CN103970850B - Site information recommends method and system - Google Patents
Site information recommends method and system Download PDFInfo
- Publication number
- CN103970850B CN103970850B CN201410185036.1A CN201410185036A CN103970850B CN 103970850 B CN103970850 B CN 103970850B CN 201410185036 A CN201410185036 A CN 201410185036A CN 103970850 B CN103970850 B CN 103970850B
- Authority
- CN
- China
- Prior art keywords
- attribute information
- website platform
- grade
- platform
- website
- 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.)
- Active
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
- G06F16/9535—Search customisation based on user profiles and personalisation
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)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
Abstract
A kind of site information recommends method, including：Each website platform is docked into one or more threads, the attribute information of each website platform object to be recommended is obtained；Grade sequence is carried out to website platform according to the platform grade sequence of setting rule, website platform rank list is obtained；Scanned for according to the searching request of user from the attribute information of the object to be recommended of acquisition, the grade weights in the website platform rank list are ranked up to each object in search result；Recommendation information is generated according to the ranking results, the recommendation information is issued to client and is shown.This programme also provides a kind of site information commending system.The present invention program realizes the polymerization to multiple website platform object informations by the way that each website platform is docked into a thread.And the object in search result is ranked up according to the website platform rank list of foundation, so that recommendation information be shown, it is to avoid user opens multiple website contrasts, greatly improves the search efficiency of user.
Description
Technical field
The present invention relates to Internet technical field, more particularly to a kind of site information recommends method and system.
Background technology
Developing rapidly and popularizing with internet, ecommerce has obtained quick development, various electronics business in the whole world
Business website platform occurs in succession, and Internet user's number is sharply increased, and increasing user is also gradually being adapted in website
Information needed for upper search.
In conventional art, each website platform is ranked up to respective object information to be recommended according to certain rule.Receiving
To after user's search instruction, search result is subjected to recommendation displaying according to rule set in advance, screens and checks for user.
However, as website platform is more and more, the selective object of searching of each website platform is also more and more, Yong Hu
Selection is contrasted when searching the information of object, it is necessary to open multiple websites simultaneously, then will search the information one one of object
Individual to be checked, this may result in, and the search consuming time is long, and search efficiency is low.There is the recommendation in time limit especially for some
Breath, because user can not quickly search content to be checked so that lead to miss check some have the recommendation information in time limit without
Method is searched, and also reduces the information content that user can search.
The content of the invention
Based on this, when carrying out object search for multiple websites, the problem of search efficiency is low, the present invention provides a kind of website
Information recommendation method and system.
A kind of site information recommends method, including：
Each website platform is docked into one or more threads, the attribute information of each website platform object to be recommended is obtained；
Grade sequence is carried out to website platform according to the platform grade sequence of setting rule, website platform grade row are obtained
Table；
Scanned for according to the searching request of user from the attribute information of the object to be recommended of acquisition, according to the website
Grade weights in platform rank list are ranked up to each object in search result；
Recommendation information is generated according to the ranking results, the recommendation information is issued to client and is shown.
A kind of site information commending system, including：
Attribute information acquisition module, for each website platform to be docked into one or more threads, obtains each website platform
The attribute information of object to be recommended；
Website platform rank list determining module, is carried out for the platform grade sequence rule according to setting to website platform
Grade sequence, obtains website platform rank list；
Search module, is searched for the searching request according to user from the attribute information of the object to be recommended of acquisition
Rope；
Order module, for the grade weights in the website platform rank list in search result each is right
As being ranked up；
First display module, for generating recommendation information according to the ranking results, visitor is issued to by the recommendation information
Family end is shown.
Above-mentioned site information recommends method and system, by the way that each website platform is docked into one or more threads, obtains
To the attribute information of each website platform object to be recommended, the polymerization to multiple website platform object informations is realized.And set up a web site
Platform rank list, is ranked up according to website platform rank list to the object in search result, and generation recommendation information is gone forward side by side
Row displaying, it is to avoid user opens multiple website contrasts, greatly improves the search efficiency of user.Especially for the displaying time limit
Website platform, user can be with fast search to object, and information content will not be reduced, and improves Consumer's Experience.
Brief description of the drawings
Fig. 1 is the schematic flow sheet that site information of the present invention recommends embodiment of the method；
Fig. 2 is the SNA schematic diagram in concrete application example of the present invention；
Fig. 3 is the platform sequence schematic diagram in concrete application example of the present invention；
Fig. 4 is the brand sequence schematic diagram in concrete application example of the present invention；
Fig. 5 is the commodity sequence schematic diagram in concrete application example of the present invention；
Fig. 6 is the structural representation of site information commending system embodiment of the present invention.
