CN107220852A - Method, device and server for determining target recommended user - Google Patents

Method, device and server for determining target recommended user Download PDF

Info

Publication number
CN107220852A
CN107220852A CN201710385717.6A CN201710385717A CN107220852A CN 107220852 A CN107220852 A CN 107220852A CN 201710385717 A CN201710385717 A CN 201710385717A CN 107220852 A CN107220852 A CN 107220852A
Authority
CN
China
Prior art keywords
user
label
trade company
seed
candidate
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
Application number
CN201710385717.6A
Other languages
Chinese (zh)
Inventor
李泽中
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Beijing Xiaodu Information Technology Co Ltd
Original Assignee
Beijing Xiaodu Information Technology Co Ltd
Priority date (The priority date 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 date listed.)
Filing date
Publication date
Application filed by Beijing Xiaodu Information Technology Co Ltd filed Critical Beijing Xiaodu Information Technology Co Ltd
Priority to CN201710385717.6A priority Critical patent/CN107220852A/en
Publication of CN107220852A publication Critical patent/CN107220852A/en
Pending legal-status Critical Current

Links

Classifications

    • 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
    • G06Q30/06—Buying, selling or leasing transactions
    • G06Q30/0601—Electronic shopping [e-shopping]
    • G06Q30/0631—Item recommendations
    • 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
    • G06Q50/00—Systems or methods specially adapted for specific business sectors, e.g. utilities or tourism
    • G06Q50/10—Services
    • G06Q50/12—Hotels or restaurants
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L67/00—Network arrangements or protocols for supporting network services or applications
    • H04L67/50—Network services
    • H04L67/535—Tracking the activity of the user

Abstract

This application discloses for the method for determining target recommended user, device and server.One embodiment of this method includes：The user profile of user's collection in platform where acquisition target trade company, user's collection includes the seed user of target trade company and user to be recommended, and user profile is included with the History Order record for presetting the label in tag set；Concentrate the History Order record of each user to be counted according to label to user, obtain the tag attributes feature that user concentrates each user；The user similar to seed user is screened from user to be recommended based on tag attributes feature, the candidate user of target trade company is used as；The tag attributes feature of user profile and candidate user based on each seed user generates the recommendation degree information of each candidate user；The target recommended user of target trade company is determined from candidate user according to recommendation degree information.The embodiment improves the accuracy of target recommended user positioning.

Description

Method, device and server for determining target recommended user
Technical field
The application is related to field of computer technology, and in particular to internet data digging technology field, more particularly, to Determine method, device and the server of target recommended user.
Background technology
With the development of e-commerce technology, shopping platform is purchased on line for increasing user's selection.On line Shopping platform can obtain the user data of magnanimity, include base attribute information, order data, evaluation information, the logistics of user Information etc..Based on these user data, the portrait of each user can be constructed, including age of user, hobby, consumption energy Power, purchasing habits etc..
At present, shopping platform can draw a portrait according to user the potential user of the trade company filtered out recommending trade company on line. Conventional method meets the user of screening rule and recommended to define the screening rule of trade company, and filtering out.Such as a certain meal The screening rule in the Room is " pre-capita consumption valency is differed within 10% with head store, and the style of cooking liked matches, patronized 2 within nearly three months With other dining rooms of the style of cooking more than secondary ", then the user that portrait meets the screening rule can be filtered out, target recommended user is used as Recommend the dining room.But the above method lacks the quantization to user profile, it is impossible to provide sense of the user to trade company exactly emerging Interesting degree, and the screening rule quantity and coverage rate of Manual definition are limited, the accuracy that potential user recommends has to be hoisted.
The content of the invention
In order to solve one or more technical problems of above-mentioned background section, the embodiment of the present application, which is provided, to be used for really Set the goal method, device and the server of recommended user.
In a first aspect, the embodiment of the present application provides a kind of method for determining target recommended user, including：Obtain mesh The user profile of user's collection in platform where mark trade company, user's collection includes the seed user and use to be recommended of target trade company Family, user profile includes the History Order record with the label in default tag set；The history of each user is concentrated to user Order record is counted according to label, obtains the tag attributes feature that user concentrates each user；Based on tag attributes feature from The user similar to seed user is screened in user to be recommended, the candidate user of target trade company is used as；Based on each seed user The tag attributes feature of user profile and candidate user generates the recommendation degree information of each candidate user；According to recommendation degree information from time From the target recommended user that target trade company is determined in family.
In certain embodiments, above-mentioned tag attributes feature includes each member in label characteristics vector, label characteristics vector Element is the characteristic value corresponding to each label in default tag set；The History Order of each user is concentrated to record according to mark to user Label are counted, and obtain the tag attributes feature that user concentrates each user, including：The history of each user concentrated according to user Order record, is counted to the frequency that places an order that each user corresponds to each label, generates the label characteristics vector of each user.
In certain embodiments, above-mentioned History Order record includes the order generation time of each bar History Order record；Root The History Order record of each user concentrated according to user, is counted to the frequency that places an order that each user corresponds to each label, The label characteristics vector of each user is generated, including：To each label in tag set, label is corresponded to based on each user Each bar History Order record the order generation time, with default time decay factor determine each bar History Order record etc. The number of times that places an order is imitated, the equivalent number of times that places an order that each bar History Order is recorded is summed, user's placing an order corresponding to label is obtained The statistical result of the frequency；Correspond to the statistical result of the frequency that places an order of each label in default tag set based on user, generation is used The label characteristics vector at family.
In certain embodiments, the statistics knot of the frequency that places an order of each label in default tag set is corresponded to based on user Really, the label characteristics vector of generation user, including：The frequency that places an order of each label corresponded to user in default tag set Statistical result is normalized, using after normalized place an order frequency statistics result as label characteristics vector in it is each right Answer the characteristic value of element.
In certain embodiments, the use similar to seed user is screened from user to be recommended based on tag attributes feature Family, as the candidate user of target trade company, including：Calculate the vectorial label with each user to be recommended of label characteristics of seed user The similarity of characteristic vector, is used as seed user and the similarity of user to be recommended；Candidate is filtered out according to the sequence of similarity User.
In certain embodiments, the tag attributes feature generation of user profile and candidate user based on each seed user is each The recommendation degree information of candidate user, including：Closed according to the History Order of seed user record statistics seed user with target trade company The number of times that places an order of connection；The number of times that places an order that candidate user is associated with the similarity and seed user of seed user with target trade company Product as candidate user the score corresponding to seed user；The score for corresponding to each seed user to candidate user is summed, The score of candidate user is obtained, the recommendation degree information of candidate user is used as.
In certain embodiments, the step of above method also includes determining the seed user of target trade company, including：By user The History Order record that the History Order record of concentration is associated with target trade company and associated with target trade company meets preparatory condition User as target trade company seed user.
Second aspect, the embodiment of the present application provides a kind of device for being used to determine target recommended user, including：Obtain single Member, the user profile for the user's collection for being configured to obtain in the platform where target trade company, user's collection includes the kind of target trade company Child user and user to be recommended, user profile include the History Order record with the label in default tag set；Statistics is single Member, is configured to concentrate user the History Order record of each user to count according to label, obtains user and concentrate each user Tag attributes feature；Screening unit, is configured to screen and seed user from user to be recommended based on tag attributes feature Similar user, is used as the candidate user of target trade company；Generation unit, be configured to user profile based on each seed user and The tag attributes feature of candidate user generates the recommendation degree information of each candidate user；Determining unit, is configured to according to recommendation degree Information determines the target recommended user of target trade company from candidate user.
In certain embodiments, above-mentioned tag attributes feature includes each member in label characteristics vector, label characteristics vector Element is the characteristic value corresponding to each label in default tag set；Statistic unit is further configured to right as follows User concentrates the History Order record of each user to be counted according to label：The History Order of each user concentrated according to user Record, is counted to the frequency that places an order that each user corresponds to each label, generates the label characteristics vector of each user.
In certain embodiments, above-mentioned History Order record includes the order generation time of each bar History Order record；System Meter unit is further configured to as follows count the frequency that places an order that each user corresponds to each label, generates The label characteristics vector of each user：To each label in tag set, each bar for corresponding to label based on each user is gone through The order generation time of history order record, the equivalent lower single of each bar History Order record is determined with default time decay factor Number, sums to the equivalent number of times that places an order that each bar History Order is recorded, and obtains system of the user corresponding to the frequency that places an order of label Count result；Correspond to the statistical result of the frequency that places an order of each label in default tag set based on user, generate the label of user Characteristic vector.
In certain embodiments, statistic unit be further configured to generate as follows the label characteristics of user to Amount：The statistical result of the frequency that places an order of each label corresponded to user in default tag set is normalized, and will return Characteristic value of the frequency statistics result as each corresponding element in label characteristics vector that place an order after one change processing.
In certain embodiments, screening unit is further configured to screen similar to seed user as follows User, is used as the candidate user of target trade company：The label characteristics for calculating seed user are vectorial special with each user to be recommended label The similarity of vector is levied, seed user and the similarity of user to be recommended is used as；Candidate is filtered out according to the sequence of similarity to use Family.
In certain embodiments, generation unit is further configured to generate the recommendation of each candidate user as follows Spend information：The number of times that places an order that statistics seed user is associated with target trade company is recorded according to the History Order of seed user；By candidate The product for the number of times that places an order that user associates with the similarity and seed user of seed user with target trade company is as candidate user Corresponding to the score of seed user；The score for corresponding to each seed user to candidate user is summed, and obtains the score of candidate user, It is used as the recommendation degree information of candidate user.
In certain embodiments, said apparatus also includes the unit for determining the seed user of target trade company, is configured to：Will The History Order record that the History Order record that user concentrates is associated with target trade company and associated with target trade company meets default bar The user of part as target trade company seed user.
The third aspect, the embodiment of the present application provides a kind of server, including：One or more processors；Storage device, For storing one or more programs, when said one or multiple programs are by said one or multiple computing devices so that on State one or more processors and realize the above-mentioned method for determining target recommended user.
Method, device and the server for determining target recommended user that the application is provided, by obtaining target trade company The user profile for including recording with the History Order for presetting the label in tag set of user's collection on the platform of place, then Concentrate the History Order record of each user to be counted according to label to user, obtain the tag attributes feature of each user, then Screened for the similar user of seed user, used as the candidate of target trade company from user to be recommended based on tag attributes feature Family, the tag attributes feature of user profile and candidate user afterwards based on each seed user generates the recommendation degree of each candidate user Information, the target recommended user of target trade company is determined finally according to recommendation degree information from candidate user, can utilize seed User is analyzed the relevance between user and target trade company more comprehensively, exactly, and is effectively and reasonably quantified to be recommended User is to the interest-degree of target trade company, so as to improve the accuracy of target recommended user positioning.
Brief description of the drawings
Non-limiting example is described in detail with reference to what the following drawings was made by reading, other features, Objects and advantages will become more apparent upon：
Fig. 1 is that the application can apply to exemplary system architecture figure therein；
Fig. 2 is the flow chart for being used to determine one embodiment of the method for target recommended user according to the application；
Fig. 3 is the flow chart for being used to determine another embodiment of the method for target recommended user according to the application；
Fig. 4 is a kind of schematic stream for implementing scene for being used for shown in Fig. 3 determining the method for target recommended user Cheng Tu；
Fig. 5 is the effect diagram for being used to determine the method for target recommended user according to the application；
Fig. 6 is the structural representation for being used to determine one embodiment of the device of target recommended user of the application；
Fig. 7 is adapted for the structural representation of the computer system of the server for realizing the embodiment of the present application.
Embodiment
The application is described in further detail with reference to the accompanying drawings and examples.It is understood that this place is retouched The specific embodiment stated is used only for explaining related invention, rather than the restriction to the invention.It also should be noted that, in order to Be easy to description, illustrate only in accompanying drawing to about the related part of invention.
It should be noted that in the case where not conflicting, the feature in embodiment and embodiment in the application can phase Mutually combination.Describe the application in detail below with reference to the accompanying drawings and in conjunction with the embodiments.
Fig. 1 is shown can be using the method for being used to determine target recommended user of the application or for determining that target is recommended The exemplary system architecture 100 of the embodiment of the device of user.
As shown in figure 1, system architecture 100 can include trade company 110 use terminal device 101,102, user 120, 130th ... terminal device 103,104 ..., network 105 and server 106.Network 105 be used to terminal device 101,102, The 103rd, 104 ... the medium of communication link is provided between server 106.Network 105 can include various connection types, for example Wired, wireless communication link or fiber optic cables etc..
Trade company 110 can be interacted with using terminal equipment 101,102 by network 105 with server 106, to receive or send Message.Terminal device 101,102 can be provided with the application that the service provided with server 106 is associated, class application of for example doing shopping.
User 120,130 ... can also using terminal equipment 103,104 ... interacted by network 105 with server 106, To receive or send message.Terminal device 103,104 ... various telecommunication customer end applications, such as web page browsing can be installed Device application, the application of shopping class, social software etc..
Terminal device 101,102,103,104 .. can be with display screen and support the various of Data Communication in Computer Networks Electronic equipment, including but not limited to smart mobile phone, tablet personal computer, E-book reader, MP3 player (Moving Picture Experts Group Audio Layer III, dynamic image expert's compression standard audio aspect 3), MP4 (Moving Picture Experts Group Audio Layer IV, dynamic image expert's compression standard audio aspect 4) player, knee Mo(u)ld top half pocket computer and desktop computer etc..
Server 106 can be for the terminal device 101,102 of trade company 110 and user 120,130 ... terminal device 103rd, 104 the server of same data, services ... is provided, the background server for class application of for example, doing shopping.Shopping class application Background server can receive user 120,130 ... terminal device 103,104 ... request of data, and to request of data Sent after the processing such as being analyzed, being stored to the terminal device 101,102 of trade company 110, and by the terminal device 101 of trade company 110, 102 return feedback informations analyzed and processed after send to user 120,130 ... terminal device 103,104 ....
It should be noted that the method for being used to determine target recommended user that the embodiment of the present application is provided is general by servicing Device 106 is performed, correspondingly, for determining that the device of target recommended user is generally positioned in server 106.
It should be understood that the number of the terminal device, network and server in Fig. 1 is only schematical.According to realizing need Will, can have any number of terminal device, network and server.
With continued reference to Fig. 2, one embodiment for being used to determine the method for target recommended user according to the application is shown Flow 200.This is used for the method for determining target recommended user, comprises the following steps：
Step 201, the user profile of user's collection in the platform where acquisition target trade company.
In the present embodiment, the above-mentioned method for being used to determine target recommended user is run with electronic equipment thereon (for example Server 106 shown in Fig. 1) user profile of user's collection in platform where target trade company can be obtained.Herein, platform On can have multiple trade companies, target trade company is one or more of trade companies.In platform where user's collection includes target trade company All users, include seed user and the user to be recommended of target trade company.Seed user can be to have shadow to target trade company The user for ringing power, for example can be target trade company consuming frequency is higher, geographical position and the address location phase of target trade company Closely, the preferable user of prestige, or can be the user for performing the predetermined registration operation associated with target trade company, for example can be will The relevant information of target trade company shares the user into social platform.User to be recommended can be other in addition to seed user User, or the user in addition to user excessively single under target trade company.
Above-mentioned user profile includes the History Order record with the label in default tag set.Specifically, preset Tag set includes multiple default labels, can be the Commercial goods labelses (the vegetable label in such as restaurant) in trade company, trade company Geographical position label (such as commercial circle label), item price label.Every History Order records once placing an order for correspondence user Operation, then can in association store with above-mentioned default label by lower single operation of the user within a period of time, obtain above-mentioned History Order record with the label in default tag set.In the present embodiment, above-mentioned electronic equipment can be from local The History Order record of each user on platform is transferred in memory, can also receive each user's from remote equipment by network History Order is recorded.
In some optional implementations, above-mentioned user profile can also include the base attribute information of user, including The information such as age, sex, occupation, hobby, custom, the geographical position of user.These base attribute information can be by user actively Typing, such as age, geographical location information；The operation behavior data that user can also be combined in platform are obtained, for example can be by The time that places an order of user, the mode that places an order, geographical position determine the information such as the occupation of user, hobby.Ordered in the history for obtaining user , can be according to the mark of user, while obtaining the above-mentioned base attribute information of user during unirecord.
Step 202, concentrate the History Order record of each user to be counted according to label to user, obtain user and concentrate each The tag attributes feature of user.
Every History Order record of above-mentioned each user is respectively provided with label, in the present embodiment, can be according to label pair The History Order record of each user is counted, and regard statistical result as the tag attributes feature for corresponding to user.Herein, mark It can be the user property feature based on label to sign attributive character, in other words, that is, the user represented with tag feature Attribute.
Specifically, the tag attributes feature of user can be represented using various ways.For example, user A history Order record includes the tag attributes feature of 3, the order with label a and 1, the order, then user A with label b A3+b1, or a3＆b1 can be expressed as.
In some optional implementations of the present embodiment, the number of labels in above-mentioned default tag set is more, can By labeling, the label in tag set is divided into multiple label classifications, then for each label classification pair first The History Order record of user is counted, i.e., statistics belongs to the quantity of the History Order record of each label classification, as The tag attributes feature of each user.Follow-up operand can be so reduced, shortens operation time, improves and determines that target recommends to use The efficiency at family.
Step 203, the user similar to seed user is screened from user to be recommended based on tag attributes feature, is used as mesh Mark the candidate user of trade company.
, can be by the seed user of target trade company and tag attributes feature after the tag attributes feature of each user is obtained It is compared, filters out similar to the tag attributes feature of seed user to be recommended with the tag attributes feature of user to be recommended User as target trade company candidate user.
When the tag attributes feature of seed user and the tag attributes feature of user to be recommended are compared, it can adopt A variety of methods are used, such as can be by the tag attributes feature data mode of each user (such as side character string, vector, matrix Formula) represent, characteristic value or the characteristic strong point of tag attributes feature are then extracted, using characteristic value or characteristic strong point to two Tag attributes feature is matched, if the characteristic value of two tag attributes features or the matching degree at characteristic strong point exceed it is default Threshold value, then can determine that corresponding two users are similar, therefore deduce that one or more similar use of each seed user Family.
It should be noted that the similar users of the different seed users of target trade company can have overlapping, that is to say, that one User to be recommended can be the similar users of the different seed user of two or more, then the user to be recommended can be target business Family corresponds to the candidate user of the different seed user of two or more.
Step 204, the tag attributes feature of user profile and candidate user based on each seed user generates each candidate and used The recommendation degree information at family.
In the present embodiment, can be based on the seed similar to the candidate user to each candidate user of target trade company The user profile of user and the tag attributes feature of the candidate user generate its recommendation degree information, for above-mentioned electronic equipment according to Recommendation degree information determines target recommended user.Recommendation degree information can be used to indicate that candidate user to the potential of target trade company Interest-degree, or for represent by candidate user recommend after target trade company candidate user performed in target trade company browse, place an order, The possibility of the operations such as concern, evaluation.Specifically, recommendation degree information can include recommending index.Herein, each candidate user Tag attributes to the influence power and candidate user of target trade company in itself of recommendation degree information and the seed user similar to its Feature is related.Seed user is bigger to the influence power of target trade company, then the similar users of the seed user feel emerging to target trade company The possibility of interest is bigger, and the recommendation index of the corresponding candidate user of the seed user is higher.Also, candidate user and seed user Similarity it is higher, then candidate user possibility interested in target trade company is also bigger, then the recommendation index of the candidate user Also it is higher.In addition, if a certain candidate user is similar to multiple seed users of target trade company, the candidate user is to target trade company Interest-degree can be more than the only interest-degree of the candidate user similar to a seed user to target trade company.It therefore, it can comprehensive Close the recommendation index that factors above generates each candidate user.
Specifically, influence of the seed user to target trade company can be determined according to the user profile of seed user first The power factor, then by the tag attributes characteristic quantification of the corresponding candidate user of the seed user, by seed user to target trade company Effetiveness factor candidate user corresponding with the seed user tag attributes feature quantized value be multiplied can obtain the time Recommendation degree index from family relative to the seed user, afterwards by recommendation index of the candidate user to variant seed user It is added or is averaged, obtain the recommendation index of the candidate user, that is, generates the recommendation degree information of the candidate user.Further Ground, when quantifying the tag attributes feature of candidate user, it is possible to use candidate user and the similarity of seed user are used as its amount Change value；Or the tag attributes feature of candidate user and the tag attributes feature of target trade company can be matched, by matching degree It is used as quantized value.The tag attributes feature of goal trade company can be set in advance for trade company, can include target trade company Price label, type of merchandise label, geographical position label, scoring label etc..For example, being used as a certain dining room X of target trade company Default tag attributes feature includes " Sichuan cuisine, per capita 80-100 members, Zhong Guan-cun ", when candidate user Y tag attributes feature bag When including " spicy ", " Zhong Guan-cun commercial circle ", candidate user Y and dining room X to match angle value higher.
In some optional implementations, recommendation degree information can also be generated using methods such as machine learning. It can for example build and carry out target recommended user screening in recommendation degree information generation model, acquisition platform and carried out pushing away for correlation The feedback information of the trade company of activity is recommended, whether such as target recommended user carried out lower single operation and these business in these trade companies Sequence information, evaluation information of family association etc.；Similarity, the seed user of target recommended user and seed user can also be obtained The information such as History Order record data, utilize this information as training sample set, recommendation degree information generation model carried out Training, the recommendation degree information generation model after being trained.So, can by the user profile of the seed user of target trade company and Recommendation degree information generation model after the tag attributes feature input training of candidate user, draws the recommendation degree letter of candidate user Breath.
Step 205, the target recommended user of target trade company is determined from candidate user according to recommendation degree information.
The recommendation degree information that above-mentioned electronic equipment can be generated according to step 204 for example may be used come selection target recommended user To be ranked up according to each candidate user of recommendation degree exponent pair according to rule from high to low, selected and sorted is in preceding presetting digit capacity Candidate user as target trade company target recommended user；Or candidate use of the recommendation degree index more than predetermined threshold value can be selected Family as target trade company target recommended user.
In some optional implementations of the present embodiment, the use of user's collection in platform where obtaining target trade company It is above-mentioned to be used to determine that the method for target recommended user include the step for determining the seed user of target trade company after the information of family Suddenly, including：The History Order record that the History Order record that user is concentrated is associated with target trade company and associated with target trade company The user for meeting preparatory condition is used as the seed user of target trade company.That is, above-mentioned electronic equipment can be first from user Concentrate to filter out and record the user associated with target trade company with History Order, and judge the History Order record of user filtered out Whether preparatory condition is met, if, it is determined that the user filtered out is the seed user of target trade company.Herein, history is ordered It can have the order record occurred in target trade company during History Order is recorded that unirecord is associated with target trade company.Preparatory condition can To be that quantity on order is more than default quantity, or quantity on order in preset time is not small with default quantity etc..
Further, while seed user is determined, it may be determined that go out the user to be recommended of target trade company, such as may be used So that user to be concentrated to the other users in addition to seed user as user to be recommended, or it can filter out not in target trade company Single user was descended as user to be recommended.
The method for determining target recommended user that the above embodiments of the present application are provided, obtains target trade company place first The user profile of the History Order record included with the label in default tag set of user's collection on platform, then to Family concentrates the History Order record of each user to be counted according to label, obtains the tag attributes feature of each user, is then based on Tag attributes feature is screened from user to be recommended for the similar user of seed user, as the candidate user of target trade company, The tag attributes feature of user profile and candidate user afterwards based on each seed user generates the recommendation degree letter of each candidate user Breath, the target recommended user of target trade company is determined finally according to recommendation degree information from candidate user, can be used using seed Family is analyzed the relevance between user and target trade company more comprehensively, exactly, and effectively and reasonably quantifies use to be recommended Family is to the interest-degree of target trade company, so as to improve the accuracy of target recommended user positioning.
With continued reference to Fig. 3, it illustrates the flow of another embodiment of the method for determining target recommended user 300.This is used for the flow 300 for determining the method for target recommended user, comprises the following steps：
Step 301, the user profile of user's collection in the platform where acquisition target trade company.
In the present embodiment, for determining that the method for target recommended user runs electronic equipment (such as Fig. 1 institutes thereon The server 106 shown) can from it is local obtain or receive the platform that remote equipment is sent in user's collection user profile.Its In, all users in platform where user's collection includes target trade company include seed user and the user to be recommended of target trade company, User profile includes the History Order record with the label in default tag set.
Step 302, the History Order record of each user concentrated according to user, corresponds to each label to each user The frequency that places an order is counted, and generates the label characteristics vector of each user.
The tag attributes of each user are characterized as the user property feature based on label, in the present embodiment, can with to The mode of amount come represent the user property feature based on label, i.e. tag attributes feature can include label characteristics vector.Label Each element in characteristic vector is the characteristic value corresponding to each label in default tag set, is also to correspond to each mark The quantized value of the user property feature of label.
In the present embodiment, each user's that above-mentioned electronic equipment can be according to acquired in step 301 has default label The History Order record of label in set, is counted to the frequency that places an order that each user corresponds to each label, so as to generate The label characteristics vector of each user.Specifically, under above-mentioned history in unirecord, every record all has one or more marks Label, then can with the bar number of unirecord under the history with each label of counting user, or counting user with each label History under unifrequency (such as single time all over the world per x), be used as the corresponding characteristic value of each label in label characteristics vector.
In some optional implementations of the present embodiment, label characteristics vector is drawn in statistical history order record When, it is also contemplated that the interest-degree of user changes with time, such as order record in History Order record places an order Time gap current time farther out when, the importance that this order record is assessed user interest degree places an order the time less than another The importance of the order record nearer apart from current time.Specifically, in some optional implementations, above-mentioned steps 302 It can perform as follows：
Step 3021, to each label in tag set, each bar History Order of label is corresponded to based on each user The order generation time of record, the equivalent number of times that places an order of each bar History Order record is determined with default time decay factor, it is right The equivalent number of times that places an order of each bar History Order record is summed, and obtains statistics knot of the user corresponding to the frequency that places an order of label Really.
In the present embodiment, when being counted to user for the frequency that places an order of one of label, can according to Order generation time and the distance of current time that the History Order that family corresponds to the label is recorded, determine time decay factor pair Place an order the influence powers of frequency statistics, such as default time decay factor is α (0 ＜ α ＜ 1, such as α=0.95), History Order The order generation time gap current time of record t days, then influence power of the time decay therefore to the frequency statistics that place an order is αt, should The equivalent number of times that places an order of bar History Order record is αt.So, every History Order with same label is recorded etc. The number of times addition that places an order is imitated, that is, obtains statistical result of the user corresponding to the frequency that places an order of the label.
Step 3022, the statistical result of the frequency that places an order of each label in default tag set, generation are corresponded to based on user The label characteristics vector of user.
After the completion of the frequency statistics that place an order to each label, user can be corresponded to the frequency that places an order of each label Statistical result is used as the characteristic value for corresponding to each label in the label characteristics vector of the user.
In further implementation, above-mentioned steps 3022 can be realized in the following way：User is corresponded to pre- If the statistical result of the frequency that places an order of each label in tag set is normalized, by the lower single-frequency after normalized Secondary statistical result as label characteristics vector in each corresponding element characteristic value.That is, can be to the statistics of the frequency that places an order As a result it is normalized, regard the frequency statistics result that places an order after normalization as the feature of corresponding label in label characteristics vector Value.
With the user on platform that makes a reservation, as an example, this is made a reservation, default tag set includes multiple vegetable marks in platform Label, such as " braised in soy sauce ", " the meat clip Mo ", " vinegar-pepper ", " boiled dumpling ", the vegetable label of each vegetable of each trade company can by Obtained with the tag set.User i label characteristics vector viFollowing formula (1) such as can be used to represent：
Wherein,
count(dish_tagj) be j-th of vegetable label in tag set the frequency that places an order statistical result, j is just Integer and 1≤j≤K, K are the number of labels in default tag set.
So, can be by the tag attributes feature of each user with an one-dimensional label characteristics vector representation, and consider Influence of the time to user interest, so as to more accurately depict dynamic user's portrait, lifting user's is recent emerging Interest it is determined that influence power during target recommended user, can further lift being directed to for the target recommended user that determines Property.
Step 303, the user similar to seed user is screened from user to be recommended based on label characteristics vector, is used as mesh Mark the candidate user of trade company.
In above-mentioned steps 302, the tag attributes feature of each user is represented using label characteristics vector, then between user Similarity can be represented by the similarity of its characteristic vector.In the present embodiment, can calculate the label characteristics of seed user to Amount and the similarity of the label characteristics vector of each user to be recommended, as seed user and the similarity of user to be recommended, and then It can be shone according to the sequence of similarity and select candidate user.Alternatively, the similarity of two label characteristics vectors can use Europe The existing similarity calculating methods such as family name's distance, cosine similarity, Pearson correlation coefficients are drawn.In screening candidate user When, the similar users that position is preset before sequencing of similarity can be regard as candidate user.
In some optional implementations of the present embodiment, in order to reduce screening scope, can calculate similarity it It is preceding to treat recommended user using some anticipation conditions and filtered, geographical position can for example be differed with seed user too far, Or average order value differs excessive user with seed user and rejected.Afterwards in the user to be recommended for reducing screening scope Candidate user is filtered out by sequencing of similarity.Amount of calculation can so be reduced, accelerate arithmetic speed, and then be rapidly mesh Mark the selection result that trade company provides target recommended user.
Step 304, the tag attributes feature of user profile and candidate user based on each seed user generates each candidate and used The recommendation degree information at family.
In the present embodiment, the recommendation degree information of each candidate user and the seed user similar to its are to target trade company The tag attributes feature of influence power and candidate user in itself is related.Here tag attributes feature includes label characteristics vector.
The similarity of seed user and each candidate user can be calculated in above-mentioned steps 303.Herein, Hou Xuanyong The recommendation degree information at family can be represented using the score of candidate user.Pushing away for each candidate user can be generated in the following way Degree of recommending information：The number of times that places an order that statistics seed user is associated with target trade company is recorded according to the History Order of seed user first, The product for the number of times that places an order that candidate user is associated with the similarity and seed user of seed user with target trade company afterwards as The score corresponding to seed user of candidate user, the score that each seed user is corresponded to candidate user is summed, and obtains candidate The score of user, is used as the recommendation degree information of candidate user.
Specifically, the score of each candidate user can be calculated according to following formula (3)：
Wherein, uqFor q-th of seed user of target trade company；UpFor p-th of candidate user；score(Up) waited for p-th From the score at family；order_num(uq) for q-th of seed user in preset time period (such as in one month) in target business The number of times that family places an order；If p-th of candidate user UpTo be filtered out in step 303 and q-th of seed user uqSimilar user, Then sim (Up, uq) it is p-th of candidate user U that step 303 is calculatedpWith q-th of seed user u of target trade companyqBetween Similarity, otherwise sim (Up, uq)=0.Q be and candidate user UpThe quantity of the seed user of similar target trade company.
Utilize above-mentioned formula (3), it can be deduced that the score of each candidate user, from formula (3) as can be seen that working as candidate user For multiple seed users similar users when, its score can add up, namely when candidate user is similar to multiple seed users, The candidate user is higher to the potential interest-degree of target trade company.
Step 305, the target recommended user of target trade company is determined from candidate user according to recommendation degree information.
, can be according to the row of score after recommendation degree information of the score of each candidate user as each candidate user is drawn The threshold value whether sequence or score exceed setting determines target recommended user.For example can be to candidate user according to score progress descending Sequence, the individual target recommended user as target trade company of predetermined number before taking.
Step 301, step 305 in above method flow respectively with the step 201 in previous embodiment, step 205 phase Together, here is omitted.
From figure 3, it can be seen that compared with the corresponding embodiments of Fig. 2, being used in the present embodiment determines that target recommends use The flow 300 of the method at family utilizes label characteristics by being label characteristics vector by the tag attributes characteristic quantification of user Similarity between vector characterizes the similarity between seed user and user to be recommended, can obtain more accurately target recommendation The selection result of user.
In some optional implementations of the present embodiment, the above-mentioned method flow for being used to determine target recommended user 300 are additionally may included in after the user profile of user's collection in the platform got where target trade company, concentrate true from user Set the goal trade company seed user the step of, the step specifically includes the History Order record for concentrating user and closed with target trade company The user that the History Order record for joining and being associated with target trade company meets preparatory condition is used as the seed user of target trade company.For example Can descending single and be more than the user of preset value in the lower unifrequency of target trade company and be used as the kind of target trade company in target trade company Child user.
Fig. 4 shows that being used for shown in Fig. 3 determines a kind of signal for implementing scene of the method for target recommended user Property flow chart.As shown in figure 4, first, in step 401, obtaining the user profile in platform, user profile here includes going through History order record, every History Order record has corresponding label；Then, in step 402, mesh is determined according to user profile The seed user of trade company is marked, while the user to be recommended of target trade company can be determined；Then, in step 403, according to above-mentioned The label characteristics vector of each user in user profile calculating platform；Afterwards, in step 404, the kind determined according to step 402 Child user and user to be recommended, calculate vectorial similar of the vectorial label characteristics with user to be recommended of label characteristics of seed user Degree, M are used as candidate user before sequencing of similarity；Afterwards, in step 405, the score of candidate user is calculated, is needed exist for Consider the importance degree and candidate user and the similarity of seed user of seed user；Descending sort finally is carried out to score, The candidate user of selected and sorted top N is target recommended user.Herein, M, N be can positive integer set in advance.
With further reference to Fig. 5, it illustrates the effect for being used to determine the method for target recommended user according to the application Fruit schematic diagram, namely show the effect diagram of an application scenarios of method shown in Fig. 2 or Fig. 3.
As shown in figure 5, on the platform that XX takes out, target trade company " * * chop house " selects " drawing new " service at vendor end Afterwards, the back-end server that XX takes out can obtain the History Order record of all users on line, and be recorded according to History Order In every time lower single user order the label of vegetable and determine the tag attributes feature of each user, alternatively, can also from it is useful Filter out the seed user of trade company " * * chop house " in family in advance, such as in user H, and other users based on user H on line Similar user is selected, the recommendation degree information of similar user is generated, then determines that target draws new use according to recommendation degree information Family, including user A, user B, user C, user D etc., afterwards the result newly screened will be drawn to push in the client of target trade company And present.Score and some detail informations, the taste preference of such as user, commercial circle, phase that target draws new user can also be provided Order record of pass etc., new the selection result is accurately drawn so that trade company end can be obtained, and carry out targetedly commercial product recommending or Action message is pushed.
With further reference to Fig. 6, as the realization to method shown in above-mentioned each figure, it is used to determine mesh this application provides one kind One embodiment of the device of recommended user is marked, the device embodiment is corresponding with the embodiment of the method shown in Fig. 6, device tool Body can apply in various electronic equipments.
As shown in fig. 6, the device 600 for determining target recommended user of the present embodiment includes：Acquiring unit 601, system Count unit 602, screening unit 603, generation unit 604 and determining unit 605.Wherein, acquiring unit 601 is configured to obtain The user profile of user's collection in platform where target trade company, wherein, user's collection includes the seed user of target trade company and treated Recommended user, user profile includes the History Order record with the label in default tag set；The configuration of statistic unit 602 is used In concentrating the History Order record of each user to be counted according to label to user, the tag attributes that user concentrates each user are obtained Feature；Screening unit 603 is configured to screen the use similar to seed user from user to be recommended based on tag attributes feature Family, is used as the candidate user of target trade company；Generation unit 604 is configured to user profile and candidate's use based on each seed user The tag attributes feature at family generates the recommendation degree information of each candidate user；Determining unit 605 is configured to according to recommendation degree information The target recommended user of target trade company is determined from candidate user.
In the present embodiment, acquiring unit 601 can transfer out the user profile of each user on platform from local storage, Or can by wired connection mode or radio connection each user on receiving platform from other servers user Information.Here user profile can also include the base attribute information such as age, geographical position, the occupation of user.
The user profile for each user that statistic unit 602 can be obtained to acquiring unit 601 is counted based on label, The corresponding statistics of each label is obtained, so as to draw the tag attributes feature of each user.
The tag attributes feature that screening unit 603 can count obtained each user according to statistic unit 602 is used seed Family and user to be recommended are compared, and filter out the user similar to seed user, are used as the candidate user of target trade company.
The candidate user filtered out to screening unit 603, generation unit 604 can utilize the user of corresponding seed user Information determines importance degree of the seed user to target trade company, and based on candidate user label characteristics attribute and target trade company The degree of association between attribute or the degree of similarity between seed user generate the recommendation degree information of candidate user.Pushing away here Degree of recommending information can represent potential interest-degree of the candidate user to target trade company.
Determining unit 605 can select the higher candidate user of the potential interest-degree to target trade company according to recommendation degree information It is used as the target recommended user of target trade company.Alternatively, if the recommendation degree Information Pull that generation unit 604 is generated recommends index Represent, then the candidate user of position can be preset before selected and sorted according to recommending the descending of index to be ranked up each candidate user For the target recommended user of target trade company.
In certain embodiments, above-mentioned tag attributes feature includes each in label characteristics vector, label characteristics vector Element is the characteristic value corresponding to each label in default tag set；Then statistic unit 602 can further be configured to by The History Order record that each user is concentrated to user according to following manner is counted：The history of each user concentrated according to user Order record, is counted to the frequency that places an order that each user corresponds to each label, generates the label characteristics vector of each user. Herein, the characteristic value of corresponding label is the statistics knot that user corresponds to the frequency that places an order of the label in label characteristics vector Really.
In a further embodiment, when above-mentioned History Order record includes the order generation of each bar History Order record Between；Then statistic unit 602 can further be configured to as follows correspond to each user the lower single-frequency of each label It is secondary to be counted, so as to generate the label characteristics vector of each user：To each label in tag set, based on each user The time is generated corresponding to the order of each bar History Order record of label, determines that each bar history is ordered with default time decay factor The equivalent number of times that places an order of unirecord, sums to the equivalent number of times that places an order that each bar History Order is recorded, obtains user and correspond to The statistical result of the frequency that places an order of label；Correspond to the statistics knot of the frequency that places an order of each label in default tag set based on user Really, the label characteristics vector of generation user.
Further, above-mentioned statistic unit 602 can further be configured to correspond in default tag set user The statistical result of the frequency that places an order of each label be normalized, the frequency statistics result that places an order after normalized is made For the characteristic value of each corresponding element in label characteristics vector.
Specifically, time decay factor can be set as α (0 ＜ α ＜ 1, such as α=0.95), History Order record ordered Single generation time gap current time t days, then influence power of the time decay therefore to the frequency statistics that place an order is αt, this history orders The equivalent number of times that places an order of unirecord is αt.User i label characteristics vector viIt can use as following formula is represented：
Wherein,
count(dish_tagj) be j-th of label in tag set the frequency that places an order statistical result, K is pre- bidding Number of labels in label set.
In a further embodiment, screening unit 603 is further configured to screen as follows and used with seed The similar user in family is used as the candidate user of target trade company：The label characteristics for calculating seed user are vectorial with each user to be recommended Label characteristics vector similarity, be used as the similarity of seed user and the user to be recommended；According to the sequence of similarity Filter out candidate user.Here similarity can be calculated using modes such as Euclidean distance, cosine similarity, Pearson's coefficients, Similarity is higher, then candidate user is bigger to the interest-degree of target trade company.
Further, above-mentioned generation unit 604 is further configured to generate pushing away for each candidate user as follows Degree of recommending information：The number of times that places an order that statistics seed user is associated with target trade company is recorded according to the History Order of seed user；It will wait The product of the number of times that places an order associated from family with the similarity and seed user of seed user with target trade company is as candidate user The score corresponding to seed user；The score for corresponding to each seed user to candidate user is summed, and obtains obtaining for candidate user Point, it is used as the recommendation degree information of candidate user.Specific calculation is, each candidate user UpScore score (Up) be：
Wherein, uqFor q-th of seed user of target trade company；UpFor p-th of candidate user；order_num(uq) it is q The number of times that individual seed user places an order in preset time period (such as in one month) in target trade company；If p-th of candidate user Up To be filtered out in step 303 and q-th of seed user uqSimilar user, then sim (Up, uq) calculated for step 303 P-th of candidate user UpWith q-th of seed user u of target trade companyqBetween similarity, otherwise sim (Up, uq)=0.Q be with Candidate user UpThe quantity of the seed user of similar target trade company.
In certain embodiments, device 600 can also include the unit for being configured to determine the seed user of target trade company, It is configured specifically for：The History Order that the History Order record that user is concentrated is associated with target trade company and associated with target trade company The user that record meets preparatory condition is used as the seed user of target trade company.It is configured to determine the seed user of target trade company Unit can extract History Order and record user related to target trade company and meeting above-mentioned preparatory condition, such as in target Trade company descended singly, and the number of times that places an order exceedes the user of predetermined threshold value, is used as the seed user of target trade company.Alternatively, while can To regard other users as user to be recommended as user to be recommended, or using user excessively not single under target trade company.
It should be appreciated that all units described in device 600 and each step phase in the method referring to figs. 2 and 3 description Correspondence.Thus, the unit that the operation and feature described above with respect to method is equally applicable to device 600 and wherein included, herein Repeat no more.
The device 600 for determining target recommended user that the embodiment of the present application is provided, utilizes the tag attributes of each user Feature Selection goes out the user similar to the seed user of target trade company, and based on the tag attributes of the user similar to seed user Feature and the user profile of seed user generate the recommendation degree information of the user similar to seed user, according to recommendation degree information come Target recommended user is determined, can effectively and reasonably quantify user and the interest-degree of target trade company, lifting target recommended user are determined The accuracy of position.
Below with reference to Fig. 7, it illustrates suitable for the computer system 700 for the server of realizing the embodiment of the present application Structural representation.Server shown in Fig. 7 is only an example, to the function of the embodiment of the present application and should not use range band Carry out any limitation.
As shown in fig. 7, computer system 700 includes CPU (CPU) 701, it can be read-only according to being stored in Program in memory (ROM) 702 or be loaded into program in random access storage device (RAM) 703 from storage part 708 and Perform various appropriate actions and processing.In RAM 703, the system that is also stored with 700 operates required various programs and data. CPU 701, ROM 702 and RAM 703 are connected with each other by bus 704.Input/output (I/O) interface 705 is also connected to always Line 704.
I/O interfaces 705 are connected to lower component：Importation 706 including keyboard, mouse etc.；Penetrated including such as negative electrode The output par, c 707 of spool (CRT), liquid crystal display (LCD) etc. and loudspeaker etc.；Storage part 708 including hard disk etc.； And the communications portion 709 of the NIC including LAN card, modem etc..Communications portion 709 via such as because The network of spy's net performs communication process.Driver 710 is also according to needing to be connected to I/O interfaces 705.Detachable media 711, such as Disk, CD, magneto-optic disk, semiconductor memory etc., are arranged on driver 710, in order to read from it as needed Computer program be mounted into as needed storage part 708.
Especially, in accordance with an embodiment of the present disclosure, the process described above with reference to flow chart may be implemented as computer Software program.For example, embodiment of the disclosure includes a kind of computer program product, it includes being carried on computer-readable medium On computer program, the computer program includes the program code for being used to perform above-mentioned flow chart 2 or the method shown in Fig. 3. In such embodiments, the computer program can be downloaded and installed by communications portion 709 from network, and/or from Detachable media 711 is mounted.When the computer program is performed by CPU (CPU) 701, the side of the application is performed The above-mentioned functions limited in method.It should be noted that computer-readable medium described herein can be computer-readable letter Number medium or computer-readable recording medium either the two any combination.Computer-readable recording medium for example may be used System, device or the device of --- but being not limited to --- electricity, magnetic, optical, electromagnetic, infrared ray or semiconductor to be, or it is any with On combination.The more specifically example of computer-readable recording medium can include but is not limited to：With one or more wires Electrical connection, portable computer diskette, hard disk, random access storage device (RAM), read-only storage (ROM), erasable type can compile Journey read-only storage (EPROM or flash memory), optical fiber, portable compact disc read-only storage (CD-ROM), light storage device, magnetic Memory device or above-mentioned any appropriate combination.In this application, computer-readable recording medium can be any includes Or the tangible medium of storage program, the program can be commanded execution system, device or device using or in connection make With.And in this application, computer-readable signal media can be included in a base band or as carrier wave part propagation Data-signal, wherein carrying computer-readable program code.The data-signal of this propagation can take various forms, bag Include but be not limited to electromagnetic signal, optical signal or above-mentioned any appropriate combination.Computer-readable signal media can also be Any computer-readable medium beyond computer-readable recording medium, the computer-readable medium can send, propagate or Transmit for being used or program in connection by instruction execution system, device or device.Computer-readable medium On the program code that includes any appropriate medium can be used to transmit, include but is not limited to：Wirelessly, electric wire, optical cable, RF etc., Or above-mentioned any appropriate combination.
Flow chart and block diagram in accompanying drawing, it is illustrated that according to the system of the various embodiments of the application, method and computer journey Architectural framework in the cards, function and the operation of sequence product.At this point, each square frame in flow chart or block diagram can generation The part of one module of table, program segment or code, the part of the module, program segment or code is used comprising one or more In the executable instruction for realizing defined logic function.It should also be noted that in some realizations as replacement, being marked in square frame The function of note can also be with different from the order marked in accompanying drawing generation.For example, two square frames succeedingly represented are actually It can perform substantially in parallel, they can also be performed in the opposite order sometimes, this is depending on involved function.Also to note Meaning, the combination of each square frame in block diagram and/or flow chart and the square frame in block diagram and/or flow chart can be with holding The special hardware based system of function or operation as defined in row is realized, or can use specialized hardware and computer instruction Combination realize.
Being described in unit involved in the embodiment of the present application can be realized by way of software, can also be by hard The mode of part is realized.Described unit can also be set within a processor, for example, can be described as：A kind of processor bag Include acquiring unit, statistic unit, screening unit, generation unit and determining unit.Wherein, the title of these units is in certain situation Under do not constitute restriction to the unit in itself, for example, acquiring unit is also described as " obtaining flat where target trade company The unit of the user profile of user's collection in platform ".
As on the other hand, present invention also provides a kind of computer-readable medium, the computer-readable medium can be Included in device described in above-described embodiment；Can also be individualism, and without be incorporated the device in.Above-mentioned calculating Machine computer-readable recording medium carries one or more program, when said one or multiple programs are performed by the device so that should Device：The user profile of user's collection in platform where acquisition target trade company, user's collection includes the seed of target trade company User and user to be recommended, the user profile include the History Order record with the label in default tag set；To institute Stating user concentrates the History Order record of each user to be counted according to the label, obtains the user and concentrates each described The tag attributes feature of user；Screened and the seed user phase from the user to be recommended based on the tag attributes feature As user, be used as the candidate user of the target trade company；User profile and the candidate based on each seed user are used The tag attributes feature at family generates the recommendation degree information of each candidate user；Used according to the recommendation degree information from the candidate The target recommended user of the target trade company is determined in family.
The embodiment of the present application discloses A1, a kind of method for determining target recommended user, and methods described includes：Obtain mesh Mark the user profile of user's collection in the platform where trade company, user collection includes the seed user of target trade company and to be recommended User, the user profile includes the History Order record with the label in default tag set；Each is concentrated to the user The History Order record of the user is counted according to the label, obtains the label category that the user concentrates each user Property feature；The user similar to the seed user is screened from the user to be recommended based on the tag attributes feature, made For the candidate user of the target trade company；The tag attributes of user profile and the candidate user based on each seed user Feature generates the recommendation degree information of each candidate user；Institute is determined from the candidate user according to the recommendation degree information State the target recommended user of target trade company.
In A2, the method as described in A1, the tag attributes feature includes label characteristics vector, the label characteristics vector In each element be corresponding to each label in the default tag set characteristic value；It is described to concentrate each described to the user The History Order record of user is counted according to the label, is obtained the user and is concentrated the tag attributes of each user special Levy, including：The History Order record of each user concentrated according to the user, corresponds to each label to each user The frequency that places an order is counted, the label characteristics vector of each user of generation.
In A3, the method as described in A2, when the History Order record includes the order generation of each bar History Order record Between；The History Order of each user concentrated according to the user is recorded, and corresponds to each label to each user The frequency that places an order is counted, the label characteristics vector of each user of generation, including：To each mark in the tag set Label, based on each user correspond to the label each bar History Order record order generate the time, with it is default when Between decay factor determine the equivalent number of times that places an order of each bar History Order record, the equivalent number of times that places an order recorded to each bar History Order Summed, obtain statistical result of the user corresponding to the frequency that places an order of the label；Institute is corresponded to based on the user The statistical result of the frequency that places an order of each label in default tag set is stated, the label characteristics vector of the user is generated.
It is described that each mark in the default tag set is corresponded to based on the user in A4, the method as described in A3 The statistical result of the frequency that places an order of label, generates the label characteristics vector of the user, including：The user is corresponded to described pre- If the statistical result of the frequency that places an order of each label in tag set is normalized, under after normalized Single-frequency time statistical result as each corresponding element in label characteristics vector characteristic value.
In A5, the method as described in A2, it is described based on the tag attributes feature screened from the user to be recommended with The similar user of the seed user, as the candidate user of the target trade company, including：Calculate the label of the seed user The similarity of the label characteristics vector of characteristic vector and each user to be recommended, as the seed user with it is described to be recommended The similarity of user；The candidate user is filtered out according to the sequence of the similarity.
In A6, the method as described in A5, the user profile based on each seed user and the candidate user Tag attributes feature generates the recommendation degree information of each candidate user, including：Remembered according to the History Order of the seed user The number of times that places an order that the record statistics seed user is associated with the target trade company；By the candidate user and the seed user The product for the number of times that places an order that similarity and the seed user are associated with the target trade company is used as the corresponding of the candidate user In the score of the seed user；The score for corresponding to each seed user to the candidate user is summed, and obtains the time From family score as the candidate user recommendation degree information.
A7, A1 into A6 it is any as described in method in, methods described also includes determining that the seed of the target trade company is used The step of family, including：The History Order record that the user is concentrated associated with the target trade company and with the target trade company The user that the History Order record of association meets preparatory condition is used as the seed user of the target trade company.
The embodiment of the present application discloses B1, a kind of device for being used to determine target recommended user, and described device includes：Obtain single Member, the user profile for the user's collection for being configured to obtain in the platform where target trade company, user's collection includes target trade company Seed user and user to be recommended, the user profile includes the History Order note with the label in default tag set Record；Statistic unit, is configured to concentrate the user History Order record of each user to unite according to the label Meter, obtains the tag attributes feature that the user concentrates each user；Screening unit, is configured to be based on the tag attributes Feature screens the user similar to the seed user from the user to be recommended, is used as the candidate of the target trade company Family；Generation unit, is configured to the tag attributes feature of user profile and the candidate user based on each seed user Generate the recommendation degree information of each candidate user；Determining unit, is configured to according to the recommendation degree information from the candidate The target recommended user of the target trade company is determined in user.
In B2, the device as described in B1, the tag attributes feature includes label characteristics vector, the label characteristics vector In each element be corresponding to each label in the default tag set characteristic value；The statistic unit further configures use Counted in concentrating the History Order record of each user to the user as follows according to the label：According to The History Order record for each user that the user concentrates, the frequency that places an order that each label is corresponded to each user is carried out Statistics, the label characteristics vector of each user of generation.
In B3, the device as described in B2, when the History Order record includes the order generation of each bar History Order record Between；The statistic unit is further configured to as follows correspond to each user the frequency that places an order of each label Counted, the label characteristics vector of each user of generation：To each label in the tag set, based on each institute The order generation time that user records corresponding to each bar History Order of the label is stated, is determined with default time decay factor The equivalent number of times that places an order of each bar History Order record, sums to the equivalent number of times that places an order that each bar History Order is recorded, obtains The user corresponds to the statistical result of the frequency that places an order of the label；The default tag set is corresponded to based on the user In each label the frequency that places an order statistical result, generate the label characteristics vector of the user.
In B4, the device as described in B3, the statistic unit is further configured to generate the use as follows The label characteristics vector at family：The system of the frequency that places an order of each label corresponded to the user in the default tag set Meter result is normalized, using the frequency statistics result that places an order after normalized as each in label characteristics vector The characteristic value of corresponding element.
In B5, the device as described in B2, the screening unit be further configured to screen as follows with it is described The similar user of seed user, is used as the candidate user of the target trade company：Calculate the label characteristics vector of the seed user With the similarity of the label characteristics vector of each user to be recommended, the seed user and the phase of the user to be recommended are used as Like degree；The candidate user is filtered out according to the sequence of the similarity.
In B6, the device as described in B5, the generation unit is further configured to generate as follows each described The recommendation degree information of candidate user：
The seed user is counted according to the History Order of seed user record to associate down with the target trade company Single number；
The candidate user is associated with the similarity and the seed user of the seed user with the target trade company The number of times that places an order product as the candidate user the score corresponding to the seed user；
The score for corresponding to each seed user to the candidate user is summed, and obtains the score of the candidate user, It is used as the recommendation degree information of the candidate user.
B7, B1 into B6 it is any as described in device in, described device also includes determining that the seed of the target trade company is used The unit at family, is configured to：The History Order record that the user is concentrated associated with the target trade company and with the target The user that the History Order record of trade company's association meets preparatory condition is used as the seed user of the target trade company.
The embodiment of the present application discloses C1, a kind of server, including：One or more processors；Storage device, for storing One or more programs, when one or more of programs are by one or more of computing devices so that it is one or Multiple processors realize A1 into A7 it is any as described in method.
The embodiment of the present application discloses D1, a kind of computer-readable recording medium, is stored thereon with computer program, the program When being executed by processor realize A1 into A7 it is any as described in method.
Above description is only the preferred embodiment of the application and the explanation to institute's application technology principle.People in the art Member should be appreciated that invention scope involved in the application, however it is not limited to the technology of the particular combination of above-mentioned technical characteristic Scheme, while should also cover in the case where not departing from foregoing invention design, is carried out by above-mentioned technical characteristic or its equivalent feature Other technical schemes formed by any combination.Such as features described above has similar work(with (but not limited to) disclosed herein The technical characteristic of energy carries out technical scheme formed by replacement mutually.

Claims (10)

1. a kind of method for determining target recommended user, it is characterised in that methods described includes：
The user profile of user's collection in platform where acquisition target trade company, the seed that user's collection includes target trade company is used Family and user to be recommended, the user profile include the History Order record with the label in default tag set；
Concentrate the History Order record of each user to be counted according to the label to the user, obtain user's collection In each user tag attributes feature；
The user similar to the seed user is screened from the user to be recommended based on the tag attributes feature, institute is used as State the candidate user of target trade company；
Each candidate of tag attributes feature generation of user profile and the candidate user based on each seed user uses The recommendation degree information at family；
The target recommended user of the target trade company is determined from the candidate user according to the recommendation degree information.
2. according to the method described in claim 1, it is characterised in that the tag attributes feature includes label characteristics vector, institute It is the characteristic value corresponding to each label in the default tag set to state each element in label characteristics vector；
The History Order record that each user is concentrated to the user is counted according to the label, obtains the use The tag attributes feature of each user is concentrated at family, including：
The History Order record of each user concentrated according to the user, places an order to each user corresponding to each label The frequency is counted, the label characteristics vector of each user of generation.
3. method according to claim 2, it is characterised in that the History Order record includes each bar History Order record Order generation the time；
The History Order of each user concentrated according to the user is recorded, and corresponds to each label to each user The frequency that places an order is counted, the label characteristics vector of each user of generation, including：
To each label in the tag set, each bar History Order for corresponding to the label based on each user is remembered The order generation time of record, the equivalent number of times that places an order of each bar History Order record is determined with default time decay factor, to each The equivalent number of times that places an order of bar History Order record is summed, and obtains system of the user corresponding to the frequency that places an order of the label Count result；
Correspond to the statistical result of the frequency that places an order of each label in the default tag set based on the user, generate institute State the label characteristics vector of user.
4. method according to claim 3, it is characterised in that described that the default tally set is corresponded to based on the user The statistical result of the frequency that places an order of each label in conjunction, generates the label characteristics vector of the user, including：
The statistical result of the frequency that places an order of each label corresponded to the user in the default tag set is returned One change is handled, and regard the frequency statistics result that places an order after normalized as the spy of each corresponding element in label characteristics vector Value indicative.
5. method according to claim 2, it is characterised in that it is described based on the tag attributes feature from described to be recommended The user similar to the seed user is screened in user, as the candidate user of the target trade company, including：
The similarity of the vectorial label characteristics vector with each user to be recommended of label characteristics of the seed user is calculated, is made For the seed user and the similarity of the user to be recommended；
The candidate user is filtered out according to the sequence of the similarity.
6. method according to claim 5, it is characterised in that the user profile and institute based on each seed user The tag attributes feature for stating candidate user generates the recommendation degree information of each candidate user, including：
Recorded according to the History Order of the seed user and count the lower single that the seed user is associated with the target trade company Number；
The candidate user is associated down with the similarity and the seed user of the seed user with the target trade company The product of single number as the candidate user the score corresponding to the seed user；
The score for corresponding to each seed user to the candidate user is summed, and obtains the score of the candidate user as institute State the recommendation degree information of candidate user.
7. the method according to claim any one of 1-6, it is characterised in that methods described also includes determining the target business The step of seed user at family, including：
The history that the History Order record that the user is concentrated is associated with the target trade company and associated with the target trade company The user that order record meets preparatory condition is used as the seed user of the target trade company.
8. a kind of device for being used to determine target recommended user, it is characterised in that described device includes：
Acquiring unit, the user profile for the user's collection for being configured to obtain in the platform where target trade company, the user Ji Bao The seed user of target trade company and user to be recommended are included, the user profile includes going through with the label in default tag set History order record；
Statistic unit, is configured to concentrate the user History Order record of each user to unite according to the label Meter, obtains the tag attributes feature that the user concentrates each user；
Screening unit, is configured to screen from the user to be recommended and the seed user based on the tag attributes feature Similar user, is used as the candidate user of the target trade company；
Generation unit, is configured to the tag attributes feature of user profile and the candidate user based on each seed user Generate the recommendation degree information of each candidate user；
Determining unit, is configured to determine the mesh of the target trade company from the candidate user according to the recommendation degree information Mark recommended user.
9. a kind of server, it is characterised in that including：
One or more processors；
Storage device, for storing one or more programs,
When one or more of programs are by one or more of computing devices so that one or more of processors are real The existing method as described in any in claim 1-7.
10. a kind of computer-readable recording medium, is stored thereon with computer program, it is characterised in that the program is by processor The method as described in any in claim 1-7 is realized during execution.
CN201710385717.6A 2017-05-26 2017-05-26 Method, device and server for determining target recommended user Pending CN107220852A (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN201710385717.6A CN107220852A (en) 2017-05-26 2017-05-26 Method, device and server for determining target recommended user

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN201710385717.6A CN107220852A (en) 2017-05-26 2017-05-26 Method, device and server for determining target recommended user

Publications (1)

Publication Number Publication Date
CN107220852A true CN107220852A (en) 2017-09-29

Family

ID=59946733

Family Applications (1)

Application Number Title Priority Date Filing Date
CN201710385717.6A Pending CN107220852A (en) 2017-05-26 2017-05-26 Method, device and server for determining target recommended user

Country Status (1)

Country Link
CN (1) CN107220852A (en)

Cited By (55)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN107665383A (en) * 2017-10-26 2018-02-06 北京拉勾科技有限公司 A kind of construction method and computing device of resume processing delay model
CN107742245A (en) * 2017-10-31 2018-02-27 北京小度信息科技有限公司 A kind of merchant information recommends method, apparatus and equipment
CN107844584A (en) * 2017-11-14 2018-03-27 北京小度信息科技有限公司 Usage mining method, apparatus, electronic equipment and computer-readable recording medium
CN107886354A (en) * 2017-10-31 2018-04-06 广州云移信息科技有限公司 A kind of method and system for determining marketing target colony
CN108153824A (en) * 2017-12-06 2018-06-12 阿里巴巴集团控股有限公司 The determining method and device of targeted user population
CN108537567A (en) * 2018-03-06 2018-09-14 阿里巴巴集团控股有限公司 A kind of determination method and apparatus of targeted user population
CN108764371A (en) * 2018-06-08 2018-11-06 Oppo广东移动通信有限公司 Image processing method, device, computer readable storage medium and electronic equipment
CN108876470A (en) * 2018-06-29 2018-11-23 腾讯科技（深圳）有限公司 Tagging user extended method, computer equipment and storage medium
CN109118296A (en) * 2018-09-04 2019-01-01 南京星邺汇捷网络科技有限公司 Movable method for pushing, device and electronic equipment
CN109190807A (en) * 2018-08-15 2019-01-11 上海交通大学 A kind of cost minimization propagation optimization method of object-oriented group
CN109218390A (en) * 2018-07-12 2019-01-15 北京比特智学科技有限公司 User's screening technique and device
CN109325806A (en) * 2018-09-20 2019-02-12 北京小度信息科技有限公司 A kind of processing method and processing device of user information
CN109447700A (en) * 2018-10-23 2019-03-08 广州致轩服饰有限公司 A kind of commodity feedback method and device based on storage user
CN109447748A (en) * 2018-10-23 2019-03-08 广州致轩服饰有限公司 A kind of commodity feedback method and device based on increment user
CN109509051A (en) * 2018-09-12 2019-03-22 北京奇艺世纪科技有限公司 A kind of item recommendation method and device
CN109753994A (en) * 2018-12-11 2019-05-14 东软集团股份有限公司 User's portrait method, apparatus, computer readable storage medium and electronic equipment
CN109858344A (en) * 2018-12-24 2019-06-07 深圳市珍爱捷云信息技术有限公司 Love and marriage object recommendation method, apparatus, computer equipment and storage medium
CN109961311A (en) * 2017-12-26 2019-07-02 中国移动通信集团四川有限公司 Lead referral method, apparatus calculates equipment and storage medium
CN109977979A (en) * 2017-12-28 2019-07-05 中国移动通信集团广东有限公司 Position method, apparatus, electronic equipment and the storage medium of seed user
CN109992632A (en) * 2019-01-14 2019-07-09 江苏智途科技股份有限公司 A kind of spatial data intelligence distribution method of servicing based on big data
CN110008973A (en) * 2018-11-23 2019-07-12 阿里巴巴集团控股有限公司 A kind of model training method, the method and device that target user is determined based on model
CN110033324A (en) * 2019-04-11 2019-07-19 上海拉扎斯信息科技有限公司 Data processing method, device, electronic equipment and computer readable storage medium
CN110109901A (en) * 2018-02-02 2019-08-09 北京京东尚科信息技术有限公司 The method and apparatus for screening target object
CN110162692A (en) * 2018-12-10 2019-08-23 腾讯科技（深圳）有限公司 User tag determines method, apparatus, computer equipment and storage medium
CN110503498A (en) * 2018-05-16 2019-11-26 北京三快在线科技有限公司 A kind of order recommended method and device
CN110517001A (en) * 2019-08-29 2019-11-29 江苏德赢软件科技有限公司 A kind of timeliness guest room commodity go the method, apparatus and system of inventory
CN110517099A (en) * 2018-05-22 2019-11-29 北京京东尚科信息技术有限公司 Method and apparatus for determining joint supply side
CN110738521A (en) * 2019-10-10 2020-01-31 深圳市云积分科技有限公司 multi-vendor brand customer selling method and device
CN110781395A (en) * 2019-10-25 2020-02-11 广东小天才科技有限公司 Question recommendation method and device and terminal equipment
CN110852830A (en) * 2019-09-30 2020-02-28 口碑(上海)信息技术有限公司 Target user information acquisition method and service subscription system
CN110851729A (en) * 2019-11-19 2020-02-28 深圳前海微众银行股份有限公司 Resource information recommendation method, device, equipment and computer storage medium
CN110888945A (en) * 2019-11-28 2020-03-17 上海风秩科技有限公司 User behavior prediction method and device, electronic equipment and storage medium
CN111061979A (en) * 2018-10-17 2020-04-24 北京字节跳动网络技术有限公司 User label pushing method and device, electronic equipment and medium
CN111127060A (en) * 2018-10-31 2020-05-08 百度在线网络技术（北京）有限公司 Method and device for determining promotion user of service
CN111222923A (en) * 2020-01-13 2020-06-02 秒针信息技术有限公司 Method and device for judging potential customer, electronic equipment and storage medium
CN111259281A (en) * 2020-01-20 2020-06-09 腾讯科技（深圳）有限公司 Method and device for determining merchant label and storage medium
CN111292164A (en) * 2020-01-21 2020-06-16 上海风秩科技有限公司 Commodity recommendation method and device, electronic equipment and readable storage medium
CN111309774A (en) * 2018-12-11 2020-06-19 北京嘀嘀无限科技发展有限公司 Data processing method and device, electronic equipment and storage medium
CN111353688A (en) * 2020-02-05 2020-06-30 口碑(上海)信息技术有限公司 User resource allocation method and device
CN111352962A (en) * 2018-12-24 2020-06-30 网智天元科技集团股份有限公司 Client portrait construction method and device
CN111581518A (en) * 2020-05-14 2020-08-25 北京易数科技有限公司 Information pushing method and device
CN111582906A (en) * 2020-03-26 2020-08-25 口碑(上海)信息技术有限公司 Target user information acquisition method and device and electronic equipment
CN111598622A (en) * 2020-05-21 2020-08-28 深圳市元征科技股份有限公司 Method, device, equipment and storage medium for generating qualification right data
CN111651669A (en) * 2020-05-20 2020-09-11 拉扎斯网络科技（上海）有限公司 Information recommendation method and device, electronic equipment and computer-readable storage medium
CN111723289A (en) * 2020-06-08 2020-09-29 北京声智科技有限公司 Information recommendation method and device
CN112035611A (en) * 2020-08-28 2020-12-04 康键信息技术（深圳）有限公司 Target user recommendation method and device, computer equipment and storage medium
CN112052388A (en) * 2020-08-20 2020-12-08 深思考人工智能科技(上海)有限公司 Method and system for recommending gourmet stores
CN112287236A (en) * 2020-11-19 2021-01-29 每日互动股份有限公司 Text message pushing method and device, computer equipment and storage medium
CN112328866A (en) * 2019-08-05 2021-02-05 四川大学 Specific user group mining method in network space security field
CN112534424A (en) * 2018-08-03 2021-03-19 脸谱公司 Neural network based content distribution in online systems
CN112818255A (en) * 2021-02-05 2021-05-18 深圳市枣孖健康科技有限责任公司 Label recommendation method based on user portrait
CN113312513A (en) * 2021-07-28 2021-08-27 贝壳找房(北京)科技有限公司 Object recommendation method and device, electronic equipment and storage medium
CN113379511A (en) * 2021-07-02 2021-09-10 北京沃东天骏信息技术有限公司 Method and apparatus for outputting information
CN114429371A (en) * 2022-04-06 2022-05-03 新石器慧通（北京）科技有限公司 Unmanned vehicle-based commodity marketing method and device, electronic equipment and storage medium
CN116501972A (en) * 2023-05-06 2023-07-28 兰州柒禾网络科技有限公司 Content pushing method and AI intelligent pushing system based on big data online service

Citations (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN101441667A (en) * 2008-12-29 2009-05-27 北京搜狗科技发展有限公司 Music recommend method and apparatus
CN104732422A (en) * 2015-02-27 2015-06-24 湖南大学 Mobile user tag based personalized food recommendation method
CN104751354A (en) * 2015-04-13 2015-07-01 合一信息技术（北京）有限公司 Advertisement cluster screening method
CN105023175A (en) * 2015-07-24 2015-11-04 金鹃传媒科技股份有限公司 Online advertisement classified pushing method and system based on consumer behavior data analysis and classification technology
CN105528395A (en) * 2015-11-30 2016-04-27 苏州大学 Method and system for recommending potential consumers
CN105574216A (en) * 2016-03-07 2016-05-11 达而观信息科技（上海）有限公司 Personalized recommendation method and system based on probability model and user behavior analysis

Patent Citations (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN101441667A (en) * 2008-12-29 2009-05-27 北京搜狗科技发展有限公司 Music recommend method and apparatus
CN104732422A (en) * 2015-02-27 2015-06-24 湖南大学 Mobile user tag based personalized food recommendation method
CN104751354A (en) * 2015-04-13 2015-07-01 合一信息技术（北京）有限公司 Advertisement cluster screening method
CN105023175A (en) * 2015-07-24 2015-11-04 金鹃传媒科技股份有限公司 Online advertisement classified pushing method and system based on consumer behavior data analysis and classification technology
CN105528395A (en) * 2015-11-30 2016-04-27 苏州大学 Method and system for recommending potential consumers
CN105574216A (en) * 2016-03-07 2016-05-11 达而观信息科技（上海）有限公司 Personalized recommendation method and system based on probability model and user behavior analysis

Cited By (73)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN107665383A (en) * 2017-10-26 2018-02-06 北京拉勾科技有限公司 A kind of construction method and computing device of resume processing delay model
CN107742245A (en) * 2017-10-31 2018-02-27 北京小度信息科技有限公司 A kind of merchant information recommends method, apparatus and equipment
CN107886354A (en) * 2017-10-31 2018-04-06 广州云移信息科技有限公司 A kind of method and system for determining marketing target colony
CN107742245B (en) * 2017-10-31 2021-04-27 北京星选科技有限公司 Merchant information recommendation method, device and equipment
CN107844584A (en) * 2017-11-14 2018-03-27 北京小度信息科技有限公司 Usage mining method, apparatus, electronic equipment and computer-readable recording medium
TWI709927B (en) * 2017-12-06 2020-11-11 開曼群島商創新先進技術有限公司 Method and device for determining target user group
CN108153824A (en) * 2017-12-06 2018-06-12 阿里巴巴集团控股有限公司 The determining method and device of targeted user population
WO2019109698A1 (en) * 2017-12-06 2019-06-13 阿里巴巴集团控股有限公司 Method and apparatus for determining target user group
CN109961311A (en) * 2017-12-26 2019-07-02 中国移动通信集团四川有限公司 Lead referral method, apparatus calculates equipment and storage medium
CN109961311B (en) * 2017-12-26 2023-04-18 中国移动通信集团四川有限公司 Client recommendation method and device, computing equipment and storage medium
CN109977979B (en) * 2017-12-28 2021-12-07 中国移动通信集团广东有限公司 Method and device for locating seed user, electronic equipment and storage medium
CN109977979A (en) * 2017-12-28 2019-07-05 中国移动通信集团广东有限公司 Position method, apparatus, electronic equipment and the storage medium of seed user
CN110109901A (en) * 2018-02-02 2019-08-09 北京京东尚科信息技术有限公司 The method and apparatus for screening target object
CN108537567A (en) * 2018-03-06 2018-09-14 阿里巴巴集团控股有限公司 A kind of determination method and apparatus of targeted user population
TWI743428B (en) * 2018-03-06 2021-10-21 開曼群島商創新先進技術有限公司 Method and device for determining target user group
CN110503498B (en) * 2018-05-16 2021-07-20 北京三快在线科技有限公司 Order recommendation method and device
CN110503498A (en) * 2018-05-16 2019-11-26 北京三快在线科技有限公司 A kind of order recommended method and device
CN110517099A (en) * 2018-05-22 2019-11-29 北京京东尚科信息技术有限公司 Method and apparatus for determining joint supply side
CN108764371A (en) * 2018-06-08 2018-11-06 Oppo广东移动通信有限公司 Image processing method, device, computer readable storage medium and electronic equipment
CN108876470A (en) * 2018-06-29 2018-11-23 腾讯科技（深圳）有限公司 Tagging user extended method, computer equipment and storage medium
CN109218390A (en) * 2018-07-12 2019-01-15 北京比特智学科技有限公司 User's screening technique and device
CN112534424A (en) * 2018-08-03 2021-03-19 脸谱公司 Neural network based content distribution in online systems
CN109190807B (en) * 2018-08-15 2021-06-04 上海交通大学 Cost minimization propagation optimization method for target group
CN109190807A (en) * 2018-08-15 2019-01-11 上海交通大学 A kind of cost minimization propagation optimization method of object-oriented group
CN109118296A (en) * 2018-09-04 2019-01-01 南京星邺汇捷网络科技有限公司 Movable method for pushing, device and electronic equipment
CN109509051A (en) * 2018-09-12 2019-03-22 北京奇艺世纪科技有限公司 A kind of item recommendation method and device
CN109325806A (en) * 2018-09-20 2019-02-12 北京小度信息科技有限公司 A kind of processing method and processing device of user information
CN111061979A (en) * 2018-10-17 2020-04-24 北京字节跳动网络技术有限公司 User label pushing method and device, electronic equipment and medium
CN111061979B (en) * 2018-10-17 2023-04-21 北京字节跳动网络技术有限公司 User tag pushing method and device, electronic equipment and medium
CN109447748A (en) * 2018-10-23 2019-03-08 广州致轩服饰有限公司 A kind of commodity feedback method and device based on increment user
CN109447700A (en) * 2018-10-23 2019-03-08 广州致轩服饰有限公司 A kind of commodity feedback method and device based on storage user
CN111127060B (en) * 2018-10-31 2023-08-08 百度在线网络技术（北京）有限公司 Method and device for determining popularization users of service
CN111127060A (en) * 2018-10-31 2020-05-08 百度在线网络技术（北京）有限公司 Method and device for determining promotion user of service
CN110008973A (en) * 2018-11-23 2019-07-12 阿里巴巴集团控股有限公司 A kind of model training method, the method and device that target user is determined based on model
CN110008973B (en) * 2018-11-23 2023-05-02 创新先进技术有限公司 Model training method, method and device for determining target user based on model
CN110162692A (en) * 2018-12-10 2019-08-23 腾讯科技（深圳）有限公司 User tag determines method, apparatus, computer equipment and storage medium
CN111309774A (en) * 2018-12-11 2020-06-19 北京嘀嘀无限科技发展有限公司 Data processing method and device, electronic equipment and storage medium
CN109753994A (en) * 2018-12-11 2019-05-14 东软集团股份有限公司 User's portrait method, apparatus, computer readable storage medium and electronic equipment
CN111352962A (en) * 2018-12-24 2020-06-30 网智天元科技集团股份有限公司 Client portrait construction method and device
CN111352962B (en) * 2018-12-24 2024-03-29 网智天元科技集团股份有限公司 Customer portrait construction method and device
CN109858344A (en) * 2018-12-24 2019-06-07 深圳市珍爱捷云信息技术有限公司 Love and marriage object recommendation method, apparatus, computer equipment and storage medium
CN109992632A (en) * 2019-01-14 2019-07-09 江苏智途科技股份有限公司 A kind of spatial data intelligence distribution method of servicing based on big data
CN110033324A (en) * 2019-04-11 2019-07-19 上海拉扎斯信息科技有限公司 Data processing method, device, electronic equipment and computer readable storage medium
CN112328866A (en) * 2019-08-05 2021-02-05 四川大学 Specific user group mining method in network space security field
CN110517001A (en) * 2019-08-29 2019-11-29 江苏德赢软件科技有限公司 A kind of timeliness guest room commodity go the method, apparatus and system of inventory
CN110852830A (en) * 2019-09-30 2020-02-28 口碑(上海)信息技术有限公司 Target user information acquisition method and service subscription system
CN110738521B (en) * 2019-10-10 2022-04-15 深圳市云积分科技有限公司 Client selling method and device for multi-merchant brand
CN110738521A (en) * 2019-10-10 2020-01-31 深圳市云积分科技有限公司 multi-vendor brand customer selling method and device
CN110781395A (en) * 2019-10-25 2020-02-11 广东小天才科技有限公司 Question recommendation method and device and terminal equipment
CN110851729A (en) * 2019-11-19 2020-02-28 深圳前海微众银行股份有限公司 Resource information recommendation method, device, equipment and computer storage medium
CN110888945A (en) * 2019-11-28 2020-03-17 上海风秩科技有限公司 User behavior prediction method and device, electronic equipment and storage medium
CN111222923A (en) * 2020-01-13 2020-06-02 秒针信息技术有限公司 Method and device for judging potential customer, electronic equipment and storage medium
CN111222923B (en) * 2020-01-13 2023-12-15 秒针信息技术有限公司 Method and device for judging potential clients, electronic equipment and storage medium
CN111259281B (en) * 2020-01-20 2023-04-07 腾讯科技（深圳）有限公司 Method and device for determining merchant label and storage medium
CN111259281A (en) * 2020-01-20 2020-06-09 腾讯科技（深圳）有限公司 Method and device for determining merchant label and storage medium
CN111292164A (en) * 2020-01-21 2020-06-16 上海风秩科技有限公司 Commodity recommendation method and device, electronic equipment and readable storage medium
CN111353688B (en) * 2020-02-05 2024-02-27 口碑(上海)信息技术有限公司 User resource allocation method and device
CN111353688A (en) * 2020-02-05 2020-06-30 口碑(上海)信息技术有限公司 User resource allocation method and device
CN111582906A (en) * 2020-03-26 2020-08-25 口碑(上海)信息技术有限公司 Target user information acquisition method and device and electronic equipment
CN111581518A (en) * 2020-05-14 2020-08-25 北京易数科技有限公司 Information pushing method and device
CN111651669A (en) * 2020-05-20 2020-09-11 拉扎斯网络科技（上海）有限公司 Information recommendation method and device, electronic equipment and computer-readable storage medium
CN111598622A (en) * 2020-05-21 2020-08-28 深圳市元征科技股份有限公司 Method, device, equipment and storage medium for generating qualification right data
CN111723289A (en) * 2020-06-08 2020-09-29 北京声智科技有限公司 Information recommendation method and device
CN111723289B (en) * 2020-06-08 2024-02-02 北京声智科技有限公司 Information recommendation method and device
CN112052388A (en) * 2020-08-20 2020-12-08 深思考人工智能科技(上海)有限公司 Method and system for recommending gourmet stores
CN112035611A (en) * 2020-08-28 2020-12-04 康键信息技术（深圳）有限公司 Target user recommendation method and device, computer equipment and storage medium
CN112287236A (en) * 2020-11-19 2021-01-29 每日互动股份有限公司 Text message pushing method and device, computer equipment and storage medium
CN112818255A (en) * 2021-02-05 2021-05-18 深圳市枣孖健康科技有限责任公司 Label recommendation method based on user portrait
CN113379511A (en) * 2021-07-02 2021-09-10 北京沃东天骏信息技术有限公司 Method and apparatus for outputting information
CN113312513A (en) * 2021-07-28 2021-08-27 贝壳找房(北京)科技有限公司 Object recommendation method and device, electronic equipment and storage medium
CN114429371A (en) * 2022-04-06 2022-05-03 新石器慧通（北京）科技有限公司 Unmanned vehicle-based commodity marketing method and device, electronic equipment and storage medium
CN116501972A (en) * 2023-05-06 2023-07-28 兰州柒禾网络科技有限公司 Content pushing method and AI intelligent pushing system based on big data online service
CN116501972B (en) * 2023-05-06 2024-01-05 广州市巨应信息科技有限公司 Content pushing method and AI intelligent pushing system based on big data online service

Similar Documents

Publication Publication Date Title
CN107220852A (en) Method, device and server for determining target recommended user
CN107247786A (en) Method, device and server for determining similar users
US20210182938A1 (en) Item recommendation system, item recommendation method and program
CN107105031A (en) Information-pushing method and device
CN108960992A (en) A kind of information recommendation method and relevant device
CN108090162A (en) Information-pushing method and device based on artificial intelligence
CN107203894A (en) Information-pushing method and device
CN106708821A (en) User personalized shopping behavior-based commodity recommendation method
CN107273436A (en) The training method and trainer of a kind of recommended models
CN108335177A (en) Shopping recommendation method, user terminal, server-side, equipment and storage medium
CN106649890A (en) Data storage method and device
CN107301592A (en) The method and device excavated for commodity substitute
CN105279672A (en) Lead recommendations
CN107644364A (en) Object filter method and system
CN107679898A (en) A kind of Method of Commodity Recommendation and device
CN107944956A (en) Method and apparatus for generating information
CN107885784A (en) The method and apparatus for extracting user characteristic data
CN106776716A (en) A kind of intelligent Matching marketing consultant and the method and apparatus of user
CN110020162A (en) User identification method and device
CN107305677A (en) Product information method for pushing and device
CN110674391A (en) Product data pushing method and system based on big data and computer equipment
CN107911449A (en) Method and apparatus for pushed information
CN107741967A (en) Method, apparatus and electronic equipment for behavioral data processing
CN107742128A (en) Method and apparatus for output information
CN107545491A (en) A kind of data processing method and device of recommendation information

Legal Events

Date Code Title Description
PB01 Publication
PB01 Publication
SE01 Entry into force of request for substantive examination
SE01 Entry into force of request for substantive examination
CB02 Change of applicant information
CB02 Change of applicant information

Address after: Building N3, building 12, No. 27, Jiancai Chengzhong Road, Haidian District, Beijing 100096

Applicant after: Beijing Xingxuan Technology Co.,Ltd.

Address before: 100085 Beijing, Haidian District on the road to the information on the ground floor of the 1 to the 3 floor of the 2 floor, room 11, 202

Applicant before: Beijing Xiaodu Information Technology Co.,Ltd.

RJ01 Rejection of invention patent application after publication
RJ01 Rejection of invention patent application after publication

Application publication date: 20170929