CN110109901A - The method and apparatus for screening target object - Google Patents
The method and apparatus for screening target object Download PDFInfo
- Publication number
- CN110109901A CN110109901A CN201810105294.2A CN201810105294A CN110109901A CN 110109901 A CN110109901 A CN 110109901A CN 201810105294 A CN201810105294 A CN 201810105294A CN 110109901 A CN110109901 A CN 110109901A
- Authority
- CN
- China
- Prior art keywords
- data
- characteristic
- product
- association
- degree
- 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.)
- Granted
Links
- 238000012216 screening Methods 0.000 title claims abstract description 94
- 238000000034 method Methods 0.000 title claims abstract description 58
- 238000012545 processing Methods 0.000 claims abstract description 80
- 238000004458 analytical method Methods 0.000 claims abstract description 16
- 238000012549 training Methods 0.000 claims description 18
- 238000003860 storage Methods 0.000 claims description 15
- 230000002159 abnormal effect Effects 0.000 claims description 14
- 230000000737 periodic effect Effects 0.000 claims description 13
- 238000011156 evaluation Methods 0.000 claims description 11
- 238000009499 grossing Methods 0.000 claims description 10
- 238000004590 computer program Methods 0.000 claims description 7
- 238000005516 engineering process Methods 0.000 abstract description 3
- 239000000047 product Substances 0.000 description 92
- 230000006399 behavior Effects 0.000 description 78
- 238000010586 diagram Methods 0.000 description 16
- 230000006870 function Effects 0.000 description 15
- 238000004422 calculation algorithm Methods 0.000 description 13
- 230000006854 communication Effects 0.000 description 6
- 230000003542 behavioural effect Effects 0.000 description 5
- 238000004891 communication Methods 0.000 description 5
- 230000003993 interaction Effects 0.000 description 5
- 238000010276 construction Methods 0.000 description 4
- 230000008030 elimination Effects 0.000 description 4
- 238000003379 elimination reaction Methods 0.000 description 4
- 230000008901 benefit Effects 0.000 description 3
- 238000004364 calculation method Methods 0.000 description 3
- 230000000694 effects Effects 0.000 description 3
- 238000012986 modification Methods 0.000 description 3
- 230000004048 modification Effects 0.000 description 3
- 230000003287 optical effect Effects 0.000 description 3
- 230000008569 process Effects 0.000 description 3
- 239000013065 commercial product Substances 0.000 description 2
- 238000007405 data analysis Methods 0.000 description 2
- 238000013461 design Methods 0.000 description 2
- 238000009826 distribution Methods 0.000 description 2
- 230000005484 gravity Effects 0.000 description 2
- 238000010801 machine learning Methods 0.000 description 2
- 230000005291 magnetic effect Effects 0.000 description 2
- 238000004519 manufacturing process Methods 0.000 description 2
- 239000004065 semiconductor Substances 0.000 description 2
- 238000006467 substitution reaction Methods 0.000 description 2
- 241001269238 Data Species 0.000 description 1
- 241000406668 Loxodonta cyclotis Species 0.000 description 1
- 230000009471 action Effects 0.000 description 1
- 238000007635 classification algorithm Methods 0.000 description 1
- 238000004140 cleaning Methods 0.000 description 1
- 230000002596 correlated effect Effects 0.000 description 1
- 230000000875 corresponding effect Effects 0.000 description 1
- 230000005611 electricity Effects 0.000 description 1
- 239000000835 fiber Substances 0.000 description 1
- 210000003127 knee Anatomy 0.000 description 1
- 239000004973 liquid crystal related substance Substances 0.000 description 1
- 230000006855 networking Effects 0.000 description 1
- 239000013307 optical fiber Substances 0.000 description 1
- 238000005192 partition Methods 0.000 description 1
- 238000003672 processing method Methods 0.000 description 1
- 238000007873 sieving Methods 0.000 description 1
- 238000012706 support-vector machine Methods 0.000 description 1
- 230000002123 temporal effect Effects 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/20—Information retrieval; Database structures therefor; File system structures therefor of structured data, e.g. relational data
- G06F16/21—Design, administration or maintenance of databases
- G06F16/215—Improving data quality; Data cleansing, e.g. de-duplication, removing invalid entries or correcting typographical errors
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/90—Details of database functions independent of the retrieved data types
- G06F16/95—Retrieval from the web
- G06F16/953—Querying, e.g. by the use of web search engines
- G06F16/9535—Search customisation based on user profiles and personalisation
-
- 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
- G06Q30/06—Buying, selling or leasing transactions
- G06Q30/0601—Electronic shopping [e-shopping]
- G06Q30/0631—Item recommendations
Landscapes
- Engineering & Computer Science (AREA)
- Databases & Information Systems (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- Business, Economics & Management (AREA)
- General Physics & Mathematics (AREA)
- Accounting & Taxation (AREA)
- Finance (AREA)
- General Engineering & Computer Science (AREA)
- Data Mining & Analysis (AREA)
- General Business, Economics & Management (AREA)
- Strategic Management (AREA)
- Quality & Reliability (AREA)
- Marketing (AREA)
- Economics (AREA)
- Development Economics (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The invention discloses the method and apparatus of screening target object, are related to field of computer technology.One specific embodiment of this method includes: to carry out analysis to primary data to generate characteristic, and carry out characteristic processing to characteristic, and wherein primary data includes object information data and product information data；Utilize the degree of association of characteristic computing object and product after characteristic processing；According to the degree of association, screening meets the target object of preset condition.The embodiment carries out characteristic processing to the characteristic of generation, then the degree of association of computing object and product, and then screens satisfactory target object, can be improved the accuracy of screening target object.
Description
Technical field
The present invention relates to field of computer technology more particularly to a kind of method and apparatus for screening target object.
Background technique
In recent years, constantly bringing forth new ideas and improving with information technology, the purchasing and selling commodities of user, offer or receive service and
It is engaged in the also gradually interconnection networking comprehensively of other business activities modes, how rule is found from user behavior data, predicts user
Following demand, to be that big data applies the critical issue in precision marketing Products Show to the user needed.Its
In, Products Show may be considered to the user needed, target object is screened according to target product.
The prior art is screened in the method for target object, according to the artificially defined screening target object of staff's historical experience
Condition, for example, Products Show to all purchases or was browsed the objects of the product.
In realizing process of the present invention, at least there are the following problems in the prior art for inventor's discovery: sieving in the prior art
Select target object mode be it is subjective, target object is confirmed by the essential attribute of object and behavior, accuracy is low, Bu Nengda
To good screening effect, efficiency is lower, is easy to miss Best Times.
Summary of the invention
In view of this, the embodiment of the present invention provides a kind of method and apparatus for screening target object, to the characteristic of generation
According to characteristic processing is carried out, the then degree of association of computing object and product, and then the satisfactory target object of screening be can be improved
Screen the accuracy of target object.
To achieve the above object, according to an aspect of an embodiment of the present invention, a kind of side for screening target object is provided
Method.
A kind of method of screening target object of the embodiment of the present invention includes: to carry out analysis to primary data to generate characteristic
According to, and characteristic processing is carried out to the characteristic, the primary data includes object information data and product information data；Benefit
With the degree of association of characteristic computing object and product after characteristic processing；According to the degree of association, screening meets preset condition
Target object.
