the following steps that：
3-1, the typical awareness character of the sample product outward appearance obtained according to step (2) and each sample product outward appearance exist
Score on typical awareness character, sets up awareness character matrix；
3-2, sample product outward appearance is randomly divided into test set and training set；
3-3, by support vector regression, using training set in sample product outward appearance digital coding and its corresponding allusion quotation
Type awareness character builds composite model；
3-4, the digital coding using the sample product outward appearance in test set and its corresponding typical awareness character checking step
The accuracy of the composite model that rapid 3-3 is obtained, obtains the relational model of final product appearance and compound awareness character.
In order to improve the accuracy of model, in step (1), the product appearance form at least includes：Structure composition, color
And material.
Present invention also offers product appearance and awareness character reponse system, including above-mentioned product appearance and compound perception
The relational model of feature.
In order to improve the practicality of the present invention, it is preferred that product appearance and awareness character reponse system also include：
Awareness character feedback unit, for importing product appearance and compound perception spy by the attribute of product appearance and property value
The relational model that levies, exports the typical awareness character score of the product appearance；
Product appearance attribute feedback unit, for typical awareness character to be imported the pass of product appearance and compound awareness character
It is model, the attribute and its property value of output products outward appearance.
In order to further improve the practicality of the present invention, product appearance and awareness character reponse system also include：
Product figure deconstruction unit, by attribute and property value that product graphics decomposition is product appearance.
The method that the decomposition method or software of product figure destructing module can be broken up using form in morphological analysis,
The morphological feature of product is quantified, the various pieces of composition product appearance is concluded and numbered, morphological feature set
X={ X1, X2 ..., Xi } can be expressed as.
Beneficial effects of the present invention：
The method of the present invention set up product appearance and awareness character relational model and with the anti-of the relational model
Feedback system, connects corresponding with user psychology demand for product appearance, by analyzing to product appearance, just gives at the design initial stage and sets
The guidance of meter teacher's design direction, and corresponding design inspiration can be provided according to product design demand, improve the design of designer
Efficiency.
The method of the present invention set up product appearance and awareness character relational model and with the anti-of the relational model
Feedback system, only product design does not provide data support, additionally it is possible to help product in the design phase just to the possible emotion of user
Experience is made prediction and is assessed, and increases successful probability after product enters into market, reduces the operating cost of product, improves design work
The efficiency of work.
Description of the drawings
Fig. 1 is the wire frame flow chart of the method for building up of the product appearance of the present invention and awareness character relational model.
Specific embodiment
As shown in figure 1, the method for building up of the product appearance of the present embodiment and awareness character relational model, including following step
Suddenly：
(1) product appearance form is deconstructed, is obtained the property value that each attribute after product appearance is decomposed is different from,
Digital coding is carried out to attribute and property value；Obtain product appearance attribute and disassemble matrix；Detailed process is as follows：
From the electric kettle picture that each website, magazine etc. collect 324 different manufacturers and pattern.Sieved by expert
Choosing, the profile excessively eccentric or form hot-water bottle that excessively complexity is difficult to conclude is removed, while simplifying style excessively thunder
Same picture, obtains 54 kinds of electric kettle patterns.
Morphological analysis are carried out to the outward appearance of all products respectively, each attribute after product appearance is decomposed is obtained corresponding to which
Property value.The characteristics such as product color, material are simultaneously introduced, its property value in this dvielement is extracted.Will destructing attribute with
Property value carries out digital coding, obtains product appearance attribute and disassembles matrix.As, as a example by electric kettle, disassembled matrix such as following table：
(2) sample data of the awareness character of sample product outward appearance is obtained, awareness character is sorted out and is processed, extracted
The typical awareness character of product appearance and each score of the sample product outward appearance on typical awareness character；Concrete steps are such as
Under：
Similar to the acquisition of product appearance combinations of attributes in step (1), by network selling platform, product official website, product
The adjective with regard to electric kettle form is collected in magazine, outdoor products advertisement etc. as far as possible, after screening and cluster, tentatively obtains 12
The primary experiential word pair of group：Health-harmful, unique-common, streamline-geometry, pure and fresh-dull, letter
About-complicated, modern-classic, hale and hearty-soft, make widely known-containing, easy-to-use-difficult, professional
- sparetime, innervation-static state, exquisiteness-coarse.
