CN107092629A - Recommend method and device - Google Patents
Recommend method and device Download PDFInfo
- Publication number
- CN107092629A CN107092629A CN201710040010.1A CN201710040010A CN107092629A CN 107092629 A CN107092629 A CN 107092629A CN 201710040010 A CN201710040010 A CN 201710040010A CN 107092629 A CN107092629 A CN 107092629A
- Authority
- CN
- China
- Prior art keywords
- recommendation
- user
- content
- position location
- content recommendation
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/90—Details of database functions independent of the retrieved data types
- G06F16/95—Retrieval from the web
- G06F16/953—Querying, e.g. by the use of web search engines
- G06F16/9537—Spatial or temporal dependent retrieval, e.g. spatiotemporal queries
Landscapes
- Engineering & Computer Science (AREA)
- Databases & Information Systems (AREA)
- Theoretical Computer Science (AREA)
- Data Mining & Analysis (AREA)
- Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
Abstract
The embodiment of the present application provides a kind of recommendation method and device.Recommendation method includes:In response to recommending trigger event, the region scene of the historical interest point for recommending trigger event correspondence user, position location and position location association is obtained;The region scene associated according to the historical interest of user point, position location and position location, determines the recommendation order between at least one content recommendation and at least one content recommendation;With the recommendation order between at least one content recommendation, at least one content recommendation is recommended to user.The embodiment of the present application can recommend to more conform to the content of user's request to user.
Description
Technical field
The application is related to Internet technical field, more particularly to a kind of recommendation method and device.
Background technology
With the development of Internet technology, the application based on internet is more and more.Increasing application is begun to focus on
Recommended technology, and service quality is lifted by recommended technology, increase number of users.
By taking the application of e-commerce field as an example, it is usually according to being browsed before user, the behavior such as buys, searches for, evaluating
Recognize the historical interest point of user;Historical interest point based on user, from the trade company that position location is recalled, determines that user's sense is emerging
The trade company of interest;When the operation such as scanning for, click on as user, the trade company determined in advance is recommended into user.This recommendation side
Formula can understand user's request in time, efficiently, accurately recommend interest trade company to user, can greatly lift Consumer's Experience,
So as to keep user here.
The content of the invention
Based on being browsed before user, the behavior such as buy, search for, evaluating, the historical interest point of user being recognized, based on user's
Historical interest point recommends trade company to user, to a certain extent it will be seen that user's request, efficiently, accurately recommends interest to user
Trade company, is conducive to lifting Consumer's Experience.But, during the actual use this way of recommendation, often occur user not
The situation for recommending trade company is browsed or clicks on, this explanation recommends trade company to be unsatisfactory for user's current demand, and recommendation effect is undesirable,
Consumer's Experience is poor, it can be seen that, the way of recommendation is further improved.
For above-mentioned user do not browse or click on recommend trade company in the case of, present inventor is analyzed and researched,
It was found that:When above-mentioned situation occurs, the position of user is typically changed.Based on the clue, present inventor has done one
Step analysis is found:When above-mentioned situation occurs, the position of user is typically changed, but customer location when changing not necessarily
Above-mentioned situation occurs, then releases:Customer location change is presentation, is not the real original for causing above-mentioned situation occur
Cause.
Based on above-mentioned analysis result, change in location when present inventor occurs to above-mentioned situation done deeper into grind
Study carefully, find:These change in location are not simple change in location, but why the scene in these positions differs user
Sample, for ease of describing the scene that user is in certain position being referred to as region scene.As can be seen here, region scene can also influence to use
Family trade company interested.For example, user is when company goes to work, fast food is more inclined to, but when user travels to somewhere, may be to working as
Ground speciality is interested, if according to the historical interest of user, recommending fast food to user, it is clear that be no longer complies with user at that time
Demand.
Based on above-mentioned discovery, present inventor provides the suggested design after a kind of improvement, and cardinal principle is:In combination with
The region scene of the historical interest point of user, position location and position location association, to user's content recommendation, so as to user
Recommendation more conforms to the content of user's request, improves recommendation effect, improves user experience.
Based on a kind of above-mentioned, recommendation method of the embodiment of the application one offer, including:
In response to recommending trigger event, the historical interest point for recommending trigger event correspondence user, position location are obtained
And the region scene of the position location association;
The region scene associated according to the historical interest of user point, the position location and the position location,
Determine the recommendation order between at least one content recommendation and at least one described content recommendation;
With the recommendation order between at least one described content recommendation, recommend to the user at least one described recommendation
Hold.
In an optional embodiment, the obtaining step of the region scene of the position location association, including:Obtain described
The corresponding scene labeled data in position location or at least one non-scene labeled data;According to the scene labeled data or at least
A kind of non-scene labeled data, recognizes the region scene of the position location association.
Alternatively, at least one non-scene labeled data includes:It is the type of the position location affiliated area, described
User the residence time of the position location affiliated area, the user the position location affiliated area movable model
Enclose, the wireless signal that the user uses in the position location is identified and/or current time feature.
In an optional embodiment, the identification step of the region scene of the position location association, including:
Residence time for the user in position location meets temporal regularity when being in, belonging to the position location
The type in region be apartment/residential quarter, and the user used in the position location wireless signal mark with the user
The wireless signal mark identical situation used at multiple weekends, determines the region scene of the position location association to be in;
Residence time for the user in position location meets the temporal regularity in company, the position location institute
Belong to the type in region for shopping centre/other non-residential areas, and the wireless signal mark that the user uses in the position location
Identical situation is identified with the user wireless signal that the time uses on weekdays, the region of the position location association is determined
Scene is in company;
Belong to festivals or holidays for the current time, the position location belongs to other cities and its type is apartment/live
Quarter, and the wireless signal that is used in the position location of the user identifies the wireless signal commonly used with before and identifies and differ
Situation, the region scene of position location association is determined to be on home leave;
Type for the position location affiliated area is Tour region, and the user is in the affiliated area in the position location
It is movable in each sight spot in domain, the wireless signal that the user commonly uses in the wireless signal mark that the position location is used with before
Mark is differed, the feelings of temporal regularity of the user when the residence time of the position location affiliated area meets tourism
Condition, the region scene for determining the position location association is tourism;
For other situations, determine the region scene of the position location association to go on business.
In an optional embodiment, between at least one described content recommendation and at least one described content recommendation
The determination step of recommendation order, including:According to the historical interest of user point and the position location, at least one time is obtained
Select content;From at least one described alternating content, it is described that the region scene that selection is associated with the position location matches
At least one content recommendation;The matching of the region scene associated according at least one described content recommendation with the position location
Degree, it is determined that the recommendation order between at least one described content recommendation.
In an optional embodiment, between at least one described content recommendation and at least one described content recommendation
The determination step of recommendation order, including:The region scene associated according to the position location with the position location, is obtained at least
One alternating content;According to the historical interest of user point, from least one described alternating content, at least one described in selection
Individual content recommendation;The matching degree of the region scene associated according at least one described content recommendation with the position location, it is determined that
Recommendation order between at least one described content recommendation.
In an optional embodiment, methods described also includes:In response to the user-association in it is described at least one push away
The operation of content is recommended, the region scene that the daily record data produced according to the operation is associated with the position location updates described
At least one content recommendation.
In an optional embodiment, the renewal step of at least one content recommendation, including:According to the daily record number
According to the region scene associated with the position location, obtain at least one new content recommendation and at least one described new recommendation
Recommendation order between appearance;According to pushing away between at least one described new content recommendation and at least one described new content recommendation
Order is recommended, is updated at least one described content recommendation not by browsed content recommendation.
It is described not by the renewal step of browsed content recommendation, including following at least one in an optional embodiment
Kind:
In browsed content recommendation, do not rejected interior at least one new content recommendation described in being different from from described
Hold;
Described in being different from into described not at least one new content recommendation described in browsed content recommendation, adding extremely
The content of a few content recommendation;
According to the recommendation order between at least one described new content recommendation, not by browsed content recommendation described in adjustment
Between recommendation order.
In an optional embodiment, at least one described new content recommendation and at least one described new content recommendation it
Between recommendation order obtaining step, including:The region scene associated according to the daily record data with the position location, generation
At least one interest contents list;According to each self-corresponding content accounting of at least one described interest contents list, from it is described to
In a few interest contents list, at least one described new content recommendation is obtained;According at least one new content recommendation
User accesses value, determines the recommendation order between the new content recommendation.
In an optional embodiment, the generation step of at least one interest contents list, including following at least one
Plant generation operation:
According to action type of the user recorded in the daily record data at least one content recommendation, from institute
State in the content that user was operated with the action type, obtain user and access the content that value meets sets requirement, to generate
First interest contents list;
According to the content in the first interest contents list, in the region scene progress associated with reference to the position location
Hold the recommendation of dimension, to generate the second interest contents list;
According to the mark of the content operated by the user recorded in the daily record data, closed with reference to the position location
The region scene of connection carries out the recommendation of user's dimension, to generate the 3rd interest contents list;
According to the historical behavior data of the user, user historical content interested is obtained, it is emerging to generate the 4th
Interesting contents list;
According to the content in the 4th interest contents list, used with reference to the region scene that the position location is associated
The recommendation of family dimension, to generate the 5th interest contents list.
In an optional embodiment, the first interest contents list, the second interest contents list, the described 3rd
Interest contents list, the 4th interest contents list and the corresponding content accounting of the 5th interest contents list are passed successively
Subtract, and plus and be 1.
In an optional embodiment, the obtaining step of at least one new content recommendation, in addition to:If it is described at least
One new content recommendation includes content of multimedia, according to the equipment performance and/or network performance of the user, selection resolution ratio symbol
Close desired content of multimedia.
In an optional embodiment, operation of the user at least one content recommendation, including it is following at least
It is a kind of:Click on, browse, pulling up, gliding, for a long time stop, search, comment, thumb up, collection, forward, share, not liking and
Browse end.
In an optional embodiment, the recommendation trigger event includes following at least one:Log in, browse, gliding, point
Hit and search for.
Correspondingly, the embodiment of the present application also provides a kind of recommendation apparatus, including:
Acquiring unit, in response to recommending trigger event, obtaining the history for recommending trigger event correspondence user emerging
The region scene of interesting point, position location and position location association;
Determining unit, is closed for the historical interest point according to the user, the position location and the position location
The region scene of connection, determines the recommendation order between at least one content recommendation and at least one described content recommendation;
Recommendation unit, for the recommendation order between at least one described content recommendation, recommending to the user described
At least one content recommendation.
In an optional embodiment, the acquiring unit specifically for:Obtain the corresponding scene mark in the position location
Note data or at least one non-scene labeled data;According to the scene labeled data or at least one non-scene labeled data,
Recognize the region scene of the position location association.
In an optional embodiment, the acquiring unit specifically for:
Residence time for the user in position location meets temporal regularity when being in, belonging to the position location
The type in region be apartment/residential quarter, and the user used in the position location wireless signal mark with the user
The wireless signal mark identical situation used at multiple weekends, determines the region scene of the position location association to be in;
Residence time for the user in position location meets the temporal regularity in company, the position location institute
Belong to the type in region for shopping centre/other non-residential areas, and the wireless signal mark that the user uses in the position location
Identical situation is identified with the user wireless signal that the time uses on weekdays, the region of the position location association is determined
Scene is in company;
Belong to festivals or holidays for the current time, the position location belongs to other cities and its type is apartment/live
Quarter, and the wireless signal that is used in the position location of the user identifies the wireless signal commonly used with before and identifies and differ
Situation, the region scene of position location association is determined to be on home leave;
Type for the position location affiliated area is Tour region, and the user is in the affiliated area in the position location
It is movable in each sight spot in domain, the wireless signal that the user commonly uses in the wireless signal mark that the position location is used with before
Mark is differed, the feelings of temporal regularity of the user when the residence time of the position location affiliated area meets tourism
Condition, the region scene for determining the position location association is tourism;
For other situations, determine the region scene of the position location association to go on business.
In an optional embodiment, the determining unit specifically for:According to the historical interest of user point and institute
Position location is stated, at least one alternating content is obtained;From at least one described alternating content, selection is closed with the position location
At least one described content recommendation that the region scene of connection matches;According at least one described content recommendation and the sprocket bit
The matching degree of the region scene of association is put, it is determined that the recommendation order between at least one described content recommendation.
In an optional embodiment, the determining unit specifically for:According to the position location and the sprocket bit
The region scene of association is put, at least one alternating content is obtained;According to the historical interest of user point, from it is described at least one
In alternating content, at least one described content recommendation of selection;Closed according at least one described content recommendation and the position location
The matching degree of the region scene of connection, it is determined that the recommendation order between at least one described content recommendation.
In an optional embodiment, described device also includes:Updating block, in response to the user-association in institute
The operation of at least one content recommendation is stated, the regional field that the daily record data produced according to the operation is associated with the position location
Scape, updates at least one described content recommendation.