Optionally, before carrying out analysis to primary data and generating characteristic, the method also includes: obtain initial number
According to wherein the object information data includes: object behavior data and object base data, the product information data includes production
Judge valence mumber evidence and product basic data；The primary data is cleaned, to reject the abnormal data in primary data.
Optionally, primary data is carried out analysis to generate characteristic including: the difference according to dimension, to the initial number
The characteristic is established according to being combined.
Optionally, carrying out characteristic processing to the characteristic includes: to be carried out based on time cumulation attribute to characteristic
Attenuation processing；And/or it is based on carrying out smoothing techniques to characteristic with degree of association negative correlation attribute；And/or it is based on the period
Property attribute sliding-model control is carried out to characteristic, the characteristic that same period is belonged to after sliding-model control is closed
And.
Optionally, the characteristic after the characteristic processing includes: fisrt feature data and second feature data, wherein institute
Stating fisrt feature data and the second feature data is mutually isostructural data, but the data time of the fisrt feature data
Earlier than the data time of the second feature data.
It optionally, include: to utilize described the using the degree of association of characteristic computing object and product after characteristic processing
One characteristic trains computation model, wherein the input of the computation model is characteristic, output is object and product
The degree of association；Using the computation model, the degree of association of object and product in second feature data is calculated.
Optionally, according to the degree of association, the target object that screening meets preset condition includes: sieve according to target product
Select demand, screening is greater than the target object of preset threshold with the degree of association of the target product, wherein the target product is the
There is the product of screening requirements in two characteristics.
To achieve the above object, according to another aspect of an embodiment of the present invention, a kind of dress for screening target object is provided
It sets.
A kind of device of screening target object of the embodiment of the present invention includes: processing module, for carrying out to primary data
Analysis generate characteristic, and to the characteristic carry out characteristic processing, the primary data include object information data and
Product information data；Computing module, for the degree of association using characteristic computing object and product after characteristic processing；Screening
Module, for according to the degree of association, screening to meet the target object of preset condition.
Optionally, described device further include: module is obtained, for obtaining primary data, wherein the object information data
It include: object behavior data, object base data, the product information data includes product evaluation data, product basic data；
The primary data is cleaned, to reject the abnormal data in primary data.
Optionally, the processing module is also used to: according to the difference of dimension, being combined to the primary data and is established institute
State characteristic.
Optionally, the processing module is also used to: carrying out attenuation processing to characteristic based on time cumulation attribute；With/
Or, based on smoothing techniques are carried out to characteristic with degree of association negative correlation attribute；And/or based on periodic nature to feature
Data carry out sliding-model control, and the characteristic that same period is belonged to after sliding-model control is merged.
Optionally, the characteristic after the characteristic processing includes: fisrt feature data and second feature data, wherein institute
Stating fisrt feature data and the second feature data is mutually isostructural data, but the data time of the fisrt feature data
Earlier than the data time of the second feature data.
Optionally, the computing module is also used to: using fisrt feature data training computation model, wherein the meter
The input for calculating model is characteristic, output be object and product the degree of association；Using the computation model, second is calculated
The degree of association of object and product in characteristic.
Optionally, the screening module is also used to: according to the screening requirements of target product, screening and the target product
The degree of association is greater than the target object of preset threshold, wherein the target product is the production for having screening requirements in second feature data
Product.
To achieve the above object, according to an embodiment of the present invention in another aspect, providing a kind of electronic equipment.
The a kind of electronic equipment of the embodiment of the present invention includes: one or more processors；Storage device, for storing one
Or multiple programs, when one or more programs are executed by one or more processors, so that one or more processors realize this
The method of the screening target object of inventive embodiments.
To achieve the above object, another aspect according to an embodiment of the present invention, provides a kind of computer-readable medium.
A kind of computer-readable medium of the embodiment of the present invention, is stored thereon with computer program, and program is held by processor
The method of the screening target object of the embodiment of the present invention is realized when row.
One embodiment in foregoing invention has the following advantages that or the utility model has the advantages that the characteristic to generation carries out feature
It handles, then the degree of association of computing object and product, and then satisfactory target object is filtered out according to the degree of association, so as to
To improve the accuracy of screening target object；The primary data of acquisition is cleaned in the embodiment of the present invention, so as to incite somebody to action
Abnormal data elimination improves the accuracy of algorithm model；According to the difference of dimension in the embodiment of the present invention, primary data is carried out
Construction feature data are combined, so as to ensure the diversification of characteristic；From the time of characteristic in the embodiment of the present invention
Accumulative attribute carries out characteristic processing to characteristic with multiple angles such as degree of association negative correlation attribute and periodic natures, from
And it can be further improved the accuracy for training the characteristic of computation model；In the embodiment of the present invention after characteristic processing
Characteristic includes fisrt feature data and second feature data, their structure is identical, the time is different, so as to utilize the
One characteristic trains computation model, using trained computation model, calculates being associated with for object and product in second feature data
Degree；According to the screening requirements of target product in the embodiment of the present invention, the target object of preset condition is met using degree of association screening,
So as to which preset condition is arranged according to specific business need, the target object met is selected.
Further effect possessed by above-mentioned non-usual optional way adds hereinafter in conjunction with specific embodiment
With explanation.
Detailed description of the invention
Attached drawing for a better understanding of the present invention, does not constitute an undue limitation on the present invention.Wherein:
Fig. 1 is the schematic diagram of the key step of the method for screening target object according to an embodiment of the present invention；
Fig. 2 is adapted for realizing the signal of the main frame of the system of the method for the screening target object of the embodiment of the present invention
Figure；
Fig. 3 is the attenuation function curve synoptic diagram of the method for screening target object according to an embodiment of the present invention；
Fig. 4 and Fig. 5 is the schematic diagram of the method for screening target object according to an embodiment of the present invention；
Fig. 6 is the schematic diagram of the main flow of the method for screening target object according to an embodiment of the present invention；
Fig. 7 is the schematic diagram of the main modular of the device of screening target object according to an embodiment of the present invention；
Fig. 8 is that the embodiment of the present invention can be applied to exemplary system architecture figure therein；
Fig. 9 is adapted for the structural representation of the computer system for the terminal device or server of realizing the embodiment of the present invention
Figure.
Specific embodiment
Below in conjunction with attached drawing, an exemplary embodiment of the present invention will be described, including the various of the embodiment of the present invention
Details should think them only exemplary to help understanding.Therefore, those of ordinary skill in the art should recognize
It arrives, it can be with various changes and modifications are made to the embodiments described herein, without departing from scope and spirit of the present invention.Together
Sample, for clarity and conciseness, descriptions of well-known functions and structures are omitted from the following description.
Fig. 1 is the schematic diagram of the key step of the method for screening target object according to an embodiment of the present invention, such as Fig. 1 institute
Show, the method for the screening target object of the embodiment of the present invention mainly comprises the steps that
Step S101: analysis is carried out to primary data and generates characteristic, and characteristic processing is carried out to characteristic.Its
In, primary data may include object information data and product information data.The purpose of the present invention is according to target product, screening
Satisfactory target object out.It therefore, include object information data and product information data in the primary data of use.This hair
In bright in order to improve screening target object accuracy rate, characteristic processing has been carried out to the characteristic of generation.
