CN105335386B - A kind of method and device that navigation tag is provided - Google Patents

A kind of method and device that navigation tag is provided Download PDF

Info

Publication number
CN105335386B
CN105335386B CN201410311241.8A CN201410311241A CN105335386B CN 105335386 B CN105335386 B CN 105335386B CN 201410311241 A CN201410311241 A CN 201410311241A CN 105335386 B CN105335386 B CN 105335386B
Authority
CN
China
Prior art keywords
combination
information
information labels
navigation
tag
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
Application number
CN201410311241.8A
Other languages
Chinese (zh)
Other versions
CN105335386A (en
Inventor
谢朋峻
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Alibaba Group Holding Ltd
Original Assignee
Alibaba Group Holding Ltd
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Alibaba Group Holding Ltd filed Critical Alibaba Group Holding Ltd
Priority to CN201410311241.8A priority Critical patent/CN105335386B/en
Publication of CN105335386A publication Critical patent/CN105335386A/en
Priority to HK16106712.6A priority patent/HK1218793A1/en
Application granted granted Critical
Publication of CN105335386B publication Critical patent/CN105335386B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Landscapes

  • Information Retrieval, Db Structures And Fs Structures Therefor (AREA)

Abstract

This application discloses a kind of method and devices providing navigation tag, including:At least two information labels are established for each business object in business object set;According to the information labels of each business object, determine that information labels combine;Wherein, each information labels combination includes at least two information labels, and the information labels for including are appeared in jointly in the information labels of different business object;Determine the incidence relation between the different information labels in described information tag combination;By the incidence relation between different information labels be mutually related information labels combination be determined as the first navigation tag combination;First navigation tag combination is presented on navigation engine interface.Using scheme provided by the embodiments of the present application, the tag combination with complete entity meaning actively can be presented to user, and user is without inputting keyword in the case where not depending on manual operation, more human resources can not only be saved, also there is stronger flexibility.

Description

A kind of method and device that navigation tag is provided
Technical field
This application involves Internet technical field and field of computer technology more particularly to a kind of sides providing navigation tag Method and device.
Background technology
In existing Internet technology, some business objects can be generally issued on website, for logging in the user of the website Browsing, and the further post-treatment operations for being directed to specific transactions object.For example, by taking e-commerce website as an example, business Object can be specifically the product of seller user's publication, and the information of business object can be specifically the description information of product, product Attribute information and product purchase information etc..The user for logging in e-commerce website can be by browsing institute's release product Various information, understand the details of the product, and can further execute collection, buy or recommend other users etc. Processing operation.By taking community website as an example, business object can be specifically the model of community users publication, the information tool of business object Body can be the description information of model, the content information etc. of model.The browsing user of login community website can be sent out by browsing The various information of the model of cloth understand the details of the model, and can further execute collection, money order receipt to be signed and returned to the sender or recommend it The processing operations such as his user.
With the continuous development of Internet technology, the business object issued on website is usually all magnanimity, therefore, in order to It allows users to fast and accurately find oneself desired business object from the business object of magnanimity, saves user time, it is existing Have and propose the method for providing a user navigation tag in technology, so that user is gone forward side by side based on its clear query intention of navigation tag One step browses and inquiry business object, enables a user to the business object for quickly finding itself needs.
Presently mainly after user inputs keyword, according to certain attribute of the business object in query result, wind The factors such as lattice or material generate navigation tag, and are presented to the user.Navigation tag under this mode is as a kind of fine-grained It describes (such as certain attribute, certain style, certain material), does not have complete entity meaning (such as " sweet ", " stamp " usually Deng), therefore could only be applied in the case where user inputs special scenes as keyword.
In order to can also provide a user navigation tag under the scene that user does not input keyword, in the prior art It is proposed a kind of method that the tag combination with complete entity meaning is provided.This method exists a variety of fine-grained description combinations Tag combination (such as " stamp thickening one-piece dress ") is obtained together, and current this tag combination is all to rely on manual operation to obtain , although having complete entity meaning, need to expend more human resources, and flexibility is poor.
It there is no more effective solution at present.
Invention content
In view of this, the embodiment of the present application provides a kind of method and device providing navigation tag, for solving existing skill When providing a user the tag combination with complete entity meaning present in art, need to expend more human resources, and spirit The poor problem of activity.
The embodiment of the present application is achieved through the following technical solutions:
The embodiment of the present application provides a kind of method providing navigation tag, including:
At least two information labels are established for each business object in business object set;
According to the information labels of each business object, determine that information labels combine;Wherein, each information labels are incorporated into Include two information labels less, and the information labels for including are appeared in jointly in the information labels of different business object;
Determine the incidence relation between the different information labels in described information tag combination;
By the incidence relation between different information labels be mutually related information labels combination be determined as the first navigation Tag combination;
First navigation tag combination is presented on navigation engine interface.
The embodiment of the present application provides a kind of device providing navigation tag, including:
Information labels establish unit, for establishing at least two information marks for each business object in business object set Label;
Information labels combine determination unit, for establishing each business object established in unit according to described information label Information labels, determine information labels combine;Wherein, each information labels combination includes at least two information labels, and wraps The information labels contained are appeared in jointly in the information labels of different business object;
Incidence relation determination unit, in the information labels combination for determining the determination of described information tag combination determination unit Different information labels between incidence relation;
First navigation tag combines determination unit, the different information marks for will be determined in the incidence relation determination unit Incidence relation between label be mutually related information labels combination be determined as the first navigation tag combination;
First display unit, the first navigation tag for determining first navigation tag combination determination unit combine It is presented on navigation engine interface.
It is each business pair in business object set in above-mentioned at least one technical solution provided by the embodiments of the present application As establishing at least two information labels, and according to the information labels of each business object, determines that the first navigation tag combines, present On navigation engine interface, wherein the combination of each first navigation tag includes at least two information labels, including all information Label is belonging respectively at least two business objects of business object set, and any two information labels are interrelated.Above-mentioned In technical solution, the label with complete entity meaning actively can be presented to user in the case where not depending on manual operation Combination, and user can not only save more human resources, also have stronger flexibility without inputting keyword.
Other features and advantage will illustrate in the following description, also, partly become from specification It obtains it is clear that being understood by implementing the application.The purpose of the application and other advantages can be by the explanations write Specifically noted structure is realized and is obtained in book, claims and attached drawing.
Description of the drawings
Attached drawing is used for providing further understanding of the present application, and a part for constitution instruction, implements with the application Example for explaining the application, does not constitute the limitation to the application together.In the accompanying drawings:
Fig. 1 is the flow chart of the method provided by the embodiments of the present application for providing navigation tag;
Fig. 2 is the place provided by the embodiments of the present application being presented on the combination of multiple first navigation tags on navigation engine interface Manage flow chart;
Fig. 3 is the incidence relation between the different information labels in determining information labels combination provided by the embodiments of the present application Process chart;
Fig. 4 is the processing stream of the method for providing navigation tag based on the user's input information provided by the embodiments of the present application Cheng Tu;
Fig. 5 is provided by the embodiments of the present application in practical applications for realizing the system for the method for providing navigation tag Structural schematic diagram;
Fig. 6 is a kind of structural schematic diagram for the device providing navigation tag provided by the embodiments of the present application.
Specific implementation mode
The embodiment of the present application provides a kind of method and device providing navigation tag, which both can be applied to The process of the recommendation results comprising navigation tag is provided a user when user does not input keyword, can also be applied to Family provides a user the process of the search result comprising navigation tag when inputting keyword.Below in conjunction with Figure of description to this Shen Preferred embodiment please illustrates, it should be understood that and preferred embodiment described herein is only used for describing and explaining the application, It is not used to limit the application.And in the absence of conflict, the features in the embodiments and the embodiments of the present application can be with It is combined with each other.
The embodiment of the present application provides a kind of method providing navigation tag, as shown in Figure 1, including step 11-15.
Step 11, at least two information labels are established for each business object in business object set;
The object that the business object, that is, user scans for, including described in computer or storage information, such as Merchandise news, news or audio-video frequency content in Webpage etc..
Described information label includes the characteristic information of identification service object, such as indicates the style or attribute letter of business object The descriptor of breath.When business object is dress ornament class commodity, easy dress or business wear can be divided into according to style, then described information Label may include " easy dress " or " business wear ".It sees from different perspectives, a business object can be there are many attribute and/or wind Lattice, therefore, each business object can have multiple information labels, for example can be certain according to its target user and decorated style One dress ornament class commodity stamp multiple information labels such as " woman style ", " easy dress ".
Wherein, step 11 may be used but be not limited under type such as and realizes:
Mode 1 obtains information labels by the own attribute of business object.
Mode 2 carries out participle and attribute forecast acquisition information labels by the title to business object;
Mode 3 obtains information labels by user-generated content (User Generated Content, UGC) data packet.
Still by taking business object is dress ornament class commodity as an example, when commodity are issued, commodity owner can fill out the commodity similar In the information labels such as brand, the place of production, material (mode 1);Commodity on shopping guide website when selling, from the title of the commodity Extract the information labels (mode 2) such as brand, the place of production, material;Comment information, the description information of the commodity are directed to from shopping guide website Text in extraction information labels (mode 3).
In the embodiment of the present application, information labels can be, but not limited to be marked using following form:
p1:V1, p2:V2, p3:V3, p4:V4 (such as style:It is sweet, brand:Nike, element:It embroiders).Here p1, P2, p3, p4 refer to the attribute (style, brand, element) of each information labels, and v1, v2, v3, v4 refer to each information labels Attribute value (sweet, Nike, embroidery).
Step 12, according to the information labels of each business object, determine that information labels combine;
Wherein, each information labels combination includes at least two information labels, and each information labels combine the letter for including Breath label is appeared in jointly in the information labels of different business object of business object set;
Information labels due to meeting above-mentioned condition combine may be very more, and the calculating cost in specific implementation compares Greatly, therefore the step 12 is in order to determine that information labels combine, and can be carried out frequent according to the information labels of each business object Item excavates, and then determines qualified information labels combination.
In when progress frequent-item, Apriori algorithm or FP-Growth algorithms are may be used but are not limited to.In reality Apriori algorithm or FP-Growth algorithms can be flexibly selected according to scene in the application of border.When each information that scene needs When information labels number in tag combination is less, Apriori algorithm can be selected, advantage is that algorithm is simple, the disadvantage is that needing Interative computation, it is very time-consuming on distributed platform when data volume is larger.Information in scene needs each information labels to combine When the number of label is more, it is necessary to use FPGrowth algorithms under parallel frame.
In addition, when frequent item set mining, the information labels combination preset a minimum support threshold value, and determined is needed It needs to meet:The support between different information labels in information labels combination is all higher than the preset minimum support threshold Value.
Wherein, the support between the different information labels in information labels combination is:Information labels combine Co-occurrence probabilities of the information labels in the information labels of each business object.Shown in the following formula of formula (1) of support:
Support (A, B)=P (A ∪ B) (1)
Wherein, Support (A, B) indicates that the support of A and B, P (A ∪ B) indicate the probability that A and B occurs simultaneously, A and B Two information labels in information labels combination are indicated respectively.If A and B occur simultaneously, the probability is relatively small, illustrate A and B it Between be mutually unrelated certainly, if the probability that A and B occurs simultaneously is bigger, illustrate to may be mutual pass between A and B Connection.
After screening based on minimum support threshold value, so that it may it is combined with obtaining all qualified information labels, it is optional , each information labels can also be obtained and combine the corresponding business object quantity (business pair combined containing the information labels The quantity of elephant).
In one embodiment of the application, the set of the information labels of a business object can be expressed as: (tag1tag2tag3tag4….tagn).And it is 5% that can preset minimum support threshold value in the embodiment of the present invention, it is determined that The navigation tag combination gone out can be expressed as:(tag1tag3tag5):N1 indicates that tri- information labels of tag1, tag3, tag5 are total Occur in N1 business object.
For example, by taking business object is to dress ornament class commodity as an example, business object collection is combined into 10000 dress ornament class commodity, including The commodity one of information labels " crew neck " share 8000, including the commodity of information labels " lapel " have 7000, and include simultaneously The commodity of " crew neck " and " lapel " have 500, then support=500/ between " crew neck " and " lapel " the two information labels 10000=5% meets the requirement about support threshold, it may be determined that is combined for information labels.
Step 13, the incidence relation between the different information labels in each information labels combination is determined;
Might not mutually close between the different information labels that the information labels combination obtained by step 12 includes Connection.Such as some popular information labels " Korea Spro's version " " commuting " etc. can all occur in many commodity, still " Korea Spro edition " and its The correlation of its most information label is not high, therefore can further utilize confidence level after information labels combination is determined It is screened with promotion degree, wherein confidence level and promotion degree can reflect the incidence relation between information labels.Concrete implementation Process can be discussed in detail in subsequent embodiment, and details are not described herein.
Step 14, it is that the information labels combination that is mutually related is determined as the by the incidence relation between different information labels One navigation tag combines;
Optionally, when information labels include the attribute and attribute value of business object, for example p1 (attribute) is used:V1 (belongs to Property value) this form when, step 14 can specifically include:
It is the information that is mutually related from the incidence relation between different information labels according to preassigned combinations of attributes In tag combination, the information labels combination only combined comprising attribute value corresponding with the combinations of attributes is obtained, is determined as first Navigation tag combines.
Specifically, after obtaining all qualified information labels combinations in the embodiment of the present application, can according to it includes Information labels attributes extraction it is different types of be combined into come.Such as can in attribute include " material " label, one All combined sortings of a " element " label and " product word " label, which come out, can be obtained by the similar combination such as the following table 1.
Table 1:
Material:Fleece Element:Letter Product:Defend clothing
Material:Lamb wool Element:Slide fastener Product:Defend clothing
Material:Knitting Element:Jacquard weave Product:Half body skirt
Material:Corduroy Element:Umbrella Product:Half body skirt
Then the attribute of each information labels is removed, reserved property value can obtain the information labels of table 2 such as and combine.
Table 2:
Fleece letter defends clothing
Lamb wool slide fastener defends clothing
Knitted jacquard half body skirt
Corduroy umbrella half body skirt
Finally, by " fleece letter defends clothing " " lamb wool slide fastener defends clothing " " knitted jacquard half body skirt " " corduroy umbrella half body Skirt " is determined as the combination of the first navigation tag.
Step 15, determining the first navigation tag combination is presented on navigation engine interface.
Because different business objects may have the information labels of synonymous or close justice, can cause to finally obtain very in this way First navigation tag of mostly synonymous or close justice combines.Still by taking business object is dress ornament class commodity as an example, such as " pullover oxeye needle Knit shirt " and the registration of " grandson pari's oxeye sweater " on the dress ornament class commodity of covering it is just very high, for synonymous first navigation Therefore tag combination is not recommend to be worth to user if the two first navigation tags are combined while showing user , therefore the embodiment of the present application can also be before showing to user, the first navigation tag group that can also be first high to registration It closes and carries out normalizing.
Method as shown in Figure 2 specifically may be used, the combination of the first navigation tag is presented on navigation engine interface, Specifically include following processing step:
Step 21, the registration of the first navigation tag of any two combination in the combination of the first navigation tag is determined;
The coincidence that minhash algorithms calculate the combination of the first navigation tag of any two may be used in the embodiment of the present application Degree, the registration refer to the registration of the business object of the first navigation tag combined covering.
Step 22, two the first navigation tags combination for registration being more than to default minimum registration threshold value carries out normalizing, Obtain the combination of the first navigation tag after normalizing;
Step 23, the first navigation tag combination after normalizing is presented on navigation engine interface.
Specifically, the quantity of the first navigation tag combination after normalizing may be still relatively more, therefore the embodiment of the present application In, the specify information of corresponding all business objects can be combined according to the first navigation tag each of after normalizing, determine normalizing Then each of the afterwards temperature score of the first navigation tag combination is pressed the first navigation tag combination after the normalizing of specified quantity It is presented on navigation engine interface from high to low sequence according to temperature score.
Wherein, still by business object be dress ornament class commodity for, specify information may include Sales Volume of Commodity, issuing time and The information such as comment on commodity.
Below in conjunction with the accompanying drawings, the specific implementation process of above-mentioned steps 13 is described in detail with specific embodiment.
Method as shown in Figure 3 may be used, determine that the association between the different information labels in information labels combination is closed System, specifically includes following processing step:
Step 31, the confidence level and promotion degree between any two information labels in information labels combination are determined;
The formula of confidence level is:Confidence (A → B)=P (A | B) (2)
Confidence (B → A)=P (B | A) (3)
Wherein, Confidence (A → B) indicates confidence levels of the A for B, and when P (A | B) indicates that A occurs, B also occurs simultaneously Probability;Confidence (B → A) indicates confidence levels of the B for A, and when P (B | A) indicates that B occurs, A also occurs general simultaneously Rate.
The formula of promotion degree is:Lift (A, B)=P (B | A)/P (B) (4)
Wherein, Lift (A, B) indicates the promotion degree of A and B, when P (B | A) indicates that B occurs, the probability that A also occurs simultaneously, and P (B) probability that B occurs is indicated.
In general, when promotion degree is equal to 1, illustrate that A and B does not have any association;If promotion degree is less than 1, illustrate that A and B is phase Mutually repel.Therefore, preset minimum lift degree threshold value could be provided as 1 in the embodiment of the present application.
Step 32, when the confidence level of any two information labels in information labels combination is all higher than preset minimum confidence Spend threshold value, when promotions degree is all higher than preset minimum lift degree threshold value, determine information labels combine in different information labels it Between it is interrelated.
Step 33, when the confidence level of any two information labels in information labels combination is no more than preset minimum confidence Spend threshold value, and/or promotions degree be not more than preset minimum lift degree threshold value when, determine information labels combine in different information marks It is mutually unrelated between label.
By taking two information labels A and B including during information labels combine as an example, A is needed to set the confidence level and B of B to A When reliability is all higher than the promotion degree of minimal confidence threshold and A and B more than minimum lift degree threshold value, the information labels are determined A and B in combination is interrelated.
Still with (crew neck, lapel) for:
Crew neck is 500/8000 to be approximately equal to 7% to the confidence level of lapel, smaller;
Lapel is 500/7000 to be approximately equal to 7% to the confidence level of crew neck, smaller;
The promotion degree of crew neck and lapel is that (500/8000)/(7000/10000) is approximately equal to 0.09 less than 1;
So crew neck and lapel onrelevant.
The method of above-mentioned offer navigation tag can be applied under the scene of the input information of not user.
Once receiving the input information of user, this method can further include processing procedure as shown in Figure 4:
Step 41, after receiving the input information of user, the information mark that the input information with user matches is determined Label;
Wherein, the input information of user may include query word input by user.Match with the input information of user Information labels may include information labels identical with query word input by user.In general, user can be according to business As querying condition into the search of row information, query word input by user often can also be used as the attribute or style and features of object The descriptor of information labels.According to accurate matched mode, can obtain and the matched information labels of query word input by user. When there is no information labels matched with query word input by user, it can be matched by the meaning of word or other modes are determined and used The inquiry highest information labels of Word similarity of family input are as matched information labels.
Query word input by user can be that user is inputted by keyboard, can also include user selection navigation word or Recommend word, such as in commercial articles searching, the recommended keywords of user or the page with commodity navigation feature is supplied to provide Navigation word.
Step 42, from determining the first navigation tag combination, the first of the matched information labels of input information will be included Navigation tag combination is determined as the combination of the second navigation tag;
Specifically, the step can be, but not limited to include:
To include matched information labels of input information and predetermined simultaneously from the combination of above-mentioned first navigation tag The first navigation tag combination of the preference label of the user is determined as the combination of the second navigation tag;
Wherein, the preference label of the user can be determined according to the specified operation of the user, specifically, the finger of user Fixed operation recycles word frequency-to return document frequency (Term Frequency-Inverse in conjunction with the information labels of business object Document Frequency, TF-IDF) algorithm (a kind of common weighting technique for information search and information excavating) To polymerize the preference label of the user.And for the less user of behavior, the first navigation tag determined can be combined In include information labels as its preference label.
Wherein, specified operation can be flexibly arranged according to the actual characteristic for the business object issued on website, for example, In e-commerce website, specified operation can be the operations such as the click to business object (i.e. commodity), collection and purchase.
Step 43, determining the second navigation tag combination is presented on the search result generated according to the input information of user In the page.
The method of above-mentioned offer navigation tag provided by the embodiments of the present application, in practical applications can be in following Fig. 5 institutes It is realized in the system shown:
The system can distinguish two large divisions, when off-line calculation part, including commodity mark module, structured tag group It closes and excavates module, tag combination normalizing module, tag combination point counting module and user preference label computing module.Wherein:
Commodity mark module, for establishing at least two information labels for each business object in business object set;
Module is excavated in structured tag combination, for according to the information in commodity mark module being each business object foundation Label determines that the first navigation tag combines;
Tag combination normalizing module is combined for combining the first navigation tag excavated and determined in module to structured tag It is normalized;
Tag combination point counting module is led for calculating obtained after tag combination normalizing module is normalized first The temperature score for tag combination of navigating;
User preference label computing module, the preference label for calculating user.
Second is that application on site part, may include two kinds of application scenarios:
1, in the case where searching for scene, after user inputs a search term, the corresponding first navigation tag group of the search term is counted It closes, and the combination of corresponding first navigation tag is recommended out to screen for user according to temperature score in the form of theme.In heat The customized information of user, such as preference label can also be combined in real time by spending in score.
2, in the case where recommending scene, directly first can be recommended to lead to user in the case where user does not input any search term Boat tag combination.
Based on same inventive concept, according to the method for the offer navigation tag that the above embodiments of the present application provide, correspondingly, The embodiment of the present application also provides a kind of devices providing navigation tag, and structural schematic diagram is as shown in fig. 6, specifically include:
Information labels establish unit 61, for establishing at least two information for each business object in business object set Label;
Information labels combine determination unit 62, for establishing in unit 61 each business established according to described information label The information labels of object determine that information labels combine;Wherein, each information labels combination includes at least two information labels, And the information labels for including are appeared in jointly in the information labels of different business object;
Incidence relation determination unit 63, the information labels group determined for determining described information tag combination determination unit 62 The incidence relation between different information labels in conjunction;
First navigation tag combines determination unit 64, the different letters for will be determined in the incidence relation determination unit 63 Breath label between incidence relation be mutually related information labels combination be determined as the first navigation tag combination;
First display unit 65, the first navigation tag for determining first navigation tag combination determination unit 64 Combination is presented on navigation engine interface.
Optionally, described information tag combination determination unit 62, is specifically used for:
According to the information labels of each business object, frequent item set mining is carried out, determines to include at least two information marks The information labels of label combine;Wherein, the information labels that each information labels combination includes appear in different business object jointly In information labels, and the support between the different information labels in each information labels combination is all higher than preset minimum support Spend threshold value.
Optionally, the incidence relation determination unit 63, specifically includes:
Confidence level and promotion degree determining module 631, for determining any two information mark in described information tag combination Confidence level between label and promotion degree;
Incidence relation determining module 632 determines described information for working as the confidence level and promotion degree determining module 631 The confidence level of any two information labels in tag combination is all higher than preset minimal confidence threshold, and promotion degree is all higher than pre- If minimum lift degree threshold value when, determine interrelated between the different information labels in described information tag combination;When described The confidence level and promotion degree determining module 631 determine the confidence of any two information labels in described information tag combination Degree be not more than preset minimal confidence threshold, and/or promotions degree be not more than preset minimum lift degree threshold value when, determination described in It is mutually unrelated between different information labels in the information labels combination filtered out.
Optionally, described information label includes the attribute and attribute value of business object;
First navigation tag combines determination unit 64, is specifically used for:
It is the information that is mutually related from the incidence relation between different information labels according to preassigned combinations of attributes In tag combination, the information labels combination only combined comprising attribute value corresponding with the combinations of attributes is obtained, is determined as first Navigation tag combines.
Optionally, first display unit 65, specifically includes:
Registration determining module 651, for determining the first clearing mark of any two in the first navigation tag combination Sign the registration of combination;
Normalizing module 652, the registration for will be determined in registration determining module 651 are more than default minimum registration threshold Two the first navigation tags combination of value carries out normalizing, obtains the combination of the first navigation tag after normalizing;
First is presented module 653, draws for the first navigation tag combination after 652 normalizing of normalizing module to be presented on navigation It holds up on interface.
Optionally, described first module 653 is presented, is specifically used for:
The specify information that corresponding all business objects are combined according to the first navigation tag each of after normalizing, determines normalizing Each of the afterwards temperature score of the first navigation tag combination;The first navigation tag after the normalizing of specified quantity is combined according to heat Degree score is presented on from high to low sequence on navigation engine interface.
Optionally, described device further includes:
Receiving unit 66, the input information for receiving user determine and the matched information labels of the input information;
Second navigation tag combines determination unit 67, and first for being determined from the first navigation tag combination determination unit 64 In navigation tag combination, second will be determined as comprising being combined with the first navigation tag of the matched information labels of the input information Navigation tag combines;
Second display unit 68, the second navigation tag for determining the second navigation tag combination determination unit 67 combine It is presented in the result of page searching generated according to the input information of user.
Optionally, second navigation tag combines determination unit 67, is specifically used for:
To include described defeated simultaneously from the first navigation tag combination that the first navigation tag combination determination unit 64 determines The the first navigation tag combination for entering the information labels of information matches and the preference label of the predetermined user is determined as the Two navigation tags combine;Wherein, the preference label of the user is determined according to the specified operation of the user.
It should be understood by those skilled in the art that, embodiments herein can be provided as method, system or computer program Product.Therefore, complete hardware embodiment, complete software embodiment or reality combining software and hardware aspects can be used in the application Apply the form of example.Moreover, the application can be used in one or more wherein include computer usable program code computer The computer program production implemented in usable storage medium (including but not limited to magnetic disk storage, CD-ROM, optical memory etc.) The form of product.
The application is with reference to method, the flow of equipment (system) and computer program product according to the embodiment of the present application Figure and/or block diagram describe.It should be understood that can be realized by computer program instructions every first-class in flowchart and/or the block diagram The combination of flow and/or box in journey and/or box and flowchart and/or the block diagram.These computer programs can be provided Instruct the processor of all-purpose computer, special purpose computer, Embedded Processor or other programmable data processing devices to produce A raw machine so that the instruction executed by computer or the processor of other programmable data processing devices is generated for real The device for the function of being specified in present one flow of flow chart or one box of multiple flows and/or block diagram or multiple boxes.
These computer program instructions, which may also be stored in, can guide computer or other programmable data processing devices with spy Determine in the computer-readable memory that mode works so that instruction generation stored in the computer readable memory includes referring to Enable the manufacture of device, the command device realize in one flow of flow chart or multiple flows and/or one box of block diagram or The function of being specified in multiple boxes.
These computer program instructions also can be loaded onto a computer or other programmable data processing device so that count Series of operation steps are executed on calculation machine or other programmable devices to generate computer implemented processing, in computer or The instruction executed 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 a box or multiple boxes.
Although the preferred embodiment of the application has been described, created once a person skilled in the art knows basic Property concept, then additional changes and modifications may be made to these embodiments.So it includes excellent that the following claims are intended to be interpreted as It selects embodiment and falls into all change and modification of the application range.
Obviously, those skilled in the art can carry out the application essence of the various modification and variations without departing from the application God and range.In this way, if these modifications and variations of the application belong to the range of the application claim and its equivalent technologies Within, then the application is also intended to include these modifications and variations.

Claims (16)

1. a kind of method providing navigation tag, which is characterized in that including:
At least two information labels are established for each business object in business object set;
According to the information labels of each business object, determine that information labels combine;Wherein, each information labels combination is at least wrapped Containing two information labels, and the information labels for including are appeared in jointly in the information labels of different business object;
Determine the incidence relation between the different information labels in described information tag combination;
By the incidence relation between different information labels be mutually related information labels combination be determined as the first navigation tag Combination;
First navigation tag combination is presented on navigation engine interface.
2. the method as described in claim 1, which is characterized in that according to the information labels of each business object, determine information Tag combination specifically includes:
According to the information labels of each business object, frequent item set mining is carried out, determines to include at least two information labels Information labels combine;Wherein, the information labels that each information labels combination includes appear in the information of different business object jointly In label, and the support between the different information labels in each information labels combination is all higher than preset minimum support threshold Value.
3. method as claimed in claim 1 or 2, which is characterized in that determine the different information marks in described information tag combination Incidence relation between label, specifically includes:
Determine the confidence level and promotion degree between any two information labels in described information tag combination;
When the confidence level of any two information labels in described information tag combination is all higher than preset minimal confidence threshold, When promotion degree is all higher than preset minimum lift degree threshold value, phase between the different information labels in described information tag combination is determined Mutual correlation;
When any two information labels in described information tag combination confidence level be not more than preset minimal confidence threshold, And/or promotion degree be not more than preset minimum lift degree threshold value when, determine the different information labels in described information tag combination Between it is mutually unrelated.
4. the method as described in claim 1, which is characterized in that described information label includes the attribute and attribute of business object Value;
By the incidence relation between different information labels be mutually related information labels combination be determined as the first navigation tag Combination, specifically includes:
It is the information labels that are mutually related from the incidence relation between different information labels according to preassigned combinations of attributes In combination, the information labels combination only combined comprising attribute value corresponding with the combinations of attributes is obtained, the first navigation is determined as Tag combination.
5. the method as described in claim 1, which is characterized in that first navigation tag combination is presented on navigation engine circle On face, specifically include:
Determine the registration of the first navigation tag of any two combination in the first navigation tag combination;
Two the first navigation tags combination that registration is more than to default minimum registration threshold value carries out normalizing, after obtaining normalizing First navigation tag combines;
The first navigation tag combination after normalizing is presented on navigation engine interface.
6. method as claimed in claim 5, which is characterized in that the first navigation tag combination after normalizing is presented on navigation and is drawn It holds up on interface, specifically includes:
The specify information that corresponding all business objects are combined according to the first navigation tag each of after normalizing, after determining normalizing The temperature score of each first navigation tag combination;
The first navigation tag after the normalizing of specified quantity is combined and is presented on navigation from high to low sequence according to temperature score In engine interface.
7. the method as described in claim 1, which is characterized in that the method further includes:
The input information of user is received, is determined and the matched information labels of the input information;
From first navigation tag combination, the first navigation tag with the matched information labels of the input information will be included Combination is determined as the combination of the second navigation tag;
Second navigation tag combination is presented in the result of page searching generated according to the input information of user.
8. the method for claim 7, which is characterized in that from first navigation tag combination in, will include with it is described The first navigation tag combination of the matched information labels of input information is determined as the combination of the second navigation tag, specifically includes:
To include matched information labels of the input information and predetermined simultaneously from first navigation tag combination The first navigation tag combination of the preference label of the user is determined as the combination of the second navigation tag;Wherein, the user's is inclined Good label is determined according to the specified operation of the user.
9. a kind of device providing navigation tag, which is characterized in that including:
Information labels establish unit, for establishing at least two information labels for each business object in business object set;
Information labels combine determination unit, the letter for establishing each business object established in unit according to described information label Label is ceased, determines that information labels combine;Wherein, each information labels combination includes at least two information labels, and includes Information labels are appeared in jointly in the information labels of different business object;
Incidence relation determination unit, in the information labels combination for determining the determination of described information tag combination determination unit not With the incidence relation between information labels;
First navigation tag combine determination unit, for will in the incidence relation determination unit determine different information labels it Between incidence relation be mutually related information labels combination be determined as the first navigation tag combination;
First display unit, the first navigation tag for determining first navigation tag combination determination unit, which combines, to be presented On navigation engine interface.
10. device as claimed in claim 9, which is characterized in that described information tag combination determination unit is specifically used for:
According to the information labels of each business object, frequent item set mining is carried out, determines to include at least two information labels Information labels combine;Wherein, the information labels that each information labels combination includes appear in the information of different business object jointly In label, and the support between the different information labels in each information labels combination is all higher than preset minimum support threshold Value.
11. the device as described in claim 9 or 10, which is characterized in that the incidence relation determination unit specifically includes:
Confidence level and promotion degree determining module, for determining between any two information labels in described information tag combination Confidence level and promotion degree;
Incidence relation determining module is determined for working as the confidence level and promotion degree determining module in described information tag combination The confidence levels of any two information labels be all higher than preset minimal confidence threshold, promotion degree is all higher than preset minimum and carries When liter degree threshold value, determine interrelated between the different information labels in described information tag combination;When the confidence level and carry Liter degree determining module determine the confidence levels of any two information labels in described information tag combination no more than it is preset most Small confidence threshold value, and/or promotion degree be not more than preset minimum lift degree threshold value when, determine in described information tag combination It is mutually unrelated between different information labels.
12. device as claimed in claim 9, which is characterized in that described information label includes the attribute and attribute of business object Value;
First navigation tag combines determination unit, is specifically used for:
It is the information labels that are mutually related from the incidence relation between different information labels according to preassigned combinations of attributes In combination, the information labels combination only combined comprising attribute value corresponding with the combinations of attributes is obtained, the first navigation is determined as Tag combination.
13. device as claimed in claim 9, which is characterized in that first display unit specifically includes:
Registration determining module, for determining the first navigation tag of any two combination in the first navigation tag combination Registration;
Normalizing module, the registration for that will be determined in registration determining module are more than two of default minimum registration threshold value the The combination of one navigation tag carries out normalizing, obtains the combination of the first navigation tag after normalizing;
First is presented module, for the first navigation tag combination after normalizing module normalizing to be presented on navigation engine interface.
14. device as claimed in claim 13, which is characterized in that described first is presented module, is specifically used for:
The specify information that corresponding all business objects are combined according to the first navigation tag each of after normalizing, after determining normalizing The temperature score of each first navigation tag combination;By the first navigation tag combination after the normalizing of specified quantity according to temperature point Number is presented on from high to low sequence on navigation engine interface.
15. device as claimed in claim 9, which is characterized in that described device further includes:
Receiving unit, the input information for receiving user determine and the matched information labels of the input information;
Second navigation tag combines determination unit, the first navigation tag for being determined from the first navigation tag combination determination unit In combination, the second navigation tag will be determined as comprising being combined with the first navigation tag of the matched information labels of the input information Combination;
Second display unit, the second navigation tag combination for determining the second navigation tag combination determination unit are presented on root In the result of page searching generated according to the input information of user.
16. device as claimed in claim 15, which is characterized in that second navigation tag combines determination unit, specific to use In:
To include the input information simultaneously from the first navigation tag combination that the first navigation tag combination determination unit determines The combination of first navigation tag of matched information labels and the preference label of the predetermined user is determined as the second navigation Tag combination;Wherein, the preference label of the user is determined according to the specified operation of the user.
CN201410311241.8A 2014-07-01 2014-07-01 A kind of method and device that navigation tag is provided Active CN105335386B (en)

Priority Applications (2)

Application Number Priority Date Filing Date Title
CN201410311241.8A CN105335386B (en) 2014-07-01 2014-07-01 A kind of method and device that navigation tag is provided
HK16106712.6A HK1218793A1 (en) 2014-07-01 2016-06-12 Method for providing navigation tabs and device thereof

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN201410311241.8A CN105335386B (en) 2014-07-01 2014-07-01 A kind of method and device that navigation tag is provided

Publications (2)

Publication Number Publication Date
CN105335386A CN105335386A (en) 2016-02-17
CN105335386B true CN105335386B (en) 2018-10-16

Family

ID=55285925

Family Applications (1)

Application Number Title Priority Date Filing Date
CN201410311241.8A Active CN105335386B (en) 2014-07-01 2014-07-01 A kind of method and device that navigation tag is provided

Country Status (2)

Country Link
CN (1) CN105335386B (en)
HK (1) HK1218793A1 (en)

Families Citing this family (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN105912654A (en) * 2016-04-08 2016-08-31 南方电网科学研究院有限责任公司 Case retrieval method and system for power equipment defect faults
CN108804454B (en) * 2017-04-28 2021-12-10 华为技术有限公司 Group portrait method, group portrait device and server
CN107484038A (en) * 2017-08-22 2017-12-15 北京奇艺世纪科技有限公司 A kind of generation method of video subject, device and electronic equipment
CN108460131B (en) * 2018-03-02 2021-11-26 北京奇艺世纪科技有限公司 Classification label processing method and device
CN113051344A (en) * 2020-09-15 2021-06-29 卢霞浩 Information pushing method and information pushing system based on cloud computing and big data

Citations (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN102592223A (en) * 2011-01-18 2012-07-18 卓望数码技术(深圳)有限公司 Commodity recommending method and commodity recommending system
CN103678335A (en) * 2012-09-05 2014-03-26 阿里巴巴集团控股有限公司 Method and device for identifying commodity with labels and method for commodity navigation

Family Cites Families (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US7840548B2 (en) * 2007-12-27 2010-11-23 Yahoo! Inc. System and method for adding identity to web rank

Patent Citations (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN102592223A (en) * 2011-01-18 2012-07-18 卓望数码技术(深圳)有限公司 Commodity recommending method and commodity recommending system
CN103678335A (en) * 2012-09-05 2014-03-26 阿里巴巴集团控股有限公司 Method and device for identifying commodity with labels and method for commodity navigation

Non-Patent Citations (2)

* Cited by examiner, † Cited by third party
Title
Personalized Recommendation in Folksonomies using a Joint Probabilistic Model of;Muzaffer Ege Alper 等;《2012 11th International Conference on Machine Learning and Applications》;20130111;第368-373页 *
基于标签共现的查询扩展研究;晋松 等;《中国计算机语言学研究前沿进展(2007-2009)》;20090724;第452-457页 *

Also Published As

Publication number Publication date
CN105335386A (en) 2016-02-17
HK1218793A1 (en) 2017-03-10

Similar Documents

Publication Publication Date Title
CN103914492B (en) Query word fusion method, merchandise news dissemination method and searching method and system
CN105335386B (en) A kind of method and device that navigation tag is provided
CN109670591B (en) Neural network training method and image matching method and device
CN106055710A (en) Video-based commodity recommendation method and device
KR102072339B1 (en) Image feature data extraction and use
CN105247566B (en) Method, apparatus, and medium for selectively disclosing information based on context
CN103678335B (en) The method of method, apparatus and the commodity navigation of commodity sign label
US9727906B1 (en) Generating item clusters based on aggregated search history data
CN106202317A (en) Method of Commodity Recommendation based on video and device
JP6212013B2 (en) Product recommendation device and product recommendation method
CN106933867B (en) Image query method and device
US11195227B2 (en) Visual search, discovery and attribution method, system, and computer program product
CN109242537A (en) Advertisement placement method, device, computer equipment and storage medium
CN104615721B (en) For the method and system based on return of goods related information Recommendations
CN103136257A (en) Information providing method and device thereof
JP6511204B1 (en) INFORMATION PROCESSING SYSTEM, INFORMATION PROCESSING DEVICE, SERVER DEVICE, PROGRAM, OR METHOD
US11972466B2 (en) Computer storage media, method, and system for exploring and recommending matching products across categories
US20200358847A1 (en) Fashion item analysis based on user ensembles in online fashion community
US9799065B1 (en) Associating items based at least in part on physical location information
KR20210131198A (en) Method, apparatus and computer program for advertising recommended product
CN111815404A (en) Virtual article sharing method and device
CN105205684A (en) Recommended display method of matched products and apparatus
KR102575382B1 (en) AI-based online clothing retail system
CN103942693A (en) Method and device for recognizing variety shop and method and system for searching for shop
CN105868299A (en) Data search method and device

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