Embodiment
The present invention is described in further detail with reference to embodiment and accompanying drawing, but embodiments of the present invention are not limited to
This.
As shown in figure 1, recommend the schematic flow sheet of embodiment of the method for site information of the present invention, including step：
Step S101：Each website platform is docked into one or more threads, each website platform object to be recommended is obtained
Attribute information；
The attribute information for obtaining website platform object to be recommended can be real-time acquisition, can also be according to website platform more
New frequency determines the renewal time of the website platform or updates interval time, and being obtained according to renewal time or renewal interval time should
The attribute information of website platform object to be recommended.In this way, it is possible to achieve update synchronous.Meanwhile, pass through multithreading
Customizable collection rule upgrades in time the corresponding database of each website platform, gathers the object of polymerization site platform.Wherein, originally
The object to be recommended that step is referred to, description is directly referred to as object for convenience in subsequent step.
Step S102：Grade sequence is carried out to website platform according to the platform grade sequence of setting rule, website is obtained and puts down
Platform rank list；
Platform grade sequence rule can be pre-set, so as to obtain website platform rank list.Platform grade
Ordering rule can also obtain rule.Such as, the ALEXA rankings (world rankings of website) or B2C of each website platform are obtained
Industry ranking etc. obtains website platform rank list.Wherein, the grade of each website platform in being recorded in website platform rank list
Weights.
Step S103：Scanned for according to the searching request of user from the attribute information of the object to be recommended of acquisition, root
The each object in search result is ranked up according to the grade weights in the website platform rank list.
Searching request can client issue, such as keyword etc..It is may search for out by searching request a series of
Object and its attribute information.User quickly checks for convenience, can obtain the grade weights of the website platform of each object, root
Each object is ranked up according to the grade weights of website platform.
Step S104：Recommendation information is generated according to the ranking results, the recommendation information is issued into client is carried out
Displaying.
Can be ranked up the attribute information of each object according to ranking results in the present embodiment, obtain recommendation information,
The attribute information after sequence directly is issued into client to be shown.
This embodiment scheme, by the way that each website platform is docked into a thread, gets each website platform object to be recommended
Attribute information, realize polymerization to multiple website platform object informations.And the platform rank list that sets up a web site, it is flat according to website
Platform rank list is ranked up to the object in search result, is generated recommendation information and is shown, it is to avoid user opens multiple
Website is contrasted, and greatly improves the search efficiency of user.Especially for the website platform with the displaying time limit, user can be quick
Object is searched, and information content will not be reduced, and improve Consumer's Experience.
In one of the embodiments, because the object properties type of different web sites platform is variant, on some platforms
The attribute that object has, object is not necessarily present the attribute on other website platforms, therefore, for unified attribute, further carries
High search efficiency, the present embodiment is by setting up the form of label, the attribute type of unified each object for each object.Specifically such as
Under：
After step S101, in addition to：
The corresponding attribute information of each Property Name in default label form is identified from the attribute information of the object of acquisition,
Wherein, the attribute information of the object includes word generic attribute information and/or picture category attribute information, the default label form
Record multiple Property Names；
The attribute information of acquisition is recorded under the corresponding Property Name of label form, the label of the object is obtained；
It is described to generate recommendation information according to the ranking results, the recommendation information is issued to client and is shown step
Suddenly include：
The tag sorting order of object is determined according to ranking results, the label of sequence is issued to client and is shown.
Default label form in the present embodiment can be by manually pre-setting, can also system according to certain rule
Automatically generate.Default label form only has Property Name, does not have specific object information.Specific object information is needed according to each right
The attribute information of elephant is write accordingly, and attribute information is write into the label that corresponding label form becomes the object.
The corresponding attribute information of each Property Name in default label form is identified from the attribute information of the object of acquisition
Step is specifically included：
When attribute information is literal type attribute information, the category of literal type attribute information and label form is directly judged
Whether whether property title matches, or matched with the attribute information included by the Property Name, so as to identify default label list
The corresponding attribute information of each Property Name in lattice.Such as, when Property Name is " color ", then it may determine that literal type attribute is believed
With the presence or absence of " X colors " character etc. in breath.
, can be by recognizing the profile in figure, color etc., so as to know when attribute information is graph style attribute information
Do not go out the corresponding attribute information of each Property Name in default label form.
In this way, the attribute of object has been unified.User inquires about for convenience, can also unify keyword input and press
Button, the i.e. Property Name in default label form set up corresponding keyword load button.User can by selection and
Composite key load button realizes input keyword search instruction, so as to the object to be checked of fast search to user.
