CN107209897A - User data is handled and analyzes to determine keyword quality - Google Patents
User data is handled and analyzes to determine keyword quality Download PDFInfo
- Publication number
- CN107209897A CN107209897A CN201580073843.7A CN201580073843A CN107209897A CN 107209897 A CN107209897 A CN 107209897A CN 201580073843 A CN201580073843 A CN 201580073843A CN 107209897 A CN107209897 A CN 107209897A
- Authority
- CN
- China
- Prior art keywords
- article
- user
- keyword
- website
- packet
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- 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
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q30/00—Commerce
- G06Q30/06—Buying, selling or leasing transactions
- G06Q30/0601—Electronic shopping [e-shopping]
Landscapes
- Business, Economics & Management (AREA)
- Accounting & Taxation (AREA)
- Finance (AREA)
- Strategic Management (AREA)
- Engineering & Computer Science (AREA)
- Development Economics (AREA)
- General Physics & Mathematics (AREA)
- Economics (AREA)
- Marketing (AREA)
- Physics & Mathematics (AREA)
- General Business, Economics & Management (AREA)
- Theoretical Computer Science (AREA)
- Game Theory and Decision Science (AREA)
- Entrepreneurship & Innovation (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
Abstract
The method and apparatus for attracting potential customer to access website.On the one hand, user's participation activity of polymerization site visitor.User's participation activity of polymerization is compared with least one preassigned.If activity meets one or more standards, activity can be parsed for recognizing potential keyword from the search inquiry for producing Query Result.The potential keyword value score of each potential keyword can be calculated.Potential keyword with potential keyword value score similar enough is grouped to provide packet value score, the information that the packet then is worth into score and the bid costs on those keywords is compared.Then, the comparison of score is worth in response to packet, the decision bidded to the potential keyword in packet is made whether.
Description
Cross-reference to related applications
This application claims being submitted on January 20th, 2015, entitled " processing and analysis user data are to determine keyword matter
Amount (Processing And Analysis Of User Data To Determine Keyword Quality) ", number it is
The priority and rights and interests of 14/601,114 U. S utility patent application, the U. S utility patent application are incorporated by this with it
Text.
Technical field
The each side of the disclosure is related to is attracted to website by potential customer, and joins more particularly to using the user of website
Keyword is recognized with related data to attract more customers.
Background technology
Profit model, which concentrates on the company that exists online many energy are placed on into the more customers of acquisition, carrys out company's site's purchase
Thing.In Internet world, browse more much more than buying.For major company, actually done shopping on whom and
It is generally very few so that nonsensical so as to the data of shopping how buyer is attracted to online trade company.
In theory, enough advertising expenditures are spent online, for example, keyword are bidded, and can produce significant sites conversion rate
(that is, the customer of actual shopping after accessing the website).But, spend advertising expenditure to be apparently not a kind of profit without limitation
Business model.
The sites conversion rate of improvement company can be searched out without spending the method for the too high amount of money have in online advertisement
.
Brief description of the drawings
Fig. 1 shows each side of the general frame of the system according to one embodiment.
Fig. 2 is the flow chart for each side for showing the technical operation described herein according to one embodiment.
Fig. 3 shows each side of user's participation activity according to various embodiments.
Fig. 4 shows each side of hardware embodiment, its can be used for Fig. 1 described in one or more servers or its
In its hardware.
Embodiment
User Activity in system, method and medium monitoring and inspection online marketplace.Participated in response to specific user
Standard, makes decision to buy the particular keywords on network search engines on the one hand, optimizes and counts as search engine
The part drawn.
In the following description, term " website " and " network presence " can be used for different places.Two terms are not intended to
Purport is confined to specific webpage.On the contrary, these term intentions represent any webpage that company may have.Thus, for example, in
In line market, when seller publication product be used for sell when, either buyer purchase product when or product the confession time buying it is expired
When, individual other webpage may occur in which and disappear.The citation intention of company " website " or " network presence " is covered herein all this
The webpage of sample, the webpage is related to the basic URL of company in some way.
Although the major part of this paper specifications is provided in the context of online marketplace, the techniques described herein and adjoint
Structure it is not limited, and apply to its operator wish drive flow any network exist.
Fig. 1 is network, is shown according to one embodiment with the client computer-clothes being configured to by network exchange data
The network system 100 of business device framework.Data can relate to the various functions related to network system 100 and its user (for example, online
Shopping) and each side (such as Content Management).Although as shown herein is client-server framework, other embodiments can
Including other network architectures, such as peer-to-peer or distributed network environment.
For example the data interchange platform 102 including market applied procedure 120 can by network 104 (for example, internet) to
One or more client computer provide the function of server side.One or more client computer may include using network system 100 and
More specifically using market applied procedure with will pass through network 104 exchange data user.The utilization may include that data will be sent
(communication) data are received to network system 100, from network system 100, and handle relevant content and the user of network system 100
Data.Data may include but be not limited to content-data, such as goods attribute, price and description information, with buyer and seller
Associated item lists, and auction are bidded.Data may also include the number of deals for being related to and collecting and paying among other
According to the real-time synchronization of, delivery transaction, and financial daily record.In one embodiment, network system 100 is suitable in online marketplace
Buy and sell article.
In various embodiments, the data exchange in network system 100 may depend on by one or more client computer
Or user's selection function that user interface (UI) is provided.UI can be associated with client machine, such as using Web client
106 client machine 110.Web client 106 can pass through network 104 and market applied procedure 120 via Web server 116
Communication.UI also can be associated with the client machine 112 using application clients 108 (such as client applications), or
It is associated with the third-party server 130 with third party application 128.It can be appreciated that in various embodiments, visitor
Family machine machine 110,112 or third-party server 130 can be carried with buyer, seller, third-party E-commerce platform, payment services
Associated for business, service of delivering goods provider or financial institution system, wherein each communicates with data interchange platform 102 and can
Selection of land communicates with one another.Buyer and seller can be any one in individual, trade company or service provider.
Turn more specifically to market applied procedure 120, application programming interfaces (API) server 114 and Web server 116
One or more application servers 118 are coupled to, and routine interface and web interface are supplied to one or more applications respectively
Server 118.The one or more market applied procedures 120 of the trustship of application server 118.Application server 118 and then it can couple
To one or more database servers 124, database server helps to access one or more databases 126.
In one embodiment, Web server 116 and API server 114 send and receive on list, transaction and
The data of goods attribute.For example, Web server 116 can be transmitted data in client machine (for example, client machine
110) webpage on browser application (for example, Web client 106) that is run on and data are received from above-mentioned webpage.
API server 114 can transmit data to the application run on another client machine (for example, client machine 112)
Program (for example, application clients 108) and from above-mentioned application program (for example, application clients 108) receive data.
In one embodiment, market applied procedure 120 is easy to publish and fixed a price, so that seller and buyer are in online city
Item Listing on field, article is bidded and/or article is bought.
Referring still further to Fig. 1, in one embodiment, keyword data is aggregated to one or more polymerization keyword numbers
According in server 150, as will be as well known to those of ordinary skill in the art.In one embodiment, at one or many
Aggregation engine in individual server 150 is used(extra large dupp), but other embodiments can be used it is other suitable
Aggregation engine.Platform 102 can carry out the communication that website inquires about data with one or more keyword data servers 150 that polymerize.
In one embodiment, website inquiry data may include the data related to user (potential customer) participation of platform 102.So
Data will be referred to herein as user participate in data.The essence of such data and assessment will discussed in further detail herein.
It will be appreciated that the data that one or more polymerization keyword data servers 150 are provided will be historical data.
The data are by including the information on all kinds user's participation activity.Among other reasons, some in the data are in spy
The article or the particular type of goods that timing is waited or chosen for user can be more popular.The certain kinds of data
The popularity of type (such as displaying, click-through-rate) can will have much weights to the data of those types compared with other data
Produce influence.
Search engine 160 (can be third party's search engine, such as(Google) or(must
Should)) the various data on paid search can be supplied to one by one or more paid search data modules 162
Individual or multiple polymerization keyword data servers 150.Paid search data may include following data, such as clicks on and receives every time
Enter (RPC, Revenue Per Click);Each clicking cost (CPC, Cost Per Click), also referred to as each pay-per-click
(PPC, Pay Per Click);(RPI, Revenue Per Impression) is taken in displaying every time;Expense is shown every time
(CPI, Cost Per Impression);And click-through-rate percentage (%CTR, Percent Click Through
Ratio).Some search engines provide (subjectivity) data of themselves, sometimes referred to as quality score, the quality score and key
Word is relevant with the degree of correlation with the quality of PPC advertisements.Search engine companies can determine advertisement charging with use quality score.Pass through
Such search engine, quality score can be improved by improving keyword relevance, and can reduce advertising expenditure.
As will be as discussing in further detail herein, one or more keyword evaluating servers 170, including training mould
Block 172 and evaluation module 174, are received using from search engine 160 and one or more polymerization keyword data servers 150
Data select the keyword bidded.Training module 172 and evaluation module 174 are shown to single module in Fig. 1, but
It is that they need not be single module.
One or more keyword evaluating servers 170 can receive various analyses, including the search from search engine 160
Object information, and those analyses can be assessed in the context that website participates in data to develop potential keyword.One or
Paid search data can also be used in multiple keyword evaluating servers 170, and it is directly from search engine 160 or by one or many
Individual polymerization keyword data server 150 is handled, to develop keyword bidding strategies.
Training module 172 can receive the polymerization keyword data from one or more servers 150 and (including for example go through
History data) to develop a kind of model, in one embodiment, the user that the model prediction is attracted to company's site will be actual
Therefore upper shopping produces the possibility of income for company.So income is produced from the customer for accessing website and be referred to alternatively as conversion, such as
The implication that the term is previously used.The model knows the weight of different types of data, and correspondingly can distribute in a model
Weight.
Evaluation module 174 can receive the information on transformation model from training module 172;From one or more servers
150 polymerization keyword data;And the feedback based on market from search engine 160, and such as keyword can be produced
Value score.These value scores for example can with from search engine 160 receive on particular keywords bid costs
Information is compared.The comparative result, which can cause to produce one or more keywords, bids.In one embodiment, on key
Word bids the information (budget that total amount or the keyword for specific products or specific products classification are distributed) of budget can
As whether producing a part for the decision that keyword is bidded and use.
Fig. 1 shows one or more polymerization keyword data servers 150, its with search engine 160 and one or many
Platform 102 is connected on the individual identical network of keyword evaluating server 170.However, structure as described herein and technology are not limited
In this.One or more servers 150 can be connected more directly to platform 102, or can pass through the network different from network 104
It is connected to platform 102.One or more servers 150 can be by arriving with one or more identical network connections of server 170
Platform 102.
Fig. 1 shows one or more keyword evaluating servers 170, and it is polymerizeing keyword data clothes with one or more
Platform 102 is connected on business device 150 and the identical network of search engine 160.However, structure as described herein and technology are not limited to
This.One or more servers 170 can be connected more directly to platform 102, or can be connected by the network different from network 104
It is connected to platform 102.One or more servers 170 can by with one or more identical network connections of server 150 to flat
Platform 102.
Generally from the point of view of the interacting of user and website, potential customer (user) can be by inputting company in a browser
URL is directly entered the website of company.Alternatively, one or more links that potential customer can be presented by clicking on the page
And entering company's site, search keyword is input to network search engines in response to potential customer and presented by the link.Chain
It can be search result to connect, or they can be paid for a part for advertisement, and it is placed by being bidded to keyword.Point
Analysis may be notified that how company potential customer enters company's site.
Once potential customer comes company's site, the essence that potential customer interacts with website can show that what can promote to company
Enter the more various flow of company's site.Article buy, as a part of flow at company's site, be typically considered to it is very few and
Many information are not provided.However, can provide thicker with interact (even if the interaction does not cause purchase) of website before purchase
Close data and what is more reliably seen clearly customer can be attracted to website.Specifically, see clearly can cause it is competing to keyword
The more intelligent selection of valency.
For the welcome high sales volume product with a variety of Sales Channels, many entities (including competitor) may be willing to
Meaning is bidded to particular keywords.But company is also desirable to bid to those keywords, so as to ensure the webpage of company with
Notable position is appeared in the relevant search result of those products.
For less welcome product, competition may not be such fierce.Certainly, many products assign to this
In class.Just this product how is being searched for by monitoring customer and this product is checked in company's site, and company can excavate gained
To data and formulate enhanced Keyword Selection and bidding strategies.
Can be by the actually countless access from potential customer with the company existed online.In order to potential customer
Become customer, how analysis customer interacts with website is helpful.As will be discussed herein, customer's interaction can be such as
The inquiry (certainly using various words and word combination) carried out based on customer on website, indicate the higher of potential keyword or
The lower successful possibility of person.For example, potential customer and ad hoc inquiry are (for example, checking Query Result, checking specific in the result
Article) it is interactive must be more, then customer be used for edit inquiry word indicate keyword be suitable for the possibility of website context just
It is bigger.
Customer queries are not perfect data sources.Some customers broadly can interact with the inquiry that it is finally abandoned.
For example, as the enough results for checking ad hoc inquiry, customer would know that inquiry may need to be changed to for example use
Less keyword;Identical keyword is used in mutual different relations;Use different keywords;Or these three can
Any combinations of energy property.This, which is enumerated, is not intended it is detailed;Other inquiries change
Significantly.
If customer spends few time or relatively little of time to inquiry, more limited interaction can notify assessment to draw
Hold up, show that user has found that the inquiry or its content are improper.If customer takes relatively long period of time to specific inquiry, so
After abandon the inquiry, and it is preferred that another inquiry with different keywords, then such interaction can be by the signal of mixing
It is supplied to evaluation module 174.On the one hand, customer has done many to an inquiry really, shows keyword in the inquiry
Grade of fit.On the other hand, customer abandons the inquiry and it is preferred that another is inquired about, then may indicate that one or more keywords really
Be not suitable for.
In order to compensate the mixed signal of type just described, in one embodiment, training module 172 can customer in website
On overall session context in check the use of inquiry.For example, longer session, perhaps using one or more overlapping or
The multiple search that person's same keyword is carried out may indicate that is used for one or more same keywords of inquiry in ession for telecommunication customer
With bigger overall desirability.Using the independent inquiry of overlapping keyword set can be compared to each other assess one or
Multiple overlapping or same keyword grade of fit.
A kind of method for checking each session is, as the vote of confidence, in more or less degree, to see that customer thinks net
Stand content matching its inquiry degree how.Thus, for example, checking single thing in Query Result list when customer stops
During product, this can be considered as positive signal.This article that customer checks is more, and user's sense of participation is by just more positive.Customer
Check that the time that one or more articles are spent is longer, then user's sense of participation is by just more positive.
In addition to devoting a tremendous amount of time and checking one or more results of ad hoc inquiry, customer can make other things
With indicate positive user's sense of participation by.For example, in online marketplace, customer may just search for be occurred in a group polling result
Special article, customer can take contact seller the step of.Customer may be inquired on product, on its delivery, on seller
The problem of returns policy etc..Ad hoc inquiry result on the product can be placed on " hope " list or " concern " list by customer
In.In " hope " list, as one of ordinary skill in the well-known, customer can place that she or he may be interested in
In the future some when the goods bought.The price of goods may be changed over time." hope " list can be used in online marketplace
Content is to customer notifications price change." concern " list can similarly work.Additionally or alternatively, in customer to goods
In the online marketplace bidded (such as in online auction), the content of " concern " list can be used to notify customer specific for online marketplace
The remaining time in auction.Online marketplace can also be related to the auction of same or similar goods to customer notifications.
Different users can have different interactions from website, even when searching for identical article.For example, it is contemplated that following
Situation, particular brand, model, color and configuration of the two of which user in search smart mobile phone.One user can be inputted to intelligence
The general inquiry of energy mobile phone, for example, only name and model, and can obtain hundreds and thousands of search results.In online city
In, for example, there is dozens of or hundreds of sellers, it issues the details page of specific intelligence mobile phone model, or single seller
Multiple details pages can be issued, each details page is used for a kind of available smart mobile phone model of the seller.Website can be carried to user
For one or more filters so as to his/her search of constriction, for example, by color and/or configuration, by price, possibly
Searched for by cellular phone operators come constriction.If user is very interested in purchase smart mobile phone, user can be according to
One or more Query Results are checked as original presentation or according to as filtering.If initial query is returned very
The result (displaying) of big quantity, and user clicks on some in those results, it is believed that and the inquiry has relatively low point
Hit percent of pass.
Each example of filtering can be considered as different inquiries, or can be considered as one of identical initial query
Point.If constriction Query Result (displaying) is filtered, and user clicks on some in those results, then is clicked on user from just
The identical result for beginning to inquire about is compared, and click-through-rate will be higher.
If there is enough interest to one or more Query Results, user can propose problem to seller.With seller's
Interaction can promote user to be bidded to article (if online marketplace includes auction function and seller sells production using auction function
Product);Concern article (to check to bid how to be carried out on article);Or article is listed, for example, indicating that the article is user
Want in the article sometime bought.User can determine article to put online shopping in-car into buy, or simply straight
Connecing the purchase article, (if for example, function is not auctioned in online marketplace, or if auction function includes leaving bid process simultaneously
And the option of set price is paid for article).
Alternatively, the same user can determine that his/her initial query is excessively wide in range, and executable different narrower
Inquiry.Narrower inquiry may be by shared at least one keyword as initial query so that inquiry can usefully be divided
It is grouped together, and is considered as a part for same session.Or, user can determine that the smart mobile phone that she or he is just searching for is not
Suitably, or simply determine to need to search for different smart mobile phones.Inquiry for different intelligent mobile phone can be used one
Or multiple different keywords, therefore can be considered as different sessions.
Another user can input same brand and more details of model smart mobile phone, because the user may decide
Specific color, configuration and operator.The inquiry of the user by produce may be more relevant with the user less result.The user
Can follow with first user's identical operate, and check one or more Query Results, interacted with one or more sellers, it is right
Article is bidded, pays close attention to article, article is placed in list, article is put into shopping cart, directly purchase etc..
If user determines purchase smart mobile phone, there is purchase decision and come from the income information of purchase.
Multiple users can simultaneously or on the time closely or within the specific period for identical overall brand and
The smart mobile phone of version is scanned for.For example in the case of smart mobile phone, listed in specific smart mobile phone model and version
In some period afterwards, such search will be more popular.The intelligent hand is may indicate that in less search of more late period
The lower popularity of machine, or certain type of user interest are reduced.Check all these from what is carried out in proper time period
The total data polymerizeing in user's search, can learn the adaptability on particular keywords.
The degree that user in the aggregation group actually buys smart mobile phone will indicate Revenue, the Revenue and then can
Compared with keyword bid information, not only to determine that the keyword of special article classification is bidded budget, also determine to such
Not interior one or more special articles spent on keyword how much.For itself, user purchase data may it is very few and
Without meaning statistically.However, with reference to other data on User Activity, income data is probably useful.
The price of article can be included in income data.Certainly, compared with low price article, expensive articles can produce phase with less sale
With income.The quantity for selling article is carried out with the number of times that user is looked back Query Result the number of times of article interrogation or user
Compare, can also learn the value for the keyword that can be bidded.
Above-mentioned discussion concentrates on paid search, and such as user is scanned for using search engine 160, and in search result
It is middle when there is the advertisement of company's site.However, it will be recognized by those of ordinary skill in the art that one or more skills described herein
Art and structure can be applicable to permission or issue user and can click on any website of advertisement.Social media site is (for example(facebook),(pushing away spy),(Google+) and(spelling interest)) will be
Example.That is, instead of being bidded on a search engine to keyword, user can be bidded on social media site so as to will be with pass
The corresponding goods advertisement of keyword is placed on social media site.Keyword Selection result will be equally applicable.
Fig. 2 shows the flow chart of the embodiment of keyword bid process 200.
201, one or more polymerization keyword servers 150 can be in the user conversation at platform 102 on monitoring website.
At this point, session may include inquiry, or may include clicking directly on for the article seen to user when entering website.
202, one or more polymerization keyword servers 150 can monitor user's inquiry on website.As described above,
Depending on the property of inquiry, user conversation may include multiple queries, or may include single query.
After user is inquired about, 203, one or more servers 150 can monitor user how with Query Result
Interaction.Such interaction is discussed above with certain details, and will be discussed in detail below.For Fig. 2
The example of the purpose of the shown discussion, user mutual or participation may include to interact with article seller;Article is bidded;
The purchase of article;Look back the plenty of time that ad hoc inquiry result is spent;To same or similar article or goods categories
Input of multiple queries etc..
204, one or more polymerization keyword servers 150 merge or polymerize relevant with article or goods categories
User's participation activity.Some activities may be favourable, and may point to the identification or selection of one or more keywords.Some
Activity may be less favourable or unfavorable, and is not directed to the identification or selection of one or more keywords.By total
Amount, the data of polymerization, which should be pointed to, to be correctly oriented.
In one embodiment, user's participation activity may be parsed in more detail, for example, to attempt clearly to filter
The data that make mistake (such as one or more users input significantly different extraneous probes during single website visiting), therefore can
Can twisting data.
205, one or more keyword evaluating servers 170 are by least one standard come comparative item or article
The aggregated data of classification.In one embodiment, a specific standard, such as shows, can be focus.If there is no foot
Enough matchings, then flow may return to Fig. 2 beginning.If between user's participation activity and at least one preassigned
In the presence of enough matchings, then 206, one or more servers 170 can parse the aggregate number of user's participation activity in more detail
Just identification is suitable to article or the potential keyword of goods categories according to this.
207, one or more servers 170 can be calculated for each potential keyword identified in 206 and are worth
Score.In one embodiment, value score may have with the quantity or magnitude of one or more users participation class of activity
Close.208, keyword identify in 206 and with value score similar enough can be grouped.At one
In embodiment, such packet can be limited to the keyword with identical or almost identical value score.It is attached in characteristic value
The variable quantity of near value score is probably any amount from 0% to 3%.The scope of change or the amount of change are selected to the greatest extent
It is possible small, while providing enough data to avoid the very few problem of data discussed above.Therefore, in some cases,
There can be the keyword of relatively small number in a packet, and in other cases, there can be the pass of relatively large amount in a packet
Keyword.Keyword quantity in a packet should provide the data of sufficient amount, while avoiding significant change as far as possible.At one
In embodiment, value score can be determined based on the relation with click income (RPC) every time.Other income or expense can be used
Value, including but not limited to above-mentioned RPI, CPC or CPI.Keyword bid budget the point can also be taken into account so as to
Bid value score.In one embodiment, 209 or after 209 budget can taken into account, as described below.
209, keyword packet value score is compared with keyword bid costs.In one embodiment, should
Expense is the keyword bid costs at search engine 160, and it can provide CPC, CPI and other expenses as indicated above
Use data.In one embodiment, if keyword is bidded, budget is including 208 are not accounted for, and the budget can be in 209 quilts
Take into account.Alternatively, if packet value score is more than keyword bid costs, it is competing to check that subsequent step may be present
Whether valency, which meets keyword, is bidded budget.
210, if packet value score is more than keyword bid costs, it can be bidded.Such as that pointed out just now
Sample, budget is not met if bidded, and the decision bidded can be changed, and how score is worth but regardless of packet.
Fig. 3 shows the example of user's participation activity 300.To a certain extent, these examples can be with example provided above
It is overlapping, but described herein in more succinct form so as to organizational goal.Following discussion purpose, which is not lain in, to be participated in a class user to live
It is dynamic to bring up on another kind of user's participation activity, therefore movable order in the discussion is unessential.
One element of user's participation activity that may be related is that user spends the (meeting of cost of time taking overall length on website
The words time 302).User-accessible website is to start session.In the ession for telecommunication, user can input it is one or more inquiry with
Just one or more articles are searched for.Alternatively, go out on one or more webpages that the browsable user at website of user is accessed
Existing one or more articles, without input inquiry.Under normal circumstances, the session of user is longer, then can be more relevantly by website
It is considered that user is interested.
Either goods categories are determined or session can be by user on website the special article that session can be searched for by user
The total time of cost determines.In the context of the techniques described herein and structure, to the associative key for bidding
Identification can be relevant with specific article or goods categories.Therefore, in one embodiment, to a certain extent, specific
During website visiting, user searches for different articles, and every group polling of special article or goods categories can be considered to constitute
One session.
User's participation activity another may be related element with as produced by search inquiry of the user on website
Show that quantity is relevant (displaying quantity 304).During when user input query to search for special article or goods categories, if there is
A large amount of search results comprising the article or goods categories, the then one or more keywords used in the inquiry can be considered as
It is more relevant, it is thus possible to more valuable.
Another element that may be related of user's participation activity can be click on percent of pass (CTR) 306.User clicks on inquiry
As a result number of times, as the percentage of Query Result (displaying) quantity, can be shown that one or more articles as inquiry theme
The higher degree of correlation, therefore one or more keywords for bidding can be learned.
As indicated above, user can carry out multiple queries (number of searches 308) while website is accessed.Inquiry
Entirely different product can be directed to, or inquiry can be directed to identical product or product category.It is in different products
In the case of difference inquiry theme, each inquiry or inquiry group for being related to corresponding product can be considered as a session, and
The degree of correlation of the keyword identified in each such session can be considered separately.Looked into the difference using different keywords
In the case that inquiry is related to like products or product category, the user of difference inquiry can be shown that in ession for telecommunication is more early inquired about
Keyword has the relatively low degree of correlation.
The result inquired about as user on article, user can make certain and determine without buying article or to article
Bid.For example, user can determine concern article (310).If the article on website can be used for pin within the limited period
Sell, then user may wish to know that purchase or the available remaining time for bidding to article are being reduced.It is used for pin without concern
Article or goods categories can be added to product list (312), product of changing products in one embodiment by the special article sold, user
List is wish list, and it shows that this product is the product that user has some interest and in the future may be interested at present.Perhaps
As the result for clicking through and looking back one or more inquiries, user can determine to add the item to online shopping car
(314).Alternatively or additionally, user can determine the problem of inquiry article seller is on article (316).
310 to 316 all show that user has certain direct interest to article or goods categories, and also will be user's ginsengs
With the associated exemplary of activity, it can be taken into account to make one or more keywords for bidding.
May have following situations, wherein user sends inquiry, and user plane Query Result is done anything (as
Non-limiting example, for example click through, article concern, item lists, add the item to shopping cart, inquiry article seller ask
Topic) can disappear is longer than the normal period before to a certain extent.Such longer period (318) can be shown that user
More carefully consider Query Result, therefore can be shown that potential higher correlation of one or more keywords in user inquires about
Degree, and the corresponding potential desirability that one or more keywords are bidded.
Query Result may span across multiple pages.The page quantity that user looks back from ad hoc inquiry result (320) can be with user
The level interested for the one or more articles being related to inquiry is related.The page quantity of review is more, then one or more to look into
Ask item higher as the potentially relevant degree of one or more keywords for bidding.
In the online marketplace with auction function, user may it is interested enough in the article that occurs in Query Result from
And the article is bidded (322).User can even determine to buy the article (324), or because the online marketplace is not online
Function is auctioned, or because the online marketplace allows user to bypass bid process and agree to pay certain fund to buy the thing
Product.
The result of one or more articles is bought during specific website visiting as user, online marketplace will produce receipts
Enter.The amount received produced from specific bought article also can be shown that the desirability that the keyword in inquiring about user is bidded.
It should be noted that user can perform and will often perform indicated by more than one above-mentioned discussion such as Fig. 3
Those actions.It is used as non-limiting example, the executable inquiry to article of user;One or more results can be clicked through;Certainly
It is fixed that article is placed in concern list;Seller is inquired on article;Put article into shopping cart;Article is bidded;And/or purchase
Article.
In some previous discussions, it is compared user's participation is movable with predetermined standard, specifically, with displaying number
Amount (304 in Fig. 3) is compared.One or more of the other standard can be used to replace showing quantity, or with showing quantity
It is combined.For example, the Session Time (302 in Fig. 3) spent;Click-through-rate (306 in Fig. 3);Or one of concern
Or multiple articles (310 in Fig. 3) may make up such standard.This enumerate and previous discussion be intended that it is exemplary, and simultaneously
The non-exhaustive of being intended that.
The meter of the value score shown in one or more keyword evaluating servers 170 and Fig. 2 shown in Fig. 1 is checked in return
Calculate, it has been noted that the weight of the particular instance of user's participation activity can be the function of the amount of these active instances.Algorithm is such as
The various classifications what finally assigns a weighting to user's participation activity are not crucial for the techniques described herein and structure.Take
Certainly in product or product category or product price or product quantity, and to user's online browse, choose and buy habit
Used analysis, the particular type of user's participation activity in some cases can be more relevant with keyword value determination, and other
In the case of it is less relevant.For example, specific product type can have conversion ratio (if i.e., someone finds the product online, to be somebody's turn to do
People more likely actually buys the product).Relatively large weight can be allocated so as to the conversion ratio of the product type.
Some embodiments are described herein as including logic or multiple parts, module or mechanism.Module can shape
Into software module (for example, on non-transitory machine readable media (1) or in the transmission signal (2) carrying code) or
The module that hardware is implemented.The module that hardware is implemented is able to carry out specific operation and can configure or arrange in a specific way
Tangible unit.In the exemplary embodiment, one or more computer systems are (for example, independent computer system, client computer
Or server computer system) or one or more processors can be by software (for example, the portion of application program or application program
Point) module of hardware implementation is configured as, it operates to perform specific operation as described herein.
In various embodiments, the module that hardware is implemented can mechanically or electronically be implemented.For example, what hardware was implemented
Module may include special circuit or logic, and it is permanently configured (for example, as application specific processor, such as field-programmable
Gate array (FPGA) or application specific integrated circuit (ASIC)) into perform specific operation.The module that hardware is implemented also includes programmable
Logic or circuit (for example, as including in general processor or other programmable processors), it is temporary by software
When property is configured to perform specific operation.It will be appreciated that may be determined due to expense and time consideration special and permanent
Mechanically implement the mould of hardware implementation in the circuit of configuration or in the circuit (for example, by software merit rating) of temporary configuration
Block.
Therefore, term " module that hardware is implemented " is understood to cover tangible entity, i.e., following entities, and it is by thing
Reason ground builds, permanently configures (for example, rigid line connection), or interim or temporary configuration (for example, programming) so as to
Ad hoc fashion is operated and/or to perform specific operation as described herein.The module implemented in view of wherein hardware is matched somebody with somebody temporarily
Put the embodiment of (for example, programming), the module that each hardware is implemented need not the time it is any when inscribe and be configured or real
Exampleization.For example, in the case of general processor of the module including the use of software merit rating that hardware is implemented, general processor can be
The different time can be respectively configured to the module that different hardware are implemented.Therefore such as software can be inscribed in one of time
Configuration processor makes it constitute the module that specific hardware is implemented, and constitutes different hardware realities in the not lower in the same time of time
The module applied.
The module that hardware is implemented can provide information into the module that other hardware are implemented and the mould implemented from other hardware
Block receive information.Therefore, the module that the hardware is implemented can be considered as communicatedly to couple.Exist concurrently with multiple this hardware
In the case of the module of implementation, it can be transmitted by signal (such as proper circuit and bus by connecting the module that hardware is implemented)
To realize communication.Multiple hardware implement module in the embodiment that different time is configured or instantiates, may be implemented in
Communication between the module that such hardware is implemented, for example, passing through the addressable memory knot of the module implemented in multiple hardware
Structure memory storage and retrieve information to realize.For example, the executable operation of module that a hardware is implemented, and the output of operation is deposited
Storage is in the memory devices coupled with communicating with.Then after the module that another hardware is implemented accessible storage device equipment with
Just retrieve and handle stored output.Hardware implement module also can initiate and input or output equipment communication, and
And resource (for example, information aggregate) can be operated.
The various operations of illustrative methods as described herein can be performed by one or more processors at least in part, described
Processor is by provisional configuration (for example, by software) or the permanent operation for being configured to perform correlation.No matter it is interim or permanent
Configuration, such processor may make up the module of processor implementation, its be operable to perform one or more operations or
Function.In certain embodiments, module as referred to herein is probably what processor was implemented.
Similar, method described herein can be that processor is implemented at least in part.For example, can be by one or more
Module that processor or one or more processors are implemented performs at least some operations of method.The execution of specific operation can
It is distributed in one or more processors, does not only reside in individual machine, but is deployed in multiple machines.In some examples
Property embodiment in, one or more processors can be located at single position (for example, in home environment, in office environment, or
Person is used as server farm), and in other embodiments, processor may extend across multiple position distributions.
One or more processors be also operable to " in cloud computing environment or as " software service (SaaS,
Software as a Service) " support associative operation execution.For example, at least some operations can pass through one group of computer
(being used as the example of the machine including processor) performs, and wherein these operations can be by network 104 (for example, internet, referring to figure
1) and by one or more suitable interfaces (for example, API) access.
Embodiment can be realized in Fundamental Digital Circuit or in computer hardware, firmware, software or in its combination.
The computer-readable recording medium of non-transitory can be used to realize (for example, being visibly carried on the meter in information carrier for embodiment
Calculation machine program, for example, the computer program in the machine readable media of non-transitory, for being performed by data processing equipment
Or the operation of control data processing equipment, data processing equipment is, for example, programmable processor, a computer or many
Platform computer).
Computer program can be write with any type of programming language, including compiler language or interpretative code, and can
Dispose in any form, including it is other as stand-alone program or as module, subprogram or suitable for computing environment
Unit.Computer program can be deployed on one computer or a position across multiple position distributions and by logical
Performed in the multiple stage computers that communication network 104 is interconnected.
In embodiment, operation can be performed by one or more programmable processors, and programmable processor performs computer
Program carrys out perform function will pass through to operate input data and produce output.Method operation also can be by special logic electricity
Road (such as FPGA or ASIC) is performed, and is embodied as dedicated logic circuit, such as FPGA according to the equipment of exemplary embodiment
Or ASIC.
Computing system may include client-server.Client-server is generally remote from each other and generally passed through
Communication network (such as network 104) is interacted.The relation of client-server on respective computer by running simultaneously
And there is the computer program of client-server relationship to produce relative to each other.In the reality of deployment programmable computer system
Apply in example, it will be appreciated that hardware and software framework is all worthy of consideration.Specifically, it will be appreciated that on by forever
Hardware (for example, ASIC), the hardware (for example, combination of software and programmable processor) of provisional configuration of long property configuration, also
The selection combined to implement specific function for being the hardware of permanent and temporary configuration can be the selection in design.Now will
Description can dispose hardware (for example, machine) and software architecture in various embodiments.
Fig. 4 is the schematic diagram of exemplary computer system 400, in its interior executable instruction set so that machine performs sheet
Literary disclosed any one or more of method.Computer system 400 can be presented as any as shown in Figure 1 whole or in part
Server or other machines.
In certain embodiments, machine is operated as autonomous device, or can be connected (e.g., networked) to other machines.
In the deployment of networking, machine can in client server network environment as server or client machine 110 or
112 operations, or operated in equity (or distributed) network environment as peer machines.Machine can be personal computer
(PC), tablet personal computer, set top box (STB), personal digital assistant (PDA), smart mobile phone, Web appliance, network router, are exchanged
Machine either bridge or is capable of any machine of execute instruction collection 424 (in order or otherwise), and it is specified by the machine
The action that device is taken.In addition, although individual machine is only shown, but term " machine " also will be understood as one that including any of machine
Set, it performs (or multiple) instruction set 424 to perform any one or more of side as described herein respectively or jointly
Method.
In one embodiment, computer system 400 may include processor 402 (for example, CPU (CPU), figure
Shape processing unit (GPU or both)), main storage 404 and static memory 406, it is in communication with each other by bus 408.Meter
Calculation machine system 400 may also include video display unit 410 (for example, liquid crystal display (LCD), plasma display, light-emitting diodes
Manage (LED) display, or cathode-ray tube (CRT) display).Computer system 400 may also include alphanumeric input and set
Standby 412 (for example, keyboard or keypads), UI navigation equipments 414 (for example, mouse or touch pad), disc driver or
Other storage (such as solid condition apparatus or SSD) units 416, signal generating apparatus 418 (for example, loudspeaker), and network connect
Jaws equipment 420.
Disk drive unit 416 may include machine readable media 422, one or more instruction set and data structure (examples
Such as, 424) instruction is stored on machine readable media 422, and it embodies any one or more of method or function as described herein
Or used by any one or more of method or function as described herein.During being performed by computer system 400, refer to
Making 424 also can completely or at least partially reside in main storage 404 and/or in processor 402, wherein main storage
404 and processor 402 also constitute machine readable media 422.
Instruction 424 also passes through network using any one of a variety of well-known host-host protocols (for example, HTTP)
Interface equipment 420 is transmitted or received by network 426 (it is not a part for computer system 400).
Although machine readable media 422 is shown as single medium in one embodiment, term " machine readable media " should
This is understood as that including single medium or medium (for example, centralization or distributed data base, and/or associated height
Speed caching and server), it stores one or more instruction set 424.Term " machine readable media " be also understood that into including
Any medium, it can store, encode or carry instruction set 424, and the instruction set is performed by machine, and machine can be caused to hold
Any one or more of method of the row disclosure, or it can store, encodes or carry data structure, the data structure by
Such instruction set 424 is utilized or associated with such instruction set 424.Therefore term " machine readable media " should be understood that
Into including but not limited to solid-state memory, optical medium and magnetizing mediums.
As noted previously, the particular order of operation is to provide for convenience's sake in various embodiments
, but be not intended to imply needed in whole embodiments identical order.Wherein subsequent operation is not dependent on prior operation
Any operation order is all rightly included in the scope of the present disclosure.As non-limiting example, with reference to Fig. 3 to different types of
User's participation activity is illustrated.Worth considered sequence is not implied that to the order that those Activity Types are illustrated, is said
Bright order does not imply that the order of importance yet.
The summary of the disclosure provides into the regulation for following 37C.F.R. § 1.72 (b), and the regulation requires to provide summary, and it will
Allow reader's essence disclosed in determination technology rapidly.It should be understood that summary will not be used to interpret or limit the scope of claim
Or implication.In addition, it can be seen that in some cases, various features can be grouped together in above-mentioned embodiment
To simplify the disclosure.This open method is not necessarily to be construed as reflecting intention so that the embodiment proposed is clearly every
More features is required outside the feature referred in individual claim.On the contrary, following claims reflect theme, it is included than list
The less feature of whole features of individual disclosed embodiment.Therefore, following claims are attached to embodiment hereby
In, wherein each claim itself can be used as different embodiments.
Although describing the disclosure with reference to specific embodiment, it is readily apparent that do not depart from the disclosure it is broader spirit and
In the case of scope various modifications and variations can be carried out to embodiment.Therefore, specification and drawings will be considered as it is exemplary and
It is nonrestrictive.
Claims (20)
1. a kind of computer-implemented search engine marketing method, it is used to attract potential customer to access website, methods described bag
Include following using computing device:
User's participation activity of polymerization site visitor, the polymerization includes:
The search inquiry that monitoring visitor inputs on website;
Monitor user participation activity of the visitor for the Query Result at website;And
Merge user's participation activity to provide user's participation activity of polymerization;
Methods described also includes:
User's participation activity of the polymerization and at least one preassigned are compared;
If user's participation activity of the polymerization meets one or more of at least one described preassigned,:
User's participation activity of polymerization is parsed to recognize potential keyword from the search inquiry for producing Query Result;
Calculate the potential keyword value score of each potential keyword;
Potential keyword with potential keyword value score similar enough is grouped to provide packet value score;
The information that the packet is worth into score and the bid costs on those keywords is compared;And
In response to the comparison of packet value score, decide whether to bid to the potential keyword in packet.
2. according to the method described in claim 1, it is characterised in that user's participation activity include selected from following one or
Multiple standards:
The time span spent during the visit on website;
The time span spent during the visit on the webpage of website;
Displaying quantity in Query Result;
The click-through-rate of displaying in Query Result;
During the visit in the webpage quantity of website visiting;
For specific group polling result, the quantity of the Query Result through checking;
Check the time span that the Query Result is spent;
The time span spent in the specifically result through checking;
The time span spent before new inquiry is entered on current queries;
Potential customer interacts with article seller's;
Article is placed in shopping cart;
Article is placed in concern list;
Article is placed in wish list;
Article is bidded;
Buy article;And
It is used as the amount received produced by the result of purchase article.
3. method according to claim 2, it is characterised in that user's participation activity is marked including two or more
It is accurate.
4. according to the method described in claim 1, it is characterised in that at least one preassigned includes value, described value is more than choosing
From the analog value of following one or more standards:
The time span spent during the visit on website;
The time span spent during the visit on the webpage of website;
Displaying quantity in Query Result;
The click-through-rate of displaying in Query Result;
During the visit in the webpage quantity of website visiting;
For specific group polling result, the quantity of the Query Result through checking;
Check the time span that the Query Result is spent;
The time span spent on the specifically Query Result through checking;
The time span spent before new inquiry is entered on current queries;
Potential customer interacts with article seller's;
Article is placed in shopping cart;
Article is placed in concern list;
Article is placed in wish list;
Article is bidded;
Buy article;And
It is used as the amount received produced by the result of purchase article.
5. method according to claim 4, it is characterised in that at least one preassigned includes two or more values,
Analog value of described two or more the values more than one or more standards.
6. according to the method described in claim 1, it is characterised in that the expense bidded and received from third party's search engine
Information is related.
7. according to the method described in claim 1, it is characterised in that it is by using one to calculate potential keyword value score
Or multiple standards, come what is performed, the standard is selected from following:Every time click on income (RPC), every time displaying income (RPI), every time
Clicking cost (CPC), and displaying expense (CPI) every time.
8. according to the method described in claim 1, it is characterised in that described to decide whether to bid to the potential keyword in packet
Including packet value score and keyword budget of bidding are compared.
9. method according to claim 8, it is characterised in that if also including being worth score in budget limitations if packet
Potential keyword in packet is bidded.
10. according to the method described in claim 1, it is characterised in that include the packet of potential keyword to potential keyword valency
It is worth dividing the potential keyword in the 10% of predetermined value to be compared.
11. a kind of equipment, it includes:
Memory, it stores the executable instruction of processor;And
Processor, it performs the executable instruction of processor, and the processor is configured to:
Pass through user's participation activity of following polymerization site visitors:
The search inquiry that monitoring visitor inputs on website;
Monitor user participation activity of the visitor for the Query Result at website;And
Merge user's participation activity to provide user's participation activity of polymerization;
The processor is configured to:
User's participation activity of the polymerization and at least one preassigned are compared;
If user's participation activity of the polymerization meets one or more of at least one described preassigned,:
User's participation activity of polymerization is parsed to recognize potential keyword from the search inquiry for producing Query Result;
Calculate the potential keyword value score of each potential keyword;
Potential keyword with potential keyword value score similar enough is grouped to provide packet value score;
The information that the packet is worth into score and the bid costs on those keywords is compared;And
In response to the comparison of packet value score, decide whether to bid to the potential keyword in packet.
12. equipment according to claim 11, it is characterised in that user's participation activity includes being selected from following one
Or multiple standards:
The time span spent during the visit on website;
The time span spent during the visit on the webpage of website;
Displaying quantity in Query Result;
The click-through-rate of displaying in Query Result;
During the visit in the webpage quantity of website visiting;
For specific group polling result, the quantity of the Query Result through checking;
Check the time span that the Query Result is spent;
The time span spent in the specifically result through checking;
The time span spent before new inquiry is entered on current queries;
Potential customer interacts with article seller's;
Article is placed in shopping cart;
Article is placed in concern list;
Article is placed in wish list;
Article is bidded;
Buy article;And
It is used as the amount received produced by the result of purchase article.
13. equipment according to claim 12, it is characterised in that user's participation activity is marked including two or more
It is accurate.
14. equipment according to claim 11, it is characterised in that at least one preassigned includes value, described value is more than
Analog value selected from following one or more standards:
The time span spent during the visit on website;
The time span spent during the visit on the webpage of website;
Displaying quantity in Query Result;
The click-through-rate of displaying in Query Result;
During the visit in the webpage quantity of website visiting;
For specific group polling result, the quantity of the result through checking;
Check the time span that the Query Result is spent;
The time span spent in the specifically result through checking;
The time span spent before new inquiry is entered on current queries;
Potential customer interacts with article seller's;
Article is placed in shopping cart;
Article is placed in concern list;
Article is placed in wish list;
Article is bidded;
Buy article;And
It is used as the amount received produced by the result of purchase article.
15. equipment according to claim 14, it is characterised in that at least one preassigned includes two or more
It is worth, analog value of described two or more the values more than one or more standards.
16. equipment according to claim 11, it is characterised in that the expense bidded from third party's search engine with receiving
Information it is related.
17. equipment according to claim 11, it is characterised in that the processor is come by using one or more standards
Potential keyword value score is calculated, the standard is selected from following:Every time click on income (RPC), every time displaying income (RPI),
Each clicking cost (CPC), and displaying expense (CPI) every time.
18. equipment according to claim 11, it is characterised in that the processor by packet by being worth score and key
Word budget of bidding is compared to decide whether to bid to the potential keyword in packet.
19. equipment according to claim 18, it is characterised in that if the processor is configured to packet value score
Then the potential keyword in packet is bidded in budget limitations.
20. equipment according to claim 11, it is characterised in that the processor is configured to by potential keyword valency
It is worth dividing the potential keyword in the 10% of predetermined value to be compared to potential keyword being grouped.
Applications Claiming Priority (3)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
US201514601114A | 2015-01-20 | 2015-01-20 | |
US14/601,114 | 2015-01-20 | ||
PCT/US2015/067477 WO2016118284A1 (en) | 2015-01-20 | 2015-12-22 | Processing and analysis of user data to determine keyword quality |
Publications (1)
Publication Number | Publication Date |
---|---|
CN107209897A true CN107209897A (en) | 2017-09-26 |
Family
ID=56417573
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201580073843.7A Pending CN107209897A (en) | 2015-01-20 | 2015-12-22 | User data is handled and analyzes to determine keyword quality |
Country Status (4)
Country | Link |
---|---|
EP (1) | EP3230936A4 (en) |
KR (1) | KR101942834B1 (en) |
CN (1) | CN107209897A (en) |
WO (1) | WO2016118284A1 (en) |
Cited By (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN111639255A (en) * | 2019-03-01 | 2020-09-08 | 北京字节跳动网络技术有限公司 | Search keyword recommendation method and device, storage medium and electronic equipment |
Families Citing this family (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US10165064B2 (en) * | 2017-01-11 | 2018-12-25 | Google Llc | Data packet transmission optimization of data used for content item selection |
KR101926350B1 (en) | 2017-03-24 | 2018-12-10 | 주식회사 에코마케팅 | System and method for bidding automatically based on score of keyword group in keyword advertisement, and bidding management server |
Family Cites Families (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US7788131B2 (en) * | 2005-12-15 | 2010-08-31 | Microsoft Corporation | Advertising keyword cross-selling |
US7809740B2 (en) * | 2006-03-29 | 2010-10-05 | Yahoo! Inc. | Model for generating user profiles in a behavioral targeting system |
US20110035276A1 (en) * | 2009-08-05 | 2011-02-10 | Arpita Ghosh | Automatic Campaign Optimization for Online Advertising Using Return on Investment Metrics |
US20110145058A1 (en) * | 2009-12-15 | 2011-06-16 | Yun Liu | Method and a system for keyword valuation |
US20140114750A1 (en) * | 2011-06-03 | 2014-04-24 | Jin-Woo Jung | Effective keyword selection system using keyword advertisement for internet search and an effective keyword selection method thereof |
-
2015
- 2015-12-22 CN CN201580073843.7A patent/CN107209897A/en active Pending
- 2015-12-22 EP EP15879259.8A patent/EP3230936A4/en not_active Withdrawn
- 2015-12-22 KR KR1020177019813A patent/KR101942834B1/en active IP Right Grant
- 2015-12-22 WO PCT/US2015/067477 patent/WO2016118284A1/en active Application Filing
Cited By (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN111639255A (en) * | 2019-03-01 | 2020-09-08 | 北京字节跳动网络技术有限公司 | Search keyword recommendation method and device, storage medium and electronic equipment |
CN111639255B (en) * | 2019-03-01 | 2023-12-29 | 北京字节跳动网络技术有限公司 | Recommendation method and device for search keywords, storage medium and electronic equipment |
Also Published As
Publication number | Publication date |
---|---|
KR101942834B1 (en) | 2019-01-29 |
EP3230936A1 (en) | 2017-10-18 |
WO2016118284A1 (en) | 2016-07-28 |
EP3230936A4 (en) | 2018-05-23 |
KR20170098881A (en) | 2017-08-30 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
US8656035B2 (en) | System and method for enabling user requested channels in an IP marketplace | |
US9852477B2 (en) | Method and system for social media sales | |
KR102297669B1 (en) | System for providing matching service for connecting between manufacturer and distributor | |
US11216850B2 (en) | Predictive platform for determining incremental lift | |
Ilieva et al. | Big data based system model of electronic commerce | |
CN105488711A (en) | Multi-Agent based network direct sales system | |
Yang et al. | Customer management in Internet-based platform firms: review and future research directions | |
Alquhtani et al. | Development of Effective Electronic Customer Relationship Management (ECRM) Model by the Applications of Web Intelligence Analytics | |
CN107209897A (en) | User data is handled and analyzes to determine keyword quality | |
Raghavan et al. | Object-oriented design of a distributed agent-based framework for e-Procurement | |
Wei et al. | Online shopping behavior analysis for smart business using big data analytics and blockchain security | |
KR102463084B1 (en) | System for providing block chain and block chain payment platform based commercial service on wired and wireless network | |
Wang | Innovation of e-commerce marketing model under the background of big data and artificial intelligence | |
Bediako | The application of Big Data Analytics in improving eCommerce processes. The Retail sector user experience | |
KR20100037754A (en) | Method for providing company public relations and advertisement through the internet web site | |
Li et al. | [Retracted] Big Data Mining Method of E‐Commerce Consumption Pattern Based on Mobile Platform | |
Panniello et al. | Incorporating profit margins into recommender systems: A randomized field experiment of purchasing behavior and consumer trust | |
Xiao et al. | Analysis of influencing factors and enterprise strategy of online consumer behavior decision based on association rules and mobile computing | |
KR20120011763A (en) | Personalized social networking service system and method | |
Zheng et al. | Recommending What to Search: Sales Volume and Consumption Diversity Effects of a Query Recommender System | |
Li et al. | Lean persuasive design of electronic word-of-mouth (e-WOM) indexes for e-commerce stores based on fogg behavior model | |
Mikulová | E-business technology | |
Nagaty | New advances in E-Commerce | |
Xiao et al. | A study on C2B e-commerce model and how it succeeds from the | |
Allen et al. | Technology and Inequality Case Study: Search |
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 | ||
WD01 | Invention patent application deemed withdrawn after publication | ||
WD01 | Invention patent application deemed withdrawn after publication |
Application publication date: 20170926 |