CN110517067A - Method, apparatus, system and the memory that vehicle is recommended in search box - Google Patents
Method, apparatus, system and the memory that vehicle is recommended in search box Download PDFInfo
- Publication number
- CN110517067A CN110517067A CN201910730605.9A CN201910730605A CN110517067A CN 110517067 A CN110517067 A CN 110517067A CN 201910730605 A CN201910730605 A CN 201910730605A CN 110517067 A CN110517067 A CN 110517067A
- Authority
- CN
- China
- Prior art keywords
- user
- vehicle
- search box
- portrait
- drop
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
- 238000000034 method Methods 0.000 title claims abstract description 29
- 230000015654 memory Effects 0.000 title claims abstract description 12
- 238000012216 screening Methods 0.000 abstract 1
- 230000006399 behavior Effects 0.000 description 23
- 239000011159 matrix material Substances 0.000 description 12
- 238000010586 diagram Methods 0.000 description 10
- 230000007774 longterm Effects 0.000 description 6
- 230000003542 behavioural effect Effects 0.000 description 3
- 235000014510 cooky Nutrition 0.000 description 3
- 230000006835 compression Effects 0.000 description 2
- 238000007906 compression Methods 0.000 description 2
- 238000005516 engineering process Methods 0.000 description 2
- 238000006243 chemical reaction Methods 0.000 description 1
- 235000013399 edible fruits Nutrition 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 238000011156 evaluation Methods 0.000 description 1
- 238000000605 extraction Methods 0.000 description 1
- 230000004927 fusion Effects 0.000 description 1
- 230000010354 integration Effects 0.000 description 1
- 230000002452 interceptive effect Effects 0.000 description 1
- 239000000463 material Substances 0.000 description 1
- 238000002360 preparation method Methods 0.000 description 1
- 238000003672 processing method Methods 0.000 description 1
- 238000005201 scrubbing Methods 0.000 description 1
- 210000003813 thumb Anatomy 0.000 description 1
- 230000001960 triggered effect Effects 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/90—Details of database functions independent of the retrieved data types
- G06F16/95—Retrieval from the web
- G06F16/953—Querying, e.g. by the use of web search engines
- G06F16/9535—Search customisation based on user profiles and personalisation
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q30/00—Commerce
- G06Q30/02—Marketing; Price estimation or determination; Fundraising
- G06Q30/0241—Advertisements
- G06Q30/0251—Targeted advertisements
- G06Q30/0269—Targeted advertisements based on user profile or attribute
- G06Q30/0271—Personalized advertisement
Landscapes
- Engineering & Computer Science (AREA)
- Business, Economics & Management (AREA)
- Databases & Information Systems (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- Accounting & Taxation (AREA)
- Development Economics (AREA)
- Finance (AREA)
- Strategic Management (AREA)
- General Physics & Mathematics (AREA)
- Entrepreneurship & Innovation (AREA)
- General Business, Economics & Management (AREA)
- Marketing (AREA)
- Economics (AREA)
- Game Theory and Decision Science (AREA)
- Data Mining & Analysis (AREA)
- General Engineering & Computer Science (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
Abstract
The invention discloses method, apparatus, system and memories that vehicle in search box is recommended.This method comprises: user's placement focus in the search box but in the case where not inputting information, inquire inline cache, and the first user's vehicle interest portrait inquired is shown in the drop-down list in search box, realize that vehicle is recommended, wherein inline cache is drawn a portrait comprising first user's vehicle interest；And/or in the case that user inputs information in the search box, second user vehicle interest portrait is generated according to the information of user's input, and shows the second user vehicle interest portrait of generation in the drop-down list of search box, realizes that vehicle is recommended.The present invention provides the personalized competing product recommendation items of sheet in search box drop-down list inventory, recommendation results are proposed from search result list preposition to search box drop-down list, accurately recommend screening twice by drop-down list and recommendation results, user's vehicle search intention is positioned more accurately, is promoted and the personalized search of abundant user is experienced.
Description
The application be application No. is 201810338063.6, it is entitled that " vehicle is recommended in search box method, apparatus is
The divisional application of the Chinese patent application of system and memory ".
Technical field
This application involves more particularly to a kind of search box in vehicle recommend method, apparatus, system and memory.
Background technique
It searches for and is widely present in webpage and mobile application software as the main means for obtaining the network information.With neck
The subdivision in domain, from traditional search engine at leisure to personalized search evolution.For different search service providers,
Used technical solution is different, the mode of displaying also difference.The application is illustrated so that vehicle is searched for as an example.
Fig. 1 shows the schematic diagram using Baidu search Audi vehicle brand, and Fig. 2 and Fig. 4 show easy vehicle net and automobile
Family's homepage search box Drop Down Menu Choices, Fig. 3 and Fig. 5 show vehicle and easily search and the search homepage drop-down menu choosing of the family of automobile
.
In the prior art, search box is based on doing the matching of prefix maximum based on search key, uses Google more
The true vehicle search intention of searchers, recommendation and advertisement are not recognized accurately for Suggest and AutoComplete technology
The efficiency of delivery is not high.
It is as follows that Fig. 6 shows its front and back end interactive step in the Technical Architecture of Google Suggest:
User inputs vehicle model information, such as " Chrysler " in search box, and browser will request asynchronous hair by AJAX
It send to Web server, while the frequency that request is submitted in continuous input can be controlled, request is submitted in delay；
After server receives request, memcached cache cluster can be first forwarded requests to, uses compression search tree
(trie) either with or without, if so, returning result to server, being passed through by server with matched prefix in judgement caching
AJAX returns to browser front end；
It, can be from obtaining data in RDBMS and being saved in memcached, together if there is no corresponding data in caching
When data are returned into browser front end by server.
If rear end can be more complicated using the AutoComplete in universal search engine, but if only vehicle
The auto-complete of information, it is because data volume is very small, then slow with memcached completely without using cache cluster
It deposits server or directly also can satisfy demand using xml document.
In the prior art, these technologies used in universal search engine are often not towards specific area, field
Exclusive feature is not strong, and the particularity that not prominent vehicle is recommended is not optimized for vehicle recommendation；At present for Google
The application of Suggest is not bound with any searchers's characteristic information, so can only accomplish " thousand people one side ", and can not accomplish " thousand
Thousand face of people "；The prior art is only the vehicle model information inputted in search box have been done completion either only to make limited similar vehicle
Type is recommended, and lacks various competing product vehicles and heat searches the recommendation of vehicle；In the prior art, the advertisement of drop-down list box is not highlighted
Bit attribute, it is often non-advertisement attributes information that information, which is presented, needless to say is combined with accurate advertisement；In addition, the prior art
It is vehicle text information in middle drop-down list box, lacks the lively media formats such as intuitive vehicle figure, Dynamic Graph.
Summary of the invention
The embodiment of the present application provides method, apparatus, system and the memory that vehicle is precisely recommended in a kind of search box, to
It solves the problems, such as to recommend vehicle according to user property and behavior.
This application provides the methods that vehicle in a kind of search box is recommended, comprising:
User's placement focus in the search box but in the case where not inputting information, inquires inline cache, and searching for
The first user's vehicle interest portrait inquired is shown in the drop-down list of frame, realizes that vehicle is recommended, wherein inline cache includes
First user's vehicle interest portrait；And/or
In the case that user inputs information in the search box, second user vehicle interest is generated according to the information of user's input
Portrait, and show that the second user vehicle interest of generation is drawn a portrait in the drop-down list of search box, realize that vehicle is recommended.
Preferably, this method further include:
In the case that the option that user is hovered in search box drop-down list is advertisement item, advertisement is shown；
In the case that the option that user clicks in search box drop-down list is recommendation items, then it is corresponding to jump to recommendation items
Webpage；In the case that the option that user clicks in search box drop-down list is advertisement, then advertisement landing page is jumped to.
Preferably, this method further include:
User interest portrait is generated according to user journal.
Preferably, the first user vehicle interest is drawn a portrait according to following steps and is stored in inline cache:
Receive the user exposure of drop-down list item and click behavior in the search box；
User interest portrait is pulled, the feature vector of user is generated；
First user's vehicle interest portrait is generated according to the feature vector of user and feature-article correlation matrix；
First user's vehicle interest is drawn a portrait and is stored in inline cache.
Preferably, first user's vehicle interest is generated according to the feature vector of user and feature-article correlation matrix to draw
As including:
Initial recommendation item lists are obtained according to the feature vector of user and feature-article correlation matrix；
The vehicle for meeting relevance threshold and competing product are obtained from initial recommendation item lists according to user interest portrait
Vehicle；
The vehicle and competing product vehicle that meet relevance threshold are ranked up, first user's vehicle interest portrait is generated.
Preferably, the vehicle for meeting relevance threshold and competing product vehicle are the vehicle in candidate vehicle set.
Preferably, this method further include:
In the case where not inquiring first user's vehicle interest portrait in inline cache, first in online database is inquired
User's vehicle interest portrait, if it is present showing the first user's vehicle interest inquired in the drop-down list of search box
Portrait；
In the case where not inquiring first user's vehicle interest portrait in online database, then inquired at offline user center
First user's vehicle interest portrait then shows the first user's vehicle interest portrait inquired in the drop-down list of search box；
In the case where offline user center does not inquire first user's vehicle interest portrait, in the drop-down list of search box
Middle displaying heat searches vehicle.
Preferably, generating second user vehicle interest portrait according to the information of user's input includes:
According to the vehicle of information identification user's input of user's input；
The corresponding recommendation list of user and/or advertising listing are obtained, and utilizes the vehicle of user's input, the offline vehicle of user
Content mixing in interest portrait and the competing product table of vehicle generates second user vehicle interest portrait.
First user's vehicle interest portrait or second user vehicle interest portrait are the purchase vehicle stage portrait for having purchased automobile-used family,
For providing same brand vehicle to have purchased automobile-used family.
This application provides the devices that vehicle in a kind of search box is recommended, comprising:
First recommending module, in the case where for user's placement focus in the search box but not inputting information, inquiry exists
Line caching, and the first user's vehicle interest portrait inquired is shown in the drop-down list in search box, realize that vehicle is recommended,
Wherein inline cache is drawn a portrait comprising first user's vehicle interest；And/or
Second recommending module, in the case where inputting information in the search box for user, the information according to user's input is raw
It draws a portrait at second user vehicle interest, and shows the second user vehicle interest portrait of generation in the drop-down list of search box,
Realize that vehicle is recommended.
Preferably, which further includes advertisement module, and the option to hover in search box drop-down list for user is wide
In the case where accusing item, advertisement is shown；
Preferably, which further includes jump module, and the option clicked in search box drop-down list for user is to push away
In the case where recommending item, then the corresponding webpage of recommendation items is jumped to；The option that user clicks in search box drop-down list is advertisement
In the case where, then jump to advertisement landing page.
Preferably, which further includes user interest portrait generation module, for generating user interest according to user journal
Portrait.
Preferably, the first recommending module is also used to according to the reception user exposure of drop-down list item and click in the search box
Behavior；User interest portrait is pulled, the feature vector of user is generated；It is related according to the feature vector of user and feature-article
Matrix generates first user's vehicle interest portrait；First user's vehicle interest is drawn a portrait and is stored in inline cache.
Generating first user's vehicle interest portrait according to the feature vector of user and feature-article correlation matrix includes:
Initial recommendation item lists are obtained according to the feature vector of user and feature-article correlation matrix；According to user's Long-term Interest
The acquisition from initial recommendation item lists of drawing a portrait meets the vehicle and competing product vehicle of relevance threshold；To meeting relevance threshold
Vehicle and competing product vehicle be ranked up, generate first user's vehicle interest portrait.
Preferably, the first recommending module is also used to not inquire the feelings of first user's vehicle interest portrait in inline cache
Under condition, first user's vehicle interest portrait in online database is inquired, if it is present opening up in the drop-down list of search box
Show the first user's vehicle interest portrait inquired；Not the case where first user's vehicle interest portrait is not inquired in online database
Under, then first user's vehicle interest portrait is inquired at offline user center, then shows and inquire in the drop-down list of search box
First user's vehicle interest portrait；In the case where offline user center does not inquire first user's vehicle interest portrait, In
Show that heat searches vehicle in the drop-down list of search box.
Preferably, the second recommending module, the vehicle for the information identification user's input for being also used to be inputted according to user；It obtains and uses
The corresponding recommendation list in family and/or advertising listing, and utilize the vehicle of user's input, the offline vehicle interest portrait of user and vehicle
Content mixing in the competing product table of type generates second user vehicle interest portrait.
This application provides the system that vehicle in a kind of search box is precisely recommended, memory and processor；
Memory, for storing program；
Processor realizes the method for executing program.
This application provides a kind of memories, are stored thereon with program, and described program is used for real when being executed by processor
The existing method.
The application provides the personalized competing product recommendation items of sheet in search box drop-down list inventory, by recommendation results from search
It proposes preposition to search box drop-down list in results list, accurately recommends to screen twice by drop-down list and recommendation results,
User's vehicle search intention is positioned more accurately, is promoted and the personalized search of abundant user is experienced.The application does not have originally
Increase multiple vehicle advertisement positions in the search box drop-down list for having vehicle advertisement position, effectively expands platform in the advertisement of mobile terminal
Bit quantity provides a possibility that new for the advertisement win receipts growth of mobile platform.
Detailed description of the invention
The drawings described herein are used to provide a further understanding of the present application, constitutes part of this application, this Shen
Illustrative embodiments and their description please are not constituted an undue limitation on the present application for explaining the application.In the accompanying drawings:
Fig. 1 is the schematic diagram using Baidu search Audi vehicle brand；
Fig. 2 is the Drop Down Menu Choices schematic diagram of easy vehicle net homepage search box；
Fig. 3 is that vehicle easily searches for homepage Drop Down Menu Choices schematic diagram；
Fig. 4 is the Drop Down Menu Choices schematic diagram of family's homepage search box of automobile；
Fig. 5 is that homepage Drop Down Menu Choices schematic diagram is searched for by the family of automobile；
Fig. 6 is Google Suggest technical solution schematic diagram；
Fig. 7 is that the user interest portrait provided by the invention that provides prepares flow diagram；
Fig. 8 is personalized recommendation provided by the invention/advertising flow schematic diagram；
Fig. 9 is the method detail flowchart that vehicle is precisely recommended in search box provided by the invention；
Figure 10 is the schematic device that vehicle is precisely recommended in search box provided by the invention.
Specific embodiment
To keep the purposes, technical schemes and advantages of the application clearer, below in conjunction with the application specific embodiment and
Technical scheme is clearly and completely described in corresponding attached drawing.Obviously, described embodiment is only the application one
Section Example, instead of all the embodiments.Based on the embodiment in the application, those of ordinary skill in the art are not doing
Every other embodiment obtained under the premise of creative work out, shall fall in the protection scope of this application.
The present invention goes to realize the personalization that search box vehicle is recommended using personalized recommendation algorithm mature at present, will
The behavior of searchers and attribute and vehicle binding, which are got up, to be implemented to recommend, and is quickly found out and its search intention phase with the person of assisting search
The target vehicle matched, the realization of target effect are primarily to see clicking rate of the searchers on Drop Down Menu Choices and are promoted.
In order to achieve the object of the present invention, the present invention is mainly by the way that user interest portrait prepares, personalized recommendation/advertisement is drawn
It holds up, three parts such as recommendation/advertisement trigger mechanism are realized.
Fig. 7 is that the present invention provides user interest portrait preparation flow diagram.User interest portrait mainly passes through user
Log is analyzed to obtain user interest portrait.User interest portrait includes being generated according to behaviors such as user's history access, clicks
User's Long-term Interest portrait and the user of the behaviors generation interest portrait in short-term such as access, click immediately according to user, in short-term
Interest portrait generally requires to efficiently produce in a manner of stream process, and Long-term Interest portrait is often offline in a manner of batch processing
The next day generate.Website or application software, website of the user where accessing search box for the first time can be under the kinds of user browser end
Cookie, application software can record APP by imei/idfa and be initially opened log.Server according to user browse, access and
Buying behavior generates offline user journal.Customer center (including user behavior library and user property library) is with cookie, imei/
Idfa is unit, analyzes user journal, is formed about the ascribed characteristics of population of user, historical behavior, interest content, preference tendency, purchase
Various dimensions user interest portrait including the vehicle stage, wherein most importantly vehicle interest is drawn a portrait and the purchase vehicle stage to the application
Portrait, vehicle interest portrait are the behavior integrations such as to access, click, placing an order according to user to judge that user weighs the interest of certain vehicle
Weight, for purchase vehicle stage portrait according to the derived data by all kinds of means including data under line, comprehensive descision user is to be in interest, meaning
To, purchase, which kind of for having purchased vehicle in stage.Purchase vehicle stage portrait can be precisely captured at user with help system and Che Xiangguan
Which kind of in stage, recommend to provide different vehicles using different Generalization bounds when recommending drop-down list box to recommend, example
Competing product vehicle as much as possible such as is provided in the interest stage to select for user, and more accurate closing in is then provided in the purchase stage
Same brand vehicle it is selective, provide vehicle of purchasing on a barter basis to have purchased automobile-used family and recommend to select in combobox for user.In addition, also
Access and the hints data of all users are counted, refreshes the competing product table of vehicle and heat searches vehicle table.User behavior library includes data item
Having user to browse vehicle channel, submit clue, behaviors, the user property library such as thumb up, comment on, collecting includes user's gender, age
Section, educational background, marital status, operation class (of an amplifying stage), permanent residence, Income Classes, hobby etc..
As shown in fig. 7, the process for generating user interest portrait includes:
Step 705, determine that user uses browser or application software APP；
Step 710, if using APP, record is initially opened log；
Step 715, determine whether user passes through browser and access for the first time；
Step 720, the case where user accesses for the first time, cookie under browser kind；
Step 725, additional offline user journal；
Step 730, user interest portrait is generated；
Step 735, refresh the competing product table of vehicle；
Step 740, refresh heat and search vehicle table.
Fig. 8 is personalized recommendation provided by the invention/advertising flow figure.By the operation of user, such as user is by cursor
Drop-down list recommendation/advertisement position item exposure and the click behavior of user, pull user from customer center when putting in the search box
Interest portrait, and generate the feature vector of user.Drop-down list is unfolded to execute when recommending, when do not need to use vehicle interest as
Stereotactic conditions are for example integrally recommended to the user of certain districts and cities to do region orientation, at this moment if only using non-behavioural characteristic, namely
Using user property feature, there is no need to vehicle behavior extraction and behavioural characteristic conversion module, the output of the module is exactly user
Feature vector.Using the feature vector and pre-stored feature-article correlation matrix of user, by the feature vector of user
It is converted into initial recommendation item lists.Preferably, obvious problematic vehicle can be filtered out to recommending item lists to pre-process
Type, for example, the vehicle etc. that the vehicle of title exception, interest weight are extremely low.The user's Long-term Interest generated according to off-line data
Portrait obtains the vehicle and competing product vehicle that the degree of correlation meets threshold value, is then clicked according to trained order models prediction logical
Rate CTR is crossed, selects top n as consequently recommended as a result, obtaining vehicle interest portrait.Vehicle interest portrait can use " (< vehicle
Type title, vehicle quotation section>|<vehicle title, vehicle page URL, vehicle quotation section>|<vehicle title, advertisement type, extensively
Accuse material, advertisement land page URL, vehicle offer section >) * " form, and can periodically scrubbing brush enter inline cache.
As shown in figure 8, personalized recommendation provided by the invention/advertising flow figure includes:
Step 805, the behavioural informations such as relevant to vehicle are extracted from user behavior library, including behavior time of origin, classification,
Number etc.；
Step 810, user behavior characteristics are converted, i.e., the behaviors such as user are browsed, collected, thumbed up, buy and convert and be characterized
Namely the expression way of label；
Step 815, user characteristics vector is generated, feature vector is made of feature and feature weight, in general, when generation
Between the more behavior of closer behavior, number, more active behavior weight can be higher；
Step 820, user characteristics-vehicle correlation matrix, phase are generated according to user characteristics vector and candidate vehicle set
Close matrix record is feature-vehicle-weight three corresponding relationship, and feature may be age, gender, the color liked, purchase
Power etc. is bought, purpose existing for candidate vehicle is to guarantee that recommendation results only include the vehicle in candidate vehicle set, such as only wish
Hope provide certain competing product vehicles recommend user or it is three months nearest in the new model that lists；
Step 825, it forms initial vehicle and recommends inventory；
Step 830, processing is filtered to initial recommendation list, for example filtered out previously it is recommended that crosses is certain competing
Product vehicle, or exclude the domestic car of certain valence low-qualitys time；
Step 835, it is based on offline sort algorithm, generates user interest vehicle sorted lists, for example can analyze user elder generation
The preceding click behavior in drop-down menu recommended area, prediction user is interested in which recommendation results, this is fusion user feedback
A kind of sequence processing method；
Step 840, it introduces interpretable policing rule and strengthens ranking results, however, it would be possible to comprehensive vehicle novelty, vehicle
A variety of ordering strategies such as sales volume, acceptance of the users evaluation, time change diversity；
Step 845, consequently recommended result is formed.
Fig. 9 is the method detail flowchart that vehicle is precisely recommended in search box provided by the invention.User by focus (such as
Cursor) it puts in the search box, but when no input information, server first inquires inline cache cluster, passes through compression Trie tree and claps
It is matched, if there is matching result is then directly shown in drop-down list box.If do not inquired in inline cache cluster
It arrives, then can call online database, continue to match, if there is matched as a result, then being shown in drop-down list box.Such as
Matching result is also not present in fruit in online database, then is matched in customer center, if there is matching result, then under
It draws in list box and shows.If customer center also without matching result, shows that heat searches vehicle in drop-down list box.Work as user
When inputting information in search box, the vehicle for including can be identified according to the information of input, browser or APP believe vehicle
Breath is sent to server.Corresponding recommendation/the advertising listing of the server calls user, and it is defeated using real-time recommendation algorithm to share family
Enter vehicle, user's Long-term Interest portrait, content mixing in the competing product table of vehicle, the recommendation list of the user, directly return will be directed to
It shows in drop-down list box.Specifically, by Kafka+storm real time parsing log, according to user's acts and efforts for expediency and interest
Feature is analyzed in streaming fashion.
In the search result of return, if it is advertisement item that user's mouse-over corresponds to option in drop-down list box,
The different types such as video ads, expansible advertisement, animated advertising can be triggered according to advertisement type.If user clicks drop-down column
Respective selection is recommendation items in bezel, cluster, then can jump to the corresponding URL of vehicle page, if user clicks in drop-down list box accordingly
Option is advertisement item, then can jump to advertisement landing page.
As shown in figure 9, the method that vehicle is precisely recommended in search box specifically includes:
Step 905, search box obtains input focus；
Step 910, using search box input content as search index inline cache；
Step 915, it is determined whether there are prefix matching as a result, if it is execution step 930, it is no to then follow the steps 920；
Step 920, determine online database with the presence or absence of matching result, if it is execution step 930, otherwise execute step
Rapid 925；
Step 925, determine customer center with the presence or absence of matching result, if it is execution step 930, it is no to then follow the steps
935；
Step 930, recommendation results are shown in combobox；
Step 935, show that heat searches vehicle in combobox；
Step 940, user inputs information in search box, and backstage carries out query rewrite；
Step 945, the vehicle in identification input information, can be by front end, such as browser or APP, can also be by rear
End, for example, server come realize input information in vehicle；
Step 950, user's vehicle interest is called to draw a portrait offline from customer center；
Step 955, the corresponding competing product vehicle of identification vehicle is obtained, competing product vehicle numerical procedure is related to focus on conjunction, clue weight
The many kinds such as conjunction, the coincidence of public sentiment volume；
Step 960, search result is calculated in real time.Real-time computing engines can request offline user center to call the user corresponding
Recommendation/advertising listing, and using real-time recommendation algorithm combination user input the offline vehicle interest of vehicle, user draw a portrait, vehicle
Content mixing in competing product table.
By above-mentioned process, may be implemented to be drawn a portrait according to user interest more accurately shows recommendation results to user.
This application provides the devices that vehicle in a kind of search box is recommended, as shown in Figure 10, comprising:
First recommending module, in the case where for user's placement focus in the search box but not inputting information, inquiry exists
Line caching, and the first user's vehicle interest portrait inquired is shown in the drop-down list in search box, realize that vehicle is recommended,
Wherein inline cache is drawn a portrait comprising first user's vehicle interest；And/or
Second recommending module, in the case where inputting information in the search box for user, the information according to user's input is raw
It draws a portrait at second user vehicle interest, and shows the second user vehicle interest portrait of generation in the drop-down list of search box,
Realize that vehicle is recommended
Preferably, which further includes advertisement module, and the option to hover in search box drop-down list for user is wide
In the case where accusing item, advertisement is shown；
Preferably, which further includes jump module, and the option clicked in search box drop-down list for user is to push away
In the case where recommending item, then the corresponding webpage of recommendation items is jumped to；The option that user clicks in search box drop-down list is advertisement
In the case where, then jump to advertisement landing page.
Preferably, which further includes user interest portrait generation module, for generating user interest according to user journal
Portrait.
Preferably, the first recommending module is also used to according to the reception user exposure of drop-down list item and click in the search box
Behavior；User interest portrait is pulled, the feature vector of user is generated；It is related according to the feature vector of user and feature-article
Matrix generates first user's vehicle interest portrait；First user's vehicle interest is drawn a portrait and is stored in inline cache.
Generating first user's vehicle interest portrait according to the feature vector of user and feature-article correlation matrix includes:
Initial recommendation item lists are obtained according to the feature vector of user and feature-article correlation matrix；According to user's Long-term Interest
The acquisition from initial recommendation item lists of drawing a portrait meets the vehicle and competing product vehicle of relevance threshold；To meeting relevance threshold
Vehicle and competing product vehicle be ranked up, generate first user's vehicle interest portrait.
Preferably, the first recommending module is also used to not inquire the feelings of first user's vehicle interest portrait in inline cache
Under condition, first user's vehicle interest portrait in online database is inquired, if it is present opening up in the drop-down list of search box
Show the first user's vehicle interest portrait inquired；Not the case where first user's vehicle interest portrait is not inquired in online database
Under, then first user's vehicle interest portrait is inquired at offline user center, then shows and inquire in the drop-down list of search box
First user's vehicle interest portrait；In the case where offline user center does not inquire first user's vehicle interest portrait, In
Show that heat searches vehicle in the drop-down list of search box.
Preferably, the second recommending module, the vehicle for the information identification user's input for being also used to be inputted according to user；It obtains and uses
The corresponding recommendation list in family and/or advertising listing, and utilize the vehicle of user's input, the offline vehicle interest portrait of user and vehicle
Content mixing in the competing product table of type generates second user vehicle interest portrait.
It is described the prefered embodiments of the present invention in detail above in conjunction with attached drawing, still, the present invention is not limited to above-mentioned realities
The detail in mode is applied, within the scope of the technical concept of the present invention, a variety of letters can be carried out to technical solution of the present invention
Monotropic type, these simple variants all belong to the scope of protection of the present invention.
It is further to note that specific technical features described in the above specific embodiments, in not lance
In the case where shield, it can be combined in any appropriate way.In order to avoid unnecessary repetition, the present invention to it is various can
No further explanation will be given for the combination of energy.
In addition, various embodiments of the present invention can be combined randomly, as long as it is without prejudice to originally
The thought of invention, it should also be regarded as the disclosure of the present invention.
Claims (10)
1. a kind of method that vehicle is recommended in search box characterized by comprising
In the case that user inputs information in the search box, second user vehicle interest is generated according to the information of user's input and is drawn
Picture, and show that the second user vehicle interest of generation is drawn a portrait in the drop-down list of search box, realize that vehicle is recommended.
2. the method according to claim 1, wherein this method further include:
In the case that the option that user is hovered in search box drop-down list is advertisement item, advertisement is shown；
In the case that the option that user clicks in search box drop-down list is recommendation items, then the corresponding net of recommendation items is jumped to
Page；In the case that the option that user clicks in search box drop-down list is advertisement, then advertisement landing page is jumped to.
3. the method according to claim 1, wherein it is emerging to generate second user vehicle according to the information of user's input
Interest is drawn a portrait
According to the vehicle of information identification user's input of user's input；
The corresponding recommendation list of user and/or advertising listing are obtained, and utilizes the vehicle of user's input, the offline vehicle interest of user
Content mixing in portrait and the competing product table of vehicle generates second user vehicle interest portrait.
4. method according to claim 1 to 3, which is characterized in that second user vehicle interest portrait is to have purchased
The purchase vehicle stage at automobile-used family draws a portrait, for providing same brand vehicle to have purchased automobile-used family.
5. the device that vehicle is recommended in a kind of search box characterized by comprising
Second recommending module in the case where inputting information in the search box for user, generates the according to the information of user's input
Two user's vehicle interest portrait, and show that the second user vehicle interest of generation is drawn a portrait in the drop-down list of search box, it realizes
Vehicle is recommended.
6. device according to claim 5, which is characterized in that the device further includes advertisement module, is being searched for for user
In the case that the option to hover in frame drop-down list is advertisement item, advertisement is shown.
7. device according to claim 5, which is characterized in that the device further includes jump module, is being searched for for user
In the case that the option clicked in frame drop-down list is recommendation items, then the corresponding webpage of recommendation items is jumped to；User is in search box
In the case that the option clicked in drop-down list is advertisement, then advertisement landing page is jumped to.
8. device according to claim 5, which is characterized in that the second recommending module is also used to the letter inputted according to user
The vehicle of breath identification user's input；The corresponding recommendation list of user and/or advertising listing are obtained, and utilizes the vehicle of user's input
Content mixing in type, the offline vehicle interest portrait of user and the competing product table of vehicle, generates second user vehicle interest portrait.
9. the system that vehicle is precisely recommended in a kind of search box, which is characterized in that memory and processor；
Memory, for storing program；
Processor, for executing program to realize method according to any of claims 1-4.
10. a kind of memory, is stored thereon with program, which is characterized in that described program is used for the realization when being executed by processor
Method according to any of claims 1-4.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910730605.9A CN110517067A (en) | 2018-04-16 | 2018-04-16 | Method, apparatus, system and the memory that vehicle is recommended in search box |
Applications Claiming Priority (2)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910730605.9A CN110517067A (en) | 2018-04-16 | 2018-04-16 | Method, apparatus, system and the memory that vehicle is recommended in search box |
| CN201810338063.6A CN108537596B (en) | 2018-04-16 | 2018-04-16 | Method, device and system for recommending vehicle type in search box and memory |
Related Parent Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201810338063.6A Division CN108537596B (en) | 2018-04-16 | 2018-04-16 | Method, device and system for recommending vehicle type in search box and memory |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN110517067A true CN110517067A (en) | 2019-11-29 |
Family
ID=63480496
Family Applications (2)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201810338063.6A Active CN108537596B (en) | 2018-04-16 | 2018-04-16 | Method, device and system for recommending vehicle type in search box and memory |
| CN201910730605.9A Pending CN110517067A (en) | 2018-04-16 | 2018-04-16 | Method, apparatus, system and the memory that vehicle is recommended in search box |
Family Applications Before (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201810338063.6A Active CN108537596B (en) | 2018-04-16 | 2018-04-16 | Method, device and system for recommending vehicle type in search box and memory |
Country Status (1)
| Country | Link |
|---|---|
| CN (2) | CN108537596B (en) |
Families Citing this family (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110263066B (en) * | 2019-06-13 | 2022-11-29 | 优信拍（北京）信息科技有限公司 | Vehicle screening method and device |
| CN110807049A (en) * | 2019-10-23 | 2020-02-18 | 上海能塔智能科技有限公司 | Test driving vehicle type recommendation method and device and computer readable storage medium |
| CN110990692A (en) * | 2019-11-13 | 2020-04-10 | 上海易点时空网络有限公司 | Data processing method and device based on portrait analysis |
| CN113378034A (en) * | 2020-03-09 | 2021-09-10 | 腾讯科技（北京）有限公司 | Interest portrait expansion method, interest portrait expansion device and storage medium |
| CN111861666A (en) * | 2020-07-21 | 2020-10-30 | 上海仙豆智能机器人有限公司 | Vehicle information interaction method and device |
| CN112199465B (en) * | 2020-09-01 | 2024-02-06 | 北京达佳互联信息技术有限公司 | Information searching method and device and electronic equipment |
| CN112148984B (en) * | 2020-09-30 | 2023-11-10 | 微梦创科网络科技（中国）有限公司 | Method and system for capturing instant interests of user |
| CN117349508A (en) * | 2022-06-30 | 2024-01-05 | 华为技术有限公司 | Searching method, searching system and related device |
Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103092923A (en) * | 2011-12-29 | 2013-05-08 | 微软公司 | Menu-based advertisement of search engine |
| CN107038213A (en) * | 2017-02-28 | 2017-08-11 | 华为技术有限公司 | A kind of method and device of video recommendations |
Family Cites Families (15)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102663627A (en) * | 2012-04-26 | 2012-09-12 | 焦点科技股份有限公司 | Personalized recommendation method |
| CN103631803A (en) * | 2012-08-23 | 2014-03-12 | 百度国际科技（深圳）有限公司 | Method, device and server for advertisement orientation based on input behaviors |
| JP6246657B2 (en) * | 2013-05-20 | 2017-12-13 | パナソニック インテレクチュアル プロパティ コーポレーション オブ アメリカＰａｎａｓｏｎｉｃ Ｉｎｔｅｌｌｅｃｔｕａｌ Ｐｒｏｐｅｒｔｙ Ｃｏｒｐｏｒａｔｉｏｎ ｏｆ Ａｍｅｒｉｃａ | Information providing method and information providing apparatus |
| US20150242474A1 (en) * | 2014-02-27 | 2015-08-27 | Microsoft Corporation | Inline and context aware query box |
| CN103942279B (en) * | 2014-04-01 | 2018-07-10 | 百度(中国)有限公司 | Search result shows method and apparatus |
| CN105869001A (en) * | 2015-01-19 | 2016-08-17 | 苏宁云商集团股份有限公司 | Customized product recommendation guiding method and system |
| CN104899315A (en) * | 2015-06-17 | 2015-09-09 | 百度在线网络技术（北京）有限公司 | Method and device for pushing user information |
| CN105046531A (en) * | 2015-08-07 | 2015-11-11 | 百度在线网络技术（北京）有限公司 | Online delivery information display method and apparatus |
| CN105760508A (en) * | 2016-02-23 | 2016-07-13 | 北京搜狗科技发展有限公司 | Information push method and device and electronic equipment |
| CN106296368A (en) * | 2016-08-19 | 2017-01-04 | 北京好车轰轰电子商务有限公司 | A kind of vehicle commending system and method |
| CN107818103B (en) * | 2016-09-12 | 2023-01-24 | 阿里巴巴集团控股有限公司 | Method and device for providing data object information |
| CN107545457B (en) * | 2017-02-13 | 2020-01-21 | 北京车慧科技有限公司 | Automobile racing product type determination method and device |
| CN107180078A (en) * | 2017-04-21 | 2017-09-19 | 河海大学 | A kind of method for vertical search based on user profile learning |
| CN107506367B (en) * | 2017-07-03 | 2021-12-24 | 创新先进技术有限公司 | Method and device for determining application display content and server |
| CN107526828B (en) * | 2017-09-01 | 2020-10-16 | 阿里巴巴（中国）有限公司 | Page information recommendation method and device |
-
2018
- 2018-04-16 CN CN201810338063.6A patent/CN108537596B/en active Active
- 2018-04-16 CN CN201910730605.9A patent/CN110517067A/en active Pending
Patent Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103092923A (en) * | 2011-12-29 | 2013-05-08 | 微软公司 | Menu-based advertisement of search engine |
| CN107038213A (en) * | 2017-02-28 | 2017-08-11 | 华为技术有限公司 | A kind of method and device of video recommendations |
Also Published As
| Publication number | Publication date |
|---|---|
| CN108537596A (en) | 2018-09-14 |
| CN108537596B (en) | 2020-02-21 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN110517067A (en) | Method, apparatus, system and the memory that vehicle is recommended in search box | |
| Gunawardana et al. | Evaluating recommender systems | |
| TWI544352B (en) | System and method to facilitate matching of content to advertising information in a network | |
| US10198503B2 (en) | System and method for performing a semantic operation on a digital social network | |
| JP5878245B2 (en) | Intelligent navigation of category system | |
| US7680786B2 (en) | Optimization of targeted advertisements based on user profile information | |
| CN104679771B (en) | A kind of individuation data searching method and device | |
| US10198520B2 (en) | Search with more like this refinements | |
| US20220366456A1 (en) | System and method for purchasing advertisements associated with words and phrases | |
| US20150379571A1 (en) | Systems and methods for search retargeting using directed distributed query word representations | |
| Cheng et al. | Video ecommerce++: Toward large scale online video advertising | |
| WO2015179756A1 (en) | Systems and methods for customizing search results and recommendations | |
| JP2016146213A (en) | System and method for performing semantic operation on digital social network | |
| US20150254714A1 (en) | Systems and methods for keyword suggestion | |
| KR20130016339A (en) | Infinite browse | |
| CN102456044B (en) | The method and system of pushed information | |
| US20080201219A1 (en) | Query classification and selection of associated advertising information | |
| US20100318427A1 (en) | Enhancing database management by search, personal search, advertising, and databases analysis efficiently using core-set implementations | |
| JP4939637B2 (en) | Information providing apparatus, information providing method, program, and information recording medium | |
| US20140214548A1 (en) | User Profiling Using Submitted Review Content | |
| JP2019003610A (en) | Extraction device, extraction method, and extraction program | |
| US7814109B2 (en) | Automatic categorization of network events | |
| US20090077093A1 (en) | Feature Discretization and Cardinality Reduction Using Collaborative Filtering Techniques | |
| TW201801006A (en) | Personalized online marketing recommendation method capable of predicting the trend of preference according to factors such as browsing date, browsing time, and website visited, to thereby provide marketing materials | |
| JP6311052B1 (en) | Extraction apparatus, extraction method, and extraction program |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20191129 |