Step S102: the degree of association of characteristic computing object and product after characteristic processing is utilized.In the present invention, object
It can be adjusted according to actual scene with the degree of association of product, for example, under the scene of the buying behavior of electric business platform prediction user,
The degree of association of object and product can be a possibility that user buys some product.
Step S103: according to the degree of association, screening meets the target object of preset condition.Step is utilized in the embodiment of the present invention
The degree of association calculated in S102, screening meet the target object of preset condition, pre- so as to be arranged according to specific business need
If condition, the target object met is selected.
In the embodiment of the present invention, before carrying out analysis to primary data and generating characteristic, the side of target object is screened
Method can also include: acquisition primary data；Primary data is cleaned, to reject the abnormal data in primary data.Wherein,
Object information data may include: object behavior data and object base data, and product information data may include product evaluation
Data and product basic data.Object behavior data include time of the act data and behavior classification data.Behavior classification includes pair
Image point such as hits, browses, pays close attention to and buys at the behaviors.Object base data are the primary attribute information datas of object, such as title, class
Not, the information such as registration date and grade.Product evaluation data refer to object to the evaluation information of product, such as difference comments rate, positive rating
With accumulative comment number etc..Product basic data is the basic information data of product, such as category, brand, price and attribute information.
The primary data of acquisition is cleaned in the embodiment of the present invention, abnormal data is removed, wherein abnormal data may include: superfluous
Data and the data of attribute information exception that the data of remaining record, crawler user generate etc..Redundant recording is to utilize technological means
The redundant recording generated when obtaining data.Crawler user refers to " network robot ", is a kind of automatic by writing program
Complete the behavior of the operation to website.Attribute information refers to the Information abnormity of product or the Information abnormity of object extremely, such as produces
Product selling at exorbitant prices, the registration date of object are B.C. 100 years etc..The primary data of acquisition is carried out in the embodiment of the present invention clear
It washes, so as to improve the accuracy of algorithm model for abnormal data elimination.
In the embodiment of the present invention, primary data is carried out analysis to generate characteristic may include: the difference according to dimension,
Primary data is combined and establishes characteristic.According to the difference of dimension in the embodiment of the present invention, group is carried out to primary data
Construction feature data are closed, so as to ensure the diversification of characteristic.
In the embodiment of the present invention, carrying out characteristic processing to characteristic may include: based on time cumulation attribute to feature
Data carry out attenuation processing；And/or it is based on carrying out smoothing techniques to characteristic with degree of association negative correlation attribute；And/or
Sliding-model control is carried out to characteristic based on periodic nature, the characteristic of same period will be belonged to after sliding-model control
It merges.In the present invention, it is based on time cumulation attribute, characteristic can be divided into time cumulation characteristic and non-temporal
More days accumulative totals of accumulative characteristic, time cumulation characteristic and object have a relationship, time cumulation characteristic can because
Length apart from current time is different and leads to weighted, therefore makees attenuation processing to this category feature.With degree of association negative correlation
Attribute refers to that certain characteristics when accumulated value is excessively high, can generate negatively correlated influence, it is therefore desirable to such spy to the degree of association
The peak fractions for levying data carry out smoothing techniques, eliminate peak value.Sliding-model control convenient for explanation and delineates reality in the present invention
Object behavior data.Periodic nature refers to the periodic behavior of object, the operation carried out in a period of time, is called primary
The characteristic for belonging to the same time is merged processing by behavior.From the time cumulation of characteristic in the embodiment of the present invention
Attribute carries out characteristic processing to characteristic with multiple angles such as degree of association negative correlation attribute and periodic natures, so as to
To further increase the accuracy for training the characteristic of computation model.
In the embodiment of the present invention, the characteristic after characteristic processing may include: fisrt feature data and second feature number
According to.The degree of association using characteristic computing object and product after characteristic processing may include: to be instructed using fisrt feature data
Practice computation model, wherein the input of trained computation model is characteristic, output be object and product the degree of association；
Using computation model, the degree of association of object and product in second feature data is calculated.In the present invention, fisrt feature data and second
Characteristic is mutually isostructural data, and fisrt feature data are for training computation model, and second feature data are for predicting
The behavior of object, therefore the data time of second feature data is closer to forecast date, that is to say, that fisrt feature data
Data time of the data time earlier than second feature data.
In the embodiment of the present invention, according to the degree of association, it may include: according to target that screening, which meets the target object of preset condition,
The screening requirements of product, screening are greater than the target object of preset threshold with the degree of association of target product.Wherein, target product
There is the product of screening requirements in two characteristics.In the present invention, preset condition can be the target object of screening preset quantity,
It can be the target object of the big Mr. Yu's threshold value of the degree of association of screening and target product, certainly, the present invention limits this without excessive,
Preset condition herein can be defined flexibly according to actual needs.
It,, will by taking the buying behavior of electric business platform prediction user as an example in subsequent descriptions of the invention in order to facilitate understanding
" product " is taken as " commodity ", and " object " is taken as " user ", and " product information data " are taken as " information of goods information data ", " object information
Data " are taken as " user information data ", and " product evaluation data " are taken as " commodity evaluation data ", and " product master data " is taken as
" commodity master data ", " object behavior data " are taken as " user behavior data ", and " object base data " are taken as " user base number
According to ", " fisrt feature data " are taken as " the characteristic table of building model ", and " second feature data " are taken as " prediction purchase intention
Mark sheet ", " degree of association of object and product " is taken as the intention values of commodity " user purchase " and is described in detail, certain " quotient
Product ", " user ", " information of goods information data ", " user information data ", " commodity evaluation data ", " commodity master data ", " user
Behavioral data ", " user base data ", " mark sheet of building model ", " mark sheet of prediction purchase " and " user purchase quotient
The intention value of product " is not used to be defined the protection scope of technical solution of the present invention, the present invention in " product ", " object ",
" product information data ", " object information data ", " product evaluation data ", " product master data ", " object behavior data ",
" object base data ", " fisrt feature data ", " second feature data " and " degree of association of object and product " can basis
Specific business scenario is adaptively adjusted.The intention value that user buys commodity refers to a possibility that user buys commodity, user
The more big then intention value of a possibility that buying some commodity is higher.
In the embodiment of the present invention, screening target pair can be implemented by the system of one screening target object of design
The method of elephant.Fig. 2 is adapted for realizing the signal of the main frame of the system of the method for the screening target object of the embodiment of the present invention
Figure.As shown in Fig. 2, the system for being adapted for carrying out the method for the screening target object of the embodiment of the present invention may include: data access
Unit, behavioural analysis unit, purchase predicting unit and data outputting unit.Data access unit from big data warehouse for obtaining
Access evidence, and the data of acquisition are cleaned.Behavioural analysis unit is used to be combined according to user behavior data and basic data
The mark sheet of the building model of various dimensions is established, and according to characteristic developing algorithm model.It buys predicting unit and is used for basis
The algorithm model of building calculates the probable value that user buys commodity.Data outputting unit is used for according to selected commodity set, defeated
It is likely to purchase the user of commodity in the set out.
