Detailed description of the invention
In order to make technical problems to be solved in this application, technical scheme and beneficial effect clearer, clear,
Below in conjunction with drawings and Examples, the application is further elaborated.Should be appreciated that this place is retouched
The specific embodiment stated, only in order to explain the application, is not used to limit the application.
The commodity of high-quality can not be filtered out in order to solve e-business network site server present in prior art,
Accurately for providing suitable merchandise news as the buyer of the side of browsing, the side of browsing is made to browse the road of merchandise news
Footpath increases, and increases the problem that the operation complexity of suitable commercial product is searched by the side of browsing, and the embodiment of the present application provides
A kind of information providing method, is realized by system as shown in Figure 1, and this system includes information-providing server
1 and client 2.Wherein:
Client 2, for obtaining the search word of user's input, and the information that is supplied to by described search word provides
Server 1, and the information that information-providing server 1 provides is showed user.
The search word that information-providing server 1 inputs according to user, issues data in conjunction with search engine to information
In storehouse, Query Result and the navigation algorithm of the information that the publisher of storage issues are deposited in information gathering data base
The correlativity calculation result of the information that the side of browsing of storage clicks on, in conjunction with the priority regulation rule preset, determines
The offer priority of each information category, and provide a user with and described search word according to described offer priority
The information of coupling.
When information-providing server 1 is for e-commerce website, the information that the publisher of its storage issues is
The merchandise news that seller issues, the information that the side of browsing of storage clicks on also refers to the merchandise news that buyer clicks on.
The flow process of above-mentioned information providing method is as in figure 2 it is shown, its principle is as it is shown on figure 3, comprise the steps:
Step S11: obtain the search word of user's input.
User can use client to input search word, and the user generally as the side of browsing can be searched by input
The mode of rope word searches oneself information interested, and the search word that server side inputs according to the side of browsing user is looked into
Look for the information that publisher shows, provide for the side of browsing user.
Step S12: according to the search word obtained, in Query Information publication database, the publisher of storage issues
Information, determine that the information content of the search word comprising acquisition in each information category accounts for the search comprising acquisition
The quantitative proportion of the information total quantity of word.
The process of quantification ratio specifically includes: determine that in information publication database, the publisher of storage issues
Information in comprise the total quantity of information of search word of acquisition, and each information category comprises acquisition
First quantity of the information of search word;Calculate the ratio of each first quantity and the total quantity determined respectively,
The information content obtaining comprising search word in each information category accounts for the information sum of the search word comprising acquisition
The ratio of amount.
Wherein, in information publication database, storage has the information that all publishers issue, as shown in Figure 3,
Searching for these information by search engine inquiry, this is the true statistical to the information that publisher issues, accordingly
Can obtain embodying the industry distribution data of the incidence relation of the information that the side's of browsing search word is issued with publisher
Relation, wherein industry distribution data can be understood as certain product distribution situation in each information category.
Step S13: according to the search word obtained, the side's of browsing click of storage in Query Information acquisition database
Information, according to the clicked number of times of information of the search word comprising acquisition in each information category, determine acquisition
The relevance weight of search word and each information category.
Determine that the process of relevance weight specifically includes: determine the side's of browsing point of storage in information gathering data base
The information hit comprises the clicked total degree of the information of the search word of acquisition, and in each information category
Comprise the first clicked number of clicks of the information of the search word of acquisition;Calculate respectively determine each
One number of clicks and the ratio of total degree, obtain the information quilt of the search word comprising acquisition in each information category
Number of clicks accounts for the relevance weight of the clicked total degree of information of the search word comprising acquisition.
Wherein, in information gathering data base, storage has the collection data to the information that the side of browsing clicks on, and record is each
The various information that the individual side of browsing once clicked on.As shown in Figure 3, by navigation algorithm, according to gathering data,
Calculating the search word of the side's of browsing input and the dependency of each information category, this is to the side's of browsing operation behavior number
According to statistics, the row embodying the side's of browsing search word with the incidence relation of the side's of browsing operation behavior can be obtained accordingly
Industry distribution statistics.
Step S14: according to the above-mentioned quantitative proportion determined and relevance weight, determine the search comprising acquisition
The prediction priority of each information category of word.Wherein, it was predicted that priority is according to quantitative proportion and dependency
One priority level of each information category that weight primarily determines that puts in order, and each information category is excellent by prediction
After first level arrangement, follow-up each further according to obtaining after the information category that priority is wherein designated is adjusted
The offer priority of information category.
Determine that the process of prediction priority specifically includes: obtain first weighting corresponding with quantitative proportion preset
The factor and second weighter factor corresponding with relevance weight;The most calculated quantitative proportion and first adds
The product of weight factor, and the product of the relevance weight obtained and the second weighter factor, and to taking advantage of of obtaining
Long-pending summation, obtains the predictive value of each information category of the search word comprising acquisition;According to the predictive value obtained
Size determine prediction priority.
Wherein, the first weighter factor is the quantitative proportion the determined weight coefficient when calculating predictive value, the
Two weighter factors are the relevance weight the determined weight coefficient when calculating predictive value, the first weighter factor
Can be arranged as required to the value of the second weighter factor, it is preferred that the first weighter factor and the second weighting
The span of the factor is 0~1.
Wherein, when determining prediction priority according to the size of the predictive value obtained, can be according to each info class
Other predictive value size order, determines the prediction priority of each information category, such as: predictive value maximum
The prediction priority of information category is the highest, it was predicted that the prediction priority of the information category that value is minimum is minimum;Otherwise
Also may be used.
Such as it is shown on figure 3, be intended to computing platform by the side's of browsing industry and calculate: obtain comprises acquisition
The size of the predictive value of each information category of search word is expressed as f (x), then:
Wherein, C represents the search word of acquisition and the relevance weight of each information category, is to call navigation algorithm
The user's search word navigation algorithm anticipatory data distribution obtained;X represent the weighting corresponding with relevance weight because of
Son, for the threshold values determined according to website platform truthful data distribution situation;B/A represents each info class
The information content of the search word comprising acquisition in not accounts for the quantity of the information total quantity of the search word comprising acquisition
Ratio, B is that the information comprising search word of calling search engine acquisition is at each information category, i.e. every profession and trade
Product data distribution situation, A be calling search engine obtain the information comprising search word in all row
Overall product data distribution;(1-x) weighter factor that expression is corresponding with quantitative proportion.
Certainly, this one simply calculating predictive value can preferred form of this, it is also possible to counted by some modes of texturing
Calculate, such as: two weighter factor sums are not equal to 1, etc..
Step S15: according to the prediction priority determined and the priority regulation rule preset, determine each
The offer priority of information category.Wherein, priority regulation rule refer to configuration to the above-mentioned prediction determined
The rule that priority is adjusted, wherein can specify certain/priority of a little information categories.
Determine that the process providing priority specifically includes: the priority regulation rule that inquiry is preset, acquisition comprises
The assigned priority of selected information classification in the information category of search word;Preferential to prediction according to assigned priority
Level is adjusted, and obtains the offer priority that each information category is corresponding.
Assigned priority refer to certain/a little information categories can configure its priority, such as, specify certain information
The priority of classification is 1, and the priority of another information category is 5 etc..According to assigned priority to prediction
Priority is adjusted referring to, on the basis of prediction priority arrangement order, to which specify priority
Information category is adjusted to corresponding priority position.
Configuration intervention rule arranges and can realize arranging priority regulation rule as shown in Figure 3, Fig. 3 institute
The intervention rule comprising search word shown realizes obtaining selected information classification in the information category comprising search word
Assigned priority.
Step S16: provide a user with the information mated with search word according to the offer priority determined.
As shown in Figure 3, the visitor that information-providing server uses to user according to the offer priority determined
Family end provides the information mated with search word, client present to user.According to provide priority to
When family provides information, can be according to providing priority order from high to low by complete for the information mated with search word
Portion shows user.The categorical measure of the information being supplied to user can also be specified, when specifying categorical measure
Time, order the most from high to low determines the information providing a user with which information category.Such as only
There is provided prediction priority bit in front three, the information category of first five or first eight bits information to user, specifically
It is that former position determines according to the categorical measure specified.
As shown in Figure 4, input as a example by search word mp3 by user, describe above-mentioned information providing method in detail
Implement process.
After user inputs search word mp3, server gets this search word, calls offer mass data and looks into
The search engine asking service obtains the side's of browsing search word product information industry data distribution and the product of single industry
Product informational capacity;I.e. obtain total quantity and each information of the information that the publisher comprising search word mp3 issues
Classification comprises the information content of search word mp3.
Such as shown in Fig. 4, information category or say that product information industry includes mobile phones, car
Video, car dvd player ... etc..(Mp3, car dvd player, information content) in such as figure
Represent that (side's of browsing search word is looked in this product information industry for the side's of browsing search word, product information industry
Ask the information content arrived), the data set order that search engine returns is can according to information content from high to low
Arrangement.
After user inputs search word mp3, server gets this search word, calls navigation algorithm part,
Navigation algorithm part can browse catalogue according to the industry that user's search word provides relatedness strong, obtains the side of browsing
Search word expection industry data distribution and corresponding weight.I.e. obtain the side's of browsing click comprises search word mp3
The total quantity of information and the side of browsing click each information category in comprise the information content of search word mp3.
The most such as shown in Fig. 4, navigation algorithm part obtains (mp3, mobile phones, weight)
Represent that (side's of browsing search word is relevant to this product information industry for the side's of browsing search word, product information industry
Property weight), navigation algorithm part return data set order can also be arranged from high to low by weighted value.
This part actually obtains the side's of browsing industry intent data distribution.
Then it is intended to computing terrace part by the side's of browsing industry and carries out conformity calculation, provide according to search engine
User's search word navigation algorithm anticipatory data that the distribution of product industry truthful data provides with navigation algorithm part
Distribution carries out conformity calculation, such as, can carry out conformity calculation according to the computing formula of above-mentioned f (x).
Such as: search engine provide data include (mp3, mobile phones, product quantity A),
(mp3, car video, product quantity B), (mp3, car dvd player, product quantity C), (mp3,
Navigation&GPS, product quantity D), (mp3, mp3 players, product quantity E), (mp3, speaker,
Product quantity F), (mp3, digital photo frame, product quantity G), (mp3, car mp3 player,
Product quantity H)---}.I.e. search engine provides above-mentioned product industry (the such as mobile phones enumerated
Deng) in comprise the quantity of information of search word mp3.
The data that navigation algorithm part provides include: (mp3, mobile phones, weight a), (and mp3,
Mp3players, weight b), (mp3, navigation&GPS, weight c), (mp3, car mp3 player,
Weight d), (mp3, mp4 players, weight e), (mp3, car video, weight f), (mp3, car
Dvd player, weight g), and (mp3, speaker, weight h)---}.I.e. navigation algorithm provides above-mentioned enumerating
Product industry (such as mobile phones etc.) in comprise the information of search word mp3 and search word mp3
Relevance weight.
Configuration Interference service platform part, intervenes the distribution of navigation algorithm anticipatory data, increases data more reasonable and correct
Property distribution, configuration intervene high priority data consider, f (x) data arrange rearward, and carry out duplicate removal merging and number
According to rearranging, such as: configuration intervene the regulation rule that sets as data acquisition system (mp3, mp3 players,
2), (mp3, mp4 players, 6), (mp3, digital photo frame, 0)---}, wherein, 2,6
The ordinal position that expression is to be arranged, 0 represents that the related data of this data industry to be deleted, and number is intervened in configuration
According to being a small amount of, can not rank commensurability with f (x), the i.e. assigned priority of a predetermined fraction information category.
The f (x) being intended to obtain after computing platform calculates by the side's of browsing industry is { (mp3, mobile
Phones), (mp3, car video), (mp3, mp3 players), (mp3, car dvd player), (mp3,
Mp4 players), (mp3, navigation&GPS), (mp3, speaker), (mp3, car mp3 player),
(mp3, digital photo frame)----after, the result intervened in conjunction with configuration, the number after the merging obtained
{ (mp3, mp3 players, 2), (mp3, mp4 players, 6), (mp3, digital is included according to set
Photo frame, 0), (mp3, mobile phones), (mp3, car video), (mp3, mp3 players),
(mp3, car dvd player), (mp3, mp4 players), (mp3, navigation&GPS), (mp3,
Speaker), (mp3, car mp3 player), (mp3, digital photo frame)----,
After the first step merges, then the duplicate removal carrying out second step processes and each information of the position arrangement of the 3rd step
The offer priority arrangement order of classification is: (mp3, mobile phones), and (mp3, mp3 players),
(mp3, car video), (mp3, car dvd player), (mp3, navigation&GPS), (mp3,
Mp4 players), (mp3, speaker), (mp3, car mp3 player) }.
Based on above-mentioned information providing method, a kind of information provider unit that the embodiment of the present application also provides for, this dress
Putting in the server in the information providing system can being arranged on as shown in Figure 5, the structure of this device such as schemes institute
Show, including: ratio-dependent module 10, weight determination module 20, prediction Priority Determination module 30 with carry
For Priority Determination module 40.
Ratio-dependent module 10, for the search word according to the user's input obtained, Query Information issues data
The information that in storehouse, the publisher of storage issues, determines the information of the search word comprising acquisition in each information category
Quantity accounts for the quantitative proportion of the information total quantity of the search word comprising acquisition.
Weight determination module 20, for the search word according to the user's input obtained, Query Information gathers data
The information that in storehouse, the side of browsing of storage clicks on, according to the information of the search word comprising acquisition in each information category
Clicked number of times, determines the search word of acquisition and the relevance weight of each information category.
Prediction Priority Determination module 30, for each information category determined according to ratio-dependent module 10
Quantitative proportion and the relevance weight of each information category determined of weight determination module 20, determine and comprise
The prediction priority of each information category of the search word obtained.
There is provided Priority Determination module 40, for the prediction determined according to prediction Priority Determination module 30
Priority and the priority regulation rule preset, determine the offer priority of each information category, and according in advance
Survey the Priority Determination module 30 offer priority determined to provide a user with and mate with the search word obtained
Information.
Preferably, aforementioned proportion determines module 10, specifically for determining that store in information publication database sends out
The information that cloth side issues comprises the total quantity of the information of the search word of acquisition, and each information category wraps
First quantity of the information containing search word;Calculate the first quantity that each information category determined is corresponding respectively
With the ratio of above-mentioned total quantity, the information content of the search word obtaining comprising acquisition in each information category accounts for bag
Ratio containing the information total quantity of the search word obtained.
Preferably, above-mentioned weight determination module 20, specifically for determining in information gathering data base the clear of storage
The information that the side of looking at clicks on comprises the clicked total degree of the information of the search word of acquisition, and each information
Classification comprises the first clicked number of clicks of the information of search word;Calculate each letter determined respectively
Cease the first number of clicks corresponding to classification and the ratio of total degree, obtain each information category comprises acquisition
The clicked number of times of information of search word accounts for the dependency of the clicked total degree of information of the search word comprising acquisition
Weight.
Preferably, above-mentioned prediction Priority Determination module 30 is corresponding specifically for obtaining the quantitative proportion preset
The first weighter factor second weighter factor corresponding with relevance weight;Calculate each information category respectively corresponding
The quantitative proportion relevance weight that each information category is corresponding with the sum of products of the first weighter factor add with second
The product of weight factor, and to the product summation obtained, obtain each information category of the search word comprising acquisition
Predictive value;Size according to the predictive value obtained determines each information category of the search word comprising acquisition
Prediction priority.
Preferably, above-mentioned offer Priority Determination module 40, the search word of acquisition is comprised specifically for acquisition
The assigned priority of selected information classification in information category;According to the assigned priority obtained to prediction priority
It is adjusted, the offer priority of each information category of the search word comprising acquisition is provided.
Above-mentioned information provider unit, it is also possible to regard the side's of browsing industry as and be intended to determine platform, can arrange
In the server of network side, it is also possible to being separately provided is an independent equipment.
As shown in Figure 6, in server side, the side's of browsing industry be intended to determine platform and search engine module,
The priority regulation rule module settings that the data statistics module of navigation algorithm and configuration are intervened acts on jointly, poly-
Collection product industry truthful data (search engine), navigation algorithm anticipatory data (navigation algorithm), manually run
Data (configuration is intervened) carry out comprehensive computing, it is achieved to the determination providing priority.Wherein, configuration is intervened
Priority regulation rule arrange to provide interface be set, it is allowed to operation personnel depaly is intervened, input
The assigned priority of some selected information classification, priority regulation rule is inquired about, and inquires about the search of user's input
The regulation rule that word is corresponding, such as assigned priority etc..The side's of browsing industry is intended to determine that platform provides to each
The output interface of e-commerce website.
As shown in Figure 6, the side's of browsing industry is intended to determine that platform provides serviceization mode, and has opening
OpenAPI interface mode, to provide any application call, such as, www.aliexpress.com applies,
Search.aliexpress.com applies, and www.alibaba.com applies, and partner site connects with OpenAPI
Mouth mode is called, and such as, navigates for keyword search results page classification in aliexpress.com,
The keyword search results page presses sales volume sequence etc.;
As long as need to judge the application that the side's of browsing industry is intended to, all directly invoke this platform, it is not necessary to each
Application is controlled oneself and is the most all developed these codes, resource reuse, safeguards and updates conveniently, low cost.
Said method is according to natural Search Results and the comprehensive operation result of user click data, in conjunction with actual products
Result is intervened in the distribution of information industry data and expection industry data distribution and artificial operation, obtains rational data
Distribution, more precisely determines the side's of browsing intention, reduces the side's of browsing browse path, strengthens website viscosity, filters out
The product of more high-quality, prevents swindle, it is provided that a uniform service platform, intervenes with artificial operation data,
And come into force in real time, it is provided that serviceization mode, can be called flexibly, safeguard and extend, improve resource reuse
Property.
Wherein, search engine data merge with navigation algorithm data, can effectively prevent user from clicking maliciously
Swindle, disturbs normal data distribution, configuration intervention data to merge and comes into force in real time, some data can be avoided dilute
In the case of Shuing, accurate data problem, ultimately form a unified platform, in service mode, provide flexibly
To external call.
Described above illustrate and describes the preferred embodiment of the application, but as previously mentioned, it should be understood that this Shen
Please be not limited to form disclosed herein, be not to be taken as the eliminating to other embodiments, and can be used for
Other combinations various, amendment and environment, and can be in invention contemplated scope described herein, by above-mentioned religion
Lead or the technology of association area or knowledge are modified.And the change that those skilled in the art are carried out and change do not take off
From spirit and scope, the most all should be in the protection domain of the application claims.