Using semantic difference scale collecting sample in 12 primary semantic scores of the experiential word on.Carry out factorial analysis
(12 words being integrated according to score), summarizes extraction common factor and obtains typical experiential word pair：Low-grade-high-grade, stiff
- active, streamline-geometry.
(3) the sample product outward appearance of step (2) is deconstructed according to step (1), is obtained each sample product outward appearance
Attribute and the digital coding of property value, the typical awareness character of the product appearance obtained in conjunction with step (2) and each sample are produced
Score of the product outward appearance on typical awareness character sets up the relational model of product appearance and compound awareness character, and concrete steps are such as
Under：
To 54 electric kettle sample images of initial screening in " low-grade-high-grade ", " stiff-active ", " stream
Line-geometry " 3 experiential words are to above carrying out semantic difference assessment.39 samples are selected as training set, 15 sample conducts
Test set., used as independent variable, three experiential words are to carrying out machine learning as dependent variable for property value using sample fractionation.
SVR model parameters are adjusted in test process, ideal model is obtained, and model is surveyed with 15 samples of test set
Examination.As shown in the table, the root-mean-square error (RMSE) of test set is 0.004,0.008,0.003, extremely close to 0；Square with
Coefficient (R2) is 0.946,0.931,0.936, is in close proximity to 1.The model evaluation result explanation test shows that system is created as
Work(, shows the better performances of the SVMs semantic knowledge model set up by experimental basis perception semantic data, can use
To in the middle of the semantic classification prediction of the perception of actual product.
In the embodiment of the present application, unit can be integrated in one, it is also possible to be deployed separately, and system can be used for specific
On the awareness character system construction of product, it is also possible to create the awareness character system of multi-product by the upgrading of feature database.Each
Unit can be distributed in the system of embodiment according to embodiment description, it is also possible to carried out respective change and be disposed other than the application
In one or more systems of embodiment.
Those skilled in the art should be understood that the embodiment of the present application can be provided becomes system, method or corresponding program
Product.
The each unit of the embodiment of the present application or each flow process can be realized with general computing device, additionally, the application reality
Apply each unit of example alternatively, they can be realized with the executable program code of computing device, such that it is able to deposit them
Storage is executed by computing device in the storage device, or they is fabricated to each integrated circuit modules respectively, or by it
In multiple modules or step be fabricated to single integrated circuit module to realize.Therefore, the embodiment of the present application is not restricted to appoint
What specific hardware and software is combined.
Disclosed above is only specific embodiment of the invention, but the present invention is not limited to this, the technology of this area
Personnel the present invention can be carried out various change and modification without departing from the spirit and scope of the present invention.Obviously these changes and change
Type all should belong in the protection domain protection of application claims.
Claims (7)
1. the method for building up of product appearance and awareness character relational model, it is characterised in that comprise the following steps：
(1) product appearance form is deconstructed, is obtained the property value that each attribute after product appearance is decomposed is different from, to category
Property with property value carry out digital coding；
(2) sample data of the awareness character of sample product outward appearance is obtained, awareness character is sorted out and is processed, extract product
The typical awareness character of outward appearance and each score of the sample product outward appearance on typical awareness character；
(3) the sample product outward appearance of step (2) is deconstructed according to step (1), is obtained the attribute of each sample product outward appearance
With the digital coding of property value, outside the typical awareness character and each sample product of the product appearance obtained in conjunction with step (2)
See the relational model that the score on typical awareness character sets up product appearance and compound awareness character.
2. the method for building up of product appearance as claimed in claim 1 and awareness character relational model, it is characterised in that step
(2), in, that awareness character sorted out and is processed comprises the following steps that：
2-1, the similarity degree according to sample data, sort out to product awareness character, extract primary awareness character；
2-2, to the compound semantic feature matrix of the primary awareness character construction of the sample product outward appearance of step 2-1, poor using semanteme
Tendency of the product appearance in different awareness character dimensions assessed by different scale, obtains score of the product on primary awareness character；
The score of 2-3, the sample product outward appearance obtained by step 2-2 on primary awareness character carries out factorial analysis, extracts allusion quotation
Type awareness character and each score of the sample product outward appearance on typical awareness character.
3. the method for building up of product appearance as claimed in claim 1 and awareness character relational model, it is characterised in that step
(3), in, opening relationships model adopts multi-variate statistical analysis, support vector regression method.
4. the method for building up of product appearance as claimed in claim 1 and awareness character relational model, it is characterised in that step
(3), in, opening relationships model adopts support vector regression method, comprises the following steps that：
3-1, the typical awareness character of the sample product outward appearance obtained according to step (2) and each sample product outward appearance are in typical case
Score on awareness character, sets up awareness character matrix；
3-2, sample product outward appearance is randomly divided into test set and training set；
3-3, by support vector regression, using training set in sample product outward appearance digital coding and its corresponding typical case's sense
Property feature construction composite model；
3-4, the digital coding using the sample product outward appearance in test set and its corresponding typical awareness character verification step 3-3
The accuracy of the composite model for obtaining, obtains the relational model of final product appearance and compound awareness character.
5. the method for building up of product appearance as claimed in claim 1 and awareness character relational model, it is characterised in that step
(1), in, the product appearance form at least includes：Structure composition, color and material.
6. a kind of product appearance and awareness character reponse system, it is characterised in that include that such as the arbitrary right of Claims 1 to 5 will
The product appearance that asks and the relational model of compound awareness character.
7. product appearance as claimed in claim 6 and awareness character reponse system, it is characterised in that also include：
Awareness character feedback unit, for importing product appearance and compound awareness character by the attribute of product appearance and property value
Relational model, exports the typical awareness character score of the product appearance；
Product appearance attribute feedback unit, for typical awareness character to be imported the relation mould of product appearance and compound awareness character
Type, the attribute and its property value of output products outward appearance.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201610971240.5A CN106503917A (en) | 2016-10-31 | 2016-10-31 | The method for building up and reponse system of product appearance and awareness character relational model |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201610971240.5A CN106503917A (en) | 2016-10-31 | 2016-10-31 | The method for building up and reponse system of product appearance and awareness character relational model |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN106503917A true CN106503917A (en) | 2017-03-15 |
Family
ID=58323083
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201610971240.5A Pending CN106503917A (en) | 2016-10-31 | 2016-10-31 | The method for building up and reponse system of product appearance and awareness character relational model |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN106503917A (en) |
Cited By (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107680437A (en) * | 2017-10-18 | 2018-02-09 | 商丘师范学院 | A kind of demo system for Layout Design teaching |
| CN108470306A (en) * | 2018-03-23 | 2018-08-31 | 汕头大学 | A method of predict client to product feature preference based on sales data |
| CN108681591A (en) * | 2018-05-15 | 2018-10-19 | 东华大学 | A kind of color textile fabric is based on Affective Evaluation and product library matching process |
| CN109948278A (en) * | 2019-03-29 | 2019-06-28 | 山东建筑大学 | Building Design aesthetic measure appraisal procedure and system based on width study |
| CN109993556A (en) * | 2017-12-30 | 2019-07-09 | 中国移动通信集团湖北有限公司 | User behavior analysis method, apparatus calculates equipment and storage medium |
| CN110852329A (en) * | 2019-10-21 | 2020-02-28 | 南京航空航天大学 | Method for defining product appearance attribute |
| CN112016608A (en) * | 2020-08-21 | 2020-12-01 | 四川大学 | Garment perceptual intention classification method based on convolutional neural network, classification model and construction method thereof |
| CN112085404A (en) * | 2020-09-17 | 2020-12-15 | 辽宁工程技术大学 | Method for screening perceptual engineering product samples |
-
2016
- 2016-10-31 CN CN201610971240.5A patent/CN106503917A/en active Pending
Cited By (11)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107680437A (en) * | 2017-10-18 | 2018-02-09 | 商丘师范学院 | A kind of demo system for Layout Design teaching |
| CN107680437B (en) * | 2017-10-18 | 2019-12-31 | 商丘师范学院 | Demonstration system for layout design teaching |
| CN109993556A (en) * | 2017-12-30 | 2019-07-09 | 中国移动通信集团湖北有限公司 | User behavior analysis method, apparatus calculates equipment and storage medium |
| CN109993556B (en) * | 2017-12-30 | 2021-06-08 | 中国移动通信集团湖北有限公司 | User behavior analysis method and device, computing equipment and storage medium |
| CN108470306A (en) * | 2018-03-23 | 2018-08-31 | 汕头大学 | A method of predict client to product feature preference based on sales data |
| CN108681591A (en) * | 2018-05-15 | 2018-10-19 | 东华大学 | A kind of color textile fabric is based on Affective Evaluation and product library matching process |
| CN109948278A (en) * | 2019-03-29 | 2019-06-28 | 山东建筑大学 | Building Design aesthetic measure appraisal procedure and system based on width study |
| CN110852329A (en) * | 2019-10-21 | 2020-02-28 | 南京航空航天大学 | Method for defining product appearance attribute |
| CN110852329B (en) * | 2019-10-21 | 2021-06-15 | 南京航空航天大学 | Method for defining product appearance attribute |
| CN112016608A (en) * | 2020-08-21 | 2020-12-01 | 四川大学 | Garment perceptual intention classification method based on convolutional neural network, classification model and construction method thereof |
| CN112085404A (en) * | 2020-09-17 | 2020-12-15 | 辽宁工程技术大学 | Method for screening perceptual engineering product samples |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN106503917A (en) | The method for building up and reponse system of product appearance and awareness character relational model | |
| CN105279495B (en) | A kind of video presentation method summarized based on deep learning and text | |
| CN104809142A (en) | Trademark inquiring system and method | |
| CN108734184B (en) | Method and device for analyzing sensitive image | |
| CN109740154A (en) | A kind of online comment fine granularity sentiment analysis method based on multi-task learning | |
| CN107391547B (en) | Human object data matching method and system | |
| US20080097937A1 (en) | Distributed method for integrating data mining and text categorization techniques | |
| CN107633007A (en) | A kind of comment on commodity data label system and method based on stratification AP clusters | |
| CN104866831B (en) | The face recognition algorithms of characteristic weighing | |
| CN111008333B (en) | Shopping system and method based on user characteristic information and product component information | |
| CN110309303A (en) | A kind of judicial dispute data visualization analysis method based on Weighted T F-IDF | |
| CN109213853A (en) | A kind of Chinese community's question and answer cross-module state search method based on CCA algorithm | |
| CN110113634A (en) | A kind of information interaction method, device, equipment and storage medium | |
| CN105653522A (en) | Non-classified relation recognition method for plant field | |
| CN107577655A (en) | Name acquiring method and apparatus | |
| CN106649334A (en) | Conjunction word set processing method and device | |
| CN110334724A (en) | Remote sensing object natural language description and multiple dimensioned antidote based on LSTM | |
| CN109241379A (en) | A method of across Modal detection network navy | |
| Gu et al. | Data driven webpage color design | |
| CN107896335A (en) | Video detection and ranking method based on big data technology | |
| WO2020007989A1 (en) | Method for co-clustering senders and receivers based on text or image data files | |
| TW201820172A (en) | System, method and non-transitory computer readable storage medium for conversation analysis | |
| CN108154156A (en) | Image Ensemble classifier method and device based on neural topic model | |
| CN109815478A (en) | Medicine entity recognition method and system based on convolutional neural networks | |
| CN109471930B (en) | Emotional board interface design method for user emotion |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | 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 |
Application publication date: 20170315 |
|
| RJ01 | Rejection of invention patent application after publication |