In data access unit, the data of acquisition may include: user base data, commodity basic data, commodity evaluation
Data and user behavior data.User base data mainly from electric business platform website user's registration and consumption statistics,
It may include: age, gender, user gradation, registration date etc..Commodity basic data adopts pin mainly from the commodity of electric business platform
Data may include: attribute, category, brand of commodity etc..Commodity evaluate data mainly from user to the commodity bought
Whether comment and marking summarize data, may include: to have difference to comment, difference comments rate, add up to comment on number etc..User behavior data can wrap
Include: user clicks, browsing, concern, shopping cart is added, deletes the behaviors such as shopping cart, purchase classification and executes these behaviors flower
The time taken, i.e. time of the act.User clicks and browsing data bury point from the front end of electric business platform website and (bury and a little refer to use
Technological means acquires user behavior) data on flows that generates afterwards, the source for acquiring data on flows may include: WEB (World
Wide Web, global wide area network) website, cell phone application (Application, application program), mobile phone WAP (Wireless
Application Protocol, Wireless Application Protocol) website or third party's entrance etc..Other users behavioral data comes from
User is in electric business platform website to the operation data of commodity.
Data can be obtained by the task schedule in big data warehouse in the embodiment of the present invention.Wherein mould is calculated for constructing
Data of type and for predicting that the data of purchase intention are derived from data in different time periods, and the number for predicting purchase intention
According to period closer to for distribution time then can choose the 1st day extremely for example, the date of marketing is the 11st day
5th day data select the 6th day to the 10th day data for predicting purchase intention for constructing computation model.
After data access unit gets data from big data warehouse, the data of acquisition are cleaned.Wherein, right
The cleaning of data, which may include: that removal is duplicate, to be clicked, browses data；Remove influence of the behavior of crawler user to data；Clearly
The exceptional value of data is washed, for example, age of user is 1000 years old, commodity price is excessively high.
User behavior data and basic data are combined according to the difference of dimension in behavior analytical unit, generated special
Levy data.Then with Pearson correlation coefficient, (Pearson Correlation Coefficient is for measuring two numbers
Whether on one wire according to set face, it is used to measure the linear relationship between spacing variable) measure the characteristic and purchase generated
The linear relationship of behavior is bought, mark sheet of the positively related characteristic related with buying behavior as building model is chosen.
The present invention can be used for constructing the feature in the mark sheet of model so that electric business platform predicts user's buying behavior as an example
May include: user be averaged the behavior period (user be averaged the behavior period refer to user behavior operation time cycle), Yong Hudian
Hit the commodity used time, user browses commodity used time, commodity and is purchased total used time, user (category is to the nearest time interval of category
Refer to commodity belonging to classification), user browse the commodity used time seniority among brothers and sisters, behavior used time of the user to commodity, (the behavior used time referred to user
Execute click, browsing, the time for concern, shopping cart being added, deleting shopping cart, the behaviors such as purchase) in user to the commodity institute
In the used time seniority among brothers and sisters of category, (behavior number refers to that user executes click, browsing, concern, addition shopping to behavior number in user seven days
Vehicle, the number for deleting shopping cart, the behaviors such as purchase), behavior number, user enliven number of days (use in user's the last time behavior period
When being more than certain time to the behavior of commodity on the day of family, it is believed that user is active this day), user's commodity click the average used time
Seniority among brothers and sisters, user buy purchase number browsing number buying rate of commodity etc..
Other than above-mentioned important feature data, following secondary feature is also the important indicator for constructing computation model: characteristic day
Behavior number of days (characteristic day section can be promotion period section, holiday time section etc.), user are initial/last in section
The average value of each commodity behavior of the hourage of behavior range prediction day, user's category, user enliven the total number of days of history, user
Continuously log in that number of days, the average commodity number operated daily of user, the identical user of interaction time enlivens number of days sequence (interaction for the first time
Time, which refers to, has the operation of more than one user's process performing in some time, this time is considered interaction time), recently three times
Active period duration, interaction time section number, interaction time interval minimax mean value median sum it up, in user's login behavior,
It is clear to have seen that the specific gravity of current category commodity, user averagely enliven the number of minutes, user's active time center of gravity, user's last time daily
Look at time and browsing time first time poor, user gradation growth rate etc..
In behavior analytical unit, characteristic processing has been carried out to the mark sheet for constructing computation model.Wherein, characteristic processing
It may include: that attenuation processing is carried out based on mark sheet of the time cumulation attribute to building computation model.In the present invention, attenuation processing
It can be and attenuation function is added to time cumulation characteristic.Wherein, the formula of attenuation function can be with are as follows:, d is distance
The time difference of statistical forecast day, γ be attenuation coefficient (obtained by experience, the value of attenuation coefficient γ 0.3 to 0.95 it
Between).Fig. 3 is the attenuation function curve synoptic diagram of the method for screening target object according to an embodiment of the present invention.Fig. 3 shows γ
When being 0.7, the parameter distribution situation curve graph of daily attenuation function.By taking user click data as an example, by user click data plus
Statistical after entering attenuation function is as follows:
Sum=d1*1.0+d2*0.78+d3*0.562+d4*0.438+ ...
Wherein, di is the user clicks apart from the marketing date i-th day, and Sum is the click in user's statistical time
Measure aggregate value, i.e., final user's click feature data.Single feature and purchase in the embodiment of the present invention, after attenuation function is added
The Pearson correlation coefficient of behavior can promote 30% or more.
In behavior analytical unit, characteristic processing may include: to be based on degree of association negative correlation attribute to building computation model
Mark sheet carry out smoothing techniques.For example, in embodiments of the present invention, having in the behaviors aggregate values such as user clicks, browses logical
The data value for the behavior that robot clicks, browsing is not bought but is crossed, therefore smoothing techniques are done to this class behavior, reduces theirs
It influences.
In behavior analytical unit, characteristic processing may include: the mark sheet based on periodic nature to building computation model
Sliding-model control is carried out, the characteristic that same period is belonged to after sliding-model control is merged.Discretization in the present invention
Processing is convenient for explanation and delineates actual user behavior data.Periodic nature refers to the periodic behavior of user, when one section
The operation of interior progress is called a behavior, for example certain user has browsed some commodity, today between evening yesterday 18-20 point
8-11 point has browsed some commodity, therefore referred to as two behavior periods.In the present invention, user front and back can be operated twice
Operating time difference less than 45 minutes (think 45 minutes under normal circumstances for the website log out-of-service time, be also possible to other times,
This is not limited by the present invention) operation be merged together, the behavior more than 45 minutes is considered primary new user behavior week
Phase.Fig. 4 and Fig. 5 is the schematic diagram of the method for screening target object according to an embodiment of the present invention.Fig. 4 shows certain user one day
The behavior of user shown in Fig. 4 is merged into two behavior periods, includes a certain number of in each period by interior click behavior
User behavior, the schematic diagram after obtaining merging treatment shown in fig. 5.In the embodiment of the present invention, time in counting user behavior period
The characteristics such as behavior number, behavior period liveness in several, the recent behavior period can effective boosting algorithm model it is accurate
Rate.
XGBOOST can be used after carrying out characteristic processing to the mark sheet of building computation model in behavior analytical unit
The two classification algorithm training computation models of (algorithmic tool of user's machine learning).Rule of thumb, XGBOOST parameter is such as
Under:
'learning_rate':0.1,'n_estimators':1000,
'max_depth':8,'min_child_weight':3,
'subsample':1.0,'colsample_bytree':0.8,
'eta':0.1,'objective':'binary:logistic',
'gamma':0.1,'lambda':550.
When training computation model, the model whether user buys can be first established, user is resettled and whether buys and refer to
Determine the model of commodity, partition building computation model can promote the accuracy of prediction in the present invention.Certainly, can also make in this hair
With other two classification machine learning algorithm, such as SVM (support vector machine, a kind of engineering that can be trained
Learning method) etc., construct computation model.The mark sheet for constructing model is subjected to model training, the result after training is stored in model
In file.
In purchase predicting unit, the structure phase of the prediction data and the data of behavioural analysis building unit algorithm model of use
Together, but the time is different.It calls identical characteristic processing method to generate the mark sheet of prediction purchase intention prediction data, will predict
The model file generated in the mark sheet input behavior analytical unit of purchase intention is calculated, and purchase of the user to commodity is exported
Intention value and ranking.
In data outputting unit, the commodity set that can be provided by operation personnel is selected and is speculating in the time and may purchase
The user group of the commodity set is bought, as long as the user group for buying the commodity set refers to that the commodity in the set may be by user group
In some user purchase.Select the method for user group that can push away by threshold decision, such as by commodity in the present invention
The user for being greater than preset threshold to the intention value for the commodity is recommended, it can also be by selecting the user of preset quantity, such as
If by some commercial product recommending in commodity set give 100 users, then by commercial product recommending to before intention value ranking 100 user.
Fig. 6 is the schematic diagram of the main flow of the method for screening target object according to an embodiment of the present invention.It is flat with electric business
For platform, basic data be it is stable, choose prediction buy 6-10 days a few days ago user behavior data be used for developing algorithm model, select
The user behavior data for taking prediction to buy 1-5 days a few days ago is for predicting purchase intention, certainly, when the present invention also can choose other
Between section data, be only ensured that for the data of developing algorithm model and the data for predicting purchase intention must be the time
Uncrossed data.
As shown in fig. 6, the main flow of the method for screening target object may include: step S601, basic data is obtained
And user behavior data, and the data of acquisition are cleaned, remove abnormal data；Step S602, according to dimension difference, to base
The user behavior data that plinth data and prediction buy 6-10 days a few days ago is combined the spy for being used to construct model for establishing various dimensions
Levy table；Step S603 carries out attenuation processing to the feature related with the more days accumulative totals of user in mark sheet；Step S604,
Smoothing techniques are carried out to the feature related with user's click, browsing behavior in mark sheet；Step S605, in mark sheet
Feature related with the user behavior period merges processing；Step S606 utilizes place according to two sorting algorithms of XGBOOST
Mark sheet training computation model after reason；Step S607, according to dimension difference, to basic data and prediction purchase 1-5 days a few days ago
User behavior data be combined establish various dimensions for predicting the mark sheet of purchase intention, and call at identical feature
Reason method carries out characteristic processing to it；The mark sheet for being used to predict purchase intention is input in computation model, obtains by step S608
To the purchase intention value and ranking of user's corresponding product；Step S609 is selected according to the commodity set that operation personnel provides
Prediction purchase is in a few days likely to purchase the user of the commodity set.
In the present invention, step S603, step S604 and step S605 is to carry out characteristic processing to mark sheet, they hold
Row sequence can adjust according to the actual situation, and this is not limited by the present invention.
The technical solution of screening target object according to an embodiment of the present invention can be seen that the characteristic progress to generation
Characteristic processing, the then degree of association of computing object and product, and then the satisfactory target object of screening, so as to improve sieve
Select the accuracy of target object；The primary data of acquisition is cleaned in the embodiment of the present invention, so as to by abnormal data
It rejects, improves the accuracy of algorithm model；According to the difference of dimension in the embodiment of the present invention, building is combined to primary data
Characteristic, so as to ensure the diversification of characteristic；In the embodiment of the present invention from the time cumulation attribute of characteristic,
Characteristic processing is carried out to characteristic with multiple angles such as degree of association negative correlation attribute and periodic nature, so as into one
Step improves the accuracy for training the characteristic of computation model；Characteristic packet in the embodiment of the present invention after characteristic processing
Fisrt feature data and second feature data are included, their structure is identical, the time is different, so as to utilize fisrt feature data
Training computation model calculates the degree of association of object and product in second feature data using trained computation model；The present invention is real
The screening requirements in example according to target product are applied, meet the target object of preset condition, using degree of association screening so as to root
Preset condition is set according to specific business need, selects the target object met.
Fig. 7 is the schematic diagram of the main modular of the device of screening target object according to an embodiment of the present invention.Such as Fig. 7 institute
Show, the device 700 of the screening target object of the embodiment of the present invention mainly comprises the following modules: processing module 701, computing module
702 and screening module 703.
Wherein, processing module 701 can be used for carrying out primary data analysis generation characteristic, and carry out to characteristic
Characteristic processing.Primary data may include: object information data and product information data.Computing module 702 can be used for utilizing spy
The degree of association of sign treated characteristic computing object and product.Screening module 703 can be used for according to the degree of association, and screening meets
The target object of preset condition.
In the embodiment of the present invention, the device 700 for screening target object can also include: to obtain module (showing in figure).
Obtaining module can be used for obtaining primary data；Primary data is cleaned, to reject the abnormal data in primary data.Its
In, object information data may include: object behavior data and object base data, and product information data may include: product
Evaluate data and product basic data.
In the embodiment of the present invention, processing module 701 can also be used in: according to the difference of dimension, be combined to primary data
Establish characteristic.
In the embodiment of the present invention, processing module 701 can also be used in: be decayed based on time cumulation attribute to characteristic
Processing；And/or it is based on carrying out smoothing techniques to characteristic with degree of association negative correlation attribute；And/or based on periodically category
Property sliding-model control is carried out to characteristic, the characteristic that same period is belonged to after sliding-model control is merged.
In the embodiment of the present invention, the characteristic after characteristic processing may include: fisrt feature data and second feature number
According to.Wherein, fisrt feature data are mutually isostructural data with second feature data, but the data time of fisrt feature data is early
In the data time of second feature data.
In the embodiment of the present invention, computing module 702 can also be used in: utilize fisrt feature data training computation model；It utilizes
Computation model calculates the degree of association of object and product in second feature data.Wherein, the input of computation model is characteristic
According to, output be object and product the degree of association.
In the embodiment of the present invention, screening module 703 can also be used in: according to the screening requirements of target product, screening with it is described
The degree of association of target product is greater than the target object of preset threshold.Wherein, target product is to have screening to need in second feature data
The product asked.