In an optional embodiment, the updating block includes:Obtain subelement, for according to the daily record data and
The region scene of position location association, obtain at least one new content recommendation and at least one described new content recommendation it
Between recommendation order;Subelement is updated, for according at least one described new content recommendation and at least one described new recommendation
Recommendation order between content, updates at least one described content recommendation not by browsed content recommendation.
In an optional embodiment, the renewal subelement is specifically for performing following at least one operation:
In browsed content recommendation, do not rejected interior at least one new content recommendation described in being different from from described
Hold;
To described not by browsed content recommendation, it is not different from least one described new content recommendation of addition described
The content of at least one content recommendation;
According to the recommendation order between at least one described new content recommendation, not by browsed content recommendation described in adjustment
Between recommendation order.
In an optional embodiment, it is described acquisition subelement specifically for:According to the daily record data and the positioning
The region scene of position association, generates at least one interest contents list;According at least one described interest contents list each
Corresponding content accounting, from least one described interest contents list, obtains at least one described new content recommendation;According to institute
The user for stating at least one new content recommendation accesses value, determines the recommendation order between the new content recommendation.
In an optional embodiment, the acquisition subelement is operated specifically for performing following at least one generation:
According to action type of the user recorded in the daily record data at least one content recommendation, from
In the content that the user was operated with the action type, obtain user and access the content that value meets sets requirement, with life
Into the first interest contents list;
According to the content in the first interest contents list, in the region scene progress associated with reference to the position location
Hold the recommendation of dimension, to generate the second interest contents list;
According to the mark of the content operated by the user recorded in the daily record data, closed with reference to the position location
The region scene of connection carries out the recommendation of user's dimension, to generate the 3rd interest contents list;
According to the historical behavior data of the user, user historical content interested is obtained, it is emerging to generate the 4th
Interesting contents list;
According to the content in the 4th interest contents list, used with reference to the region scene that the position location is associated
The recommendation of family dimension, to generate the 5th interest contents list.
In an optional embodiment, the first interest contents list, the second interest contents list, the described 3rd
Interest contents list, the 4th interest contents list and the corresponding content accounting of the 5th interest contents list are passed successively
Subtract, and plus and be 1.
In an optional embodiment, the acquisition subelement is additionally operable to:If at least one described new content recommendation includes
Content of multimedia, according to the equipment performance and/or network performance of the user, is selected in the satisfactory multimedia of resolution ratio
Hold.
In an optional embodiment, operation of the user at least one content recommendation, including it is following at least
It is a kind of:Click on, browse, pulling up, gliding, for a long time stop, search, comment, thumb up, collection, forward, share, not liking and
Browse end.
In an optional embodiment, the recommendation trigger event includes following at least one:Log in, browse, gliding, point
Hit and search for.
In a possible design, processor and memory can be included in the structure of above-mentioned recommendation apparatus, it is described to deposit
Reservoir is used for the program for storing the recommendation method for supporting recommendation apparatus to perform above-described embodiment offer, and the processor is configured as
For performing the program stored in the memory.
Optionally, the recommendation apparatus can also include communication interface, for recommendation apparatus and other equipment or communication network
Network communicates.
The embodiment of the present application additionally provides a kind of computer-readable storage medium, by storing based on used in above-mentioned recommendation apparatus
Calculation machine software instruction, it, which is included, is used to perform the recommendation apparatus that the recommendation method of above-described embodiment offer provides for above-described embodiment
Involved program.
In the embodiment of the present application, the region associated with reference to the historical interest point of user, position location and position location
Scene, to user's content recommendation, is conducive to recommending to more conform to the content of user's request to user, improves recommendation effect, improve
User experience.
Brief description of the drawings
Accompanying drawing described herein is used for providing further understanding of the present application, constitutes the part of the application, this Shen
Schematic description and description please is used to explain the application, does not constitute the improper restriction to the application.In the accompanying drawings:
The schematic flow sheet for the recommendation method that Fig. 1 provides for the embodiment of the application one;
The schematic flow sheet for the recommendation method that Fig. 2 provides for another embodiment of the application;
The structural representation for the take-away system that Fig. 3 provides for the embodiment of the application one;
The operation flow schematic diagram for the commending contents that Fig. 4 provides for the another embodiment of the application;
The interface schematic diagram for the take-away system that Fig. 5 provides for the another embodiment of the application;
The interface schematic diagram for the take-away system that Fig. 6 provides for the another embodiment of the application;
The structural representation for the recommendation apparatus that Fig. 7 provides for the another embodiment of the application;
The structural representation for the recommendation apparatus that Fig. 8 provides for the another embodiment of the application.
Embodiment
To make the purpose, technical scheme and advantage of the application clearer, below in conjunction with the application specific embodiment and
Technical scheme is clearly and completely described corresponding accompanying drawing.Obviously, described embodiment is only the application one
Section Example, rather than whole embodiments.Based on the embodiment in the application, those of ordinary skill in the art are not doing
Go out the every other embodiment obtained under the premise of creative work, belong to the scope of the application protection.
Fig. 1 provides the schematic flow sheet of recommendation method for the embodiment of the application one.As shown in figure 1, methods described include with
Lower step:
101st, in response to recommending trigger event, the historical interest point for recommending trigger event correspondence user, position location are obtained
And the region scene of position location association.
102nd, the region scene associated according to the historical interest of user point, position location and position location, it is determined that at least
Recommendation order between one content recommendation and at least one content recommendation.
103rd, with the recommendation order between at least one content recommendation, at least one content recommendation is recommended to user.
In the present embodiment, recommend trigger event to refer to can trigger or recommend the event of operation suitable for triggering.Recommend triggering
Event can be the event of user's triggering, such as logging in, browse, glide, click on or search for.Certainly, trigger event is recommended
Can be the event that system is triggered automatically, such as some timed events.
What deserves to be explained is, according to the difference of application scenarios, above-mentioned recommendation trigger event adaptability is set, however it is not limited to
The above-mentioned several situations enumerated.
In the present embodiment, recommendation trigger event can be monitored in real time, when monitoring recommendation trigger event, it is determined that needing
To user's content recommendation;Then the relevant information of commending contents institute foundation is obtained, is primarily referred to as recommending trigger event correspondence user
Historical interest point, position location and position location association region scene.
Above-mentioned recommendation trigger event correspondence user is primarily referred to as producing the user for recommending trigger event, or with recommending to trigger
Event has the user of mapping relations.
The historical interest point of above-mentioned user is used to represent interest preference of the user in time in the past.In realization, user
Historical interest point can include the information of any interest preference in time in the past that can embody user, can for example include
User's arbitrary act in time in the past section, such as buying, browse, click on, evaluating, content related thumb up, and/or, use
The interest characteristics at family etc..
Above-mentioned position location can be the position that user is currently located, and can be automatically positioned by system, but not limited to this.According to
The difference of application scenarios, position location can have different explanations.For example, in application is taken out, position location can also be use
The receipts meal address of family positioning, receipts meal address is different from user's present position.
The region scene of above-mentioned position location association mainly describes the scene that user is in position location, for instance it can be possible that on
Class, be in, travel, going on business, being on home leave.Region scene is different, and the demand of user may be different.For example, user is in public affairs
During department's working, fast food is more inclined to, therefore fast food can be recommended to user;But when user travels to somewhere, to local characteristic
Diet is interested, now should recommend speciality to user.
In the present embodiment, the regional field associated with reference to the historical interest point of user, position location and position location
Scape, it is determined that the recommendation order between needing at least one content recommendation and at least one content recommendation for recommending to user, so
Afterwards with the recommendation order between at least one content recommendation, recommend at least one content recommendation to user, different from prior art
Historical interest point and position location based on user fully demonstrate shadow of the region scene to user's request to user's content recommendation
Ring, be conducive to recommending to more conform to the content of user's request to user, improve recommendation effect, and then improve user experience.
In above-described embodiment or following embodiments, it is contemplated that in various application scenarios, user typically can be all preserved
The self-portrait of relevant information, referred to as user.According to the difference of application scenarios, the information that the self-portrait of user is included can not yet
Together.By taking the take-away class system in e-commerce field as an example, the self-portrait of user can include but is not limited to following information:
Based on attribute information:System identifier (such as ID of user:2612*****627), city (such as Beijing), use where user
The sex (such as man) at family, the facility information (such as Samsung Note3) of user, platform age (such as 13 days), hobby, food and drink
The level of consumption, receipts meal address etc.;Primary attribute information can be shown by modes such as word, charts;
Transaction attribute information:It can be obtained, for example may be used based on the data statistics in certain time scope (such as nearly 1 year)
Including complete the moon single amount trend, platform total revenue trend, it is preferential before it is objective it is monovalent, preferential after objective unit price etc.;Text can be passed through
The modes such as word, chart show transaction attribute information;
Transaction preference information:Can be obtained based on the data statistics in certain time scope (nearly 90 days), for example may include but
It is not limited to the preference of following dimension:User's concentration degree, taste, food materials, the means of payment, name of firm, consumption time, consumption
Date, commercial circle etc.;Transaction preference information can be shown by modes such as word, charts;
Preferential attribute information:Subsidy rate trend, without privileges order scaling trend, the nearly monthly average subsidy amount of money, nearly one
Always subsidize the amount of money, a nearly month without privileges order volume, nearly month without privileges order accounting etc. within individual month;Word, figure can be passed through
The modes such as table show preferential attribute information;
Viscous properties information:Finally place an order time, competing product information, the frequency that places an order, first single time etc.;Can by word,
The modes such as chart show viscous properties information;
Consumer's Experience information:Logistics list accounting trend, logistics odd number, logistics list accounting, overtime trouble ticket accounting, logistics Dan Ping
Equal food delivery duration, logistics list average distance, logistics beat reward number of times, logistics and beat money reward volume, comment on number, difference and comment name of firm, complain
Number of times, overall merit, dispatching evaluation, vegetable evaluation etc.;Consumer's Experience information can be shown by modes such as word, charts.
For example, from database or can be deposited such as phone number, ID card No. according to the identity identification information of user
The self-portrait that user is read in equipment is stored up, and then from the self-portrait of user, statistical separates out the historical interest point of user.
In above-described embodiment or following embodiments, application can provide a user scene marking Function, based on this, user
Scene mark can be carried out to position location.The situation of scene mark is carried out to position location to user, sprocket bit can be obtained
Corresponding scene labeled data is put, the region scene of position location association is recognized based on the scene labeled data.But, it is more susceptible
Under condition, user may not carry out scene mark to position location.The feelings of scene mark are not carried out to position location for user
Condition, it is considered to which different zones scene has respective characteristic, these characteristics can pass through some related non-scene labeled data of user
To embody, then the corresponding at least one non-scene labeled data in position location can be obtained, be marked according at least one non-scene
Data, the region scene of identification position location association.
Alternatively, above-mentioned at least one non-scene labeled data includes but is not limited to:The type of position location affiliated area,
User is in the residence time of position location affiliated area, user in the scope of activities of position location affiliated area, user in positioning
Wireless signal mark and/or current time feature that position is used.
Different application scene, the division of region scene can be different.Exemplified by taking out application, relatively conventional regional field
Scape includes but is not limited to:Go on business, travel, being in, being on home leave in company and other places.
With reference to above-mentioned non-scene labeled data, and several regional fields such as go on business, travel, being in, being on home leave in company, other places
Scape, the mode to the region scene of identification position location association is illustrated:
It is in scene:
In time, with certain rule, this rule can be presented as more fixed (such as evening time of having a rest in the evening
11 .-second days 5:00 AMs), it is either more fixed to leave home the work hours (such as 7 points or so of every morning) or more
It is fixed to get home the time (such as 8 points or so of every night).The rule of this time be not once in a while, it is but long-term.Therefore,
When this scene is in identification, it may be considered that this temporal rule.
In terms of wireless signal, it may have the rule more fixed, this rule can be presented as that user is at weekend etc.
In the probability higher time, it can be surfed the Net by the wireless network of family, the mark (such as SSID) of family wireless network can the cycle
Property frequently occurs in a long time.Therefore, when this scene is in identification, it may be considered that the rule in terms of wireless signal
Rule.
In addition, the corresponding address of scene of being in is generally apartment/residential quarter, if position location is a shopping centre, typically
It is not belonging to scene of being in.Therefore, when this scene is in identification, it is also contemplated that area type.
Based on above-mentioned analysis, it can be determined that whether residence time of the user in position location meets time rule when being in
Rule, whether the type for judging position location affiliated area is apartment/residential quarter, and it is wireless to judge that user uses in position location
Signal identification identifies whether identical with the wireless signal that user uses at multiple weekends;If user is in the residence time of position location
Meet temporal regularity when being in, the type of position location affiliated area is apartment/residential quarter, and user uses in position location
The wireless signal mark identical situation that is used at multiple weekends with user of wireless signal mark, then can determine that user exists
The region scene of family, i.e. position location association is to be in.
