CN109766855A - A kind of intelligent movable device sensor fingerprint identification method - Google Patents
A kind of intelligent movable device sensor fingerprint identification method Download PDFInfo
- Publication number
- CN109766855A CN109766855A CN201910040039.9A CN201910040039A CN109766855A CN 109766855 A CN109766855 A CN 109766855A CN 201910040039 A CN201910040039 A CN 201910040039A CN 109766855 A CN109766855 A CN 109766855A
- Authority
- CN
- China
- Prior art keywords
- sensing data
- intelligent movable
- data
- fingerprint
- fingerprint recognition
- 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
- 238000000034 method Methods 0.000 title claims abstract description 53
- 238000003062 neural network model Methods 0.000 claims abstract description 32
- 238000012549 training Methods 0.000 claims abstract description 31
- 238000012545 processing Methods 0.000 claims abstract description 14
- 230000015654 memory Effects 0.000 claims description 14
- 238000013136 deep learning model Methods 0.000 claims description 9
- 210000005036 nerve Anatomy 0.000 claims 1
- 230000006870 function Effects 0.000 description 3
- 230000004927 fusion Effects 0.000 description 2
- 210000004218 nerve net Anatomy 0.000 description 2
- 238000013528 artificial neural network Methods 0.000 description 1
- 230000006399 behavior Effects 0.000 description 1
- 230000009194 climbing Effects 0.000 description 1
- 238000012790 confirmation Methods 0.000 description 1
- 235000014510 cooky Nutrition 0.000 description 1
- 238000013075 data extraction Methods 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- 238000010801 machine learning Methods 0.000 description 1
- 238000007781 pre-processing Methods 0.000 description 1
- 230000011218 segmentation Effects 0.000 description 1
- 230000006403 short-term memory Effects 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
- 238000012360 testing method Methods 0.000 description 1
Landscapes
- Image Analysis (AREA)
- Image Input (AREA)
Abstract
The invention discloses a kind of intelligent movable device sensor fingerprint identification methods, step 1, data acquisition: the sensing data of acquisition intelligent movable equipment；Step 2, data prediction: carrying out slice and interpolation processing to collected sensing data, and sensing data is divided into timing length is consistent and equally distributed multistage sensing data on a timeline；Step 3, sensor fingerprint identifies: obtaining intelligent movable device sensor fingerprint to the multistage sensing data identification obtained by preparatory trained fingerprint recognition neural network model；Model training mode are as follows: fingerprint recognition training first is carried out to initial fingerprint recognition neural network model with the multistage sensing data obtained, obtains trained fingerprint recognition neural network model after training.The recognition methods is analyzed and is trained by motion sensor data, is realized and is identified to intelligent movable device-fingerprint.The method achieve identify different equipment to high-accuracy in given multiple mobile internet devices.
Description
Technical field
The present invention relates to belong to data-privacy and mobile application field more particularly to a kind of intelligent movable device sensor refers to
Line recognition methods.
Background technique
With the prosperity of mobile Internet, the identification of intelligent movable equipment becomes a urgent demand, only accurately
Intelligent movable equipment and the user for identifying user can just make various Information Mobile Services (personalized cross-platform recommendation, advertisement striding equipment
Service etc.) and data processing (data fusion, data trade etc.) be possibly realized.The existing method for obtaining device-fingerprint, it is common
Such as browser obtain the IMEI of cookie, APP request Android device and the IDFA of apple equipment, but since people are to hidden
Private more more and more intense concern, these methods receive more and more specific limitation, seek more stable, effective equipment (or people)
The recognition methods of fingerprint becomes more more and more urgent demand.
Summary of the invention
Based on the problems of prior art, the object of the present invention is to provide a kind of intelligent movable device sensor fingerprints
Recognition methods can identify different equipment to high-accuracy in given multiple mobile internet devices.
The purpose of the present invention is what is be achieved through the following technical solutions:
Embodiment of the present invention provides a kind of intelligent movable device sensor fingerprint identification method, comprising:
Step 1, data acquire: the sensing data of acquisition intelligent movable equipment；
Step 2, data prediction: the sensing data collected to the step 1 carries out slice and interpolation processing,
Sensing data is divided into timing length is consistent and equally distributed multistage sensing data on a timeline；
Step 3, sensor fingerprint identifies: being obtained by preparatory trained fingerprint recognition neural network model to the step 2
Multistage sensing data out carries out identification and obtains intelligent movable device sensor fingerprint；The trained fingerprint recognition in advance
The training method of neural network model are as follows: first with the step 1, step 2 obtain multistage sensing data as training data,
Fingerprint recognition training is carried out to initial fingerprint recognition neural network model using training data, is trained after the completion of training
Fingerprint recognition neural network model.
As seen from the above technical solution provided by the invention, intelligent movable device senses provided in an embodiment of the present invention
Device fingerprint identification method, it has the advantage that:
By the sensing data of acquisition intelligent movable equipment, multistage sensing data is divided into after specifically pre-processing,
Using the multistage sensing data separated, first initial fingerprint recognition neural network model is trained, by trained
Fingerprint recognition neural network model carries out fingerprint recognition to multistage sensing data and obtains intelligent movable device sensor fingerprint, into
And different intelligent movable equipment can be identified using sensing data, the method achieve in given multiple mobile Internets
Different equipment is identified to high-accuracy in equipment.
Detailed description of the invention
In order to illustrate the technical solution of the embodiments of the present invention more clearly, required use in being described below to embodiment
Attached drawing be briefly described, it should be apparent that, drawings in the following description are only some embodiments of the invention, for this
For the those of ordinary skill in field, without creative efforts, it can also be obtained according to these attached drawings other
Attached drawing.
Fig. 1 is the flow chart of intelligent movable device sensor fingerprint identification method provided in an embodiment of the present invention；
Fig. 2 is that more classification length of intelligent movable device sensor fingerprint identification method provided in an embodiment of the present invention is remembered in short-term
Recall the composition schematic diagram of deep learning model.
Specific embodiment
Below with reference to particular content of the invention, technical solution in the embodiment of the present invention is clearly and completely retouched
It states, it is clear that described embodiments are only a part of the embodiments of the present invention, instead of all the embodiments.Based on the present invention
Embodiment, every other embodiment obtained by those of ordinary skill in the art without making creative efforts,
Belong to protection scope of the present invention.The content being not described in detail in the embodiment of the present invention belongs to professional and technical personnel in the field
The well known prior art.
As shown in Figure 1, the embodiment of the present invention provides a kind of intelligent movable equipment (such as smart phone, Intelligent flat) sensing
Device fingerprint identification method carries out the identification of intelligent movable device sensor fingerprint by collection analysis sensing data, comprising:
Step 1, data acquire: the sensing data of acquisition intelligent movable equipment；
Step 2, data prediction: the sensing data collected to the step 1 carries out slice and interpolation processing,
Sensing data is divided into timing length is consistent and equally distributed multistage sensing data on a timeline；
Step 3, sensor fingerprint identifies: being obtained by preparatory trained fingerprint recognition neural network model to the step 2
Multistage sensing data out carries out identification and obtains intelligent movable device sensor fingerprint；The trained fingerprint recognition in advance
The training method of neural network model are as follows: first with the step 1, step 2 obtain multistage sensing data as training data,
Fingerprint recognition training is carried out to initial fingerprint recognition neural network model using training data, is trained after the completion of training
Fingerprint recognition neural network model.
In the step 1 of above-mentioned recognition methods, the sensing data of intelligent movable equipment is acquired are as follows:
By calling the interface of the application program of intelligent movable equipment or the browser of operation, intelligent movable equipment is obtained
Acceierometer sensor and gyro sensor initial data, by collection of server initial data obtained as sensing
Device data.
In the step 2 of above-mentioned recognition methods, the sensing data collected to the step 1 carries out slice and interpolation
Sensing data, is divided into that timing length is consistent and equally distributed multistage sensing data on a timeline by processing are as follows:
The collected sensing data of the step 1 is sliced to obtain multistage sensing data, multistage sensor number
According to time span be consistent, to multistage sensing data carry out PCHIP interpolation processing, make the timing of multistage sensing data
It is uniformly distributed on a timeline.
In the step 3 of above-mentioned recognition methods, fingerprint recognition neural network model used is using long short-term memory of more classifying
Deep learning model.The composition of more long memory deep learning models in short-term of classification is as shown in Figure 2, comprising: one layer of full articulamentum
With two layers long memory network in short-term；Wherein,
Memory network is sequentially connected described one layer full articulamentum in short-term with two layers of the length；
The processing mode of more long memory deep learning models in short-term of classification are as follows:
(1) for list entries by the dimension of full articulamentum reproducing sequence, the list entries is multistage sensing data；
(2) length of the second layer in short-term the last one memory unit of memory network data pass through Softmax generate output
Result (identifying which intelligent movable equipment the sensing data of input belongs to) as identification intelligent movable equipment.
In the step 3 of above-mentioned recognition methods, the multistage sensing data obtained with the step 2 is to fingerprint recognition nerve net
Network model carries out recognition training are as follows:
As a sample, each intelligent movable is set each section of sensing data obtained using after the step 2 processing
It is standby to be used as a kind of sample, fingerprint recognition training is carried out to fingerprint recognition neural network model.
In above-mentioned recognition methods, the time for stopping recognition training being determined by following conditions:
It (1) can deconditioning higher than 95% as worked as model accuracy rate.(such as inventor with 117 equipment be trained for
The accuracy rate for being higher than 95% can be used in example, also can according to need if equipment is more and is adjusted to accuracy rate).
(2) (loss function is deep learning model when the loss function value of model is lower than number of devices × 0.3%
One function of middle indispensability, the direction of representative model training) it can deconditioning.
Two above trains stop condition, meets any one and thinks to meet training stopping opportunity of the invention, Ji Keting
Only have trained.In practice, in machine learning field, model training, the opportunity that training is completed is mostly an empirical value, is doing reality
Be exactly during testing rule of thumb confirmation reach enough well can deconditioning, this is different trainer and is required according to it
Difference can be customized.
In the step 3 of above-mentioned recognition methods, if the sensing data of identification is continuous data, continuous data is divided into
A plurality of sensing data is identified with trained fingerprint recognition neural network model respectively, using most ballot sides after identification
Formula determines which intelligent movable equipment each sensing data belongs to；
If the sensing data of identification is non-continuous data, after non-continuous data is pre-processed respectively, then pass through instruction
The fingerprint recognition neural network model perfected is identified, determines whole non-continuous data categories using most ballot modes after identification
In which intelligent movable equipment.
In above-mentioned recognition methods, most ballot modes are as follows: a plurality of sensing data is used into trained fingerprint recognition respectively
After neural network model is identified, each sensing data is made to obtain a label, the corresponding mobile intelligence of a label
Energy equipment, takes the highest label of the frequency as final label, determines the sensing data segment by the final label and belong to
Which intelligent movable equipment, i.e., the corresponding intelligent movable equipment of final label, is shifting corresponding to the sensing data segment
Dynamic smart machine.
In above-mentioned recognition methods, continuous data are as follows: one section of continuous sensing data of same intelligent movable equipment；
The non-continuous data are as follows: multistage sensing data of the same intelligent movable equipment in different time.
Method of the invention is used to train fingerprint after pretreatment by the initial data of acquisition intelligent movable device sensor
Neural network model is identified, with the fingerprint recognition neural network model after the completion of training to the sensing data of intelligent movable equipment
It identifies sensor fingerprint, so as to successfully identify different equipment using sensing data, realizes multiple what is given
In mobile internet device, different equipment is identified to high-accuracy.This method is based in sensor hardware process
Heterogeneity, this heterogeneity is in equipment using being almost to stablize and be difficult to modify in life cycle, and institute is in this way than tradition
Recognition methods it is more stable.Under the conditions of the experimental check of mass data, identification can reach very high precision.
The embodiment of the present invention is specifically described in further detail below.
The embodiment of the present invention provides a kind of intelligent movable device sensor fingerprint identification method, suitable for more including what is given
The identification of intelligent movable device sensor fingerprint in a intelligent movable cluster tool is a kind of based on intelligent movable equipment moving
The method of sensor (i.e. acceierometer sensor and gyro sensor) data extraction device fingerprint, this method includes following step
It is rapid:
Step 1, acquire sensing data: by call intelligent movable equipment APP acquisition motion sensor API or
Person is obtained the initial data of acceierometer sensor and gyro sensor, is adopted by server by the browser script of operation
Collect sensing data of the initial data obtained as intelligent movable equipment；
Step 2, data prediction: carrying out slice to collected sensing data and be divided into multistage sensing data, point
The time span of multistage sensing data out is consistent, and then carries out the (segmentation three of PCHIP interpolation to multistage sensing data
Secondary Hermite polynomial interopolation), it is uniformly distributed the timing of multistage sensing data on a timeline；
Step 3, fingerprint training and identification based on neural network model: the multistage sensing data first obtained with step 2
Fingerprint recognition training is carried out to initial fingerprint recognition neural network model, specifically using: each section of sensing data all as
One sample, each intelligent movable equipment carry out fingerprint recognition training, fingerprint with neural network to identification as a kind of sample
Identify that neural network model uses the long memory deep learning models in short-term of more classification, the trained fingerprint recognition nerve net obtained
Network model；
When subsequent progress intelligent movable device sensor fingerprint recognition, multistage sensor number is obtained after handling by steps 1 and 2
According to being identified with trained fingerprint recognition neural network model to multistage sensing data, identify intelligent movable equipment
Sensor fingerprint, and then the intelligent movable equipment is determined by sensor fingerprint.
It is above-mentioned to determine the intelligent movable equipment using most ballot modes by sensor fingerprint: i.e. by most ballot sides
Formula determines which intelligent movable equipment the sensing data identified belongs to.The majority ballot mode is: by a plurality of sensor number
After being identified respectively with trained fingerprint recognition neural network model, each sensing data is made to obtain a mark
Label, the corresponding intelligent movable equipment of a label, take the highest label of the frequency as final label, are sentenced by the final label
Not Chu the sensing data segment belong to which intelligent movable equipment, i.e., the corresponding intelligent movable equipment of final label, is this
Intelligent movable equipment corresponding to sensing data segment.Accuracy of identification can be improved by most ballot modes.
Method of the invention can identify continuous data and non-continuous data, wherein continuous data refers to: same movement
One section of continuous sensing data of smart machine；Non-continuous data refers to: multistage of the same intelligent movable equipment in different time
Sensing data.
Recognition methods of the invention, is analyzed and is trained by motion sensor data, is realized to intelligent movable equipment
Fingerprint recognition.The advantage of the recognition methods is: the hardware because of sensor as an equipment, and fingerprint characteristic is stablized, and
The data for obtaining sensor do not need user's clear agreement, and insensitive to user behavior, in intelligent movable, equipment holder appoints
It when conation is made, such as remain stationary, walk, running, stair climbing, can accurately identify device sensor, and then realize
High-accuracy different equipment is identified in given multiple mobile internet devices.
Since the identification of intelligent movable equipment is a kind of important demand, the method for the present invention can be used for all kinds of intelligent movables and set
Standby sensor identification and identification realizes the intelligent movable equipment for accurately identifying user and provides various mobile clothes for user
Business (personalized cross-platform recommendation, advertisement striding equipment service etc.) and data processing (data fusion, data trade etc.) are possibly realized.
The foregoing is only a preferred embodiment of the present invention, but scope of protection of the present invention is not limited thereto,
Within the technical scope of the present disclosure, any changes or substitutions that can be easily thought of by anyone skilled in the art,
It should be covered by the protection scope of the present invention.Therefore, protection scope of the present invention should be with the protection model of claims
Subject to enclosing.
Claims (8)
1. a kind of intelligent movable device sensor fingerprint identification method characterized by comprising
Step 1, data acquire: the sensing data of acquisition intelligent movable equipment；
Step 2, data prediction: the sensing data collected to the step 1 carries out slice and interpolation processing, will pass
Sensor data are divided into that timing length is consistent and equally distributed multistage sensing data on a timeline；
Step 3, sensor fingerprint identifies: being obtained by preparatory trained fingerprint recognition neural network model to the step 2
Multistage sensing data carries out identification and obtains intelligent movable device sensor fingerprint；The fingerprint recognition nerve trained in advance
The training method of network model are as follows: show that multistage sensing data as training data, utilizes first with the step 1, step 2
Training data carries out fingerprint recognition training to initial fingerprint recognition neural network model, and trained finger is obtained after the completion of training
Line identifies neural network model.
2. intelligent movable device sensor fingerprint identification method according to claim 1, which is characterized in that the method
In step 1, the sensing data of intelligent movable equipment is acquired are as follows:
By calling the interface of the application program of intelligent movable equipment or the browser of operation, adding for intelligent movable equipment is obtained
The initial data of speedometer transducer and gyro sensor, by collection of server initial data obtained as sensor number
According to.
3. intelligent movable device sensor fingerprint identification method according to claim 1 or 2, which is characterized in that the side
In the step 2 of method, the sensing data collected to the step 1 carries out slice and interpolation processing, by sensing data
It is divided into that timing length is consistent and equally distributed multistage sensing data on a timeline are as follows:
The collected sensing data of the step 1 is sliced to obtain multistage sensing data, multistage sensing data
Time span is consistent, to multistage sensing data carry out PCHIP interpolation processing, make the timing of multistage sensing data when
Between be uniformly distributed on axis.
4. intelligent movable device sensor fingerprint identification method according to claim 1 or 2, which is characterized in that the side
In method step 3, fingerprint recognition neural network model used is the long memory deep learning model in short-term of more classification.
5. intelligent movable device sensor fingerprint identification method according to claim 4, which is characterized in that the method step
In rapid 3, more long memory deep learning models in short-term of classification include:
One layer of full articulamentum and two layers of length memory network in short-term；Wherein,
Memory network is sequentially connected described one layer full articulamentum in short-term with two layers of the length；
The processing mode of more long memory deep learning models in short-term of classification are as follows:
(1) for list entries by the dimension of full articulamentum reproducing sequence, the list entries is multistage sensing data；
(2) data of the length of the second layer the last one memory unit of memory network in short-term are generated an output as by Softmax
Identify the result of intelligent movable equipment.
6. intelligent movable device sensor fingerprint identification method according to claim 1 or 2, which is characterized in that the side
In the step 3 of method, initial fingerprint recognition neural network model is carried out using the multistage sensing data that the step 2 obtains
Fingerprint recognition training are as follows:
The each section of sensing data obtained using after the step 2 processing is made as a sample, each intelligent movable equipment
For a kind of sample, fingerprint recognition training is carried out to fingerprint recognition neural network model.
7. intelligent movable device sensor fingerprint identification method according to claim 1 or 2, which is characterized in that the side
The step 3 of method, further includes: determine which intelligent movable equipment each sensing data belongs to using most ballot modes after identification
Step.
8. intelligent movable device sensor fingerprint identification method according to claim 7, which is characterized in that most throwings
Ticket mode are as follows: after being identified a plurality of sensing data with trained fingerprint recognition neural network model respectively, make each
Sensing data obtains a label, and a label corresponds to an intelligent movable equipment, takes the highest label conduct of the frequency
Final label determines which intelligent movable equipment is the sensing data segment belong to according to the final label.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910040039.9A CN109766855A (en) | 2019-01-16 | 2019-01-16 | A kind of intelligent movable device sensor fingerprint identification method |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910040039.9A CN109766855A (en) | 2019-01-16 | 2019-01-16 | A kind of intelligent movable device sensor fingerprint identification method |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN109766855A true CN109766855A (en) | 2019-05-17 |
Family
ID=66452333
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201910040039.9A Pending CN109766855A (en) | 2019-01-16 | 2019-01-16 | A kind of intelligent movable device sensor fingerprint identification method |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN109766855A (en) |
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2021092872A1 (en) * | 2019-11-13 | 2021-05-20 | 北京数字联盟网络科技有限公司 | Device fingerprint extraction method based on smartphone sensor |
| CN113111726A (en) * | 2021-03-18 | 2021-07-13 | 浙江大学 | Vibration motor equipment fingerprint extraction and identification method based on homologous signals |
| CN113111725A (en) * | 2021-03-18 | 2021-07-13 | 浙江大学 | Vibration motor equipment fingerprint extraction identification system based on homologous signal |
| US20220318347A1 (en) * | 2019-11-13 | 2022-10-06 | Beijing Digital Union Web Science And Technology Company Limited | A device fingerprint extraction method based on smart phone sensor |
Citations (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107837087A (en) * | 2017-12-08 | 2018-03-27 | 兰州理工大学 | A kind of human motion state recognition methods based on smart mobile phone |
-
2019
- 2019-01-16 CN CN201910040039.9A patent/CN109766855A/en active Pending
Patent Citations (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107837087A (en) * | 2017-12-08 | 2018-03-27 | 兰州理工大学 | A kind of human motion state recognition methods based on smart mobile phone |
Non-Patent Citations (1)
| Title |
|---|
| HUIQI LIU等: "Finding the Stars in the Fireworks: Deep Understanding of Motion Sensor Fingerprint", 《IEEE INFOCOM 2018 - IEEE CONFERENCE ON COMPUTER COMMUNICATIONS》 * |
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2021092872A1 (en) * | 2019-11-13 | 2021-05-20 | 北京数字联盟网络科技有限公司 | Device fingerprint extraction method based on smartphone sensor |
| US20220318347A1 (en) * | 2019-11-13 | 2022-10-06 | Beijing Digital Union Web Science And Technology Company Limited | A device fingerprint extraction method based on smart phone sensor |
| CN113111726A (en) * | 2021-03-18 | 2021-07-13 | 浙江大学 | Vibration motor equipment fingerprint extraction and identification method based on homologous signals |
| CN113111725A (en) * | 2021-03-18 | 2021-07-13 | 浙江大学 | Vibration motor equipment fingerprint extraction identification system based on homologous signal |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109766855A (en) | A kind of intelligent movable device sensor fingerprint identification method | |
| CN111401433B (en) | User information acquisition method and device, electronic equipment and storage medium | |
| CN109299380B (en) | Exercise personalized recommendation method based on multi-dimensional features in online education platform | |
| CN109086756A (en) | A kind of text detection analysis method, device and equipment based on deep neural network | |
| CN111080109B (en) | Customer service quality evaluation method and device and electronic equipment | |
| CN110717324B (en) | Judgment document answer information extraction method, device, extractor, medium and equipment | |
| CN111260102A (en) | User satisfaction prediction method and device, electronic equipment and storage medium | |
| CN111191133B (en) | Service search processing method, device and equipment | |
| CN110427483A (en) | Text snippet evaluating method, device, system and evaluation and test server | |
| CN111401105B (en) | Video expression recognition method, device and equipment | |
| CN105095415A (en) | Method and apparatus for confirming network emotion | |
| CN111291187B (en) | Emotion analysis method and device, electronic equipment and storage medium | |
| CN109740567A (en) | Key point location model training method, localization method, device and equipment | |
| CN106991577A (en) | A kind of method and device for determining targeted customer | |
| CN111508480A (en) | Training method of audio recognition model, audio recognition method, device and equipment | |
| CN107766234A (en) | A kind of assessment method, the apparatus and system of the webpage health degree based on mobile device | |
| EP4371027A1 (en) | Intelligent task completion detection at a computing device | |
| CN107291775A (en) | The reparation language material generation method and device of error sample | |
| CN112559923A (en) | Website resource recommendation method and device, electronic equipment and computer storage medium | |
| Shen et al. | Learning mobile application usage-a deep learning approach | |
| CN111177500A (en) | Data object classification method and device, computer equipment and storage medium | |
| CN112992187B (en) | Context-based voice emotion detection method, device, equipment and storage medium | |
| CN113590771A (en) | Data mining method, device, equipment and storage medium | |
| CN115935076A (en) | Travel service information pushing method and system based on artificial intelligence | |
| CN112612921B (en) | Video label determination method and device |
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: 20190517 |