From the above, it can be seen that carry out characteristic processing to the characteristic of generation, then computing object and product
The degree of association, and then satisfactory target object is screened, so as to improve the accuracy of screening target object；The present invention is implemented
The primary data of acquisition is cleaned in example, so as to improve the accuracy of algorithm model for abnormal data elimination；This hair
According to the difference of dimension in bright embodiment, construction feature data are combined to primary data, so as to ensure characteristic
Diversification；From the time cumulation attribute of characteristic and degree of association negative correlation attribute and periodicity in the embodiment of the present invention
Multiple angles such as attribute carry out characteristic processing to characteristic, so as to further increase the feature for training computation model
The accuracy of data；Characteristic in the embodiment of the present invention after characteristic processing includes fisrt feature data and second feature number
According to their structure is identical, the time is different, so as to utilize trained meter using fisrt feature data training computation model
Model is calculated, the degree of association of object and product in second feature data is calculated；According to the screening of target product in the embodiment of the present invention
Demand meets the target object of preset condition using degree of association screening, so as to which default item is arranged according to specific business need
Part selects the target object met.
Fig. 8 is shown can be using the method for the screening target object of the embodiment of the present invention or the device of screening target object
Exemplary system architecture 800.
As shown in figure 8, system architecture 800 may include terminal device 801,802,803, network 804 and server 805.
Network 804 between terminal device 801,802,803 and server 805 to provide the medium of communication link.Network 804 can be with
Including various connection types, such as wired, wireless communication link or fiber optic cables etc..
User can be used terminal device 801,802,803 and be interacted by network 804 with server 805, to receive or send out
Send message etc..Various telecommunication customer end applications, such as the application of shopping class, net can be installed on terminal device 801,802,803
(merely illustrative) such as the application of page browsing device, searching class application, instant messaging tools, mailbox client, social platform softwares.
Terminal device 801,802,803 can be the various electronic equipments with display screen and supported web page browsing, packet
Include but be not limited to smart phone, tablet computer, pocket computer on knee and desktop computer etc..
Server 805 can be to provide the server of various services, such as utilize terminal device 801,802,803 to user
The shopping class website browsed provides the back-stage management server (merely illustrative) supported.Back-stage management server can be to reception
To the data such as information query request analyze etc. processing, and by processing result (such as target push information, product letter
Breath -- merely illustrative) feed back to terminal device.
It should be noted that the method for screening target object provided by the embodiment of the present invention is generally held by server 805
Row, correspondingly, the device for screening target object is generally positioned in server 805.
It should be understood that the number of terminal device, network and server in Fig. 8 is only schematical.According to realization need
It wants, can have any number of terminal device, network and server.
Below with reference to Fig. 9, it illustrates the computer systems 900 for the terminal device for being suitable for being used to realize the embodiment of the present invention
Structural schematic diagram.Terminal device shown in Fig. 9 is only an example, function to the embodiment of the present invention and should not use model
Shroud carrys out any restrictions.
As shown in figure 9, computer system 900 includes central processing unit (CPU) 901, it can be read-only according to being stored in
Program in memory (ROM) 902 or be loaded into the program in random access storage device (RAM) 903 from storage section 908 and
Execute various movements appropriate and processing.In RAM 903, also it is stored with system 900 and operates required various programs and data.
CPU 901, ROM 902 and RAM 903 are connected with each other by bus 904.Input/output (I/O) interface 905 is also connected to always
Line 904.
I/O interface 905 is connected to lower component: the importation 906 including keyboard, mouse etc.；It is penetrated including such as cathode
The output par, c 907 of spool (CRT), liquid crystal display (LCD) etc. and loudspeaker etc.；Storage section 908 including hard disk etc.；
And the communications portion 909 of the network interface card including LAN card, modem etc..Communications portion 909 via such as because
The network of spy's net executes communication process.Driver 910 is also connected to I/O interface 905 as needed.Detachable media 911, such as
Disk, CD, magneto-optic disk, semiconductor memory etc. are mounted on as needed on driver 910, in order to read from thereon
Computer program be mounted into storage section 908 as needed.
Particularly, disclosed embodiment, the process described above with reference to flow chart may be implemented as counting according to the present invention
Calculation machine software program.For example, embodiment disclosed by the invention includes a kind of computer program product comprising be carried on computer
Computer program on readable medium, the computer program include the program code for method shown in execution flow chart.?
In such embodiment, which can be downloaded and installed from network by communications portion 909, and/or from can
Medium 911 is dismantled to be mounted.When the computer program is executed by central processing unit (CPU) 901, system of the invention is executed
The above-mentioned function of middle restriction.
It should be noted that computer-readable medium shown in the present invention can be computer-readable signal media or meter
Calculation machine readable storage medium storing program for executing either the two any combination.Computer readable storage medium for example can be --- but not
Be limited to --- electricity, magnetic, optical, electromagnetic, infrared ray or semiconductor system, device or device, or any above combination.Meter
The more specific example of calculation machine readable storage medium storing program for executing can include but is not limited to: have the electrical connection, just of one or more conducting wires
Taking formula computer disk, hard disk, random access storage device (RAM), read-only memory (ROM), erasable type may be programmed read-only storage
Device (EPROM or flash memory), optical fiber, portable compact disc read-only memory (CD-ROM), light storage device, magnetic memory device,
Or above-mentioned any appropriate combination.In the present invention, computer readable storage medium can be it is any include or storage journey
The tangible medium of sequence, the program can be commanded execution system, device or device use or in connection.And at this
In invention, computer-readable signal media may include in a base band or as carrier wave a part propagate data-signal,
Wherein carry computer-readable program code.The data-signal of this propagation can take various forms, including but unlimited
In electromagnetic signal, optical signal or above-mentioned any appropriate combination.Computer-readable signal media can also be that computer can
Any computer-readable medium other than storage medium is read, which can send, propagates or transmit and be used for
By the use of instruction execution system, device or device or program in connection.Include on computer-readable medium
Program code can transmit with any suitable medium, including but not limited to: wireless, electric wire, optical cable, RF etc. are above-mentioned
Any appropriate combination.
Flow chart and block diagram in attached drawing are illustrated according to the system of various embodiments of the invention, method and computer journey
The architecture, function and operation in the cards of sequence product.In this regard, each box in flowchart or block diagram can generation
A part of one module, program segment or code of table, a part of above-mentioned module, program segment or code include one or more
Executable instruction for implementing the specified logical function.It should also be noted that in some implementations as replacements, institute in box
The function of mark can also occur in a different order than that indicated in the drawings.For example, two boxes succeedingly indicated are practical
On can be basically executed in parallel, they can also be executed in the opposite order sometimes, and this depends on the function involved.Also it wants
It is noted that the combination of each box in block diagram or flow chart and the box in block diagram or flow chart, can use and execute rule
The dedicated hardware based systems of fixed functions or operations is realized, or can use the group of specialized hardware and computer instruction
It closes to realize.
Being described in module involved in the embodiment of the present invention can be realized by way of software, can also be by hard
The mode of part is realized.Described module also can be set in the processor, for example, can be described as: a kind of processor packet
Include processing module, computing module and screening module.Wherein, the title of these modules is not constituted under certain conditions to the module
The restriction of itself, for example, processing module is also described as " carrying out analysis to primary data and generating characteristic, and to spy
Levy the module that data carry out characteristic processing ".
