CN110009045A - The recognition methods of internet-of-things terminal and device - Google Patents
The recognition methods of internet-of-things terminal and device Download PDFInfo
- Publication number
- CN110009045A CN110009045A CN201910282297.8A CN201910282297A CN110009045A CN 110009045 A CN110009045 A CN 110009045A CN 201910282297 A CN201910282297 A CN 201910282297A CN 110009045 A CN110009045 A CN 110009045A
- Authority
- CN
- China
- Prior art keywords
- data
- terminal
- internet
- identified
- things
- 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
- 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
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/25—Fusion techniques
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/285—Selection of pattern recognition techniques, e.g. of classifiers in a multi-classifier system
Landscapes
- Engineering & Computer Science (AREA)
- Data Mining & Analysis (AREA)
- Theoretical Computer Science (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Bioinformatics & Computational Biology (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Artificial Intelligence (AREA)
- Evolutionary Biology (AREA)
- Evolutionary Computation (AREA)
- Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Life Sciences & Earth Sciences (AREA)
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
- Image Analysis (AREA)
Abstract
The present invention provides recognition methods and the device of a kind of internet-of-things terminal, belongs to signal processing technology field, can solve the prior art can not identify the deficiency of the type of terminal in Internet of Things.The recognition methods of internet-of-things terminal of the invention, comprising: obtain the identification data of terminal to be identified, wherein identification data include the data that terminal to be identified is interacted with Internet of Things；The recognition result of terminal to be identified is obtained according to identification data, wherein recognition result includes the type label of terminal to be identified.
Description
Technical field
The present invention relates to signal processing technology fields, and in particular, to a kind of recognition methods of internet-of-things terminal and device.
Background technique
With the development of Internet of Things, the epoch of all things on earth interconnection have arrived, and more and more terminals are added in Internet of Things, are
Convenient for the terminal management in Internet of Things, need to identify the terminal in Internet of Things, such as the knowledge of classification belonging to terminal
Not.
However, whether the terminal being only limited in identification Internet of Things to the identification of internet-of-things terminal at present is Internet of Things end
End, does not identify out the type of terminal.
Summary of the invention
The present invention at least partly solves the problems, such as the existing deficiency that can not identify the terminal type in Internet of Things, provides
A kind of recognition methods of the internet-of-things terminal that can accurately and efficiently identify the terminal type in Internet of Things and device.
Solve the recognition methods that technical solution used by present invention problem is a kind of internet-of-things terminal, comprising:
Obtain the identification data of terminal to be identified, wherein the identification data include the terminal to be identified and Internet of Things
Interactive data；
The recognition result of the terminal to be identified is obtained according to the identification data, wherein the recognition result includes institute
State the type label of terminal to be identified.
Optionally, the recognition result that the terminal to be identified is obtained according to the identification data, comprising:
The identification data are inputted into the different classifier of multiple sorting algorithms respectively, obtain being exported by each classifier
Primary recognition result；
By all primary recognition result input results Fusion Models, the recognition result of the internet-of-things terminal is obtained.
Optionally, described to input the identification data in the different classifier of multiple sorting algorithms respectively, the classification
Algorithm includes:
Logistic regression algorithm, decision Tree algorithms, random forests algorithm, in Xgboost algorithm both at least.
Optionally, it is described obtain the recognition result of the terminal to be identified according to the identification data before, further includes:
Multiple classifiers are established based on Ensemble Learning Algorithms and the result fusion is established based on stacking algorithm
Model.
Optionally, described to establish multiple classifiers based on Ensemble Learning Algorithms and institute is established based on stacking algorithm
State result Fusion Model, comprising:
Obtain training data, the training data include in interaction data that sample terminal is interacted with Internet of Things with the sample
The relevant data of this terminal service condition；
The training data is pre-processed；
The training data is divided into multiple training sample sets and test sample collection；
Each corresponding training sample set of the classifier, uses the training sample based on Ensemble Learning Algorithms
It carries out data training and establishes the classifier；
Parameter optimization is carried out to the classifier using the test sample collection based on grid search；
The knot is established using primary recognition result of the classifier to the sample terminal based on stacking algorithm
Fruit Fusion Model.
Optionally, the identification data for obtaining terminal to be identified, comprising:
From the data that the terminal to be identified is interacted with Internet of Things, screening includes and the terminal service condition to be identified
The data of relevant field are as the identification data；
The identification data are pre-processed, wherein the pretreatment includes: data type conversion, Data Mining, category
Property specification, data normalization.
Optionally, the identification data include: the cost data of terminal to be identified, flow using data, position number
According to, internet of things sensors data.
Solve the identification device that technical solution used by present invention problem is a kind of internet-of-things terminal, comprising:
Data capture unit, for obtaining the identification data of terminal to be identified, the identification data include with described wait know
The relevant data of other terminal service condition；
As a result output unit obtains the recognition result of the terminal to be identified according to the identification data, wherein the knowledge
Other result includes the type of the terminal to be identified.
Optionally, the result output unit includes: the different classifier of multiple sorting algorithms and result Fusion Module,
In,
Each classifier exports primary recognition result for inputting the identification data；
The Fusion Module exports institute for inputting the whole primary recognition result of multiple classifier outputs
State the recognition result of terminal to be identified.
Optionally, the identification device further includes training unit, for establishing multiple described points based on Ensemble Learning Algorithms
Class device and the result Fusion Model is established based on stacking algorithm.
The recognition methods of internet-of-things terminal of the present invention and device, the service condition of different types of terminal has been in Internet of Things
Difference, therefore data relevant to the service condition of terminal are also different, it is therefore, relevant according to terminal service condition to be identified
Data are treated knowledge terminal and are identified, can accurately and efficiently identify the type of terminal to be identified.
Detailed description of the invention
Fig. 1 is a kind of flow chart of the recognition methods of internet-of-things terminal of the embodiment of the present invention 1；
Fig. 2 is the flow chart of the recognition methods of another internet-of-things terminal of the embodiment of the present invention 1；
Fig. 3 is the classifier and result Fusion Module of a kind of recognition methods of internet-of-things terminal of the embodiment of the present invention 1
Establishing process figure；
Fig. 4 is the block diagram of the identification device of the internet-of-things terminal of the embodiment of the present invention 2；
Description of symbols therein: 01, data capture unit；02, result output unit；021, the first classifier；
022, the second classifier；023, third classifier；024, the 4th classifier；03, Fusion Module.
Specific embodiment
To make those skilled in the art more fully understand technical solution of the present invention, with reference to the accompanying drawing and it is embodied
Mode further retouches a kind of display unit provided by the present invention, display base plate and its driving method and display device work in detail
It states.
Embodiment 1:
Fig. 1 to 3 is participated in, the present embodiment provides a kind of recognition methods of internet-of-things terminal, this method for example operates in Internet of Things
The identification of type is carried out in net to the terminal in Internet of Things, this approach includes the following steps.
Step 11, the identification data for obtaining terminal to be identified, wherein identification data include that terminal to be identified and Internet of Things are handed over
Mutual data；
Specifically, identification data are relevant to terminal service condition to be identified and are stored in the data in Internet of Things, example
Such as, identification data may come from the database of Internet of Things cloud platform (such as edge cloud platform, core cloud platform), based on Hadoop
Tool for Data Warehouse (hive), distributed file system (Hadoop Distributed File System, abbreviation HDFS)
Equal data warehouses.
Step 12, the recognition result that terminal to be identified is obtained according to identification data, wherein recognition result includes end to be identified
The type label at end.
Specifically, the type label of terminal to be identified here refers to according to the actual situation for the terminal in Internet of Things
It distinguishes, for example, the type label of different definition depending on the application: refrigerator, television set, air-conditioning, mobile phone etc.；According to the electricity used
The type label of different definition: High Level AC Voltage source terminal, low-voltage AC source terminal, high voltage direct current terminal etc. is changed in source.
Wherein, the service condition of the terminal to be identified of different type label is different, therefore generates relevant to service condition
Data are not also identical.By taking the service condition of refrigerator and mobile phone as an example, the position data of refrigerator is almost unchanged, however the position of mobile phone
Data often change；Refrigerator continually can upload temperature data to cloud platform, and mobile phone does not upload temperature data to cloud platform.
Therefore, in above scheme, using data relevant to terminal service condition to be identified as identification data, according to identification
The recognition result accuracy of data (service condition of terminal i.e. to be identified), the terminal to be identified obtained is high.
Optionally, above-mentioned steps 12 include:
Step 12a, identification data are inputted into the different classifier of multiple sorting algorithms respectively, obtain being exported by each classifier
Primary recognition result；
Specifically, each classifier corresponds to a kind of sorting algorithm, and sorting algorithm includes but is not limited to: logistic regression algorithm,
Decision Tree algorithms, random forests algorithm, in Xgboost algorithm both at least.
Step 12b, by all primary recognition result input results Fusion Models, the recognition result of terminal to be identified is obtained.
Specifically, as a result Fusion Model can be using the method for average (such as arithmetic mean, weighted average), ballot method (as relatively
Majority voting method, weighted voting algorithm) multiple primary recognition results are merged, obtain the recognition result of terminal to be identified.
It certainly, is the recognition result for obtaining the higher terminal to be identified of accuracy, as a result Fusion Model can be based on
Multiple primary recognition results are inputted trained result Fusion Model by the model that habit method (such as stacking algorithm) is established
In, then obtain the recognition result of terminal to be identified.
In above scheme, identification terminal is treated respectively using multiple classifiers and is identified, and exports primary recognition result,
Multiple primary recognition results are subjected to fusion and show therefore the final recognition result of terminal to be identified effectively increases to be identified
The recognition correct rate and recognition efficiency of terminal type.
Optionally, before above-mentioned steps 11, further includes:
Step 13 is established multiple classifiers based on Ensemble Learning Algorithms and establishes result fusion mould based on stacking algorithm
Type.
In above scheme, multiple classifiers are established using Ensemble Learning Algorithms, therefore improve the precision of each classifier,
To improve the accuracy of each primary recognition result, and then the end to be identified obtained is merged by these primary recognition results
The accuracy of the recognition result at end is higher.
Specifically, above-mentioned steps 10 include:
Step 10a, training data is obtained, training data includes in the data that sample terminal is interacted with Internet of Things, with sample
The relevant data of terminal service condition；
Step 10b, training data is pre-processed；
Optionally, pretreatment includes: data type conversion, Data Mining, attitude layer, data normalization.
Specifically, data type conversion refer to by the data type conversion of training data be specified data type (such as
With the identifiable data type of classifier).
Data Mining includes according to exploration as a result, handling exceptional value, missing values.
Attitude layer includes carrying out feature selecting according to the correlation of each attribute.
In addition, when the data characteristics for being used to characterize sample terminal service condition in training data is fewer, after pretreatment
Training data carry out Feature Engineering, improve the dimension of training data.
Step 10c, training data is divided into multiple training sample sets and test sample collection；
Preferably, for cross validation, training data is divided into multiple training sample sets using computing engines (spark)
And test sample collection, for example, 70% being divided into for trained data in training data, 30% is divided into the number for test
According to.
Step 10d, the corresponding training sample set of each classifier, based on Ensemble Learning Algorithms, (such as boosting is calculated
Method, bagging algorithm) it is trained using training sample progress data and establishes classifier；
Preferably, in order to improve the precision of separator, prevent over-fitting, the training sample set of each classifier is random
Extraction section data are formed from for trained data.
Step 10e, parameter optimization is carried out to classifier using test sample collection based on grid search (Grid Search)；
Step 10f, result fusion mould is established using recognition result of the classifier to sample terminal based on stacking algorithm
Type.
Optionally, above-mentioned steps 11 include:
Step 11a, from the data that terminal to be identified generates in Internet of Things, screening includes terminal service condition to be identified
The data of relevant field, as identification data；
Optionally, identification data include: the cost data (such as market purchase price) of terminal to be identified, flow using data
(such as SIM card data on flows, including local flow, inter-provincial roaming flow, international roaming flow and Hong Kong, Macao and Taiwan roaming flow etc.), institute
In position data, internet of things sensors data.
Step 11b, identification data are pre-processed, wherein pretreatment includes: data type conversion, Data Mining, category
Property specification, data normalization.
Embodiment 2
The present embodiment provides a kind of identification devices of internet-of-things terminal comprising:
Data capture unit 01, for obtaining the identification data of terminal to be identified, identification data include and terminal to be identified
The relevant data of service condition；
As a result output unit 02 obtains the recognition result of terminal to be identified according to identification data, wherein recognition result includes
The type of terminal to be identified.
Optionally, as a result output unit includes: the different classifier of multiple sorting algorithms and result Fusion Module, wherein
Each classifier exports primary recognition result for inputting identification data；
Specifically, each classifier corresponds to a kind of sorting algorithm, and sorting algorithm includes but is not limited to: logistic regression algorithm,
Decision Tree algorithms, random forests algorithm, in Xgboost algorithm both at least.Referring to fig. 4,021 operation logic of the first classifier
Regression algorithm, 022 operational decisions tree algorithm of the second classifier, the operation random forests algorithm algorithm of third classifier 023, the 4th point
Class device 024 runs Xgboost algorithm.
Fusion Module 03 exports internet-of-things terminal for inputting all primary recognition result of multiple classifier outputs
Recognition result.
Optionally, identification device further includes training unit, for establishing multiple classifiers and base based on Ensemble Learning Algorithms
Result Fusion Model is established in stacking algorithm.
Optionally, training unit can be with operating procedure 13a to 13f, to complete to merge mould to multiple classifiers and result
The data training of type.
Method more than the identification device of the internet-of-things terminal of above scheme is executable, therefore it can be efficiently to end to be identified
The type at end is identified, and the recognition result accuracy exported is high.
It is understood that the principle that embodiment of above is intended to be merely illustrative of the present and the exemplary implementation that uses
Mode, however the present invention is not limited thereto.For those skilled in the art, essence of the invention is not being departed from
In the case where mind and essence, various changes and modifications can be made therein, these variations and modifications are also considered as protection scope of the present invention.
Claims (10)
1. a kind of recognition methods of internet-of-things terminal characterized by comprising
Obtain the identification data of terminal to be identified, wherein the identification data include that the terminal to be identified is interacted with Internet of Things
Data；
Obtain the recognition result of the terminal to be identified according to the identification data, wherein the recognition result include it is described to
The type label of identification terminal.
2. the recognition methods of internet-of-things terminal according to claim 1, which is characterized in that described according to the identification data
Obtain the recognition result of the terminal to be identified, comprising:
The identification data are inputted into the different classifier of multiple sorting algorithms respectively, are obtained by the first of each classifier output
Grade recognition result；
By all primary recognition result input results Fusion Models, the recognition result of the terminal to be identified is obtained.
3. the recognition methods of internet-of-things terminal according to claim 2, which is characterized in that described by the identification data point
It does not input in the different classifier of multiple sorting algorithms, the sorting algorithm includes:
Logistic regression algorithm, decision Tree algorithms, random forests algorithm, in Xgboost algorithm both at least.
4. the recognition methods of internet-of-things terminal according to claim 2, which is characterized in that described according to the identification number
Before the recognition result for obtaining the terminal to be identified, further includes:
Multiple classifiers are established based on Ensemble Learning Algorithms and the result Fusion Model is established based on stacking algorithm.
5. the recognition methods of internet-of-things terminal according to claim 4, which is characterized in that described to be based on Ensemble Learning Algorithms
It establishes multiple classifiers and the result Fusion Model is established based on stacking algorithm, comprising:
Training data is obtained, the training data includes in the data that sample terminal is interacted with Internet of Things, with the sample terminal
The relevant data of service condition；
The training data is pre-processed；
The training data is divided into multiple training sample sets and test sample collection；
Each corresponding training sample set of the classifier, is carried out based on Ensemble Learning Algorithms using the training sample
The classifier is established in data training；
Parameter optimization is carried out to the classifier using the test sample collection based on grid search；
The result is established using primary recognition result of the classifier to the sample terminal based on stacking algorithm to melt
Molding type.
6. the recognition methods of internet-of-things terminal according to claim 1, which is characterized in that the acquisition terminal to be identified
Identify data, comprising:
From the data that the terminal to be identified is interacted with Internet of Things, screening is comprising related to the terminal service condition to be identified
The data of field are as the identification data；
The identification data are pre-processed, wherein the pretreatment includes: data type conversion, Data Mining, attribute rule
About, data normalization.
7. the recognition methods of internet-of-things terminal according to claim 1, which is characterized in that the identification data include: to
The cost data of identification terminal, flow use data, position data, internet of things sensors data.
8. a kind of identification device of internet-of-things terminal characterized by comprising
Data capture unit, for obtaining the identification data of terminal to be identified, the identification data include and the end to be identified
Hold the relevant data of service condition；
As a result output unit obtains the recognition result of the terminal to be identified according to the identification data, wherein the identification knot
Fruit includes the type of the terminal to be identified.
9. the identification device of internet-of-things terminal according to claim 8, which is characterized in that the result output unit packet
It includes: the different classifier of multiple sorting algorithms and result Fusion Module, wherein
Each classifier exports primary recognition result for inputting the identification data；
The Fusion Module exports the object for inputting the whole primary recognition result of multiple classifier outputs
The recognition result of networked terminals.
10. the identification device of internet-of-things terminal according to claim 9, which is characterized in that the identification device further includes
Training unit, for establishing multiple classifiers based on Ensemble Learning Algorithms and establishing the result based on stacking algorithm
Fusion Model.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910282297.8A CN110009045A (en) | 2019-04-09 | 2019-04-09 | The recognition methods of internet-of-things terminal and device |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910282297.8A CN110009045A (en) | 2019-04-09 | 2019-04-09 | The recognition methods of internet-of-things terminal and device |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN110009045A true CN110009045A (en) | 2019-07-12 |
Family
ID=67170589
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201910282297.8A Pending CN110009045A (en) | 2019-04-09 | 2019-04-09 | The recognition methods of internet-of-things terminal and device |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN110009045A (en) |
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110445689A (en) * | 2019-08-15 | 2019-11-12 | 平安科技（深圳）有限公司 | Identify the method, apparatus and computer equipment of internet of things equipment type |
| CN110545278A (en) * | 2019-09-05 | 2019-12-06 | 中国联合网络通信集团有限公司 | industrial internet identification analysis method and system |
| CN110738309A (en) * | 2019-09-27 | 2020-01-31 | 华中科技大学 | DDNN training method and DDNN-based multi-view target identification method and system |
| CN115277063A (en) * | 2022-06-13 | 2022-11-01 | 深圳铸泰科技有限公司 | Terminal identification device under IPV4 and IPV6 hybrid network environment |
Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20100142379A1 (en) * | 2008-12-05 | 2010-06-10 | Electronics And Telecommunications Research Institute | Method for classifying packet on mobile terminal |
| CN105138913A (en) * | 2015-07-24 | 2015-12-09 | 四川大学 | Malware detection method based on multi-view ensemble learning |
| CN107277011A (en) * | 2017-06-16 | 2017-10-20 | 深圳市盛路物联通讯技术有限公司 | The data classification method and device of terminal device |
| CN108038701A (en) * | 2018-03-20 | 2018-05-15 | 杭州恩牛网络技术有限公司 | A kind of integrated study is counter to cheat test method and system |
| CN108416369A (en) * | 2018-02-08 | 2018-08-17 | 同济大学 | Based on Stacking and the random down-sampled sorting technique of overturning, system, medium and equipment |
| CN109344809A (en) * | 2018-11-21 | 2019-02-15 | 上海交通大学 | Domestic electric appliance intelligent management system based on magnetic strength induction signal |
| CN109361529A (en) * | 2018-08-03 | 2019-02-19 | 锐捷网络股份有限公司 | A kind of terminal type recognition methods, device, electronic equipment and storage medium |
-
2019
- 2019-04-09 CN CN201910282297.8A patent/CN110009045A/en active Pending
Patent Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20100142379A1 (en) * | 2008-12-05 | 2010-06-10 | Electronics And Telecommunications Research Institute | Method for classifying packet on mobile terminal |
| CN105138913A (en) * | 2015-07-24 | 2015-12-09 | 四川大学 | Malware detection method based on multi-view ensemble learning |
| CN107277011A (en) * | 2017-06-16 | 2017-10-20 | 深圳市盛路物联通讯技术有限公司 | The data classification method and device of terminal device |
| CN108416369A (en) * | 2018-02-08 | 2018-08-17 | 同济大学 | Based on Stacking and the random down-sampled sorting technique of overturning, system, medium and equipment |
| CN108038701A (en) * | 2018-03-20 | 2018-05-15 | 杭州恩牛网络技术有限公司 | A kind of integrated study is counter to cheat test method and system |
| CN109361529A (en) * | 2018-08-03 | 2019-02-19 | 锐捷网络股份有限公司 | A kind of terminal type recognition methods, device, electronic equipment and storage medium |
| CN109344809A (en) * | 2018-11-21 | 2019-02-15 | 上海交通大学 | Domestic electric appliance intelligent management system based on magnetic strength induction signal |
Cited By (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110445689A (en) * | 2019-08-15 | 2019-11-12 | 平安科技（深圳）有限公司 | Identify the method, apparatus and computer equipment of internet of things equipment type |
| CN110445689B (en) * | 2019-08-15 | 2022-03-18 | 平安科技（深圳）有限公司 | Method and device for identifying type of equipment of Internet of things and computer equipment |
| CN110545278A (en) * | 2019-09-05 | 2019-12-06 | 中国联合网络通信集团有限公司 | industrial internet identification analysis method and system |
| CN110545278B (en) * | 2019-09-05 | 2022-02-25 | 中国联合网络通信集团有限公司 | Industrial internet identification analysis method and system |
| CN110738309A (en) * | 2019-09-27 | 2020-01-31 | 华中科技大学 | DDNN training method and DDNN-based multi-view target identification method and system |
| CN110738309B (en) * | 2019-09-27 | 2022-07-12 | 华中科技大学 | DDNN training method and DDNN-based multi-view target identification method and system |
| CN115277063A (en) * | 2022-06-13 | 2022-11-01 | 深圳铸泰科技有限公司 | Terminal identification device under IPV4 and IPV6 hybrid network environment |
| CN115277063B (en) * | 2022-06-13 | 2023-07-25 | 深圳铸泰科技有限公司 | Terminal identification device under IPV4 and IPV6 mixed network environment |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN110009045A (en) | The recognition methods of internet-of-things terminal and device | |
| CN107944874B (en) | Wind control method, device and system based on transfer learning | |
| CN110245213A (en) | Questionnaire generation method, device, equipment and storage medium | |
| CN108320171A (en) | Hot item prediction technique, system and device | |
| CN105389480B (en) | Multiclass imbalance genomics data iteration Ensemble feature selection method and system | |
| CN112037222B (en) | Automatic updating method and system of neural network model | |
| CN110909224B (en) | Sensitive data automatic classification and identification method and system based on artificial intelligence | |
| CN113590698A (en) | Artificial intelligence technology-based data asset classification modeling and hierarchical protection method | |
| CN108334805A (en) | The method and apparatus for detecting file reading sequences | |
| Hussain et al. | A Simple and Efficient Deep Learning‐Based Framework for Automatic Fruit Recognition | |
| CN110310114A (en) | Object classification method, device, server and storage medium | |
| CN114419363A (en) | Target classification model training method and device based on label-free sample data | |
| CN114359582A (en) | Small sample feature extraction method based on neural network and related equipment | |
| CN117036060A (en) | Vehicle insurance fraud recognition method, device and storage medium | |
| CN114492601A (en) | Resource classification model training method and device, electronic equipment and storage medium | |
| Xue et al. | Evolutionary feature manipulation in data mining/big data | |
| CN104598898A (en) | Aerially photographed image quick recognizing system and aerially photographed image quick recognizing method based on multi-task topology learning | |
| CN111210158A (en) | Target address determination method and device, computer equipment and storage medium | |
| CN112101125B (en) | Method and device for detecting stacking degree of express goods | |
| CN110807159B (en) | Data marking method and device, storage medium and electronic equipment | |
| CN114580761B (en) | Steam volume abnormality determination method, apparatus, electronic device, and readable storage medium | |
| CN115909335A (en) | Commodity labeling method and device | |
| CN114169623A (en) | Power equipment fault analysis method and device, electronic equipment and storage medium | |
| Wardak et al. | Noise presence detection in QR code images | |
| CN113689036A (en) | Thermal imager quality problem reason prediction method based on decision tree C4.5 algorithm |
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: 20190712 |