In company:
In time, with certain rule, this rule can be presented as more fixed commuter time (such as each work
Make the 9 points -6 pm of morning of day).The rule of this time be not once in a while, it is but long-term.Therefore, identification company this
When planting scene, it may be considered that this temporal rule.
In terms of wireless signal, it may have the rule more fixed, this rule can be presented as that user waits on weekdays
In company's probability higher time, it can be surfed the Net by wireless network in company, the mark of wireless network is (for example in company
SSID) understand periodically or frequently occur in a long time.Therefore, when identification is in this scene of company, it may be considered that nothing
Rule in terms of line signal.
In addition, company position is generally non-residential area/shopping centre etc., if position location is apartment/residential quarter, typically not
Belong in company's scene.Therefore, when identification is in this scene of company, it is also contemplated that area type.
Based on above-mentioned analysis, it can be determined that time rule of the user when whether the residence time of position location meets in company
Rule, whether the type for judging position location affiliated area is shopping centre/other non-residential areas, and judge user in position location
The wireless signal mark used identifies whether identical with user's wireless signal that the time uses on weekdays;If user is in sprocket bit
The residence time put meets the temporal regularity in company, and the type of position location affiliated area is shopping centre/other non-residential
Area's (such as mansion, Recreational places, office buildings), and the wireless signal mark that user uses in position location exists with user
The wireless signal mark identical situation that working day uses, then can determine user in the region of company, i.e. position location association
Scene is in company.
It is on home leave scene:
In general, all it is to be on home leave in the legal festivals and holidays, and is on home leave and can typically arrive different from where family and company
Other cities in city, scope of activities is usually that in apartment/residential quarter, and the wireless signal that user uses can also become
Change.
Based on above-mentioned analysis, it can be determined that whether current time belongs to festivals or holidays, judge whether position location belongs to other
City and its type are apartment/residential quarter, and judge that wireless signal mark that user uses in position location is used with before
Wireless signal identifies whether identical;If current time belongs to festivals or holidays, position location belong to other cities and its type be apartment/
Residential quarter, and the wireless signal that is used in position location of user identifies the wireless signal commonly used with before and identifies and differ, explanation
User may be on home leave in other places, therefore can determine the region scene of position location association to be on home leave.
Tourism scene:
In general, tourism can all arrive Tour region, and scope of activities can be in each sight spot, and when travelling not
Residence time with sight spot has certain regularity, will not be stayed in a place too long, be, for example, less than 1 day, and user uses
Wireless signal can also change.
Based on above-mentioned analysis, it can be determined that whether the type of position location affiliated area is Tour region, judges that user exists
Whether the scope of activities of position location affiliated area is limited in each sight spot, judges the wireless signal mark that user uses in position location
Know and identified whether with the wireless signal commonly used before identical, and judge that user is in the residence time of position location affiliated area
No temporal regularity when meeting tourism;If the type of position location affiliated area is Tour region, user is belonging to position location
Movable in each sight spot in region, the wireless signal mark that the wireless signal mark that user uses in position location is commonly used with before is not
Identical, temporal regularity of the user when the residence time of position location affiliated area meets tourism illustrates user in tourism, therefore really
The region scene for determining position location association is tourism.
For other situations outside above-mentioned scene, it may be determined that the region scene of position location association is to go on business.
What deserves to be explained is, each embodiment of the region scene of above-mentioned determination position location association is merely illustrative, and
Not limited to this.For example, being in above-mentioned determination or during company's scene, it is contemplated that wireless signal identifies this information,
But if family does not have wireless signal or company not to allow employee to surf the Net, then this information can not be considered.In another example, upper
State determination to be on home leave during scene, it is also contemplated that the times or frequency that user occurs in apartment/residential quarter is (such as suitable
Long time occurrence number be more than 2 times) etc. information, to improve accuracy of judgement degree.
For ease of understanding the embodiment of above-mentioned determination region scene, it is illustrated below by table 1:
Table 1
In above-described embodiment or following embodiments, according to the historical interest of user point, position location and position location
The region scene of association, the step of determining the recommendation order between at least one content recommendation and at least one content recommendation,
It can use but be not limited to following several ways:
In one embodiment, at least one candidate can be obtained according to the historical interest point of user and position location
Content;From at least one alternating content, at least one for selecting the region scene associated with position location to match is recommended interior
Hold;The matching degree of the region scene associated according at least one content recommendation with position location, determines at least one content recommendation
Between recommendation order.
In a kind of above-mentioned embodiment, user may be interested one can be determined according to the historical interest point of user
Context, according to position location, it is determined that apart from upper relatively suitable another context;Select the friendship of two contexts
Collection, is used as alternating content, it is ensured that alternating content meets historical interest point and the requirement of position location two of user simultaneously;So
Afterwards, it is considered to which user is in the scene of position location, it is to go on business or be in or in company etc., and then calmodulin binding domain CaM scene is by candidate
Remove in content with the unmatched content of region scene, so as to obtain content recommendation;And then according to content recommendation and region scene
Matching degree, determine the sequence (Rank) between content recommendation, i.e. recommendation order.
It is alternatively possible to be pre-configured with the standard content under different zones scene, and standard content and region scene are set
Between matching degree, be designated as matches criteria degree.On the matching degree between any content recommendation and region scene, this can be calculated
The similarity of content recommendation and each standard content under the scene of region, the corresponding matches criteria of selection similarity highest standard content
Degree, is used as the content recommendation and the matching degree of region scene;Or, can be by each standard under the content recommendation and region scene
Similarity between appearance determines the corresponding weight of each standard content as weight foundation, by each standard content and region scene it
Between matches criteria degree be weighted summation, be used as the matching degree of the content recommendation and region scene.What deserves to be explained is, here
The mode for calculating the matching degree between content recommendation and region scene is only for example, however it is not limited to this.
In another embodiment, at least one time can be obtained according to the historical interest point of user and position location
Select content;From at least one alternating content, at least one recommendation for selecting the region scene associated with position location to match
Content;According to the matching degree of the historical interest of user point and at least one content recommendation, determine between at least one content recommendation
Recommendation order.
In above-mentioned another embodiment, user may be interested one can be determined according to the historical interest point of user
Context, according to position location, it is determined that apart from upper relatively suitable another context;Select the friendship of two contexts
Collection, is used as alternating content, it is ensured that alternating content meets historical interest point and the requirement of position location two of user simultaneously;So
Afterwards, it is considered to which user is in the scene of position location, is gone on business or is in or in company etc., and then calmodulin binding domain CaM scene is by candidate
Remove in appearance with the unmatched content of region scene, so as to obtain content recommendation;And then according to content recommendation and the history of user
The matching degree of point of interest, determines the sequence between content recommendation, i.e. recommendation order.
It is with a kind of above-mentioned difference of embodiment:Determine the foundation difference of the sequence between content recommendation.
Alternatively, according to the difference of the way of realization of the historical interest of user point, the historical interest of content recommendation and user
The calculation of the matching degree of point would also vary from.
If for example, the historical interest point of user is presented as the interest characteristics of user, the feature of content recommendation can be extracted,
Calculate content recommendation feature and user interest characteristics between similarity, feature and user based on content recommendation it is emerging
Similarity between interesting feature, determines the matching degree of content recommendation and the historical interest point of user.For example, maximum phase can be selected
Like matching degree of the degree as content recommendation and the historical interest of user point, or the average value of similarity can be calculated as recommendation
Matching degree of historical interest point of content and user, etc..
In another example, if the historical interest point of user is presented as that user such as produced purchase, and clicked on, browse at the arbitrary act in the past
Historical content, then the similarity between content recommendation and historical content can be calculated, based on content recommendation and historical content
Similarity, determines the matching degree of content recommendation and the historical interest point of user.For example, maximum similarity can be selected as recommendation
The matching degree of the historical interest point of content and user, or the average value of similarity can be calculated as content recommendation and user
Matching degree of historical interest point, etc..
In another embodiment, the region scene that can be associated according to position location with position location is obtained at least
One alternating content;According to the historical interest of user point, from least one alternating content, at least one content recommendation is selected;
The matching degree of the region scene associated according at least one content recommendation with position location, is determined between at least one content recommendation
Recommendation order.
In above-mentioned another embodiment, a context of region scene according to region scene, can be determined for compliance with, according to
Position location, it is determined that apart from upper relatively suitable another context;The common factor of two contexts is selected, as in candidate
Hold, it is ensured that alternating content meets region scene and the requirement of position location two simultaneously;Then, it is considered to the historical interest of user
Point, the content that the historical interest point of user will not be met in alternating content is removed, so as to obtain content recommendation;And then according to recommendation
The matching degree of content and region scene, determines the sequence between content recommendation, i.e. recommendation order.
It is alternatively possible to be pre-configured with the standard content under different zones scene, and standard content and region scene are set
Between matching degree, be designated as matches criteria degree.Based on this, it can calculate similar between each content and standard content in content library
Degree, a context of region scene is determined for compliance with based on similarity from content library.
Alternatively, according to the difference of the way of realization of the historical interest of user point, according to the historical interest of user point, from time
The mode for screening content recommendation in content is selected to would also vary from.
If for example, the historical interest point of user is presented as the interest characteristics of user, the feature of alternating content can be extracted,
Calculate alternating content feature and user interest characteristics between similarity, feature and user based on alternating content it is emerging
Similarity between interesting feature, determines whether alternating content meets the historical interest point of user;And then going through for user will not met
The content of history point of interest is removed, so as to obtain content recommendation.For example, similarity threshold can be set, similarity is less than similar
The alternating content of degree threshold value is considered as the content for not meeting the historical interest point of user, or can set similarity dimensions, by phase
It is considered as the content of historical interest point, etc. for not meeting user like the alternating content that degree is not belonging to similarity dimensions.
In another example, if the historical interest point of user is presented as that user such as produced purchase, and clicked on, browse at the arbitrary act in the past
Historical content, then the similarity between alternating content and historical content can be calculated, based on alternating content and historical content
Similarity, determines whether alternating content meets the historical interest point of user;And then will not meet in the historical interest point of user
Appearance is removed, so as to obtain content recommendation.For example, similarity threshold can be set, similarity is less than to the candidate of similarity threshold
Content is considered as the content for not meeting the historical interest point of user, or can set similarity dimensions, and similarity is not belonging into phase
It is considered as like the alternating content for spending scope and does not meet content of historical interest point of user, etc..
On the acquisition modes of the matching degree between any content recommendation and region scene, reference can be made to a kind of above-mentioned embodiment party
Corresponding description in formula, is being repeated no more.
In another embodiment, the region scene that can be associated according to position location with position location is obtained at least
One alternating content;According to the historical interest of user point, from least one alternating content, at least one content recommendation is selected;
According to the matching degree of at least one content recommendation and the historical interest of user point, the recommendation between at least one content recommendation is determined
Sequentially.
In above-mentioned a further embodiment, a context of region scene according to region scene, can be determined for compliance with, according to
Position location, it is determined that apart from upper relatively suitable another context;The common factor of two contexts is selected, as in candidate
Hold, it is ensured that alternating content meets region scene and the requirement of position location two simultaneously;Then, it is considered to the historical interest of user
Point, the content that the historical interest point of user will not be met in alternating content is removed, so as to obtain content recommendation;And then according to recommendation
The matching degree of content and region scene, determines the sequence between content recommendation, i.e. recommendation order.
Associated description in the embodiment, reference can be made to aforementioned embodiments, will not be repeated here.
In above-described embodiment or embodiment, historical interest point, position location and the sprocket bit of user can be combined
The region scene of association is put, to user's content recommendation, influence of the region scene to user's request is taken into full account, is conducive to user
Recommendation more conforms to the content of user's request.After content recommendation to be recommended to user, user can hold for content recommendation
The various operations of row, such as clicking on, browse, evaluate, search for, also some non-operations for content recommendation are to a certain degree certainly
On can also influence content recommendation, be referred to as being associated with the operation of content recommendation.Based on this, it can be pushed away according to user-association in above-mentioned
The operation of content is recommended, upgrade in time content recommendation so that content recommendation can dynamically track the demand of user.
Another embodiment of the application provides a kind of recommendation method, as shown in Fig. 2 the recommendation method includes:
201st, in response to recommending trigger event, the historical interest point for recommending trigger event correspondence user, position location are obtained
And the region scene of position location association.
202nd, the region scene associated according to the historical interest of user point, position location and position location, it is determined that at least
Recommendation order between one content recommendation and at least one content recommendation.
203rd, with the recommendation order between at least one content recommendation, at least one content recommendation is recommended to user.
204th, in response to user-association in the operation of at least one above-mentioned content recommendation, the daily record produced according to the operation
The region scene that data are associated with position location, updates at least one content recommendation.
In the present embodiment, step 201-203 can be found in the description of step 101-103 in previous embodiment, herein no longer
Repeat.
In the present embodiment, user-association in the operation of at least one above-mentioned content recommendation include but is not limited to it is following at least
It is a kind of:Click on, browse, pulling up, gliding, for a long time stop, search, comment, thumb up, collection, forward, share, not liking and
Browse end.The renewal influence that different operating is produced at least one content recommendation is different, can be divided into positive influence and negative sense shadow
Ring.One kind citing can be found in table 2:
Table 2
In the present embodiment, by monitoring the operation of user, can understand user in time by the operation of user currently needs
Ask, and then at least one content recommendation that can upgrade in time, in order to which the content for recommending user is capable of Dynamic Matching user's
Current demand.Here renewal includes but is not limited to following effect:Reduction user browses the content do not clicked on but, for example, reduce
Its Rank;Strengthen the content that user browses and clicked on, for example, improve its Rank;Strengthen user by the content of keyword search with
And the content of the hit keyword, for example improve the Rank of these contents.
The application scenarios of above-mentioned at least one content recommendation of renewal are illustrated:
Assuming that user it is interior for the previous period repeatedly placed an order from Sichuan cuisine shop, illustrate that user is interested in Sichuan cuisine, based on use
This interest at family, is combining the region scene of the historical interest point of user, position location and position location association to user
During content recommendation, the related trade company of Sichuan cuisine can be recommended to user.If related for the Sichuan cuisine of recommendation within the ensuing time
Trade company, in user's navigation process and do not click on, the interest of this explanation user, which is changed, no longer pays close attention to Sichuan cuisine, therefore does not have
The related trade company of recommended Sichuan cuisine is clicked on, by monitoring the operation of user, it can be found that the demand of user is changed, no
Sichuan cuisine is paid close attention to again.And the method for using the present embodiment to provide, operated, can weakened related to Sichuan cuisine in content recommendation by updating
Trade company, for example reduce the Rank of the related trade company of Sichuan cuisine, it might even be possible to reject the related trade company of Sichuan cuisine, strengthen or recommend again
Meet the content of user's current demand, improve Consumer's Experience.
In above-described embodiment or following embodiments, the daily record data produced and position location is operated to associate according to user
Region scene, the step of updating at least one content recommendation, Ke Yiwei:The region associated according to daily record data with position location
Scene, obtains the recommendation order between at least one new content recommendation and at least one new content recommendation;According at least one
Newly the recommendation order between content recommendation and at least one new content recommendation, updates at least one content recommendation.
In a kind of update mode, the full content that can be directed at least one content recommendation is updated.It is described more
New paragon includes following at least one:It is every to be different from least one new content recommendation by least one content recommendation
Content is rejected;In at least one content recommendation, add at least one new content recommendation different from the interior of any content recommendation
Hold;According to the recommendation order between at least one new content recommendation, the recommendation order between at least one content recommendation is adjusted.
In another update mode, at least one content recommendation can be divided into the content browsed and do not browsed
Content.The content browsed for user, user has certain memory, in order to retain the appearance in user's memory, is easy to use
Family is returned and checked, the content browsed can not be done and updated, and only the content recommendation not browsed is done and updated.To not browsed
The renewal for the content recommendation crossed can include following at least one:Never by browsed content recommendation, reject and be different from extremely
Content in a few new content recommendation;To in browsed content recommendation, not added at least one new content recommendation not
It is same as the content of at least one content recommendation;According to the recommendation order between at least one new content recommendation, adjustment is not browsed
Recommendation order between the content recommendation crossed.
For example, under a kind of scene, new content recommendation is identical with content recommendation before, recommendation order is differed only in not
Together, then can be according to the recommendation order between new content recommendation, accommodation is not by pushing away between browsed content recommendation
Recommend order.So when user needs to browse these not by browsed content recommendation, it will be carried out according to new recommendation order
Browse.
In another example, under a kind of scene, new content recommendation is different from content recommendation before, including content recommendation is not before
Comprising content, and these new contents recommendation order before not by browsed content recommendation, illustrate these new content ratios
Relatively meet user's current demand, then these contents can be added to not by browsed content recommendation, and by these contents
Come not by the foremost of browsed content recommendation, in order to which user preferentially sees these new contents.
In another example, under a kind of scene, new content recommendation is different from content recommendation before, before the portion in content recommendation
Point content is excluded, if the content being excluded belongs to browsed content recommendation, is not processed;If be excluded
Content belongs to not by browsed content recommendation, then never can be rejected these contents in browsed content recommendation.
In above-described embodiment or following embodiments, at least one new content recommendation and at least one new content recommendation it
Between recommendation order obtaining step, Ke Yiwei:The region scene associated according to daily record data with position location, generation at least one
Individual interest contents list;According to each self-corresponding content accounting of at least one interest contents list, from least one interest content
In list, at least one new content recommendation is obtained;Accessed and be worth according to the user of at least one new content recommendation, it is determined that new recommend
Recommendation order between content.
Alternatively, the generation step of at least one interest contents list, including following at least one generation operation:
According to action type of the user recorded in daily record data at least one content recommendation, from user with action type
In the content operated, obtain user and access the content that value meets sets requirement, to generate the first interest contents list;
According to the content in the first interest contents list, the region scene associated with reference to position location carries out content dimension
Recommend, to generate the second interest contents list;
According to the mark of the content being user-operably recorded in daily record data, the region scene associated with reference to position location
The recommendation of user's dimension is carried out, to generate the 3rd interest contents list;
According to the historical behavior data of user, user's historical content interested is obtained, to generate the 4th interest content row
Table;
According to the content in the 4th interest contents list, the region scene associated with reference to position location carries out user's dimension
Recommend, to generate the 5th interest contents list.
Correspondingly, at least one above-mentioned interest contents list can include:First interest contents list, the second interest content
In list, the 3rd interest contents list, the 4th interest contents list and the 5th interest contents list any one or it is any
Several combinations.
In an embodiment, at least one above-mentioned interest contents list includes simultaneously:First interest contents list,
Second interest contents list, the 3rd interest contents list, the 4th interest contents list and the 5th interest contents list.Can be pre-
This corresponding content accounting of five interest contents lisies is first configured, the weight of this five interest contents lisies is embodied by content accounting
Spend.Here the corresponding content accounting of interest contents list refers to be selected as new content recommendation from interest contents list
The ratio of content quantity and new content recommendation sum.Wherein, the corresponding content accounting sum of each interest contents list is 1, and
First interest contents list, the second interest contents list, the 3rd interest contents list, the 4th interest contents list and the 5th are emerging
The corresponding content accounting of interesting contents list is successively decreased successively.
In above-described embodiment or following embodiments, value is accessed for the user of new content recommendation, can be according to user
The time difference of the time gap current time of new content recommendation progress last time operation is embodied.In general, the time
Difference is bigger, and the new content recommendation is smaller to the value of user, i.e., user's access value is lower.User accesses value can be according to formula
N (t)=N0*e-λtCalculate.Wherein, N (t) represents that user accesses value, is the amount relevant with time difference t, N0=N (0) is initial
Amount, the i.e. amount when the time difference is zero;λ is attenuation coefficient;E, as math constant, is the truth of a matter in natural logrithm function;T tables
Show the time and the difference of current time of user's last time operation (such as browse, click on or evaluate) new content recommendation.
At least one new respective user of content recommendation is calculated based on above-mentioned formula and accesses value, can be visited according to user
The order of value from high to low is asked, the recommendation order between at least one new content recommendation is determined.
In above-described embodiment or following embodiments, content of multimedia is potentially included at least one new content recommendation, it is many
The data volume of media content is generally large, and different resolution can be set for content of multimedia, so can contemplate user's
Equipment performance and/or network performance, select the satisfactory content of multimedia of resolution ratio, to reduce multimedia content deliveries or show
Occur the probability of the phenomenons such as not smooth or interim card during showing, improve user experience.
The method that the application the various embodiments described above are provided, available for various application systems, to recommend different content, for example may be used
Applied to processing system for video to recommend message of film and TV, the purchase system that can be applied in ecommerce with recommend extensive stock/
Service, the take-away system that can be applied in ecommerce can be applied to financing system to recommend financing to produce to recommend catering information
Product, etc..
In actual implementation process, the method that the various embodiments described above are provided can be by the client or service end in application system
Or any equipment with recommendation function is independently implemented.Or, the method that the various embodiments described above are provided also can be in application system
Client and service end coordinate implement.
With reference to the system of take-away, the recommended flowsheet implemented with the client and service end in take-away system is carried out
Explanation.
The structural representation for the take-away system that Fig. 3 provides for the embodiment of the application one.As shown in figure 3, the take-away system bag
Include:Client 10 and service end 20.
It can be wirelessly or non-wirelessly network connection between client 10 and service end 20.Optionally, if client 10 passes through
Mobile network communicates to connect with service end 20, and the network formats of the mobile network can be 2G (GSM), 2.5G (GPRS), 3G
Any one in (WCDMA, TD-SCDMA, CDMA2000, UTMS), 4G (LTE), 4G+ (LTE+), WiMax etc..Certainly, remove
Outside mobile network, client 10 can also be communicated to connect by the mode such as Wi-Fi, bluetooth, infrared and service end 20.
Client 10 is corresponding with service end 20, is normally at user side, mainly provides the user local service, generally needs
Realize following functions:To user the related information of application is shown, respond the operation of user and interacted with realizing with user, and
Communicated with service end 20.
Service end 20 is mainly what is serviced for client 10, for example, provide application related information to client 10, preserve
The data of client 10, control logic, and responsible hind computation etc. are provided to client 10.
Wherein, according to the difference of application scenarios, the function that client 10 and service end 20 are implemented can be different.Under
Face emphasis describes the flow that client 10 coordinates implementation content to recommend with service end 20.
Client 10 is mainly used in the operation of monitoring users, carries out event capture, and correlation recommendation is operated when capturing user
During trigger event, recommendation trigger event is reported to service end 20, for service end 20 in response to recommend trigger event return to
The content recommendation that family is recommended, and show the content recommendation that service end 20 is returned to user.
Correspondingly, service end 20 is mainly used in response to recommending trigger event, obtains and recommends trigger event correspondence user's
The region scene of historical interest point, position location and position location association;According to the historical interest of user point, position location with
And the region scene of position location association, determine that the recommendation between at least one content recommendation and at least one content recommendation is suitable
Sequence;With the recommendation order between at least one content recommendation, at least one content recommendation is returned into client 10, for client
End 10 sequentially shows at least one content recommendation to user.
As shown in figure 4, the operation flow of commending contents is as follows:
User logs in take-away system in client 10, browses take-away content, clicks on trade company etc., can trigger and recommends operation.Visitor
Family end 10 catches log-in events, browsing event or clicks on event, and reports service end 20.
Service end 20 is according to the subscriber identity information carried in log-in events, browsing event or click event, such as mobile phone
Number, pass id etc., go in database to obtain the self-portrait information of user, and according to the current position location of user recall it is attached
Nearly trade company, combines both and obtains user's content progress recommendation sequence interested.For example, user likes eating McDonald,
It can recommend to user comprising the fast food including McDonald;User is sensitive to favor information, can recommend currently completely to subtract to user
The larger trade company of amplitude;User is high-quality user, can recommend high-quality taste trade company to user;The viscosity of user is not
Height, belongs to the user of decline phase, then can provide a user reward voucher and recommend the applicable trade company of reward voucher to user.
Apart from the above, service end 20 also needs to combine subscriber identity information and the current position location (distribute leaflets of user
Address, the positioning address of user) region scene is determined, for instance it can be possible that going on business, being on home leave, travelling, being in, in company etc.;Consider
Region scene internally holds the influence of recommendation process, considers this factor of region scene in content recommendation process, allows commending contents to arrange
Sequence is more reasonable, more conforms to user's request.
Detailed implementation process on carrying out commending contents according to historical interest point, position location and region scene can join
See previous embodiment, will not be repeated here.
Exemplified by the interface of system is taken out shown in Fig. 5, content recommendation is showed user by client 10 with tabular form.With
Family is seen after contents list, clicks on " western young master's the meat clip Mo ", as shown in Figure 5.
Client 10 captures the clicking operation of user.In one embodiment, client 10 can capture use
After the clicking operation at family, the daily record data that clicking operation is produced is reported to service end 20.In another embodiment, client 10
Can according to setting report cycle, after report cycle is reached, by all operations captured in the report cycle produce
Daily record data report service end 20.In another way of example, client 10 can use event trigger mechanism, i.e., only
When capturing particular event, all daily record datas not reported before this can be just reported to service end 20.
Service end 20 receives the daily record data that client 10 is reported, and performs two kinds of processing logics based on the daily record data.
One kind handles logic:Based on the title of trade company A in the daily record data, the details for retrieving the trade company are pushed to user, and
Similar trade company is searched using trade company A self-portrait, recommends similar trade company to user, as shown in Figure 6.
Alternatively, the self-portrait of trade company includes but is not limited to following information:
Base attribute information:For example, trade company address, affiliated commercial circle, trade company's property, business scope, fast-selling period, industry
Qualification, electronic contract effective status, contract discount rate, important (point) trade company (KeyAccount, KA) mark, average Rank value,
Main body qualification, on-line time, business hours.
Indication information:Medium index, superior index, weak tendency index can be divided into;Superior index may include:Trade company value, it is excellent
Before favour monovalent, the average daily flowing water of visitor, week again purchase rate, order conversion ratio, net investment return rate (Return on Investment,
ROI), exposure number, light exposure, the number of visiting people, the amount of placing an order, the number that places an order, visit capacity, the single number of completion, completion list are measured, exposed
The independent visitor of light-access (unique visitor, UV) conversion ratio, exposure-accession page pageview (Page View, PV) turn
Rate, access-the UV conversion ratios that place an order, access-the PV conversion ratios that place an order, place an order-complete UV conversion ratios, place an order-complete PV conversion ratios,
Working day marketing susceptibility, day off marketing susceptibility, sticky user, preferential rear objective unit price, user's scoring, customer complaint etc.;
Weak tendency index can include:Platform income;Remaining is medium index;
Trade company's value information:For example, order conversion ratio, all purchase rates again, preferential preceding monovalent, the average daily flowing water of visitor, net ROI, mouth
Taste food materials etc.;Word, graph representation trade company value information can be passed through;
Marketing susceptibility:Can be by the relation between working day and the order volume and order subsidized price on day off come body
It is existing;Can be marketed susceptibility by word, graph representation;
Customer data:For example, the index value such as user's viscosity, user's scoring, customer complaint in different cities, commercial circle;
Word, graph representation customer data can be passed through;
Competing product comparative information:For example, the online number of days of each competing product, vegetable number, business duration, rise send valency, dispatching expense etc.
Data;Word, the competing product comparative information of graph representation can be passed through;
Marketing data:For example, old user ROI, trade company's investment proportion, trade company's subsidy rate etc.;Word, graph representation can be passed through
Marketing data.
It is another processing logic be:The trade company A (" western young master's the meat clip Mo " as shown in Figure 5) clicked on according to user, with reference to
The region scene of current position location association, is obtained between at least one new content recommendation and at least one new content recommendation
Recommendation order.
For example, service end 20 can obtain all trade companies that user clicked in a period of time, according to formula N (t)=N0*
e-λtThe user for calculating each trade company accesses value;The user of user profile, merchant information, trade company is accessed into the storage of the contents such as value
Into internal memory or external storage, database, caching etc..The selection user of service end 20 accesses value and is more than default value threshold value
Trade company, forms the first interest merchant list.
Further, the calmodulin binding domain CaM scene of service end 20, using cosine similarity, obtains each business in the first interest merchant list
The similar trade company at family, these similar trade companies form the second interest merchant list.For the trade company in the first interest merchant list,
If its user accesses, value is higher, and the quantity of the similar trade company of this trade company is relatively more, otherwise the quantity of similar trade company
It is relatively fewer.
Further, the trade company A that service end 20 is clicked on according to user, the similar users of acquisition user, calmodulin binding domain CaM scene, from
The trade company outside trade company A is selected in similar users trade company interested, the 3rd interest merchant list is formed.It is alternatively possible to sharp
Pearson came relevancy algorithm is used, the similarity between user is calculated.
Further, service end 20 obtains the historical interest trade company of user, such as has lower unirecord, receipts before being user
The trade companies such as Tibetan, thumb up, form the 4th interest merchant list.
Further, service end 20 obtains the similar users of user, calmodulin binding domain CaM according to the historical interest trade company of user
Scape, the trade company outside historical interest trade company is selected from similar users trade company interested, the 5th interest merchant list is formed.Can
Selection of land, it is possible to use Pearson came relevancy algorithm, calculates the similarity between user.
Service end 20 is according to the content accounting for being in advance the configuration of above-mentioned five interest lists, respectively from above-mentioned five lists
Select user to access the part trade company of Maximum Value, and the trade company of selection is ranked up according to user's access value.First to
The corresponding content accounting sum of 5th interest merchant list is 1, and is successively decreased successively, can so ensure to operate based on user and recommend
Trade company it is relatively many, and based on historical interest recommend trade company it is relatively fewer, to give full play to the value of user's real-time operation.
New trade company is pushed to client 10 by service end 20 according to the sequence between above-mentioned new trade company.
Client 10 receives the new trade company that service end 20 is pushed, the page where user is back to recommendation trade company from trade company A,
For example shown in Fig. 5 during the page, using the sequence between new trade company and new trade company, the recommendation trade company shown in Fig. 5 on the page is entered
Row updates.
Preferably, the trade company before being updated shown in Fig. 5 on the page can be divided into three classes by client 10, be respectively:With
Family is browsed but the not browsed trade company of sightless trade company, current visible trade company, user.
It is assumed that initial recommendations trade company includes:User is browsed but sightless trade company S0, S1, S2, current visible trade company
S3, S4, S5,12 trade companies such as user sightless trade company S6, S7, S8, S9, S10, S11.
Subsequently, user clicks trade company S4;Based on clicking operation of the user to trade company S4, service end 20 is returned to client 10
Go back to the new trade company sorted as follows:12 trade companies such as S0, S2, S4, S7, S5, S10, S22, S9, S8, S11, S23, S12.
Wherein, trade company S0, S1, S2, S3, S4, S5 is browsed by user, and trade company S3, S4, S5 are that can currently see
The trade company arrived, but S1, S3 are removed in new trade company, and also original trade company S6-S11 sequence also becomes, and S6 quilts
Eliminate, accordingly, trade company S22, S23 and S12 have been added.
Client 10 can perform following update and operate:
1) the trade company S3-S5 of current visible keeps constant (although S3 has been removed in new trade company)
2) it is browsed, but sightless trade company S0-S2, trade company S1 can be rejected, retain S0 and S2;Or,
S0-S2 can be retained constant, be checked in order to which user returns.
3) it is directed to without browsed trade company S6, S7, S8, S9, S10, S11, S6 is removed, adds S22, S23 and S12,
And the order of these trade companies is adjusted, newest trade company's order is:S7、S10、S22、S9、S8、S11、S23、S12.
It should be noted that the executive agent that above-described embodiment provides each step of method may each be same equipment,
Or, this method is also used as executive agent by distinct device.Such as, the executive agent of step 101 to step 103 can be equipment
A;Again such as, step 101 and 102 executive agent can be device A, and the executive agent of step 103 can be equipment B;Etc..
The structural representation for the recommendation apparatus that Fig. 7 provides for the another embodiment of the application.As shown in fig. 7, recommendation apparatus bag
Include:Acquiring unit 71, determining unit 72 and recommendation unit 73.
Acquiring unit 71, in response to recommending trigger event, obtaining the historical interest for recommending trigger event correspondence user
The region scene of point, position location and position location association.
Determining unit 72, the regional field associated for the historical interest point according to user, position location and position location
Scape, determines the recommendation order between at least one content recommendation and at least one content recommendation.
Recommendation unit 73, for the recommendation order between at least one content recommendation, recommending at least one to push away to user
Recommend content.
In an optional embodiment, acquiring unit 71 specifically for:Obtain the corresponding scene labeled data in position location
Or at least one non-scene labeled data;According to scene labeled data or at least one non-scene labeled data, sprocket bit is recognized
Put the region scene of association.
Alternatively, at least one non-scene labeled data includes:It is the type of the position location affiliated area, described
User the residence time of the position location affiliated area, the user the position location affiliated area movable model
Enclose, the wireless signal that the user uses in the position location is identified and/or current time feature.
In an optional embodiment, acquiring unit 71 specifically for:
Residence time for user in position location meets temporal regularity when being in, the class of position location affiliated area
Xing Wei apartments/residential quarter, and user used in position location wireless signal mark with user used at multiple weekends it is wireless
Signal identification identical situation, determines the region scene of position location association to be in;
Residence time for user in position location meets the temporal regularity in company, position location affiliated area
Type be shopping centre/other non-residential areas, and user used in position location wireless signal mark with user on weekdays when
Between use wireless signal mark identical situation, determine position location association region scene be in company;
Belong to festivals or holidays for current time, position location belongs to other cities and its type is apartment/residential quarter, and uses
The situation that the wireless signal mark that the wireless signal mark that family is used in position location is commonly used with before is differed, determines sprocket bit
The region scene of association is put to be on home leave;
Type for position location affiliated area is Tour region, and user is in each sight spot of position location affiliated area
Activity, the wireless signal that user uses in position location identifies the wireless signal mark commonly used with before and differed, and user is fixed
The situation of temporal regularity when the residence time of position position affiliated area meets tourism, determines the region scene of position location association
For tourism;
For other situations, determine the region scene of position location association to go on business.
In an optional embodiment, determining unit 72 specifically for:According to the historical interest of user point and position location,
Obtain at least one alternating content;From at least one alternating content, the region scene associated with position location is selected to match
At least one content recommendation;The matching degree of the region scene associated according at least one content recommendation with position location, it is determined that
Recommendation order between at least one content recommendation.
In an optional embodiment, determining unit 72 specifically for:The area associated according to position location with position location
Domain scene, obtains at least one alternating content;According to the historical interest of user point, from least one alternating content, selection is extremely
A few content recommendation;The matching degree of the region scene associated according at least one content recommendation with position location, it is determined that at least
Recommendation order between one content recommendation.
In an optional embodiment, as shown in figure 8, described device also includes:Updating block 74.
Updating block 74, for, in the operation of at least one content recommendation, being produced in response to user-association according to operation
The region scene that daily record data is associated with position location, updates at least one content recommendation.
In an optional embodiment, one kind of updating block 74 realizes that structure includes:Obtain subelement and update son list
Member.
Subelement is obtained, for the region scene associated according to daily record data with position location, at least one is obtained and newly pushes away
Recommend the recommendation order between content and at least one new content recommendation.
Subelement is updated, for according to the recommendation between at least one new content recommendation and at least one new content recommendation
Sequentially, update at least one content recommendation not by browsed content recommendation.
In an optional embodiment, subelement is updated specifically for performing following at least one operation:
Never by browsed content recommendation, the content being different from least one new content recommendation is rejected;
It is different in browsed content recommendation, not added at least one new content recommendation at least one recommendation
The content of appearance;
According to the recommendation order between at least one new content recommendation, adjust not by pushing away between browsed content recommendation
Recommend order.
In an optional embodiment, obtain subelement specifically for:The area associated according to daily record data with position location
Domain scene, generates at least one interest contents list;According to each self-corresponding content accounting of at least one interest contents list, from
In at least one interest contents list, at least one new content recommendation is obtained;Visited according to the user of at least one new content recommendation
Value is asked, it is determined that the recommendation order between new content recommendation.
In an optional embodiment, obtain subelement and operated specifically for performing following at least one generation:
According to action type of the user recorded in daily record data at least one content recommendation, from user with action type
In the content operated, obtain user and access the content that value meets sets requirement, to generate the first interest contents list;
According to the content in the first interest contents list, the region scene associated with reference to position location carries out content dimension
Recommendation, to generate the second interest contents list;
According to the mark of the content being user-operably recorded in daily record data, the region scene associated with reference to position location
The recommendation of user's dimension is carried out, to generate the 3rd interest contents list;
According to the historical behavior data of user, user's historical content interested is obtained, to generate the 4th interest content row
Table;
According to the content in the 4th interest contents list, the region scene associated with reference to position location carries out user's dimension
Recommend, to generate the 5th interest contents list.
In an optional embodiment, the first interest contents list, the second interest contents list, the 3rd interest content row
Table, the 4th interest contents list and the corresponding content accounting of the 5th interest contents list are successively decreased successively, and plus and are 1.
In an optional embodiment, obtain subelement and be additionally operable to:If at least one new content recommendation is included in multimedia
Hold, according to the equipment performance and/or network performance of user, select the satisfactory content of multimedia of resolution ratio.
In an optional embodiment, user-association is in the operation of at least one content recommendation, including following at least one:
Click on, browse, pulling up, gliding, stop for a long time, search, comment, thumb up, collection, forwarding, share, do not like and browse knot
Beam.
In an optional embodiment, trigger event is recommended to include following at least one:Log in, browse, gliding, click on
And search.
The recommendation apparatus that the present embodiment is provided, available for the flow for performing above method embodiment, is described in detail and can be found in
Previous embodiment, will not be repeated here.
What deserves to be explained is, if the step in above method embodiment has corresponding close with the functional module in recommendation apparatus
System, then can perform corresponding method and step by corresponding functional module;If step and recommendation apparatus in above method embodiment
In functional module fail to form corresponding relation, then can be by function and the closest functional module of this method step or any work(
Can perform, in this regard, the application is not limited, i.e., step in above method embodiment can by the module in recommendation apparatus Lai
Complete.
The recommendation apparatus that the present embodiment is provided, with reference to the association of the historical interest point of user, position location and position location
Region scene, to user's content recommendation, be conducive to recommending to more conform to the content of user's request to user, improve and recommend effect
Really, user experience is improved.
In a possible design, the structure of above-mentioned recommendation apparatus includes processor and memory, the memory
The program for supporting above-mentioned recommendation apparatus to perform the recommendation method that above method embodiment is provided for storing, the processor by with
It is set to for performing the program stored in the memory.
Described program includes one or more computer instruction, wherein, one or more computer instruction is for described
Processor calls execution.
Processor is used for:The program in memory is performed, for:In response to recommending trigger event, obtain and recommend triggering
Historical interest point, position location and the region scene of position location association of event correspondence user;It is emerging according to the history of user
The region scene of interesting point, position location and position location association, determines at least one content recommendation and at least one recommendation
Recommendation order between content;With the recommendation order between at least one content recommendation, recommend to user at least one recommendation
Hold.
Memory is also configured to the other various data of storage to support the operation on recommendation apparatus.These data
Example includes the instruction of any application program or method for being used to operate on recommendation apparatus, contact data, telephone book data,
Message, picture, video etc..
Memory can be realized by any kind of volatibility or non-volatile memory device or combinations thereof, such as quiet
State random access memory (SRAM), Electrically Erasable Read Only Memory (EEPROM), the read-only storage of erasable programmable
Device (EPROM), programmable read only memory (PROM), read-only storage (ROM), magnetic memory, flash memory, disk or light
Disk.
Optionally, the recommendation apparatus can also include communication component, for recommendation apparatus and other equipment or communication network
Network communicates.Communication component is configured to facilitate the communication of wired or wireless way between recommendation apparatus and other equipment.Recommend dress
The wireless network based on communication standard, such as WiFi, 2G or 3G can be accessed by putting, or combinations thereof.In an exemplary implementation
In example, communication component receives broadcast singal or broadcast related information from external broadcasting management system via broadcast channel.
In one exemplary embodiment, communication component also includes near-field communication (NFC) module, to promote junction service.For example, in NFC
Module can be based on radio frequency identification (RFID) technology, Infrared Data Association (IrDA) technology, ultra wide band (UWB) technology, bluetooth (BT)
Technology and other technologies are realized.
The embodiments of the invention provide a kind of computer-readable storage medium, for storing the computer used in above-mentioned recommendation apparatus
Software instruction, it includes program of the recommendation method for being used for performing the offer of above method embodiment involved by recommendation apparatus.
It should be understood by those skilled in the art that, embodiments of the invention can be provided as method, system or computer program
Product.Therefore, the present invention can be using the reality in terms of complete hardware embodiment, complete software embodiment or combination software and hardware
Apply the form of example.Moreover, the present invention can be used in one or more computers for wherein including computer usable program code
The computer program production that usable storage medium is implemented on (including but is not limited to magnetic disk storage, CD-ROM, optical memory etc.)
The form of product.
The present invention is the flow with reference to method according to embodiments of the present invention, equipment (system) and computer program product
Figure and/or block diagram are described.It should be understood that can be by every first-class in computer program instructions implementation process figure and/or block diagram
Journey and/or the flow in square frame and flow chart and/or block diagram and/or the combination of square frame.These computer programs can be provided
The processor of all-purpose computer, special-purpose computer, Embedded Processor or other programmable data processing devices is instructed to produce
A raw machine so that produced by the instruction of computer or the computing device of other programmable data processing devices for real
The device for the function of being specified in present one flow of flow chart or one square frame of multiple flows and/or block diagram or multiple square frames.
These computer program instructions, which may be alternatively stored in, can guide computer or other programmable data processing devices with spy
Determine in the computer-readable memory that mode works so that the instruction being stored in the computer-readable memory, which is produced, to be included referring to
Make the manufacture of device, the command device realize in one flow of flow chart or multiple flows and/or one square frame of block diagram or
The function of being specified in multiple square frames.
These computer program instructions can be also loaded into computer or other programmable data processing devices so that in meter
Series of operation steps is performed on calculation machine or other programmable devices to produce computer implemented processing, thus in computer or
The instruction performed on other programmable devices is provided for realizing in one flow of flow chart or multiple flows and/or block diagram one
The step of function of being specified in individual square frame or multiple square frames.
In a typical configuration, computing device includes one or more processors (CPU), input/output interface, net
Network interface and internal memory.
Internal memory potentially includes the volatile memory in computer-readable medium, random access memory (RAM) and/or
The forms such as Nonvolatile memory, such as read-only storage (ROM) or flash memory (flash RAM).Internal memory is computer-readable medium
Example.
Computer-readable medium includes permanent and non-permanent, removable and non-removable media can be by any method
Or technology come realize information store.Information can be computer-readable instruction, data structure, the module of program or other data.
The example of the storage medium of computer includes, but are not limited to phase transition internal memory (PRAM), static RAM (SRAM), moved
State random access memory (DRAM), other kinds of random access memory (RAM), read-only storage (ROM), electric erasable
Programmable read only memory (EEPROM), fast flash memory bank or other memory techniques, read-only optical disc read-only storage (CD-ROM),
Digital versatile disc (DVD) or other optical storages, magnetic cassette tape, the storage of tape magnetic rigid disk or other magnetic storage apparatus
Or any other non-transmission medium, the information that can be accessed by a computing device available for storage.Define, calculate according to herein
Machine computer-readable recording medium does not include temporary computer readable media (transitory media), such as data-signal and carrier wave of modulation.
It should also be noted that, term " comprising ", "comprising" or its any other variant are intended to nonexcludability
Comprising so that process, method, commodity or equipment including a series of key elements are not only including those key elements, but also wrap
Include other key elements being not expressly set out, or also include for this process, method, commodity or equipment intrinsic want
Element.In the absence of more restrictions, the key element limited by sentence "including a ...", it is not excluded that wanted including described
Also there is other identical element in process, method, commodity or the equipment of element.
It will be understood by those skilled in the art that embodiments herein can be provided as method, system or computer program product.
Therefore, the application can be using the embodiment in terms of complete hardware embodiment, complete software embodiment or combination software and hardware
Form.Deposited moreover, the application can use to can use in one or more computers for wherein including computer usable program code
The shape for the computer program product that storage media is implemented on (including but is not limited to magnetic disk storage, CD-ROM, optical memory etc.)
Formula.
Embodiments herein is the foregoing is only, the application is not limited to.For those skilled in the art
For, the application can have various modifications and variations.It is all any modifications made within spirit herein and principle, equivalent
Replace, improve etc., it should be included within the scope of claims hereof.
The embodiment of the present application discloses A1, a kind of recommendation method, including:
In response to recommending trigger event, the historical interest point for recommending trigger event correspondence user, position location are obtained
And the region scene of the position location association;
The region scene associated according to the historical interest of user point, the position location and the position location,
Determine the recommendation order between at least one content recommendation and at least one described content recommendation;
With the recommendation order between at least one described content recommendation, recommend to the user at least one described recommendation
Hold.
In A2, the method as described in A1, the obtaining step of the region scene of the position location association, including:
Obtain the corresponding scene labeled data in the position location or at least one non-scene labeled data;
According to the scene labeled data or at least one non-scene labeled data, the area of the position location association is recognized
Domain scene.
In A3, the method as described in A2, the identification step of the region scene of the position location association, including:
Residence time for the user in position location meets temporal regularity when being in, belonging to the position location
The type in region be apartment/residential quarter, and the user used in the position location wireless signal mark with the user
The wireless signal mark identical situation used at multiple weekends, determines the region scene of the position location association to be in;
Residence time for the user in position location meets the temporal regularity in company, the position location institute
Belong to the type in region for shopping centre/other non-residential areas, and the wireless signal mark that the user uses in the position location
Identical situation is identified with the user wireless signal that the time uses on weekdays, the region of the position location association is determined
Scene is in company;
Belong to festivals or holidays for the current time, the position location belongs to other cities and its type is apartment/live
Quarter, and the wireless signal that is used in the position location of the user identifies the wireless signal commonly used with before and identifies and differ
Situation, the region scene of position location association is determined to be on home leave;
Type for the position location affiliated area is Tour region, and the user is in the affiliated area in the position location
It is movable in each sight spot in domain, the wireless communication that the user commonly uses in the wireless signal mark that the position location is used with before
Number mark is differed, the feelings of temporal regularity of the user when the residence time of the position location affiliated area meets tourism
Condition, the region scene for determining the position location association is tourism;
For other situations, determine the region scene of the position location association to go on business.
In A4, the method as described in A1, between at least one described content recommendation and at least one described content recommendation
Recommendation order determination step, including:
According to the historical interest of user point and the position location, at least one alternating content is obtained;
From at least one described alternating content, it is described that the region scene that selection is associated with the position location matches
At least one content recommendation;
The matching degree of the region scene associated according at least one described content recommendation with the position location, it is determined that described
Recommendation order between at least one content recommendation.
In A5, the method as described in A1, between at least one described content recommendation and at least one described content recommendation
Recommendation order determination step, including:
The region scene associated according to the position location with the position location, obtains at least one alternating content;
According to the historical interest of user point, from least one described alternating content, at least one is pushed away described in selection
Recommend content;
The matching degree of the region scene associated according at least one described content recommendation with the position location, it is determined that described
Recommendation order between at least one content recommendation.
In A6, the method as described in any one of A1-A5, in addition to:
In response to the user-association in the operation of at least one content recommendation, the daily record produced according to the operation
The region scene that data are associated with the position location, updates at least one described content recommendation.
In A7, the method as described in A6, the renewal step of at least one content recommendation, including:
The region scene associated according to the daily record data with the position location, obtain at least one new content recommendation with
And the recommendation order between at least one described new content recommendation;
According to the recommendation order between at least one described new content recommendation and at least one described new content recommendation,
Update at least one described content recommendation not by browsed content recommendation.
It is described not by the renewal step of browsed content recommendation, including following at least one in A8, the method as described in A7
Kind:
In browsed content recommendation, do not rejected interior at least one new content recommendation described in being different from from described
Hold;
Described in being different from into described not at least one new content recommendation described in browsed content recommendation, adding extremely
The content of a few content recommendation;
According to the recommendation order between at least one described new content recommendation, not by browsed content recommendation described in adjustment
Between recommendation order.
In A9, the method as described in A7, at least one described new content recommendation and at least one described new content recommendation
Between recommendation order obtaining step, including:
The region scene associated according to the daily record data with the position location, generates at least one interest content row
Table;
According to each self-corresponding content accounting of at least one described interest contents list, from least one described interest content
In list, at least one described new content recommendation is obtained;
Accessed and be worth according to the user of at least one new content recommendation, determine the recommendation between the new content recommendation
Sequentially.
In A10, the method as described in A9, the generation step of at least one interest contents list, including it is following at least
One kind generation operation:
According to action type of the user recorded in the daily record data at least one content recommendation, from institute
State in the content that user was operated with the action type, obtain user and access the content that value meets sets requirement, to generate
First interest contents list;
According to the content in the first interest contents list, in the region scene progress associated with reference to the position location
Hold the recommendation of dimension, to generate the second interest contents list;
According to the mark of the content operated by the user recorded in the daily record data, closed with reference to the position location
The region scene of connection carries out the recommendation of user's dimension, to generate the 3rd interest contents list;
According to the historical behavior data of the user, user historical content interested is obtained, it is emerging to generate the 4th
Interesting contents list;
According to the content in the 4th interest contents list, used with reference to the region scene that the position location is associated
The recommendation of family dimension, to generate the 5th interest contents list.
It is the first interest contents list, the second interest contents list, described in A11, the method as described in A10
3rd interest contents list, the 4th interest contents list and the corresponding content accounting of the 5th interest contents list according to
It is secondary to successively decrease, and plus and be 1.
In A12, the method as described in A7, the obtaining step of at least one new content recommendation, in addition to:
If at least one described new content recommendation includes content of multimedia, according to the equipment performance and/or net of the user
Network performance, selects the satisfactory content of multimedia of resolution ratio.
In A13, the method as described in A6, operation of the user at least one content recommendation, including below extremely
Few one kind:
Click on, browse, pulling up, gliding, for a long time stop, search, comment, thumb up, collection, forward, share, do not like with
And browse end.
In A14, the method as described in any one of A1-A5, the recommendation trigger event includes following at least one:
Log in, browse, glide, click on and search for.
The embodiment of the present application also provides B15, a kind of recommendation apparatus, including:
Acquiring unit, in response to recommending trigger event, obtaining the history for recommending trigger event correspondence user emerging
The region scene of interesting point, position location and position location association;
Determining unit, is closed for the historical interest point according to the user, the position location and the position location
The region scene of connection, determines the recommendation order between at least one content recommendation and at least one described content recommendation;
Recommendation unit, for the recommendation order between at least one described content recommendation, recommending to the user described
At least one content recommendation.
In B16, the device as described in B15, the acquiring unit specifically for:
Obtain the corresponding scene labeled data in the position location or at least one non-scene labeled data;
According to the scene labeled data or at least one non-scene labeled data, the area of the position location association is recognized
Domain scene.
In B17, the device as described in B16, the acquiring unit specifically for:
Residence time for the user in position location meets temporal regularity when being in, belonging to the position location
The type in region be apartment/residential quarter, and the user used in the position location wireless signal mark with the user
The wireless signal mark identical situation used at multiple weekends, determines the region scene of the position location association to be in;
Residence time for the user in position location meets the temporal regularity in company, the position location institute
Belong to the type in region for shopping centre/other non-residential areas, and the wireless signal mark that the user uses in the position location
Identical situation is identified with the user wireless signal that the time uses on weekdays, the region of the position location association is determined
Scene is in company;
Belong to festivals or holidays for the current time, the position location belongs to other cities and its type is apartment/live
Quarter, and the wireless signal that is used in the position location of the user identifies the wireless signal commonly used with before and identifies and differ
Situation, the region scene of position location association is determined to be on home leave;
Type for the position location affiliated area is Tour region, and the user is in the affiliated area in the position location
It is movable in each sight spot in domain, the wireless signal that the user commonly uses in the wireless signal mark that the position location is used with before
Mark is differed, the feelings of temporal regularity of the user when the residence time of the position location affiliated area meets tourism
Condition, the region scene for determining the position location association is tourism;
For other situations, determine the region scene of the position location association to go on business.
In B18, the device as described in B15, the determining unit specifically for:
According to the historical interest of user point and the position location, at least one alternating content is obtained;
From at least one described alternating content, it is described that the region scene that selection is associated with the position location matches
At least one content recommendation;
The matching degree of the region scene associated according at least one described content recommendation with the position location, it is determined that described
Recommendation order between at least one content recommendation.
In B19, the device as described in B15, the determining unit specifically for:
The region scene associated according to the position location with the position location, obtains at least one alternating content;
According to the historical interest of user point, from least one described alternating content, at least one is pushed away described in selection
Recommend content;
The matching degree of the region scene associated according at least one described content recommendation with the position location, it is determined that described
Recommendation order between at least one content recommendation.
In B20, the device as described in any one of B15-B19, in addition to:
Updating block, in response to the user-association in the operation of at least one content recommendation, according to described
The region scene that the daily record data that operation is produced is associated with the position location, updates at least one described content recommendation.
In B21, the device as described in B20, the updating block includes:
Subelement is obtained, for the region scene associated according to the daily record data with the position location, is obtained at least
Recommendation order between one new content recommendation and at least one described new content recommendation;
Update subelement, for according at least one described new content recommendation and at least one described new content recommendation it
Between recommendation order, update at least one described content recommendation not by browsed content recommendation.
In B22, the device as described in B21, the renewal subelement is specifically for performing following at least one operation:
In browsed content recommendation, do not rejected interior at least one new content recommendation described in being different from from described
Hold;
Described in being different from into described not at least one new content recommendation described in browsed content recommendation, adding extremely
The content of a few content recommendation;
According to the recommendation order between at least one described new content recommendation, not by browsed content recommendation described in adjustment
Between recommendation order.
In B23, the device as described in B21, it is described acquisition subelement specifically for:
The region scene associated according to the daily record data with the position location, generates at least one interest content row
Table;
According to each self-corresponding content accounting of at least one described interest contents list, from least one described interest content
In list, at least one described new content recommendation is obtained;
Accessed and be worth according to the user of at least one new content recommendation, determine the recommendation between the new content recommendation
Sequentially.
In B24, the device as described in B23, the acquisition subelement is operated specifically for performing following at least one generation:
According to action type of the user recorded in the daily record data at least one content recommendation, from institute
State in the content that user was operated with the action type, obtain user and access the content that value meets sets requirement, to generate
First interest contents list;
According to the content in the first interest contents list, in the region scene progress associated with reference to the position location
Hold the recommendation of dimension, to generate the second interest contents list;
According to the mark of the content operated by the user recorded in the daily record data, closed with reference to the position location
The region scene of connection carries out the recommendation of user's dimension, to generate the 3rd interest contents list;
According to the historical behavior data of the user, user historical content interested is obtained, it is emerging to generate the 4th
Interesting contents list;
According to the content in the 4th interest contents list, used with reference to the region scene that the position location is associated
The recommendation of family dimension, to generate the 5th interest contents list.
It is the first interest contents list, the second interest contents list, described in B25, the device as described in B24
3rd interest contents list, the 4th interest contents list and the corresponding content accounting of the 5th interest contents list according to
It is secondary to successively decrease, and plus and be 1.
In B26, the device as described in B21, the acquisition subelement is additionally operable to:
If at least one described new content recommendation includes content of multimedia, according to the equipment performance and/or net of the user
Network performance, selects the satisfactory content of multimedia of resolution ratio.
The embodiment of the present application also provides C27, a kind of electronic equipment, including:Processor and memory, the memory are used for
Storage supports above-mentioned recommendation apparatus to perform the program of above-mentioned recommendation method, and the processor is configurable for performing the storage
The program stored in device, for:In response to recommending trigger event, the historical interest for recommending trigger event correspondence user is obtained
The region scene of point, position location and position location association;According to the historical interest of user point, position location and positioning
The region scene of position association, determines the recommendation order between at least one content recommendation and at least one content recommendation;With
Recommendation order between at least one content recommendation, at least one content recommendation is recommended to user.
Claims (10)
1. a kind of recommendation method, it is characterised in that including:
In response to recommending trigger event, obtain it is described recommend the trigger event correspondence historical interest point of user, position location and
The region scene of the position location association;
The region scene associated according to the historical interest of user point, the position location and the position location, it is determined that
Recommendation order between at least one content recommendation and at least one described content recommendation;
With the recommendation order between at least one described content recommendation, at least one described content recommendation is recommended to the user.
2. according to the method described in claim 1, it is characterised in that the acquisition step of the region scene of the position location association
Suddenly, including:
Obtain the corresponding scene labeled data in the position location or at least one non-scene labeled data;
According to the scene labeled data or at least one non-scene labeled data, the regional field of the position location association is recognized
Scape.
3. according to the method described in claim 1, it is characterised in that at least one described content recommendation and it is described at least one
The determination step of recommendation order between content recommendation, including:
According to the historical interest of user point and the position location, at least one alternating content is obtained;
From at least one described alternating content, select that the region scene that is associated with the position location matches it is described at least
One content recommendation;
The matching degree of the region scene associated according at least one described content recommendation with the position location, it is determined that it is described at least
Recommendation order between one content recommendation.
4. according to the method described in claim 1, it is characterised in that at least one described content recommendation and it is described at least one
The determination step of recommendation order between content recommendation, including:
The region scene associated according to the position location with the position location, obtains at least one alternating content;
According to the historical interest of user point, from least one described alternating content, at least one described recommendation of selection
Hold;
The matching degree of the region scene associated according at least one described content recommendation with the position location, it is determined that it is described at least
Recommendation order between one content recommendation.
5. the method according to claim any one of 1-4, it is characterised in that also include:
In response to the user-association in the operation of at least one content recommendation, the daily record data produced according to the operation
The region scene associated with the position location, updates at least one described content recommendation.
6. method according to claim 5, it is characterised in that the renewal step of at least one content recommendation, including:
The region scene associated according to the daily record data with the position location, obtains at least one new content recommendation and institute
State the recommendation order between at least one new content recommendation;
According to the recommendation order between at least one described new content recommendation and at least one described new content recommendation, institute is updated
State at least one content recommendation not by browsed content recommendation.
7. method according to claim 6, it is characterised in that at least one described new content recommendation and described at least one
The obtaining step of recommendation order between individual new content recommendation, including:
The region scene associated according to the daily record data with the position location, generates at least one interest contents list;
According to each self-corresponding content accounting of at least one described interest contents list, from least one described interest contents list
In, obtain at least one described new content recommendation;
Accessed and be worth according to the user of at least one new content recommendation, determine that the recommendation between the new content recommendation is suitable
Sequence.
8. method according to claim 7, it is characterised in that the generation step of at least one interest contents list,
Including following at least one generation operation:
According to action type of the user recorded in the daily record data at least one content recommendation, used from described
In the content that family was operated with the action type, obtain user and access the content that value meets sets requirement, to generate first
Interest contents list;
According to the content in the first interest contents list, the region scene associated with reference to the position location carries out content dimension
The recommendation of degree, to generate the second interest contents list;
According to the mark of the content operated by the user recorded in the daily record data, associated with reference to the position location
Region scene carries out the recommendation of user's dimension, to generate the 3rd interest contents list;
According to the historical behavior data of the user, user historical content interested is obtained, to generate in the 4th interest
Hold list;
According to the content in the 4th interest contents list, the region scene associated with reference to the position location carries out user's dimension
The recommendation of degree, to generate the 5th interest contents list.
9. a kind of recommendation apparatus, it is characterised in that including:
Acquiring unit, in response to recommend trigger event, obtain it is described recommend trigger event correspondence user historical interest point,
Position location and the region scene of position location association;
Determining unit, is associated for the historical interest point according to the user, the position location and the position location
Region scene, determines the recommendation order between at least one content recommendation and at least one described content recommendation;
Recommendation unit, for the recommendation order between at least one described content recommendation, to the user recommend described at least
One content recommendation.
10. device according to claim 9, it is characterised in that also include:
Updating block, in response to the user-association in the operation of at least one content recommendation, according to the operation
The region scene that the daily record data of generation is associated with the position location, updates at least one described content recommendation.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201710040010.1A CN107092629A (en) | 2017-01-18 | 2017-01-18 | Recommend method and device |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201710040010.1A CN107092629A (en) | 2017-01-18 | 2017-01-18 | Recommend method and device |
Publications (1)
Publication Number | Publication Date |
---|---|
CN107092629A true CN107092629A (en) | 2017-08-25 |
Family
ID=59646126
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201710040010.1A Pending CN107092629A (en) | 2017-01-18 | 2017-01-18 | Recommend method and device |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN107092629A (en) |
Cited By (29)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN108170795A (en) * | 2017-12-28 | 2018-06-15 | 百度在线网络技术(北京)有限公司 | Information-pushing method, device and equipment |
CN108280749A (en) * | 2018-02-13 | 2018-07-13 | 口口相传(北京)网络技术有限公司 | The methods of exhibiting and device of service function entrance |
CN108763378A (en) * | 2018-05-18 | 2018-11-06 | 深圳春沐源控股有限公司 | Information displaying method and information display system |
CN109084795A (en) * | 2018-08-20 | 2018-12-25 | 中国工商银行股份有限公司 | The method and device of search service facility based on Map Services |
CN109614556A (en) * | 2018-10-31 | 2019-04-12 | 拉卡拉支付股份有限公司 | Access path prediction, information-pushing method and device |
CN109615428A (en) * | 2018-12-10 | 2019-04-12 | 拉扎斯网络科技(上海)有限公司 | Trade company's recommended method, device, system and server |
CN109740493A (en) * | 2018-12-27 | 2019-05-10 | 维沃移动通信有限公司 | A kind of target object recommended method and mobile terminal |
CN109839120A (en) * | 2017-11-24 | 2019-06-04 | 北京三快在线科技有限公司 | Stroke planning method, device, medium and electronic equipment |
CN110096645A (en) * | 2019-05-07 | 2019-08-06 | 北京百度网讯科技有限公司 | Information recommendation method, device, equipment and medium |
CN110110201A (en) * | 2018-01-09 | 2019-08-09 | 苏州跃盟信息科技有限公司 | A kind of content recommendation method and system |
WO2019191914A1 (en) * | 2018-04-04 | 2019-10-10 | Beijing Didi Infinity Technology And Development Co., Ltd. | Systems and methods for on-demand services |
CN110400193A (en) * | 2019-06-13 | 2019-11-01 | 拉扎斯网络科技(上海)有限公司 | Vegetable recommended method and device, electronic equipment and storage medium |
CN110413868A (en) * | 2018-11-06 | 2019-11-05 | 腾讯科技(深圳)有限公司 | Information recommendation method, device, system and storage medium |
CN110619551A (en) * | 2018-06-19 | 2019-12-27 | 北京嘀嘀无限科技发展有限公司 | Order distribution method, order distribution system, computer device and storage medium |
CN110737690A (en) * | 2018-07-03 | 2020-01-31 | 百度在线网络技术(北京)有限公司 | User label mining method and device, computer equipment and storage medium |
CN110839346A (en) * | 2018-06-19 | 2020-02-25 | 北京嘀嘀无限科技发展有限公司 | System and method for distributing service requests |
CN111026956A (en) * | 2019-11-20 | 2020-04-17 | 拉扎斯网络科技(上海)有限公司 | Data list processing method and device, electronic equipment and computer storage medium |
CN111091415A (en) * | 2019-12-02 | 2020-05-01 | 贝壳技术有限公司 | Method and system for recommending broker |
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 |
CN111310020A (en) * | 2019-09-20 | 2020-06-19 | 拉扎斯网络科技(上海)有限公司 | Dish recommendation method, recommended content uploading method and device and electronic equipment |
CN111343223A (en) * | 2018-12-19 | 2020-06-26 | 上海博泰悦臻电子设备制造有限公司 | Information pushing method and terminal |
CN111382218A (en) * | 2018-12-29 | 2020-07-07 | 北京嘀嘀无限科技发展有限公司 | System and method for point of interest (POI) retrieval |
CN111541917A (en) * | 2020-05-08 | 2020-08-14 | 广州市百果园信息技术有限公司 | Determination method of recommended video, video playing method, device and equipment |
CN111694906A (en) * | 2019-03-14 | 2020-09-22 | 北京搜狗科技发展有限公司 | Method, device and equipment for displaying POI in map |
CN111738784A (en) * | 2019-03-25 | 2020-10-02 | 上海寻梦信息技术有限公司 | Service information display system and method, electronic device and computer storage medium |
CN112882621A (en) * | 2021-02-07 | 2021-06-01 | 微民保险代理有限公司 | Module display method, module display device, computer equipment and storage medium |
CN113223224A (en) * | 2021-05-12 | 2021-08-06 | 支付宝(杭州)信息技术有限公司 | Traffic processing method and device |
CN114186134A (en) * | 2018-03-07 | 2022-03-15 | 创新先进技术有限公司 | Method and device for recommending related content |
CN115760296A (en) * | 2022-11-24 | 2023-03-07 | 阿里巴巴(中国)有限公司 | Page data processing and browsing method, terminal device and storage medium |
Citations (6)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN103581146A (en) * | 2012-08-07 | 2014-02-12 | 腾讯科技(深圳)有限公司 | Multimedia recommendation method and corresponding server and client side |
CN104102638A (en) * | 2013-04-02 | 2014-10-15 | 北京百度网讯科技有限公司 | Position information based push method, push system and push device |
CN104240163A (en) * | 2014-09-28 | 2014-12-24 | 百度在线网络技术(北京)有限公司 | Intelligent district recommendation method and intelligent district recommendation device |
CN104853229A (en) * | 2015-05-11 | 2015-08-19 | 天脉聚源(北京)传媒科技有限公司 | Program pushing method and device thereof |
CN104899602A (en) * | 2015-06-03 | 2015-09-09 | 江苏马上游科技股份有限公司 | User cluster analysis system based on K-means algorithm |
CN105302916A (en) * | 2015-11-16 | 2016-02-03 | 北京百度网讯科技有限公司 | Information recommendation method and device |
-
2017
- 2017-01-18 CN CN201710040010.1A patent/CN107092629A/en active Pending
Patent Citations (6)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN103581146A (en) * | 2012-08-07 | 2014-02-12 | 腾讯科技(深圳)有限公司 | Multimedia recommendation method and corresponding server and client side |
CN104102638A (en) * | 2013-04-02 | 2014-10-15 | 北京百度网讯科技有限公司 | Position information based push method, push system and push device |
CN104240163A (en) * | 2014-09-28 | 2014-12-24 | 百度在线网络技术(北京)有限公司 | Intelligent district recommendation method and intelligent district recommendation device |
CN104853229A (en) * | 2015-05-11 | 2015-08-19 | 天脉聚源(北京)传媒科技有限公司 | Program pushing method and device thereof |
CN104899602A (en) * | 2015-06-03 | 2015-09-09 | 江苏马上游科技股份有限公司 | User cluster analysis system based on K-means algorithm |
CN105302916A (en) * | 2015-11-16 | 2016-02-03 | 北京百度网讯科技有限公司 | Information recommendation method and device |
Cited By (44)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN109839120B (en) * | 2017-11-24 | 2021-05-18 | 北京三快在线科技有限公司 | Route planning method, device, medium and electronic equipment |
CN109839120A (en) * | 2017-11-24 | 2019-06-04 | 北京三快在线科技有限公司 | Stroke planning method, device, medium and electronic equipment |
CN108170795B (en) * | 2017-12-28 | 2022-02-15 | 百度在线网络技术(北京)有限公司 | Information pushing method, device and equipment |
CN108170795A (en) * | 2017-12-28 | 2018-06-15 | 百度在线网络技术(北京)有限公司 | Information-pushing method, device and equipment |
CN110110201A (en) * | 2018-01-09 | 2019-08-09 | 苏州跃盟信息科技有限公司 | A kind of content recommendation method and system |
CN108280749A (en) * | 2018-02-13 | 2018-07-13 | 口口相传(北京)网络技术有限公司 | The methods of exhibiting and device of service function entrance |
WO2019158069A1 (en) * | 2018-02-13 | 2019-08-22 | 口口相传(北京)网络技术有限公司 | Method and device for displaying service function entries |
CN114186134A (en) * | 2018-03-07 | 2022-03-15 | 创新先进技术有限公司 | Method and device for recommending related content |
WO2019191914A1 (en) * | 2018-04-04 | 2019-10-10 | Beijing Didi Infinity Technology And Development Co., Ltd. | Systems and methods for on-demand services |
US11303713B2 (en) | 2018-04-04 | 2022-04-12 | Beijing Didi Infinity Technology And Development Co., Ltd. | Systems and methods for on-demand services |
CN108763378A (en) * | 2018-05-18 | 2018-11-06 | 深圳春沐源控股有限公司 | Information displaying method and information display system |
CN110839346A (en) * | 2018-06-19 | 2020-02-25 | 北京嘀嘀无限科技发展有限公司 | System and method for distributing service requests |
CN110619551A (en) * | 2018-06-19 | 2019-12-27 | 北京嘀嘀无限科技发展有限公司 | Order distribution method, order distribution system, computer device and storage medium |
CN110737690A (en) * | 2018-07-03 | 2020-01-31 | 百度在线网络技术(北京)有限公司 | User label mining method and device, computer equipment and storage medium |
CN109084795B (en) * | 2018-08-20 | 2022-02-25 | 中国工商银行股份有限公司 | Method and device for searching service facilities based on map service |
CN109084795A (en) * | 2018-08-20 | 2018-12-25 | 中国工商银行股份有限公司 | The method and device of search service facility based on Map Services |
CN109614556B (en) * | 2018-10-31 | 2020-03-13 | 拉卡拉支付股份有限公司 | Access path prediction and information push method and device |
CN109614556A (en) * | 2018-10-31 | 2019-04-12 | 拉卡拉支付股份有限公司 | Access path prediction, information-pushing method and device |
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 |
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 |
CN110413868A (en) * | 2018-11-06 | 2019-11-05 | 腾讯科技(深圳)有限公司 | Information recommendation method, device, system and storage medium |
CN110413868B (en) * | 2018-11-06 | 2023-04-07 | 腾讯科技(深圳)有限公司 | Information recommendation method, device, system and storage medium |
CN109615428A (en) * | 2018-12-10 | 2019-04-12 | 拉扎斯网络科技(上海)有限公司 | Trade company's recommended method, device, system and server |
CN111343223A (en) * | 2018-12-19 | 2020-06-26 | 上海博泰悦臻电子设备制造有限公司 | Information pushing method and terminal |
CN111343223B (en) * | 2018-12-19 | 2022-11-04 | 博泰车联网科技(上海)股份有限公司 | Information pushing method and terminal |
CN109740493A (en) * | 2018-12-27 | 2019-05-10 | 维沃移动通信有限公司 | A kind of target object recommended method and mobile terminal |
CN111382218A (en) * | 2018-12-29 | 2020-07-07 | 北京嘀嘀无限科技发展有限公司 | System and method for point of interest (POI) retrieval |
CN111382218B (en) * | 2018-12-29 | 2023-09-26 | 北京嘀嘀无限科技发展有限公司 | System and method for searching point of interest (POI) |
CN111694906A (en) * | 2019-03-14 | 2020-09-22 | 北京搜狗科技发展有限公司 | Method, device and equipment for displaying POI in map |
CN111738784A (en) * | 2019-03-25 | 2020-10-02 | 上海寻梦信息技术有限公司 | Service information display system and method, electronic device and computer storage medium |
CN110096645B (en) * | 2019-05-07 | 2022-02-25 | 北京百度网讯科技有限公司 | Information recommendation method, device, equipment and medium |
CN110096645A (en) * | 2019-05-07 | 2019-08-06 | 北京百度网讯科技有限公司 | Information recommendation method, device, equipment and medium |
CN110400193A (en) * | 2019-06-13 | 2019-11-01 | 拉扎斯网络科技(上海)有限公司 | Vegetable recommended method and device, electronic equipment and storage medium |
CN111310020A (en) * | 2019-09-20 | 2020-06-19 | 拉扎斯网络科技(上海)有限公司 | Dish recommendation method, recommended content uploading method and device and electronic equipment |
CN111026956A (en) * | 2019-11-20 | 2020-04-17 | 拉扎斯网络科技(上海)有限公司 | Data list processing method and device, electronic equipment and computer storage medium |
CN111091415B (en) * | 2019-12-02 | 2024-02-20 | 贝壳技术有限公司 | Method and system for recommending brokers |
CN111091415A (en) * | 2019-12-02 | 2020-05-01 | 贝壳技术有限公司 | Method and system for recommending broker |
CN111541917B (en) * | 2020-05-08 | 2022-06-10 | 广州市百果园信息技术有限公司 | Determination method of recommended video, video playing method, device and equipment |
CN111541917A (en) * | 2020-05-08 | 2020-08-14 | 广州市百果园信息技术有限公司 | Determination method of recommended video, video playing method, device and equipment |
CN112882621A (en) * | 2021-02-07 | 2021-06-01 | 微民保险代理有限公司 | Module display method, module display device, computer equipment and storage medium |
CN113223224A (en) * | 2021-05-12 | 2021-08-06 | 支付宝(杭州)信息技术有限公司 | Traffic processing method and device |
CN113223224B (en) * | 2021-05-12 | 2023-10-24 | 支付宝(杭州)信息技术有限公司 | Traffic processing method and device |
CN115760296A (en) * | 2022-11-24 | 2023-03-07 | 阿里巴巴(中国)有限公司 | Page data processing and browsing method, terminal device and storage medium |
CN115760296B (en) * | 2022-11-24 | 2023-10-17 | 阿里巴巴(中国)网络技术有限公司 | Page data processing and browsing method, terminal equipment and storage medium |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN107092629A (en) | Recommend method and device | |
US11657411B1 (en) | System, method, software and data structure for independent prediction of attitudinal and message responsiveness, and preferences for communication media, channel, timing, frequency, and sequences of communications, using an integrated data repository | |
CN106570722B (en) | Intelligent recommendation system and intelligent recommendation method | |
JP5596152B2 (en) | Information matching method and system on electronic commerce website | |
CN103559636B (en) | Restaurant management system based on cloud computing and intelligent analysis | |
US20160171557A1 (en) | Customer Insight System Architecture | |
CN105791085B (en) | Friend recommendation method in the social networks of position based on position and time | |
JP2019508766A (en) | System, method, and device for generating a heat map of a geographical area | |
JP6937776B2 (en) | Methods and devices for identifying geographic location attribute information | |
CN101563701A (en) | Information providing system | |
US9208444B1 (en) | Determining happiness indices | |
CN106846053A (en) | A kind of recommendation method and device of the page advertisement that is polymerized | |
CN110569432A (en) | commodity sequence calculation method and device, computer equipment and storage medium | |
CN105809475A (en) | Commodity recommendation method compatible with O2O applications in internet plus tourism environment | |
CN108550048A (en) | The method, apparatus and electronics and storage device of commodity real price are determined for user | |
US20040172379A1 (en) | Method, system and apparatus for acquiring data from a database | |
CN106599137A (en) | Novel scene content pushing system and device | |
US20200401610A1 (en) | Method and apparatus for creating and curating user collections for network search | |
US20160299639A1 (en) | User interface for providing geographically delineated content | |
CN108429776A (en) | Method for pushing, device, client, interactive device and the system of network object | |
CN109190027A (en) | Multi-source recommended method, terminal, server, computer equipment, readable medium | |
WO2010146608A1 (en) | System and method for managing advertising campaigns | |
CN107977764A (en) | Index file generation method, index file distributor information query method, index file generation device, distributor information query device and server | |
US20220036460A1 (en) | Systems and Methods for Asset Analysis | |
CN112465586A (en) | Intelligent exhibition information interaction system |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
PB01 | Publication | ||
PB01 | Publication | ||
SE01 | Entry into force of request for substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
RJ01 | Rejection of invention patent application after publication | ||
RJ01 | Rejection of invention patent application after publication |
Application publication date: 20170825 |