As on the other hand, the present invention also provides a kind of computer-readable medium, which be can be
Included in equipment described in above-described embodiment；It is also possible to individualism, and without in the supplying equipment.Above-mentioned calculating
Machine readable medium carries one or more program, when said one or multiple programs are executed by the equipment, makes
Obtaining the equipment includes: to carry out analysis to primary data to generate characteristic, and carry out characteristic processing to characteristic, wherein initially
Data include object information data and product information data；Utilize the pass of characteristic computing object and product after characteristic processing
Connection degree；According to the degree of association, screening meets the target object of preset condition.
Technical solution according to an embodiment of the present invention carries out characteristic processing to the characteristic of generation, then computing object
With the degree of association of product, and then satisfactory target object is screened, so as to improve the accuracy of screening target object；This
The primary data of acquisition is cleaned in inventive embodiments, so as to improve the standard of algorithm model for abnormal data elimination
True property；According to the difference of dimension in the embodiment of the present invention, construction feature data are combined to primary data, so as to ensure
The diversification of characteristic；In the embodiment of the present invention from the time cumulation attribute of characteristic, with degree of association negative correlation attribute with
And multiple angles such as periodic nature carry out characteristic processing to characteristic, are used for training calculating mould so as to further increase
The accuracy of the characteristic of type；Characteristic in the embodiment of the present invention after characteristic processing includes fisrt feature data and second
Characteristic, their structure is identical, the time is different, so as to utilize instruction using fisrt feature data training computation model
Experienced computation model calculates the degree of association of object and product in second feature data；According to target product in the embodiment of the present invention
Screening requirements, meet the target object of preset condition using degree of association screening, so as to be arranged according to specific business need
Preset condition selects the target object met.
Above-mentioned specific embodiment, does not constitute a limitation on the scope of protection of the present invention.Those skilled in the art should be bright
It is white, design requirement and other factors are depended on, various modifications, combination, sub-portfolio and substitution can occur.It is any
Made modifications, equivalent substitutions and improvements etc. within the spirit and principles in the present invention, should be included in the scope of the present invention
Within.
Claims (16)
1. a kind of method for screening target object characterized by comprising
Analysis is carried out to primary data and generates characteristic, and characteristic processing, the primary data are carried out to the characteristic
Including object information data and product information data；
Utilize the degree of association of characteristic computing object and product after characteristic processing；
According to the degree of association, screening meets the target object of preset condition.
2. the method according to claim 1, wherein to primary data carry out analysis generate characteristic it
Before, the method also includes:
Primary data is obtained, wherein the object information data includes: object behavior data and object base data, the product
Information data includes product evaluation data and product basic data；
The primary data is cleaned, to reject the abnormal data in primary data.
3. the method according to claim 1, wherein to primary data carry out analysis generate characteristic include:
According to the difference of dimension, the primary data is combined and establishes the characteristic.
4. the method according to claim 1, wherein including: to characteristic progress characteristic processing
Attenuation processing is carried out to characteristic based on time cumulation attribute；And/or
Based on degree of association negative correlation attribute to characteristic carry out smoothing techniques；And/or
Sliding-model control is carried out to characteristic based on periodic nature, the feature of same period will be belonged to after sliding-model control
Data merge.
5. the method according to claim 1, wherein the characteristic after the characteristic processing includes: the first spy
Data and second feature data are levied, wherein the fisrt feature data are mutually isostructural data with the second feature data,
But the data time of the fisrt feature data is earlier than the data time of the second feature data.
6. according to the method described in claim 5, it is characterized in that, using the characteristic computing object after characteristic processing and producing
The degree of association of product includes:
Using fisrt feature data training computation model, wherein the input of the computation model is characteristic, output
Be object and product the degree of association；
Using the computation model, the degree of association of object and product in second feature data is calculated.
7. according to the method described in claim 6, it is characterized in that, screening meets the mesh of preset condition according to the degree of association
Marking object includes:
According to the screening requirements of target product, screening is greater than the target object of preset threshold with the degree of association of the target product,
Wherein, the target product is the product for having screening requirements in second feature data.
8. a kind of device for screening target object characterized by comprising
Processing module generates characteristic for carrying out analysis to primary data, and carries out characteristic processing to the characteristic,
The primary data includes object information data and product information data；
Computing module, for the degree of association using characteristic computing object and product after characteristic processing；
Screening module, for according to the degree of association, screening to meet the target object of preset condition.
9. device according to claim 8, which is characterized in that described device further include: module is obtained, it is initial for obtaining
Data, wherein the object information data includes: object behavior data, object base data, the product information data is included
Product evaluation data, product basic data；The primary data is cleaned, to reject the abnormal data in primary data.
10. device according to claim 8, which is characterized in that the processing module is also used to: according to the difference of dimension,
The primary data is combined and establishes the characteristic.
11. device according to claim 8, which is characterized in that the processing module is also used to:
Attenuation processing is carried out to characteristic based on time cumulation attribute；And/or
Based on degree of association negative correlation attribute to characteristic carry out smoothing techniques；And/or
Sliding-model control is carried out to characteristic based on periodic nature, the feature of same period will be belonged to after sliding-model control
Data merge.
12. device according to claim 8, which is characterized in that the characteristic after the characteristic processing includes: the first spy
Data and second feature data are levied, wherein the fisrt feature data are mutually isostructural data with the second feature data,
But the data time of the fisrt feature data is earlier than the data time of the second feature data.
13. device according to claim 12, which is characterized in that the computing module is also used to:
Using fisrt feature data training computation model, wherein the input of the computation model is characteristic, output
Be object and product the degree of association；
Using the computation model, the degree of association of object and product in second feature data is calculated.
14. device according to claim 13, which is characterized in that the screening module is also used to:
According to the screening requirements of target product, screening is greater than the target object of preset threshold with the degree of association of the target product,
Wherein, the target product is the product for having screening requirements in second feature data.
15. a kind of electronic equipment characterized by comprising
One or more processors；
Storage device, for storing one or more programs,
When one or more of programs are executed by one or more of processors, so that one or more of processors are real
The now method as described in any in claim 1-7.
16. a kind of computer-readable medium, is stored thereon with computer program, which is characterized in that described program is held by processor
The method as described in any in claim 1-7 is realized when row.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810105294.2A CN110109901B (en) | 2018-02-02 | 2018-02-02 | Method and device for screening target object |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810105294.2A CN110109901B (en) | 2018-02-02 | 2018-02-02 | Method and device for screening target object |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN110109901A true CN110109901A (en) | 2019-08-09 |
| CN110109901B CN110109901B (en) | 2024-05-24 |
Family
ID=67483554
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201810105294.2A Active CN110109901B (en) | 2018-02-02 | 2018-02-02 | Method and device for screening target object |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN110109901B (en) |
Cited By (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110764726A (en) * | 2019-10-18 | 2020-02-07 | 网易（杭州）网络有限公司 | Target object determination method and device, terminal device and storage medium |
| CN111241081A (en) * | 2020-01-09 | 2020-06-05 | 杭州涂鸦信息技术有限公司 | IOT platform data collection method and system, readable storage medium and computer equipment |
| CN111369324A (en) * | 2020-03-12 | 2020-07-03 | 苏州大学 | Target information determination method, device, equipment and readable storage medium |
| CN112488840A (en) * | 2019-09-12 | 2021-03-12 | 京东数字科技控股有限公司 | Information output method and device |
| CN113076346A (en) * | 2021-03-31 | 2021-07-06 | 北京深演智能科技股份有限公司 | Method and device for determining target object |
| CN116561188A (en) * | 2023-04-11 | 2023-08-08 | 上海御灵树网络科技有限公司 | Big data analysis method with screening function |
| CN112015726B (en) * | 2020-08-21 | 2024-04-12 | 广州欢网科技有限责任公司 | User activity prediction method, system and readable storage medium |
Citations (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102208087A (en) * | 2010-03-30 | 2011-10-05 | 株式会社日立制作所 | Information recommendation device |
| WO2015055094A1 (en) * | 2013-10-15 | 2015-04-23 | 北京百度网讯科技有限公司 | Method and device for providing screening conditions and method and device for searching |
| CN105989004A (en) * | 2015-01-27 | 2016-10-05 | 阿里巴巴集团控股有限公司 | Information releasing pretreatment method and device |
| CN106228386A (en) * | 2016-07-12 | 2016-12-14 | 腾讯科技（深圳）有限公司 | A kind of information-pushing method and device |
| CN106296257A (en) * | 2015-06-11 | 2017-01-04 | 苏宁云商集团股份有限公司 | A kind of fixation of advertisement position put-on method based on user behavior analysis and system |
| CN106649774A (en) * | 2016-12-27 | 2017-05-10 | 北京百度网讯科技有限公司 | Artificial intelligence-based object pushing method and apparatus |
| CN107220852A (en) * | 2017-05-26 | 2017-09-29 | 北京小度信息科技有限公司 | Method, device and server for determining target recommended user |
| CN107463675A (en) * | 2017-08-03 | 2017-12-12 | 北京京东尚科信息技术有限公司 | Data processing method and its system |
| CN107644036A (en) * | 2016-07-21 | 2018-01-30 | 阿里巴巴集团控股有限公司 | A kind of method, apparatus and system of data object push |
-
2018
- 2018-02-02 CN CN201810105294.2A patent/CN110109901B/en active Active
Patent Citations (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102208087A (en) * | 2010-03-30 | 2011-10-05 | 株式会社日立制作所 | Information recommendation device |
| WO2015055094A1 (en) * | 2013-10-15 | 2015-04-23 | 北京百度网讯科技有限公司 | Method and device for providing screening conditions and method and device for searching |
| CN105989004A (en) * | 2015-01-27 | 2016-10-05 | 阿里巴巴集团控股有限公司 | Information releasing pretreatment method and device |
| CN106296257A (en) * | 2015-06-11 | 2017-01-04 | 苏宁云商集团股份有限公司 | A kind of fixation of advertisement position put-on method based on user behavior analysis and system |
| CN106228386A (en) * | 2016-07-12 | 2016-12-14 | 腾讯科技（深圳）有限公司 | A kind of information-pushing method and device |
| CN107644036A (en) * | 2016-07-21 | 2018-01-30 | 阿里巴巴集团控股有限公司 | A kind of method, apparatus and system of data object push |
| CN106649774A (en) * | 2016-12-27 | 2017-05-10 | 北京百度网讯科技有限公司 | Artificial intelligence-based object pushing method and apparatus |
| CN107220852A (en) * | 2017-05-26 | 2017-09-29 | 北京小度信息科技有限公司 | Method, device and server for determining target recommended user |
| CN107463675A (en) * | 2017-08-03 | 2017-12-12 | 北京京东尚科信息技术有限公司 | Data processing method and its system |
Cited By (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112488840A (en) * | 2019-09-12 | 2021-03-12 | 京东数字科技控股有限公司 | Information output method and device |
| CN110764726A (en) * | 2019-10-18 | 2020-02-07 | 网易（杭州）网络有限公司 | Target object determination method and device, terminal device and storage medium |
| CN110764726B (en) * | 2019-10-18 | 2023-08-22 | 网易（杭州）网络有限公司 | Target object determination method and device, terminal equipment and storage medium |
| CN111241081A (en) * | 2020-01-09 | 2020-06-05 | 杭州涂鸦信息技术有限公司 | IOT platform data collection method and system, readable storage medium and computer equipment |
| CN111369324A (en) * | 2020-03-12 | 2020-07-03 | 苏州大学 | Target information determination method, device, equipment and readable storage medium |
| CN111369324B (en) * | 2020-03-12 | 2024-01-23 | 苏州大学 | Target information determining method, device, equipment and readable storage medium |
| CN112015726B (en) * | 2020-08-21 | 2024-04-12 | 广州欢网科技有限责任公司 | User activity prediction method, system and readable storage medium |
| CN113076346A (en) * | 2021-03-31 | 2021-07-06 | 北京深演智能科技股份有限公司 | Method and device for determining target object |
| CN116561188A (en) * | 2023-04-11 | 2023-08-08 | 上海御灵树网络科技有限公司 | Big data analysis method with screening function |
| CN116561188B (en) * | 2023-04-11 | 2024-06-11 | 北京联世传奇网络技术有限公司 | Big data analysis method with screening function |
Also Published As
| Publication number | Publication date |
|---|---|
| CN110109901B (en) | 2024-05-24 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN110109901A (en) | The method and apparatus for screening target object | |
| CN107609890A (en) | A kind of method and apparatus of order tracking | |
| CN110348894B (en) | Method and device for displaying resource advertisement and electronic equipment | |
| CN107423356A (en) | The processing method and processing device of evaluation information, computer-readable medium, electronic equipment | |
| CN108632311A (en) | Information-pushing method and device | |
| CN110135878B (en) | Method and device for determining sales price | |
| CN109087162A (en) | Data processing method, system, medium and calculating equipment | |
| CN110400184A (en) | Method and apparatus for generating information | |
| CN109961198A (en) | Related information generation method and device | |
| CN113065882A (en) | Commodity processing method and device and electronic equipment | |
| CN108074003A (en) | Predictive information method for pushing and device | |
| CN112927050A (en) | Method and device for determining financial product to be recommended, electronic equipment and storage medium | |
| CN113761350A (en) | Data recommendation method, related device and data recommendation system | |
| CN110866625A (en) | Promotion index information generation method and device | |
| CN109934523A (en) | Information processing method and device | |
| CN111798261A (en) | Information updating method and device | |
| CN112749323A (en) | Method and device for constructing user portrait | |
| CN109858756A (en) | A kind of service quality defect diagnostic method and device | |
| CN110838019A (en) | Method and device for determining trial supply distribution crowd | |
| CN113837648B (en) | Enterprise relevance analysis method, associated enterprise recommendation method and device | |
| CN115936764A (en) | Product promotion method and device | |
| CN110766488A (en) | Method and device for automatically determining theme scene | |
| CN114331544A (en) | Training method of advertisement conversion rate prediction model and related equipment | |
| CN113378043A (en) | User screening method and device | |
| CN114090873A (en) | Method, device, equipment and computer readable medium for matching 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 | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |