CN108647200A - Talk with intent classifier method and device, equipment and storage medium - Google Patents
Talk with intent classifier method and device, equipment and storage medium Download PDFInfo
- Publication number
- CN108647200A CN108647200A CN201810297475.XA CN201810297475A CN108647200A CN 108647200 A CN108647200 A CN 108647200A CN 201810297475 A CN201810297475 A CN 201810297475A CN 108647200 A CN108647200 A CN 108647200A
- Authority
- CN
- China
- Prior art keywords
- information
- term vector
- classification
- probability
- vector information
- 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
- G06F40/289—Phrasal analysis, e.g. finite state techniques or chunking
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/21—Design or setup of recognition systems or techniques; Extraction of features in feature space; Blind source separation
- G06F18/214—Generating training patterns; Bootstrap methods, e.g. bagging or boosting
Abstract
A kind of dialogue intent classifier method and device of present invention offer, equipment and storage medium, this method include：Import the term vector model that training obtains in advance；The dialog information of input is segmented, participle information is obtained；Information input term vector model will be segmented, term vector information and N member term vector information are obtained；Calculate the sum average value of term vector information and N member term vector information；Dialog information, which is calculated, according to sum average value belongs to each probability for being intended to classification；Classification results are generated according to the highest intention classification of probability and are exported.The present invention utilizes the sum average value of the term vector information and N member term vector information that are generated according to dialog information to calculate the probability that dialog information belongs to each intention classification, the dialogue for dialog information is realized to be intended to carry out automatic and accurate classification, have feasibility to make chat robots carry out accurate intelligent replying, and then realizes saving human cost.
Description
Technical field
This application involves online communication technical fields, and in particular to a kind of dialogue intent classifier method and device, equipment and
Storage medium.
Background technology
In current recruitment, especially during campus recruiting, HR usually requires manually to answer the recruitment largely repeated
Relevant issues result in the need for expending higher human cost, manpower are caused to waste；In another example in scenes such as exhibition, meetings of inviting outside investment,
Equally exist the higher problem of similar human cost.
Invention content
In view of drawbacks described above in the prior art or deficiency, be intended to provide a kind of dialogue to online conversation information be intended into
Dialogue intent classifier method and device, equipment and the storage medium of row automatic and accurate classification.
In a first aspect, the present invention provides a kind of dialogue intent classifier method, including：
Import the term vector model that training obtains in advance；
The dialog information of input is segmented, participle information is obtained；
Information input term vector model will be segmented, term vector information and N member term vector information are obtained；
Calculate the sum average value of term vector information and N member term vector information；
Dialog information, which is calculated, according to sum average value belongs to each probability for being intended to classification；
Classification results are generated according to the highest intention classification of probability and are exported.
Second aspect, the present invention provide a kind of dialogue intent classifier device, including model import unit, participle unit, word
Vectorial generation unit, sum-average arithmetic unit, probability calculation unit and taxon.
Model import unit is configured to import the term vector model that training obtains in advance；
Participle unit is configured to segment the dialog information of input, obtains participle information；
Term vector generation unit is configured to that information input term vector model will be segmented, and obtains term vector information and N member words
Vector information；
Sum-average arithmetic unit is configured to calculate the sum average value of term vector information and N member term vector information；
Probability calculation unit is configured to calculate the probability that dialog information belongs to each intention classification according to sum average value；
Taxon is configured to generate classification results according to the highest intention classification of probability and export.
The third aspect, the present invention also provides a kind of equipment, including one or more processors and memory, wherein memory
Including can be by instruction that the one or more processors execute so that the one or more processors are executed according to of the invention each
The dialogue intent classifier method that embodiment provides.
Fourth aspect, the present invention also provides a kind of storage medium being stored with computer program, which makes meter
Calculation machine executes the dialogue intent classifier method provided according to various embodiments of the present invention.
Dialogue intent classifier method and device, equipment and the storage medium that many embodiments of the present invention provide are utilized according to right
The sum average value for the term vector information and N member term vector information that words information generates calculates dialog information and belongs to each intention classification
Probability realizes the dialogue for dialog information and is intended to carry out automatic and accurate classification, to make chat robots carry out accurate intelligence
It can reply and have feasibility, and then realize saving human cost.
Description of the drawings
By reading a detailed description of non-restrictive embodiments in the light of the attached drawings below, the application's is other
Feature, objects and advantages will become more apparent upon：
Fig. 1 is a kind of flow chart for dialogue intent classifier method that one embodiment of the invention provides.
Fig. 2 is the flow chart of S50 in a kind of embodiment of method shown in Fig. 1.
Fig. 3 is a kind of structural schematic diagram for dialogue intent classifier device that one embodiment of the invention provides.
Fig. 4 is a kind of structural schematic diagram of embodiment of Fig. 3 shown devices.
Fig. 5 is a kind of structural schematic diagram for equipment that one embodiment of the invention provides.
Specific implementation mode
The application is described in further detail with reference to the accompanying drawings and examples.It is understood that this place is retouched
The specific embodiment stated is used only for explaining related invention, rather than the restriction to the invention.It also should be noted that in order to
Convenient for description, is illustrated only in attached drawing and invent relevant part.
It should be noted that in the absence of conflict, the features in the embodiments and the embodiments of the present application can phase
Mutually combination.The application is described in detail below with reference to the accompanying drawings and in conjunction with the embodiments.
Fig. 1 is a kind of flow chart for dialogue intent classifier method that one embodiment of the invention provides.
As shown in Figure 1, in the present embodiment, the present invention provides a kind of dialogue intent classifier method, including：
S10：Import the term vector model that training obtains in advance；
S20：The dialog information of input is segmented, participle information is obtained；
S30：Information input term vector model will be segmented, term vector information and N member term vector information are obtained；
S40：Calculate the sum average value of term vector information and N member term vector information；
S50：Dialog information, which is calculated, according to sum average value belongs to each probability for being intended to classification；
S60：Classification results are generated according to the highest intention classification of probability and are exported.
Specifically, in the present embodiment, the application scenarios of the above method are talked with for campus recruiting on-line consulting, pair of acquisition
It is recruitment dialog information to talk about information, and term vector model concrete configuration is fastText term vector models.
In more embodiments, the above method can also be applied to exhibition on-line consulting dialogue, meeting of inviting outside investment on-line consulting
The different applications scenes such as dialogue, and obtain corresponding dialog information；And term vector model is configured to by this according to actual demand
The other common term vector models in field, as long as term vector information and N member term vector information can be generated according to participle information, in turn
Seek sum average value and each probability for being intended to classification, you can realize identical technique effect.
In step slo, trained fastText term vectors model in advance is imported.Specifically, in the present embodiment, should
Model uses hierarchical softmax algorithms in the training process, and when classification number is K, the size of word insertion vector is d
When, computation complexity can be reduced to O (d log from O (Kd)2(K)), to accelerate training process, shorten what training needs expended
Time.
In step S20, the dialog information of input is segmented.Specifically, in the present embodiment, dialog information passes through
Online input, segments the dialog information of input using jieba segmenter.In more embodiments, language can also be passed through
The different modes such as speech recognition are carried out after sound input to be inputted, and other segmenter commonly used in the art may be used and divided
Word.
For example, for dialog information " salary is how ", by participle, obtain participle information (emolument, treatment, why
Sample)；In another example for dialog information " you are robot ", by participle, obtain participle information (you, is robot,
).
In step s 30, the term vector model participle information input step S10 that step S20 is generated imported, obtains word
Vector information and N member term vector information.Specifically, in the present embodiment, N members term vector information concrete configuration is binary term vector
Information (Bi-Gram).For example, for participle information (emolument, treatment, how), word can be obtained after input word vector model
Vector information (wa1, wa2, wa3) and binary term vector information (wa12, wa23)；For participle information (you, is robot,
), term vector information (w can be obtained after input word vector modelb1, wb2, wb3, wb4) and binary term vector information (wb12,
wb23, wb34).In more embodiments, can also according to actual demand by N member term vector information configurations be ternary term vector information
(Tri-Gram) the term vector information of different first numbers such as.
In step s 40, the sum average value of term vector information and N member term vector information is calculated.For example, for term vector
Information (wa1, wa2, wa3) and binary term vector information (wa12, wa23), the first sum average value can be calculated：
ha=(wa1+wa2+wa3+wa12+wa23)/5；
In another example for term vector information (wb1, wb2, wb3, wb4) and binary term vector information (wb12, wb23, wb34),
The second sum average value can be calculated：
hb=(wb1+wb2+wb3+wb4+wb12+wb23+wb34)/7；
In more embodiments, can also according to actual classifying quality to the calculation of above-mentioned sum-average arithmetic into Mobile state
Adjustment, for example, the term vector information weight, etc. different with N member term vector information configurations.
In step s 50, dialog information is calculated according to the calculated sum average values of step S40 and belongs to each intention classification
Probability.Specifically, in the present embodiment, step S50 calculates probability using method shown in Fig. 2.
Fig. 2 is the flow chart of S50 in a kind of embodiment of method shown in Fig. 1.As shown in Fig. 2, in the present embodiment, step
Suddenly S50 includes：
S501：Output vector information is calculated according to sum average value；
S502：Dialog information, which is calculated, according to output vector information and softmax functions belongs to each probability for being intended to classification.
Specifically, output vector information is calculated using sigmoid functions in step S501, for example, flat according to the first summation
Mean value haCalculate the first output vector information za：
za=sigmoid (W0ha)；
Wherein, W0For the weight of hidden layer to output layer；
In another example according to the second sum average value hbCalculate the second output vector information zb：
zb=sigmoid (W0hb)；
In more embodiments, step S501 can also using other output functions commonly used in the art come calculate output to
Measure information.
In step S502, according to the calculated output vector information of step S501, dialogue letter is calculated using softmax functions
Breath belongs to each probability for being intended to classification.For example, according to the first output vector information zaCalculating dialog information, " how is salary
The probability that sample " belongs to chat classification is 0.08, and the probability for belonging to recruitment question and answer classification is 0.92；In another example according to the second output
Vector information zbIt is 0.95 to calculate dialog information " you are robot " to belong to the probability of chat classification, belongs to recruitment question and answer class
Other probability is 0.05.In more embodiments, other classification functions commonly used in the art can also be used to carry out probability calculation.
In step S60, classification results are generated according to the highest intention classification of probability and are exported.For example, believing for dialogue
It ceases " how is salary ", generate classification results " recruitment question and answer " and exports；It is raw for dialog information " you are robot "
Constituent class result " chat " simultaneously exports.
Above-described embodiment utilizes the sum average value of the term vector information and N member term vector information that are generated according to dialog information
It calculates dialog information and belongs to each probability for being intended to classification, realize the dialogue for dialog information and be intended to carry out automatic and accurate point
Class has feasibility to make chat robots carry out accurate intelligent replying, and then realizes saving human cost.
Fig. 3 is a kind of structural schematic diagram for dialogue intent classifier device that one embodiment of the invention provides.Dress shown in Fig. 3
It sets to correspond to and executes method shown in FIG. 1.
As shown in figure 3, in the present embodiment, the present invention provides a kind of dialogue intent classifier device, including model imports list
Member 10, participle unit 20, term vector generation unit 30, sum-average arithmetic unit 40, probability calculation unit 50 and taxon 60.
Model import unit 10 is configured to import the term vector model that training obtains in advance；
Participle unit 20 is configured to segment the dialog information of input, obtains participle information；
Term vector generation unit 30 is configured to that information input term vector model will be segmented, and obtains term vector information and N members
Term vector information；
Sum-average arithmetic unit 40 is configured to calculate the sum average value of term vector information and N member term vector information；
Probability calculation unit 50 is configured to calculate the probability that dialog information belongs to each intention classification according to sum average value；
Taxon 60 is configured to generate classification results according to the highest intention classification of probability and export.
The dialogue intent classifier principle of Fig. 3 shown devices can refer to method shown in FIG. 1, and details are not described herein again.
Fig. 4 is a kind of structural schematic diagram of embodiment of Fig. 3 shown devices.Device shown in Fig. 4 can correspond to execution Fig. 2
Shown in method.
As shown in figure 4, in a preferred embodiment, probability calculation unit 50 includes the first operation subelement 501 and second
Operation subelement 502.
First operation subelement 501 is configured to calculate output vector information according to sum average value；
Second operation subelement 502 is configured to calculate dialog information category according to output vector information and softmax functions
In each probability for being intended to classification.
The principle of classification of Fig. 4 shown devices can refer to method shown in Fig. 2, and details are not described herein again.
Fig. 5 is a kind of structural schematic diagram for equipment that one embodiment of the invention provides.
As shown in figure 5, as on the other hand, present invention also provides a kind of equipment 500, including one or more centres
Unit (CPU) 501 is managed, can be added according to the program being stored in read-only memory (ROM) 502 or from storage section 508
The program that is downloaded in random access storage device (RAM) 503 and execute various actions appropriate and processing.In RAM503, also deposit
It contains equipment 500 and operates required various programs and data.CPU501, ROM502 and RAM503 pass through the phase each other of bus 504
Even.Input/output (I/O) interface 505 is also connected to bus 504.
It is connected to I/O interfaces 505 with lower component：Importation 506 including keyboard, mouse etc.；It is penetrated including such as cathode
The output par, c 507 of spool (CRT), liquid crystal display (LCD) etc. and loud speaker etc.；Storage section 508 including hard disk etc.；
And the communications portion 509 of the network interface card including LAN card, modem etc..Communications portion 509 via such as because
The network of spy's net executes communication process.Driver 510 is also according to needing to be connected to I/O interfaces 505.Detachable media 511, such as
Disk, CD, magneto-optic disk, semiconductor memory etc. are mounted on driver 510, as needed in order to be read from thereon
Computer program be mounted into storage section 508 as needed.
Particularly, in accordance with an embodiment of the present disclosure, the dialogue intent classifier method of any of the above-described embodiment description can be by
It is embodied as computer software programs.For example, embodiment of the disclosure includes a kind of computer program product comprising visibly wrap
Containing computer program on a machine-readable medium, the computer program includes the journey for executing dialogue intent classifier method
Sequence code.In such embodiments, which can be downloaded and installed by communications portion 509 from network,
And/or it is mounted from detachable media 511.
As another aspect, present invention also provides a kind of computer readable storage medium, the computer-readable storage mediums
Matter can be computer readable storage medium included in the device of above-described embodiment；Can also be individualism, it is unassembled
Enter the computer readable storage medium in equipment.There are one computer-readable recording medium storages or more than one program, should
Program is used for executing the dialogue intent classifier method for being described in the application by one or more than one processor.
Flow chart in attached drawing and block diagram, it is illustrated that according to the system of various embodiments of the invention, method and computer journey
The architecture, function and operation in the cards of sequence product.In this regard, each box in flowchart or block diagram can generation
A part for a part for one module, program segment, or code of table, the module, program segment, or code includes one or more uses
The executable instruction of the logic function as defined in realization.It should also be noted that in some implementations as replacements, being marked in box
The function of note can also occur in a different order than that indicated in the drawings.For example, two boxes succeedingly indicated are actually
It can be basically executed in parallel, they can also be executed in the opposite order sometimes, this is depending on involved function.Also it wants
It is noted that the combination of each box in block diagram and or flow chart and the box in block diagram and or flow chart, Ke Yitong
The dedicated hardware based system of functions or operations as defined in executing is crossed to realize, or specialized hardware and calculating can be passed through
The combination of machine instruction is realized.
Being described in unit or module involved in the embodiment of the present application can be realized by way of software, can also
It is realized by way of hardware.Described unit or module can also be arranged in the processor, for example, each unit can
Can also be the hardware device being separately configured with the software program being provided in computer or intelligent movable equipment.Wherein, this
The title of a little units or module does not constitute the restriction to the unit or module itself under certain conditions.
Above description is only the preferred embodiment of the application and the explanation to institute's application technology principle.People in the art
Member should be appreciated that invention scope involved in the application, however it is not limited to technology made of the specific combination of above-mentioned technical characteristic
Scheme, while should also cover in the case where not departing from the application design, appointed by above-mentioned technical characteristic or its equivalent feature
Other technical solutions of meaning combination and formation.Such as features described above has similar functions with (but not limited to) disclosed herein
Technical characteristic replaced mutually and the technical solution that is formed.
Claims (12)
1. a kind of dialogue intent classifier method, which is characterized in that including：
Import the term vector model that training obtains in advance；
The dialog information of input is segmented, participle information is obtained；
By term vector model described in the participle information input, term vector information and N member term vector information are obtained；
Calculate the sum average value of the term vector information and the N members term vector information；
The dialog information, which is calculated, according to the sum average value belongs to each probability for being intended to classification；
Classification results are generated according to the highest intention classification of probability and are exported.
2. according to the method described in claim 1, it is characterized in that, described calculate the dialogue letter according to the sum average value
Breath belongs to each probability for being intended to classification：
Output vector information is calculated according to the sum average value；
The dialog information, which is calculated, according to the output vector information and softmax functions belongs to each probability for being intended to classification.
3. according to the method described in claim 1, it is characterized in that, the term vector model is fastText term vector models.
4. according to the method described in claim 3, it is characterized in that, the fastText term vectors model passes through
The training of hierarchical softmax algorithms obtains.
5. according to the method described in claim 1, it is characterized in that, the participle is carried out by jieba segmenter.
6. a kind of dialogue intent classifier device, which is characterized in that including：
Model import unit is configured to import the term vector model that training obtains in advance；
Participle unit is configured to segment the dialog information of input, obtains participle information；
Term vector generation unit is configured to, by term vector model described in the participle information input, obtain term vector information and N
First term vector information；
Sum-average arithmetic unit is configured to calculate the sum average value of the term vector information and the N members term vector information；
Probability calculation unit, be configured to according to the sum average value calculate the dialog information belong to it is each be intended to classification it is general
Rate；
Taxon is configured to generate classification results according to the highest intention classification of probability and export.
7. device according to claim 6, which is characterized in that the probability calculation unit includes：
First operation subelement is configured to calculate output vector information according to the sum average value；
Second operation subelement is configured to calculate the dialog information according to the output vector information and softmax functions
Belong to each probability for being intended to classification.
8. device according to claim 6, which is characterized in that the term vector model is fastText term vector models.
9. device according to claim 8, which is characterized in that the fastText term vectors model passes through
The training of hierarchical softmax algorithms obtains.
10. device according to claim 6, which is characterized in that the participle unit is configured to jieba segmenter
Carry out the participle.
11. a kind of equipment, which is characterized in that the equipment includes：
One or more processors；
Memory, for storing one or more programs,
When one or more of programs are executed by one or more of processors so that one or more of processors
Execute the method as described in any one of claim 1-5.
12. a kind of storage medium being stored with computer program, which is characterized in that realized when the program is executed by processor as weighed
Profit requires the method described in any one of 1-5.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810297475.XA CN108647200A (en) | 2018-04-04 | 2018-04-04 | Talk with intent classifier method and device, equipment and storage medium |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810297475.XA CN108647200A (en) | 2018-04-04 | 2018-04-04 | Talk with intent classifier method and device, equipment and storage medium |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN108647200A true CN108647200A (en) | 2018-10-12 |
Family
ID=63745464
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201810297475.XA Pending CN108647200A (en) | 2018-04-04 | 2018-04-04 | Talk with intent classifier method and device, equipment and storage medium |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN108647200A (en) |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109615458A (en) * | 2018-11-02 | 2019-04-12 | 深圳壹账通智能科技有限公司 | Client management method, device, terminal device and computer readable storage medium |
| CN110196930A (en) * | 2019-05-22 | 2019-09-03 | 山东大学 | A kind of multi-modal customer service automatic reply method and system |
| CN111833871A (en) * | 2020-07-07 | 2020-10-27 | 信雅达系统工程股份有限公司 | Intelligent outbound system based on intention recognition and method thereof |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20170125013A1 (en) * | 2015-10-29 | 2017-05-04 | Le Holdings (Beijing) Co., Ltd. | Language model training method and device |
| CN107193865A (en) * | 2017-04-06 | 2017-09-22 | 上海奔影网络科技有限公司 | Natural language is intended to understanding method and device in man-machine interaction |
| CN107491541A (en) * | 2017-08-24 | 2017-12-19 | 北京丁牛科技有限公司 | File classification method and device |
-
2018
- 2018-04-04 CN CN201810297475.XA patent/CN108647200A/en active Pending
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20170125013A1 (en) * | 2015-10-29 | 2017-05-04 | Le Holdings (Beijing) Co., Ltd. | Language model training method and device |
| CN107193865A (en) * | 2017-04-06 | 2017-09-22 | 上海奔影网络科技有限公司 | Natural language is intended to understanding method and device in man-machine interaction |
| CN107491541A (en) * | 2017-08-24 | 2017-12-19 | 北京丁牛科技有限公司 | File classification method and device |
Non-Patent Citations (1)
| Title |
|---|
| 范涛: "Facebook：FastText 理解和在query意图识别的应用", 《HTTPS://BLOG.CSDN.NET/HERO_FANTAO/ARTICLE/DETAILS/69487744》 * |
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109615458A (en) * | 2018-11-02 | 2019-04-12 | 深圳壹账通智能科技有限公司 | Client management method, device, terminal device and computer readable storage medium |
| CN110196930A (en) * | 2019-05-22 | 2019-09-03 | 山东大学 | A kind of multi-modal customer service automatic reply method and system |
| CN110196930B (en) * | 2019-05-22 | 2021-08-24 | 山东大学 | Multi-mode customer service automatic reply method and system |
| CN111833871A (en) * | 2020-07-07 | 2020-10-27 | 信雅达系统工程股份有限公司 | Intelligent outbound system based on intention recognition and method thereof |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Nylund et al. | Collision density: driving growth in urban entrepreneurial ecosystems | |
| CN108647239A (en) | Talk with intension recognizing method and device, equipment and storage medium | |
| CN111353033B (en) | Method and system for training text similarity model | |
| Vo et al. | Sentiment analysis of news for effective cryptocurrency price prediction | |
| Xing et al. | Discovering Bayesian market views for intelligent asset allocation | |
| CN108647200A (en) | Talk with intent classifier method and device, equipment and storage medium | |
| CN111783873B (en) | User portrait method and device based on increment naive Bayes model | |
| WO2020258994A1 (en) | Node data prediction method and device | |
| CN109710766B (en) | Complaint tendency analysis early warning method and device for work order data | |
| CN116188061B (en) | Commodity sales predicting method and device, electronic equipment and storage medium | |
| CN110955770A (en) | Intelligent dialogue system | |
| CN110046303A (en) | A kind of information recommendation method and device realized based on demand Matching Platform | |
| CN113159355A (en) | Data prediction method, data prediction device, logistics cargo quantity prediction method, medium and equipment | |
| Saravanan et al. | Forecasting Economy using Machine Learning Algorithm | |
| Windiatmoko et al. | Developing FB chatbot based on deep learning using RASA framework for university enquiries | |
| Usmani et al. | Predicting market performance with hybrid model | |
| CN112925911A (en) | Complaint classification method based on multi-modal data and related equipment thereof | |
| Okhunov et al. | Tools to support the Development and Promotion of Innovative Projects | |
| CN111353728A (en) | Risk analysis method and system | |
| CN116757835A (en) | Method and device for monitoring transaction risk in credit card customer credit | |
| CN112348590A (en) | Method and device for determining value of article, electronic equipment and storage medium | |
| Zhang et al. | Improved procedures for training primal wasserstein gans | |
| CN109961801A (en) | Intelligent Service evaluation method, computer readable storage medium and terminal device | |
| Malhotra et al. | Bitcoin Price Prediction Using Machine Learning and Deep Learning Algorithms | |
| US20200286104A1 (en) | Platform for In-Memory Analysis of Network Data Applied to Profitability Modeling with Current Market Information |
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: 20181012 |