CN110060094A - Objective group's superiority and inferiority predictor method and device, computer readable storage medium - Google Patents
Objective group's superiority and inferiority predictor method and device, computer readable storage medium Download PDFInfo
- Publication number
- CN110060094A CN110060094A CN201910232399.9A CN201910232399A CN110060094A CN 110060094 A CN110060094 A CN 110060094A CN 201910232399 A CN201910232399 A CN 201910232399A CN 110060094 A CN110060094 A CN 110060094A
- Authority
- CN
- China
- Prior art keywords
- inferiority
- superiority
- group
- timing
- visitor
- 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
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/08—Learning methods
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q30/00—Commerce
- G06Q30/02—Marketing; Price estimation or determination; Fundraising
- G06Q30/0201—Market modelling; Market analysis; Collecting market data
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q30/00—Commerce
- G06Q30/02—Marketing; Price estimation or determination; Fundraising
- G06Q30/0201—Market modelling; Market analysis; Collecting market data
- G06Q30/0202—Market predictions or forecasting for commercial activities
Landscapes
- Engineering & Computer Science (AREA)
- Business, Economics & Management (AREA)
- Strategic Management (AREA)
- Finance (AREA)
- Development Economics (AREA)
- Accounting & Taxation (AREA)
- Physics & Mathematics (AREA)
- Entrepreneurship & Innovation (AREA)
- Theoretical Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Data Mining & Analysis (AREA)
- General Business, Economics & Management (AREA)
- Marketing (AREA)
- Economics (AREA)
- Game Theory and Decision Science (AREA)
- Computational Linguistics (AREA)
- Life Sciences & Earth Sciences (AREA)
- Mathematical Physics (AREA)
- General Health & Medical Sciences (AREA)
- Biophysics (AREA)
- Biomedical Technology (AREA)
- Artificial Intelligence (AREA)
- Software Systems (AREA)
- Health & Medical Sciences (AREA)
- Evolutionary Computation (AREA)
- General Engineering & Computer Science (AREA)
- Computing Systems (AREA)
- Molecular Biology (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
A kind of visitor's group's superiority and inferiority predictor method and device, computer readable storage medium, visitor's group's superiority and inferiority predictor method, comprising: obtain the user data of the target device in preset time period；The user data is handled according to timing, obtains data sequence relevant to timing；The data sequence is input to the objective group's superiority and inferiority of progress in preset objective group's superiority and inferiority prediction model to estimate；It obtains objective group's superiority and inferiority estimation results and exports.Using the above scheme, it can be improved objective group's superiority and inferiority result accuracy.
Description
Technical field
The present embodiments relate to data analysis technique field more particularly to a kind of objective group's superiority and inferiority predictor method and device,
Computer readable storage medium.
Background technique
With becoming increasingly popular for self-oriented service, Self-Service class machine is consumed down more and more common in scene online.?
During the operation of Self-Service class machine, the data for using the user of Self-Service class machine can be accumulated.
Currently, usually simply being statisticallyd analyze to the data of user, certain several index is obtained on the whole to characterize
Objective group's superiority and inferiority.However, objective group's superiority and inferiority result accuracy that aforesaid way obtains is lower.
Summary of the invention
The technical issues of embodiment of the present invention solves is that objective group's superiority and inferiority result accuracy is lower.
In order to solve the above technical problems, the embodiment of the present invention provides a kind of objective group's superiority and inferiority predictor method, comprising: obtain default
The user data of target device in period；The user data is handled according to timing, is obtained relevant to timing
Data sequence；The data sequence is input to the objective group's superiority and inferiority of progress in preset objective group's superiority and inferiority prediction model to estimate；Obtain visitor
Group's superiority and inferiority estimation results simultaneously export.
Optionally, the user data is handled according to timing, obtains data sequence relevant to timing, comprising:
Each user identity information and its corresponding first label information are formed into a child user sequence；According to the user data,
All child user sequences are arranged according to timing, obtain user's sequence relevant to timing, user's sequence is made
For the data sequence.
Optionally, first label information comprises at least one of the following: age, gender, region, occupation, old and new customers
Classification, record of bad behavior, third party's label.
Optionally, the user data is handled according to timing, obtains data sequence relevant to timing, comprising:
According to timing, the preset time period is divided into N number of sub- period；Count the user corresponding second in each sub- period
Label information；N number of sub- period and corresponding second label information of N number of sub- period are carried out according to timing
Arrangement, obtains time series relevant to timing, using the time series as the data sequence.
Optionally, second label information comprises at least one of the following: age bracket, sex ratio, occupation, old and new customers
Classification, record of bad behavior.
Optionally, objective group's superiority and inferiority is obtained using LSTM deep learning algorithm or the training of GRU deep learning algorithm to estimate
Model.
Optionally, after exporting objective group's superiority and inferiority estimation results, further includes: according to objective group's superiority and inferiority estimation results, really
The grade of the fixed target device, the grade are used to assess the degree of risk of the target device.
Optionally, after the grade for determining the target device, further includes: according to the grade of the target device, really
The position adjustment information of the fixed target device.
Optionally, after the grade for determining the target device, further includes: when the grade for detecting the target device
When reaching predetermined level threshold value, warning reminding is exported.
The embodiment of the present invention also provides a kind of objective group's superiority and inferiority estimating device, comprising: acquiring unit is suitable for obtaining preset time
The user data of target device in section；Processing unit, suitable for the user data is handled according to timing, obtain and when
The relevant data sequence of sequence；Input unit, suitable for by the data sequence be input in preset objective group's superiority and inferiority prediction model into
Row visitor's group's superiority and inferiority is estimated；First output unit, suitable for obtaining objective group's superiority and inferiority estimation results and exporting.
Optionally, the processing unit is suitable for forming each user identity information and its corresponding first label information
One child user sequence；According to the user data, all child user sequences are arranged according to timing, is obtained and timing
Relevant user's sequence, using user's sequence as the data sequence.
Optionally, first label information comprises at least one of the following: age, gender, region, occupation, old and new customers
Classification, record of bad behavior, third party's label.
Optionally, the processing unit is suitable for that the preset time period is divided into N number of sub- period according to timing；System
Count corresponding second label information of user in each sub- period；By N number of sub- period and N number of sub- period point
Not corresponding second label information is arranged according to timing, obtains time series relevant to timing, by the time series
As the data sequence.
Optionally, second label information comprises at least one of the following: age bracket, sex ratio, occupation, old and new customers
Classification, record of bad behavior.
Optionally, objective group's superiority and inferiority estimating device, further includes: model construction unit is suitable for using LSTM deep learning
Algorithm or the training of GRU deep learning algorithm obtain objective group's superiority and inferiority prediction model.
Optionally, objective group's superiority and inferiority estimating device, further includes: the first determination unit is suitable for pre- in the objective group's superiority and inferiority of output
After estimating result, according to objective group's superiority and inferiority estimation results, the grade of the target device is determined, the grade is for assessing institute
State the degree of risk of target device.
Optionally, objective group's superiority and inferiority estimating device, further includes: the second determination unit is suitable for according to the target device
Grade, determine the position adjustment information of the target device.
Optionally, objective group's superiority and inferiority estimating device, further includes: the second output unit, suitable for determining that the target sets
After standby grade, when the grade for detecting the target device reaches predetermined level threshold value, warning reminding is exported.
The embodiment of the present invention also provides a kind of objective group's superiority and inferiority estimating device, including memory and processor, the memory
On be stored with the computer instruction that can be run on the processor, the processor executes when running the computer instruction
The step of stating any objective group's superiority and inferiority predictor method.
The embodiment of the present invention also provides a kind of computer readable storage medium, and computer readable storage medium is non-volatile
Storage medium or non-transitory storage media, are stored thereon with computer instruction, and the computer instruction executes above-mentioned when running
A kind of the step of visitor's group's superiority and inferiority predictor method.
Compared with prior art, the technical solution of the embodiment of the present invention has the advantages that
The user data of target device in preset time period is handled according to timing, obtains number relevant to timing
According to sequence, data sequence is input to the objective objective group's superiority and inferiority of group's superiority and inferiority prediction model progress and is estimated, objective group's superiority and inferiority estimation results are obtained.
Due to when objective group's superiority and inferiority is estimated, it is contemplated that the user data in a period of time, and obtained data sequence is related to timing, from
And the trend changed over time using the objective group of target device can be embodied, so as to improve objective group's superiority and inferiority estimate it is accurate
Degree.
Detailed description of the invention
Fig. 1 is the flow chart of the objective group's superiority and inferiority predictor method of one of embodiment of the present invention.
Fig. 2 is the structural schematic diagram of the objective group's superiority and inferiority estimating device of one of embodiment of the present invention.
Specific embodiment
Currently, usually simply being statisticallyd analyze to the data of user, certain several index is obtained on the whole to characterize
Objective group's superiority and inferiority.However, objective group's superiority and inferiority result accuracy that aforesaid way obtains is lower.
In the embodiment of the present invention, the user data of the target device in preset time period is handled according to timing, is obtained
To data sequence relevant to timing, data sequence is input to the objective objective group's superiority and inferiority of group's superiority and inferiority prediction model progress and is estimated, is obtained
Objective group's superiority and inferiority estimation results.Due to when objective group's superiority and inferiority is estimated, it is contemplated that the user data in a period of time, and obtained data
Sequence is related to timing, the trend that the objective group so as to embody using target device changes over time, so as to improve visitor
The accuracy that group's superiority and inferiority is estimated.
It is understandable to enable the above-mentioned purpose, feature and beneficial effect of the embodiment of the present invention to become apparent, below with reference to attached
Figure is described in detail specific embodiments of the present invention.
Referring to Fig.1, a kind of flow chart of visitor's group's superiority and inferiority predictor method in the embodiment of the present invention is given, may include as follows
Step.
Step 11, the user data of the target device in preset time period is obtained.
In specific implementation, target device can be the Self-Services class machines such as self-service business handling machine.In target device
Use process in, user data related to user can be accumulated according to chronological order.
Step 12, the user data is handled according to timing, obtains data sequence relevant to timing.
In specific implementation, user data can be handled using various ways, obtains data relevant to timing
Sequence.
In an embodiment of the present invention, each user identity information and its corresponding first label information are formed into a son
User's sequence；According to the user data, all child user sequences are arranged according to timing, are obtained relevant to timing
User's sequence, using user's sequence as the data sequence.Obtained user's sequence can be a two-dimensional array.
In specific implementation, the first label information may include following at least one: the age, gender, region, occupation, new
Frequent customer's classification, record of bad behavior, third party's label.It is understood that according to practical application scene, in the first label information also
It may include other information, details are not described herein again.
For example, child user is serial are as follows: User ID 12345, male, 28 years old, Shanghai, frequent customer, occur within nearly three months it is primary also
It borrows overdue.
In an alternative embodiment of the invention, according to timing, the preset time period is divided into N number of sub- period；Statistics
Corresponding second label information of user in each sub- period；By N number of sub- period and N number of sub- period difference
Corresponding second label information is arranged according to timing, obtains time series relevant to timing, the time series is made
For the data sequence, obtained time series can be a two-dimensional array.
In specific implementation, preset time period can be evenly dividing as N number of sub- period, it can also be non-by the period
It is even to be divided into N number of period.
In specific implementation, second label information comprises at least one of the following: age bracket, sex ratio, occupation, new
Frequent customer's classification, record of bad behavior.It is understood that can also include it in the first label information according to practical application scene
His information, details are not described herein again.
There may be one or more users within a sub- period, it is also possible to not have user.Generally according to second
Label information counts client's accounting that each label is directed within each sub- period.For example, one of them sub- period is corresponding
The second label information are as follows: 20% user is in 30~40 age brackets, and 80% user is in 20~30 age brackets；Male's accounting
70%, women accounting 30%；78% is new client.
Step 13, the data sequence the objective group's superiority and inferiority of progress in preset objective group's superiority and inferiority prediction model is input to estimate.
In specific implementation, after obtaining data sequence, it is pre- that data sequence can be input to preset objective group's superiority and inferiority
Estimate in model, carries out objective group's superiority and inferiority and estimate.
In embodiments of the present invention, long short-term memory (Long Short Term Memory, LSTM) depth can be used
Learning algorithm or the training of gating cycle (Gated Recurrent Unit, GRU) deep learning algorithm obtain objective group's superiority and inferiority
Prediction model obtains objective group's superiority and inferiority estimation results.It is understood that can also according to practical application scene and application demand,
It chooses other kinds of deep learning algorithm or algorithm relevant to timing constructs objective group's superiority and inferiority prediction model, it is not another herein
One illustrates.
It, can be using the user data of history as training sample, by history in the objective group's superiority and inferiority prediction model of training
User data is according to timing sequence process, several user's sequence user's sequences or time series that perhaps time series will obtain
It is input in LSTM deep neural network model or GRU deep neural network model and is trained as sample, obtain objective group
Superiority and inferiority prediction model.
Step 14, objective group's superiority and inferiority estimation results are obtained and are exported.
In specific implementation, available objective group's superiority and inferiority prediction model obtains objective group's superiority and inferiority estimation results are simultaneously defeated
Out.
By above scheme it is found that being handled according to timing the user data of the target device in preset time period, obtain
To data sequence relevant to timing, data sequence is input to the objective objective group's superiority and inferiority of group's superiority and inferiority prediction model progress and is estimated, is obtained
Objective group's superiority and inferiority estimation results.Due to when objective group's superiority and inferiority is estimated, it is contemplated that the user data in a period of time, and obtained data
Sequence is related to timing, the trend that the objective group so as to embody using target device changes over time, so as to improve visitor
The accuracy that group's superiority and inferiority is estimated.
In addition, can know that a target device objective group's mass within which preferable period by objective group's superiority and inferiority result,
Objective group within which second-rate period.The corresponding objective group's matter of target device can also be known by objective group's superiority and inferiority result
Amount, the quality condition of the objective group of entirety so as to infer target device institute overlay area.
In specific implementation, after exporting objective group's superiority and inferiority estimation results, knot can also be estimated according to objective group's superiority and inferiority
Fruit determines the grade of target device.The grade of target device can characterize the degree of risk of target device.
For example, the grade of target device is divided into level Four, grade is followed successively by level-one, second level, three-level and level Four from high in the end, visitor
The high-quality degree of group is higher, and corresponding target device higher grade, and the risk of target device is lower；Correspondingly, the high-quality degree of objective group
Lower, the lower grade of corresponding target device, and the degree of risk of target device is higher.Grade is that the target device of level-one is corresponding
The high-quality degree of objective group be higher than the corresponding objective high-quality degree of group of target device that grade is second level.
In embodiments of the present invention, to reinforce the management to target device, mesh can be determined according to the grade of target device
The position adjustment information of marking device.It is relatively more fixed using the objective group of target device after target device, which is placed, to be completed, usually
A certain range of visitor group.When the corresponding objective group's superiority and inferiority estimation results of the grade of target device show the high-quality degree of objective group not
Height, when the threshold value of setting is not achieved, then the high-quality degree of totality for characterizing the objective group in the region is not high, and target device is placed on
The region, target device cannot generate preferable income, and so as to provide the position adjustment information of target device, target is set
Standby position is adjusted to other higher regions of objective group's mass.
In specific implementation, when the grade for detecting target device reaches predetermined level threshold value, warning reminding is exported, with
Inform operator or monitoring backstage, within a preset time period, the high-quality degree for the objective group that target device uses is lower, exists
Greater risk, so that operator or monitoring backstage can take corresponding treatment measures in time.
For example, intermediary leads the higher client of a group risk to shops, the higher client of these risks is using in shops
One target device carries out business application.Since the risk of these clients is higher, when they are concentrated through in a period of time
When target device transacting business application, corresponding user data can be generated.To these risks higher client's transacting business time
User data in section is handled according to timing, obtains data sequence relevant to timing.Objective group is passed through using data sequence
Superiority and inferiority prediction model carries out objective group's superiority and inferiority and estimates, and obtained objective group's superiority and inferiority estimation results show that objective group is second-rate.When objective group is excellent
The grade of the corresponding target device of bad estimation results reaches predetermined level threshold value, or objective group's superiority and inferiority estimation results meet setting
When as a result, warning reminding can be exported, informs operator or monitoring backstage, the user in the period exists abnormal.
In specific implementation, in application scenes, when target device is used to the business such as audit, register, Ke Yigen
According to objective group's superiority and inferiority estimation results, the application condition requirement to client is adjusted.For example, working as objective group's superiority and inferiority knot of a certain target device
Fruit shows that the high-quality degree of objective group is higher, then characterize objective group's mass whole whithin a period of time of target device institute overlay area compared with
It is good, then it can slightly relax application condition, simplify application process.For another example, when objective group's superiority and inferiority result of a certain target device is aobvious
Show that the high-quality degree of objective group is lower, then the whole objective group for characterizing target device institute overlay area is second-rate, then can will apply
Condition or auditing flow adjustment it is harsher, with better control visitor group mass.
It in specific implementation, can be to target device using above-mentioned objective group's superiority and inferiority predictor method provided in an embodiment of the present invention
Objective group's superiority and inferiority estimated offline, can also be estimated in real time.
When being estimated using offline mode to objective group's superiority and inferiority of target device, each target device can be according to setting week
Respective user data is periodically reported to server end by the phase.For example, setting period of each target device using T+1, i-th day
(i-1)-th day user data is reported into server end.Server end protects the user data of the target device got
It deposits.The user data that server end will acquire is handled according to timing, obtain user's sequence relevant to timing or when
Between sequence.Objective group's superiority and inferiority prediction model is loaded, obtained user's sequence or data sequence are input to objective group's superiority and inferiority and estimated
Model carries out objective group's superiority and inferiority and estimates, and obtains objective group's superiority and inferiority estimation results.Objective group's superiority and inferiority estimation results can be by the way of marking
It presents, can also be presented, can also be presented by the way of label by the way of grade classification, it is such as poor, qualified, good and excellent
Deng.Obtained objective group's superiority and inferiority estimation results are exported to inline system.User data processing is carried out on server end, is gone forward side by side
Row visitor's group's superiority and inferiority is estimated, lower to the computing capability requirement of target device, and bandwidth requirement is not also high, stablizes height so as to realize
The data of effect are transmitted.
It, can be as unit of every target device, to every mesh when being estimated using real-time mode to objective group's superiority and inferiority
Objective group's superiority and inferiority of marking device is estimated.Every target device can obtain the user data in preset time period in real time, and will
Accessed user data is handled according to timing, obtains user's sequence or time series.Real-time loading visitor group's superiority and inferiority
Obtained user's sequence or time series are input to objective group's superiority and inferiority prediction model, it is pre- to obtain objective group's superiority and inferiority by prediction model
Estimate result.Obtain the grade of target device according to obtained objective group's superiority and inferiority estimation results, by obtained target device etc.
Grade is applied to inline system, and real-time is higher.
To better understand and realizing that the embodiment of the present invention, the embodiment of the present invention also provide one convenient for those skilled in the art
The objective group's superiority and inferiority estimating device of kind.
Referring to Fig. 2, a kind of structural schematic diagram of visitor's group's superiority and inferiority estimating device in the embodiment of the present invention is given.The visitor group
Superiority and inferiority estimating device 20 may include: acquiring unit 21, processing unit 22, input unit 23 and the first output unit 24, in which:
Acquiring unit 21, suitable for obtaining the user data of the target device in preset time period；
Processing unit 22 obtains data sequence relevant to timing suitable for handling the user data according to timing
Column；
Input unit 23, it is excellent suitable for the data sequence is input to the objective group of progress in preset objective group's superiority and inferiority prediction model
It is bad to estimate；
First output unit 24, suitable for obtaining objective group's superiority and inferiority estimation results and exporting.
In specific implementation, the processing unit 22 is suitable for each user identity information and its corresponding first label
Information forms a child user sequence；According to user data, all child user sequences are arranged according to timing, obtain with
The relevant user's sequence of timing, using user's sequence as the data sequence.
In specific implementation, first label information comprises at least one of the following: the age, gender, region, occupation, new
Frequent customer's classification, record of bad behavior, third party's label.
In specific implementation, the processing unit 22 is suitable for that the preset time period is divided into N number of son according to timing
Period；Count corresponding second label information of user in each sub- period；By N number of sub- period and described N number of
Sub- period, corresponding second label information was arranged according to timing, time series relevant to timing was obtained, by institute
Time series is stated as the data sequence.
In specific implementation, second label information comprises at least one of the following: age bracket, sex ratio, occupation, new
Frequent customer's classification, record of bad behavior.
In specific implementation, objective group's superiority and inferiority estimating device 20 can also include: that (Fig. 2 does not show model construction unit
Out), suitable for obtaining objective group's superiority and inferiority prediction model using LSTM deep learning algorithm or the training of GRU deep learning algorithm.It can
With understanding, can also according to practical application scene and application demand, choose other kinds of deep learning algorithm or with
The relevant algorithm of timing constructs objective group's superiority and inferiority prediction model, no longer illustrates one by one herein.
In specific implementation, objective group's superiority and inferiority estimating device 20 can also include: that (Fig. 2 does not show the first determination unit
Out), it is suitable for after exporting objective group's superiority and inferiority estimation results, according to objective group's superiority and inferiority estimation results, determines the target device
Grade, the grade is used to assess the risk of the target device.
In specific implementation, objective group's superiority and inferiority estimating device 20 can also include: that (Fig. 2 does not show the second output unit
Out), suitable for after the grade for determining the target device, when the grade for detecting the target device reaches predetermined level threshold
When value, warning reminding is exported.
In specific implementation, the working principle and workflow of objective group's superiority and inferiority estimating device 20 and this hair can be referred to
The description in objective group's superiority and inferiority predictor method provided in bright above-described embodiment, details are not described herein again.
The embodiment of the present invention also provides a kind of objective group's superiority and inferiority estimating device, including memory and processor, the memory
On be stored with the computer instruction that can be run on the processor, the processor executes sheet when running the computer instruction
The step of any of the above-described kind of objective group's superiority and inferiority predictor method that inventive embodiments provide.
The embodiment of the present invention also provides a kind of computer readable storage medium, and computer readable storage medium is non-volatile
Storage medium or non-transitory storage media, are stored thereon with computer instruction, and the computer instruction executes the present invention when running
The step of any of the above-described kind of objective group's superiority and inferiority predictor method that embodiment provides.
Those of ordinary skill in the art will appreciate that all or part of the steps in the various methods of above-described embodiment is can
It is completed with instructing relevant hardware by program, which can store in any computer readable storage medium storing program for executing, deposit
Storage media may include: ROM, RAM, disk or CD etc..
Although present disclosure is as above, present invention is not limited to this.Anyone skilled in the art are not departing from this
It in the spirit and scope of invention, can make various changes or modifications, therefore protection scope of the present invention should be with claim institute
Subject to the range of restriction.
Claims (20)
1. a kind of visitor's group's superiority and inferiority predictor method characterized by comprising
Obtain the user data of the target device in preset time period；
The user data is handled according to timing, obtains data sequence relevant to timing；
The data sequence is input to the objective group's superiority and inferiority of progress in preset objective group's superiority and inferiority prediction model to estimate；
It obtains objective group's superiority and inferiority estimation results and exports.
2. visitor's group superiority and inferiority predictor method according to claim 1, which is characterized in that by the user data according to timing into
Row processing, obtains data sequence relevant to timing, comprising:
Each user identity information and its corresponding first label information are formed into a child user sequence；
According to the user data, all child user sequences are arranged according to timing, obtains user relevant to timing
Sequence, using user's sequence as the data sequence.
3. visitor's group's superiority and inferiority predictor method according to claim 2, which is characterized in that first label information includes following
It is at least one: age, gender, region, occupation, old and new customers's classification, record of bad behavior, third party's label.
4. visitor's group superiority and inferiority predictor method according to claim 1, which is characterized in that by the user data according to timing into
Row processing, obtains data sequence relevant to timing, comprising:
According to timing, the preset time period is divided into N number of sub- period；
Count corresponding second label information of user in each sub- period；
N number of sub- period and corresponding second label information of N number of sub- period are arranged according to timing,
Time series relevant to timing is obtained, using the time series as the data sequence.
5. visitor's group's superiority and inferiority predictor method according to claim 4, which is characterized in that second label information includes following
It is at least one: age bracket, sex ratio, occupation, old and new customers's classification, record of bad behavior.
6. visitor's group superiority and inferiority predictor method according to claim 1, which is characterized in that using LSTM deep learning algorithm or
The training of GRU deep learning algorithm obtains objective group's superiority and inferiority prediction model.
7. visitor's group's superiority and inferiority predictor method according to any one of claims 1 to 6, which is characterized in that exporting objective group's superiority and inferiority
After estimation results, further includes:
According to objective group's superiority and inferiority estimation results, the grade of the target device is determined, the grade is for assessing the target
The degree of risk of equipment.
8. visitor's group's superiority and inferiority predictor method according to claim 7, which is characterized in that in the grade for determining the target device
Later, further includes:
According to the grade of the target device, the position adjustment information of the target device is determined.
9. visitor's group's superiority and inferiority predictor method according to claim 7, which is characterized in that in the grade for determining the target device
Later, further includes:
When the grade for detecting the target device reaches predetermined level threshold value, warning reminding is exported.
10. a kind of visitor's group's superiority and inferiority estimating device characterized by comprising
Acquiring unit, suitable for obtaining the user data of the target device in preset time period；
Processing unit obtains data sequence relevant to timing suitable for handling the user data according to timing；
Input unit, it is pre- suitable for the data sequence is input to the objective group's superiority and inferiority of progress in preset objective group's superiority and inferiority prediction model
Estimate；
First output unit, suitable for obtaining objective group's superiority and inferiority estimation results and exporting.
11. visitor's group's superiority and inferiority estimating device according to claim 10, which is characterized in that the processing unit, being suitable for will be every
A user identity information and its corresponding first label information form a child user sequence；According to the user data, by institute
Some child user sequences are arranged according to timing, obtain user's sequence relevant to timing, using user's sequence as institute
State data sequence.
12. visitor's group superiority and inferiority estimating device according to claim 11, which is characterized in that first label information include with
Lower at least one: age, gender, region, occupation, old and new customers's classification, record of bad behavior, third party's label.
13. visitor's group superiority and inferiority estimating device according to claim 10, which is characterized in that the processing unit, be suitable for according to
The preset time period is divided into N number of sub- period by timing；Count corresponding second label of user in each sub- period
Information；N number of sub- period and corresponding second label information of N number of sub- period are arranged according to timing
Column, obtain time series relevant to timing, using the time series as the data sequence.
14. visitor's group superiority and inferiority estimating device according to claim 13, which is characterized in that second label information include with
Lower at least one: age bracket, sex ratio, occupation, old and new customers's classification, record of bad behavior.
15. visitor's group's superiority and inferiority estimating device according to claim 10, which is characterized in that further include: model construction unit is fitted
Objective group's superiority and inferiority prediction model is obtained in training using LSTM deep learning algorithm or GRU deep learning algorithm.
16. visitor's group's superiority and inferiority estimating device according to any one of claims 10 to 15, which is characterized in that further include: first
Determination unit is suitable for after exporting objective group's superiority and inferiority estimation results, according to objective group's superiority and inferiority estimation results, determines the target
The grade of equipment, the grade are used to assess the degree of risk of the target device.
17. visitor's group's superiority and inferiority estimating device according to claim 16, which is characterized in that further include: the second determination unit is fitted
In the grade according to the target device, the position adjustment information of the target device is determined.
18. visitor's group's superiority and inferiority estimating device according to claim 10, which is characterized in that further include: the second output unit is fitted
In after the grade for determining the target device, when the grade for detecting the target device reaches predetermined level threshold value,
Export warning reminding.
19. a kind of visitor's group superiority and inferiority estimating device, including memory and processor, being stored on the memory can be in the processing
The computer instruction run on device, which is characterized in that when the processor runs the computer instruction perform claim require 1 to
The step of 9 described in any item objective group's superiority and inferiority predictor methods.
20. a kind of computer readable storage medium, computer readable storage medium is non-volatile memory medium or non-transient deposits
Storage media is stored thereon with computer instruction, which is characterized in that perform claim requires 1 to 9 when the computer instruction is run
The step of objective group's superiority and inferiority predictor method described in one.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910232399.9A CN110060094A (en) | 2019-03-26 | 2019-03-26 | Objective group's superiority and inferiority predictor method and device, computer readable storage medium |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910232399.9A CN110060094A (en) | 2019-03-26 | 2019-03-26 | Objective group's superiority and inferiority predictor method and device, computer readable storage medium |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN110060094A true CN110060094A (en) | 2019-07-26 |
Family
ID=67316336
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201910232399.9A Pending CN110060094A (en) | 2019-03-26 | 2019-03-26 | Objective group's superiority and inferiority predictor method and device, computer readable storage medium |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN110060094A (en) |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107402921A (en) * | 2016-05-18 | 2017-11-28 | 阿里巴巴集团控股有限公司 | Identify event-order serie data processing method, the apparatus and system of user behavior |
| CN108734338A (en) * | 2018-04-24 | 2018-11-02 | 阿里巴巴集团控股有限公司 | Credit risk forecast method and device based on LSTM models |
| CN109002988A (en) * | 2018-07-18 | 2018-12-14 | 平安科技（深圳）有限公司 | Risk passenger method for predicting, device, computer equipment and storage medium |
| CN109190808A (en) * | 2018-08-15 | 2019-01-11 | 拍拍信数据服务（上海）有限公司 | User's behavior prediction method, apparatus, equipment and medium |
-
2019
- 2019-03-26 CN CN201910232399.9A patent/CN110060094A/en active Pending
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107402921A (en) * | 2016-05-18 | 2017-11-28 | 阿里巴巴集团控股有限公司 | Identify event-order serie data processing method, the apparatus and system of user behavior |
| CN108734338A (en) * | 2018-04-24 | 2018-11-02 | 阿里巴巴集团控股有限公司 | Credit risk forecast method and device based on LSTM models |
| CN109002988A (en) * | 2018-07-18 | 2018-12-14 | 平安科技（深圳）有限公司 | Risk passenger method for predicting, device, computer equipment and storage medium |
| CN109190808A (en) * | 2018-08-15 | 2019-01-11 | 拍拍信数据服务（上海）有限公司 | User's behavior prediction method, apparatus, equipment and medium |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US20190087737A1 (en) | Anomaly detection and automated analysis in systems based on fully masked weighted directed | |
| JP2022532443A (en) | Techniques for identifying skin color in images with uncontrolled lighting conditions | |
| CN109034078A (en) | Training method, age recognition methods and the relevant device of age identification model | |
| CN104516808B (en) | Data prediction device and method | |
| JP7007279B2 (en) | How and equipment to recommend questions | |
| US11632304B2 (en) | Methods and systems for characterizing computing system performance using peer-derived performance severity and symptom severity models | |
| US10411969B2 (en) | Backend resource costs for online service offerings | |
| US10394631B2 (en) | Anomaly detection and automated analysis using weighted directed graphs | |
| US20200097880A1 (en) | System and method for visualizing workflows in an entire management ecosystem | |
| US20150094983A1 (en) | Feature extraction method and apparatus for use in casual effect analysis | |
| CN109302301A (en) | A kind of appraisal procedure and device of business game | |
| CN105844107B (en) | Data processing method and device | |
| CN112633542A (en) | System performance index prediction method, device, server and storage medium | |
| CN108665421A (en) | The high light component removal device of facial image and method, storage medium product | |
| CN108428138B (en) | Customer survival rate analysis device and method based on customer clustering | |
| CN109977923A (en) | Driver's sex-screening method and system based on EEG signals | |
| US20160300102A1 (en) | Method of benchmarking media content based on viewer behavior | |
| CN111062602A (en) | Method, device and storage medium for enterprise credit risk assessment | |
| CN105045906B (en) | The predictor method and device of impression information clicking rate | |
| CN110060094A (en) | Objective group's superiority and inferiority predictor method and device, computer readable storage medium | |
| CN117035848A (en) | Instant lottery sales predicting method, device, equipment and medium | |
| CN110796379B (en) | Risk assessment method, device and equipment of business channel and storage medium | |
| US20210397993A1 (en) | Generalized machine learning application to estimate wholesale refined product price semi-elasticities | |
| CN111694872B (en) | Method and device for providing service handling data scheme | |
| US20160314063A1 (en) | System and method to infer call stacks from minimal sampled profile data |
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 | ||
| TA01 | Transfer of patent application right |
Effective date of registration: 20191225 Address after: 200062 floor 6, building 21, No. 958, Zhenbei Road, Putuo District, Shanghai Applicant after: Shanghai qiaopan Technology Co., Ltd Address before: No. 10 and No. 20, Lane 999, Dangui Road, China (Shanghai) Free Trade Pilot Area, Pudong New Area, Shanghai, 200120 Applicant before: SHANGHAI PPDAI FINANCE INFORMATION SERVICE CO., LTD. |
|
| TA01 | Transfer of patent application right | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20190726 |
|
| RJ01 | Rejection of invention patent application after publication |