In one of the embodiments, the platform grade sequence rule according to setting carries out grade row to website platform
Sequence, obtains website platform rank list step, including：
The index parameter of each website platform is obtained, wherein, the index parameter covers including the use of the crowd of the website platform
It is one or more in lid rate, areal coverage, opinion rating, searching rate；
Respectively by multiplied by weight of the index parameter of each website platform with corresponding setting, and by product addition, obtain every
The grade weights of individual website platform；
Website platform is ranked up according to the grade weights size of each website platform, website platform grade row are obtained
Table.
The present embodiment provides a kind of method for obtaining website platform rank list.Website platform index parameter is including the use of this
The population coverage of website platform, areal coverage, opinion rating, searching rate, but these evaluation indexes are not limited to, can basis
User specifically needs to set other or more evaluation index.Such as, it is divided into index, index of bidding etc..
Wherein, the use of the population coverage of the website platform can be user coverage rate.The opinion rating of website platform can
To be the opinion rating obtained from some evaluation systems.The evaluation index of website platform and corresponding multiplied by weight, and will multiply
Product is added, and obtains the evaluation weights of the website platform.Each website platform carries out above-mentioned processing, you can obtain each website platform
Grade weights.The grade weights descending of website platform is arranged, you can obtain website platform rank list.
In one of the embodiments, it can also can be pushed away according to searching request recommended information according to user profile
Recommend object information.I.e. this programme also includes step：
Customer attribute information is obtained by user account；
The corresponding attribute letter of each Property Name in pre-set user label form is identified from the customer attribute information of acquisition
Breath, wherein, the customer attribute information includes word generic attribute information and/or picture category attribute information, the pre-set user mark
Label charting multiple Property Names；
The attribute information of acquisition is recorded under the corresponding Property Name of pre-set user label form, the use of the user is obtained
Family model；
The attribute information of object according to user model with obtaining is matched, according in the website platform rank list
Grade weights the object of matching is ranked up, and the attribute information of the object of sequence is subjected to recommendation displaying.
Customer attribute information can be obtained or by thread from the api interface of website by user account
From Webpage crawl.The customer attribute information can include age of user stage, user preferences color, user preferences type
Deng.The object of interest of user can also be identified by the pictorial information of acquisition.Such as, when user account is social network sites
During account, the picture of user's issue can be obtained from social network sites api interface, user is gone out according to picture recognition interested right
As.Again such as, when user account is the account of shopping website, user's shopping history can be obtained from shopping website api interface
Website, so as to extrapolate the hobby of client.The hobby and object of interest belong to the attribute information of user.According to attribute information
User model can be obtained with pre-set user label form.User tag form can be artificial default or system root
It is self-defined according to certain rule.User model is polymerize to the object properties information matches of acquisition with step S101, you can obtain phase
The matching result answered.These results are ranked up according to website platform rank list and recommend displaying.Website platform grade is arranged
Table can be obtained according to a kind of any of the above described method.
The present embodiment not only can by searching request carry out recommended information, can also after the user logs, according to
The direct recommended information of user profile, facilitates user efficiently to search for and checks.
In one of the embodiments, due in search result, being also possible that multiple objects to be recommended in each platform.
For the clooating sequence of clear and definite each object, user is facilitated quickly to search, the present embodiment provides a kind of method, right for each
As there is corresponding grade weights.I.e.：
The grade weights in the website platform rank list are arranged each object in search result
Sequence step, including：
The object group that different types of object is constituted is ranked up according to the ordering rule of setting, object group is obtained
Grade weights；
The index parameter of each object in search result is obtained, wherein, the index parameter of the object includes object place
One kind in the pageview of webpage where the grade weights and object of object group where the grade weights of website platform, object
Or it is a variety of；
Respectively by multiplied by weight of the index parameter of each object with corresponding setting, and by product addition, obtain each right
The grade weights of elephant；
The each object in search result is ranked up according to the grade weights size of the object.
Wherein, groups of objects where the grade weights of website platform, object where the index parameter of object can include object
One or more in the pageview of webpage where the grade weights and object of group, can also be not limited to These parameters parameter,
Other index parameters are may also include, are set with specific reference to user's request.
In a specific embodiment, the index parameter of the object grade weights of website platform, object institute where object
It is described according in the website platform rank list where grade weights and object in object group during the pageview of webpage
Grade weights step is ranked up to each object in search result, including：
The object group that different types of object is constituted is ranked up according to the ordering rule of setting, object group is obtained
Grade weights；
Obtain in search result the grade weights of website platform, the grade weights of place object group where each object with
And the pageview of webpage where object；
The grade weights of the object are calculated using below equation：
W=k1*W1+k2*W2+k3*W3
Wherein, W represents the grade weights of object, W1Represent the grade weights of website platform, W2Represent the grade of object group
Weights, W3Represent pageview, k1Represent the corresponding weight of website platform, k2Represent the corresponding weight of object group, k3Expression is browsed
Measure corresponding weight；
The each object in search result is ranked up according to the grade weights size of the object.
Object group is that same type of object is divided in into one group.Such as, same brand can be regarded as same class
Type, regard the object of same brand as same target group.Different types of object is constituted according to the ordering rule of setting
It can be to be ranked up brand according to the ordering rule of setting that object group, which is ranked up,.
In the present embodiment, the grade weights of website platform, the grade weights of object group and pageview are regard as object
Index parameter, can obtain relatively reasonable object order, make the object information of recommendation more reasonable.
The various embodiments described above such as, can be believed with independent assortment in the attribute of the object according to customer attribute information with obtaining
After breath is matched, the object in matching result according to the grade weights of object can be ranked up and be shown.Herein no longer
Repeat one by one.The present invention program also provides a kind of concrete application example and illustrated, such as can be used for recommended website platform
Merchandise news.
For above-mentioned application example, realized based on the SNA described in Fig. 2, in Fig. 2, aggregation platform 210 is distinguished
With each website platform 220 by network connection, aggregation platform 210 passes through network connection with each client 230 respectively.Specific side
Method flow is as follows：
A1：Aggregation platform docks a thread with multiple website platforms, obtains the attribute of each website platform commodity to be recommended
Information；
A2：The corresponding attribute letter of each Property Name in default label form is identified from the attribute information of the commodity of acquisition
Breath, the attribute information of acquisition is recorded under the corresponding Property Name of label form, the label of the commodity is obtained.Wherein, it is described
The attribute information of commodity includes word generic attribute information and/or picture category attribute information, and the default label form records many
Individual Property Name.
Wherein, preset the Property Name in label form include title, brand, color, collar, fabric, pattern, feature,
Style, classification, price, style etc..Specific such as table 1：
Table 1
A3：Grade sequence is carried out to website platform according to the platform grade sequence of setting rule, website platform grade is obtained
List；
As shown in figure 3, being platform sequence schematic diagram in application example.Sorted including natural ordering and business.Natural ordering
Including index parameter can include：Using the website platform platform coverage rate fg (such as：Population coverage, region overlay
Rate), word-of-mouth influence power kb (such as：Evaluating deg, searchable index), platform industry ranking pm (such as：B2C industries ranking, ALEXA
Ranking), the index parameter of industry ranking can include：It is divided into ratio fc (being divided into the higher grade point of ratio bigger), valency of bidding
Lattice jj (the high grade point of sequence of bidding is bigger).A algorithm can be used：A=fg*30%+kb*50%+pm*20% can also be adopted
Use B algorithms：B=fc*50%+jj*50%.Certainly, when needing, B algorithms can also be first arranged and sorted again A algorithm.Specifically
Setting as needed.
Furthermore it is also possible to which the information editing carried out again to the platform information attribute of collection storage arranges and supplements perfect.
As shown in table 2：
Table 2
During subsequent recommendation, the relevant information of each platform can be shown according to table 2.
A4：Different brands are ranked up according to the ordering rule of setting, the grade weights of brand are obtained；
Such as, as shown in figure 4, being brand sequence schematic diagram in application example.Can be by brand entity shop quantity st, product
Board popularity zm, positive rating hp etc. are as index parameter, respectively by the index parameter of each brand and corresponding multiplied by weight, and
By product addition, the grade weights A=st*30%+zm*30%+hp*40% of each brand is obtained；According to each brand etc.
Level weights size is ranked up to brand, obtains brand rank list.
Furthermore it is also possible to the information flag carried out again to the brand message attribute of collection storage arranges and supplements perfect,
Such as table 3：
Table 3
During subsequent recommendation, the relevant information of various brands can be shown according to table 3.
A5：Scanned for according to search instruction from the attribute information of the commodity to be recommended of acquisition；
A6：As shown in figure 5, being commodity sequence schematic diagram in application example.Obtain net where each commodity in search result
The grade weights of platform, the pageview of the grade weights of place commodity group and commodity place webpage, offtake, business
Page click ratio etc. where product；
The grade weights of the commodity are calculated using below equation：
W=js*40%+pp*30%+pt*30%
Wherein, W represents the grade weights of commodity, and js represents commodity sequence index, and pp represents the grade of the affiliated brand of commodity
Weights, pt represents the grade weights of the affiliated platform of commodity, and 40%, 30%, 30% is corresponding weight.
A7：Each commodity in search result are ranked up according to the grade weights size of the commodity.
When search condition is brand, then recommend the merchandise news under displaying different platform for the brand.Work as searching bar
When part is commodity, then search result is shown into the Commercial goods labelses information by grade size.
Furthermore it is also possible to interest is excavated according to the behavior of user's microblogging, navigation patterns, geographical position, occupation, age etc.,
That is customer attribute information.User model is set up by customer attribute information.According to customer attribute information, environmental characteristic, special selling business
The matching degree of product feature three is recommended.Information on special sale quality and index of selling fast are considered during recommendation, recommends temperature high, good
The high commodity of degree of commenting.Real-time recommendation, calculates recommendation results in 0.1 second, complete within 3 seconds the extraction of information on special sale, excavate, disappear again, divide
Class, calculates new user interest distribution for 5 seconds, user model is updated in 10 seconds.
This specifically uses example, and mode, the sort algorithm of modularization search, personalization are refreshed using data message multithreading
Recommended engine and natural language processing and image recognition technology.The refreshing mode of data message is realized by multithreading.Keep away
Exempt from directly to user's presentation brand and commodity in conventional art, user oneself carries out selection brand and commodity one by one.This example is
Carried out rearranging reconstruct according to the related data of brand message and merchandise news, the custom in conjunction with user scans for pushing away
Recommend, so as to conveniently recommend most suitable brand article for user.
According to the above method, the present invention also provides a kind of site information commending system, as shown in fig. 6, being website of the present invention
The structural representation of information recommendation system embodiment, including：
Attribute information acquisition module 610, for each website platform to be docked into one or more threads, obtains each website and puts down
The attribute information of platform object to be recommended；
Website platform rank list determining module 620, for the platform grade sequence rule according to setting to website platform
Grade sequence is carried out, website platform rank list is obtained；
Search module 630, is carried out for the searching request according to user from the attribute information of the object to be recommended of acquisition
Search；
Order module 640, for the grade weights in the website platform rank list to each in search result
Individual object is ranked up；
First display module 650, for generating recommendation information according to the ranking results, the recommendation information is issued to
Client is shown.
In one of the embodiments, in addition to：
Attribute information identification module, respectively belongs to for being identified from the attribute information of the object of acquisition in default label form
Property the corresponding attribute information of title, wherein, the attribute information of the object includes word generic attribute information and/or picture generic attribute
Information, the default label form records multiple Property Names；
Label determining module, for the attribute information of acquisition to be recorded under the corresponding Property Name of label form, is obtained
The label of the object；
First display module includes：
Tag sorting module, the tag sorting order for determining object according to ranking results；
Sub- display module, is shown for the label of sequence to be issued into client.
In one of the embodiments, the website platform rank list determining module, including：
Index parameter acquisition module, the index parameter for obtaining each website platform, wherein, the index parameter includes making
With one or more in the population coverage of the website platform, areal coverage, opinion rating, searching rate；
Website platform grade weights determining module, for respectively by the index parameter of each website platform and corresponding setting
Multiplied by weight, and by product addition, obtain the grade weights of each website platform；
Website platform rank list determining module, for according to the grade weights size of each website platform to website platform
It is ranked up, obtains website platform rank list.
In one of the embodiments, in addition to：
Customer attribute information acquisition module, for obtaining customer attribute information by user account；
User model sets up module, each in pre-set user label form for being identified from the customer attribute information of acquisition
The corresponding attribute information of Property Name, the attribute information of acquisition is recorded in the corresponding Property Name of pre-set user label form
Under, the user model of the user is obtained, wherein, the customer attribute information includes word generic attribute information and/or picture generic
Property information, the pre-set user label form records multiple Property Names；
Second display module, the attribute information for the object according to user model with obtaining is matched, according to described
Grade weights in website platform rank list are ranked up to the object of matching, and the attribute information of the object of sequence is carried out
Recommend displaying.
In one of the embodiments, the order module includes：
Object group grade weights determining module, is constituted different types of object for the ordering rule according to setting
Object group is ranked up, and obtains the grade weights of object group；
Acquisition module, the index parameter for obtaining each object in search result, wherein, the index parameter of the object
Webpage is clear where the grade weights and object of object group where grade weights, object including website platform where object
One or more in the amount of looking at；
Object level weight computing module, for respectively by weight phase of the index parameter of each object with corresponding setting
Multiply, and by product addition, obtain the grade weights of each object
Sub- order module, is arranged each object in search result for the grade weights size according to the object
Sequence.
The site information commending system of the present invention and the site information of the present invention recommend method to be one-to-one, above-mentioned net
Correlation technique feature and its technique effect in information recommendation method embodiment of standing are applied to site information commending system and implemented
In example, it will not be repeated here.
Embodiment described above only expresses the several embodiments of the present invention, and it describes more specific and detailed, but simultaneously
Therefore the limitation to the scope of the claims of the present invention can not be interpreted as.It should be pointed out that for one of ordinary skill in the art
For, without departing from the inventive concept of the premise, various modifications and improvements can be made, these belong to the guarantor of the present invention
Protect scope.Therefore, the protection domain of patent of the present invention should be determined by the appended claims.
Claims (8)
1. a kind of site information recommends method, it is characterised in that including：
Each website platform is docked into one or more threads, the attribute information of each website platform object to be recommended is obtained；Wherein,
The process for obtaining the object to be recommended of each website platform is synchronous with the renewal process of the website platform；
Grade sequence is carried out to website platform according to the platform grade sequence of setting rule, website platform rank list is obtained；
Scanned for according to the searching request of user from the attribute information of the object to be recommended of acquisition, according to the website platform
Grade weights in rank list are ranked up to each object in search result；
Recommendation information is generated according to the ranking results, the recommendation information is issued to client and is shown；
The grade weights in the website platform rank list are ranked up step to each object in search result
Suddenly, including：
The object group that different types of object is constituted is ranked up according to the ordering rule of setting, obtain object group etc.
Level weights；
The index parameter of each object in search result is obtained, wherein, the index parameter of the object includes website where object
One kind or many in the pageview of webpage where the grade weights and object of object group where the grade weights of platform, object
Kind；
Respectively by multiplied by weight of the index parameter of each object with corresponding setting, and by product addition, obtain each object
Grade weights；
The each object in search result is ranked up according to the grade weights size of the object.
2. site information according to claim 1 recommends method, it is characterised in that described that each website platform is docked into one
After individual or multiple threads, the attribute information step for obtaining each website platform object to be recommended, in addition to：
The corresponding attribute information of each Property Name in default label form is identified from the attribute information of the object of acquisition, its
In, the attribute information of the object includes word generic attribute information and/or picture category attribute information, the default label form note
Record multiple Property Names；
The attribute information of acquisition is recorded under the corresponding Property Name of label form, the label of the object is obtained；
It is described to generate recommendation information according to the ranking results, the recommendation information is issued to client and is shown step bag
Include：
The tag sorting order of object is determined according to ranking results, the label of sequence is issued to client and is shown.
3. site information according to claim 1 recommends method, it is characterised in that described to be arranged according to the platform grade of setting
Sequence rule carries out grade sequence to website platform, obtains website platform rank list step, including：
Obtain the index parameter of each website platform, wherein, the index parameter including the use of the website platform population coverage,
It is one or more in areal coverage, opinion rating, searching rate；
Respectively by multiplied by weight of the index parameter of each website platform with corresponding setting, and by product addition, each net is obtained
The grade weights of platform；
Website platform is ranked up according to the grade weights size of each website platform, website platform rank list is obtained.
4. site information according to claim 1 recommends method, it is characterised in that also including step：
Customer attribute information is obtained by user account；
The corresponding attribute information of each Property Name in pre-set user label form is identified from the customer attribute information of acquisition, its
In, the customer attribute information includes word generic attribute information and/or picture category attribute information, the pre-set user label form
Record multiple Property Names；
The attribute information of acquisition is recorded under the corresponding Property Name of pre-set user label form, user's mould of the user is obtained
Type；
The attribute information of object according to user model with obtaining is matched, in the website platform rank list etc.
Level weights are ranked up to the object of matching, and the attribute information of the object of sequence is carried out into recommendation displaying.
5. a kind of site information commending system, it is characterised in that including：
Attribute information acquisition module, for each website platform to be docked into one or more threads, obtains each website platform and waits to push away
Recommend the attribute information of object；Wherein, obtain the object to be recommended of each website platform process and the website platform it is updated
Journey is synchronous；
Website platform rank list determining module, grade is carried out for the platform grade sequence rule according to setting to website platform
Sequence, obtains website platform rank list；
Search module, is scanned for for the searching request according to user from the attribute information of the object to be recommended of acquisition；
Order module, enters for the grade weights in the website platform rank list to each object in search result
Row sequence；
First display module, for generating recommendation information according to the ranking results, client is issued to by the recommendation information
It is shown；
The order module includes：
Object group grade weights determining module, the object for constituting different types of object for the ordering rule according to setting
Group is ranked up, and obtains the grade weights of object group；
Acquisition module, the index parameter for obtaining each object in search result, wherein, the index parameter of the object includes
The pageview of webpage where the grade weights and object of object group where the grade weights of website platform, object where object
In one or more；
Object level weight computing module, for respectively by multiplied by weight of the index parameter of each object with corresponding setting, and
By product addition, the grade weights of each object are obtained
Sub- order module, is ranked up for the grade weights size according to the object to each object in search result.
6. site information commending system according to claim 5, it is characterised in that also include：
Attribute information identification module, for identifying each attribute-name in default label form from the attribute information of the object of acquisition
Claim corresponding attribute information, wherein, the attribute information of the object includes word generic attribute information and/or picture generic attribute is believed
Breath, the default label form records multiple Property Names；
Label determining module, for the attribute information of acquisition to be recorded under the corresponding Property Name of label form, obtains this pair
The label of elephant；
First display module includes：
Tag sorting module, the tag sorting order for determining object according to ranking results；
Sub- display module, is shown for the label of sequence to be issued into client.
7. site information commending system according to claim 5, it is characterised in that the website platform rank list is determined
Module, including：
Index parameter acquisition module, the index parameter for obtaining each website platform, wherein, the index parameter is including the use of this
It is one or more in the population coverage of website platform, areal coverage, opinion rating, searching rate；
Website platform grade weights determining module, for respectively by weight of the index parameter of each website platform with corresponding setting
It is multiplied, and by product addition, obtains the grade weights of each website platform；
Website platform rank list determining module, is carried out for the grade weights size according to each website platform to website platform
Sequence, obtains website platform rank list.
8. site information commending system according to claim 5, it is characterised in that also include：
Customer attribute information acquisition module, for obtaining customer attribute information by user account；
User model sets up module, for identifying each attribute in pre-set user label form from the customer attribute information of acquisition
The corresponding attribute information of title, the attribute information of acquisition is recorded under the corresponding Property Name of pre-set user label form, obtained
The user model of the user is obtained, wherein, the customer attribute information includes word generic attribute information and/or picture generic attribute is believed
Breath, the pre-set user label form records multiple Property Names；
Second display module, the attribute information for the object according to user model with obtaining is matched, according to the website
Grade weights in platform rank list are ranked up to the object of matching, and the attribute information of the object of sequence is recommended
Displaying.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201410185036.1A CN103970850B (en) | 2014-05-04 | 2014-05-04 | Site information recommends method and system |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201410185036.1A CN103970850B (en) | 2014-05-04 | 2014-05-04 | Site information recommends method and system |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN103970850A CN103970850A (en) | 2014-08-06 |
| CN103970850B true CN103970850B (en) | 2017-09-22 |
Family
ID=51240347
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201410185036.1A Active CN103970850B (en) | 2014-05-04 | 2014-05-04 | Site information recommends method and system |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN103970850B (en) |
Families Citing this family (13)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104915418B (en) * | 2015-06-08 | 2018-09-04 | 小米科技有限责任公司 | Recommendation of websites method and device |
| CN106598985B (en) * | 2015-10-16 | 2020-03-27 | 阿里巴巴集团控股有限公司 | Information recommendation method and device |
| CN105528392B (en) * | 2015-11-27 | 2020-06-09 | 网易传媒科技（北京）有限公司 | Classification label ordering method and device |
| CN105761106A (en) * | 2016-02-07 | 2016-07-13 | 徐璇炫 | Object ordering method based on user setting and device thereof |
| CN107632971B (en) * | 2016-07-19 | 2020-12-15 | 百度在线网络技术（北京）有限公司 | Method and device for generating multidimensional report |
| WO2018023430A1 (en) * | 2016-08-02 | 2018-02-08 | 步晓芳 | Method for pushing information when displaying search results according to purpose, and search engine |
| CN106372191A (en) * | 2016-08-31 | 2017-02-01 | 广东华邦云计算股份有限公司 | Data search method and device |
| CN107967276A (en) * | 2016-10-19 | 2018-04-27 | 阿里巴巴集团控股有限公司 | A kind of method and apparatus of recommended |
| CN107798147B (en) * | 2017-12-05 | 2021-05-25 | 深圳市敏思跃动科技有限公司 | News client and message pushing method thereof |
| CN109493199A (en) * | 2019-01-04 | 2019-03-19 | 深圳壹账通智能科技有限公司 | Products Show method, apparatus, computer equipment and storage medium |
| CN110990723A (en) * | 2019-12-23 | 2020-04-10 | 上海米哈游天命科技有限公司 | Friend recommendation method, device, equipment and storage medium |
| CN111324789B (en) * | 2020-02-13 | 2022-06-17 | 创新奇智(上海)科技有限公司 | Method for calculating network information data heat |
| CN112035201B (en) * | 2020-08-24 | 2021-08-24 | 珠海格力电器股份有限公司 | Device parameter display method and device, computer device and storage medium |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101996215A (en) * | 2009-08-27 | 2011-03-30 | 阿里巴巴集团控股有限公司 | Information matching method and system applied to e-commerce website |
| CN102486784A (en) * | 2010-12-06 | 2012-06-06 | 耶宝智慧(北京)技术发展有限公司 | Information requesting method and information providing method |
| CN102663054A (en) * | 2012-03-29 | 2012-09-12 | 奇智软件（北京）有限公司 | Method and device for determining weight of website |
| CN102915506A (en) * | 2012-10-29 | 2013-02-06 | 江苏乐买到网络科技有限公司 | System for recommending group purchasing information |
| CN102945253A (en) * | 2012-10-17 | 2013-02-27 | 北京奇虎科技有限公司 | Search server and searching method thereof |
Family Cites Families (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20110191231A1 (en) * | 2010-01-29 | 2011-08-04 | Zuver Jr Kenneth Ray | Method and System for Improving Trading and a Trader's Ranking |
| CN102880607A (en) * | 2011-07-15 | 2013-01-16 | 舆情（香港）有限公司 | Dynamic network content grabbing method and dynamic network content crawler system |
| CN102779169A (en) * | 2012-06-27 | 2012-11-14 | 江苏新瑞峰信息科技有限公司 | Extracting method and device for webpage content based on HTML (Hypertext Markup Language) label |
| CN103455636A (en) * | 2013-09-27 | 2013-12-18 | 浪潮齐鲁软件产业有限公司 | Automatic capturing and intelligent analyzing method based on Internet tax data |
-
2014
- 2014-05-04 CN CN201410185036.1A patent/CN103970850B/en active Active
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101996215A (en) * | 2009-08-27 | 2011-03-30 | 阿里巴巴集团控股有限公司 | Information matching method and system applied to e-commerce website |
| CN102486784A (en) * | 2010-12-06 | 2012-06-06 | 耶宝智慧(北京)技术发展有限公司 | Information requesting method and information providing method |
| CN102663054A (en) * | 2012-03-29 | 2012-09-12 | 奇智软件（北京）有限公司 | Method and device for determining weight of website |
| CN102945253A (en) * | 2012-10-17 | 2013-02-27 | 北京奇虎科技有限公司 | Search server and searching method thereof |
| CN102915506A (en) * | 2012-10-29 | 2013-02-06 | 江苏乐买到网络科技有限公司 | System for recommending group purchasing information |
Also Published As
| Publication number | Publication date |
|---|---|
| CN103970850A (en) | 2014-08-06 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN103970850B (en) | Site information recommends method and system | |
| CN103914492B (en) | Query word fusion method, merchandise news dissemination method and searching method and system | |
| KR102127191B1 (en) | Method, apparatus and computer program for providing shopping informations | |
| KR101511050B1 (en) | Method, apparatus, system and computer program for offering and displaying a product information | |
| KR101806169B1 (en) | Method, apparatus, system and computer program for offering a shopping information | |
| US10360623B2 (en) | Visually generated consumer product presentation | |
| US9489400B1 (en) | Interactive item filtering using images | |
| CN105718184A (en) | Data processing method and apparatus | |
| US9043351B1 (en) | Determining search query specificity | |
| CN107632984A (en) | A kind of cluster data table shows methods, devices and systems | |
| CN106327227A (en) | Information recommendation system and information recommendation method | |
| CN104679771A (en) | Individual data searching method and device | |
| JP5575270B2 (en) | Ad source and keyword set adaptation in online commerce platforms | |
| CN102411754A (en) | Personalized recommendation method based on commodity property entropy | |
| KR20200101068A (en) | A method of providing a fashion item recommendation service using a body shape and purchase history | |
| KR102309373B1 (en) | System for building the contents automatically for planning sales and method therefor | |
| US20200226168A1 (en) | Methods and systems for optimizing display of user content | |
| US9330071B1 (en) | Tag merging | |
| KR20200045668A (en) | Method, apparatus and computer program for style recommendation | |
| KR20180052489A (en) | method of providing goods recommendation for cross-border E-commerce based on user experience analysis and environmental factors | |
| CN110619559B (en) | Method for accurately recommending commodities in electronic commerce based on big data information | |
| CN102073737A (en) | Method and system for automatically providing real-time purchasing suggestion for network retailer | |
| CN106682963A (en) | Recommendation system data completion method based on convex optimization local low-rank matrix approximation | |
| US20230030560A1 (en) | Methods and systems for tagged image generation | |
| KR20220019737A (en) | Method, apparatus and computer program for fashion item recommendation |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |