CN103902597A - Method and device for determining search relevant categories corresponding to target keywords - Google Patents

Abstract

The invention aims to provide a method and a device for determining search relevant categories corresponding to target keywords. The method specifically includes: determining target keyword clusters to which the target keywords belong from one or multiple keyword clusters according to search sorting path information of the target keywords; determining the search relevant categories corresponding to the target keywords for subsequent processing according to the target keywords clusters. Compared with the prior art, the method and the device for determining search relevant categories have the advantages that the target keyword clusters to which the target keywords belong are determined, and the search relevant categories corresponding to the target keywords are determined to be used for the subsequent processing, the search relevant categories corresponding to the keywords are effectively determined, automated testing for batched keyword data is realized, reference is provided for optimizing search sorting of a search engine, and testing efficiency of search engine relevance is improved.

Description

Determine the method and apparatus of the corresponding relevance of searches classification of target keyword

Technical field

The present invention relates to Internet technical field, relate in particular to a kind of for determining the technology of the corresponding relevance of searches classification of target keyword.

Background technology

Current, along with development and the infiltration of internet, applications to user learning, work and life of Internet technology, people are more and more by Network Capture information, as inputted keyword by search engine, search engine returns to by taking certain searching order mode to determine the Search Results that user and keyword match, but the matching degree of the search sequence that the Search Results that search engine returns and user input has affected the accuracy of user's obtaining information to a great extent.Correspondingly, the matching degree of the search sequence of the Search Results that search engine returns if can improve and user's input, can significantly improve the efficiency of user's obtaining information.Therefore, need to carry out effectively assessment test to the correlativity of search engine, as keyword classified according to the matching degree of keyword and Search Results, determine the corresponding relevance of searches classification of keyword, effectively determine the corresponding relevance of searches classification of keyword, and improve the testing efficiency of search engine relevance.

Summary of the invention

The object of this invention is to provide a kind of for determining the method and apparatus of the corresponding relevance of searches classification of target keyword.

According to an aspect of the present invention, provide a kind of for determining the corresponding relevance of searches class of target keyword method for distinguishing, wherein, the method comprises the following steps:

A is according to the searching order routing information of target keyword, determines the target critical term clustering under described target keyword from one or more keyword clustering;

B, according to described target critical term clustering, determines the corresponding relevance of searches classification of described target keyword, for subsequent treatment.

According to another aspect of the present invention, also provide a kind of and determined equipment for the classification of determining the corresponding relevance of searches classification of target keyword, wherein, this classification determines that equipment comprises:

Cluster determining device for according to the searching order routing information of target keyword, is determined the target critical term clustering under described target keyword from one or more keyword clustering;

Classification determining device, for according to described target critical term clustering, determines the corresponding relevance of searches classification of described target keyword, for subsequent treatment.

According to a further aspect of the invention, also provide a kind of computer equipment, wherein, this computer equipment comprise as aforementioned according to a further aspect of the present invention for determining that the classification of the corresponding relevance of searches classification of target keyword determines equipment.

According to a further aspect of the invention, also provide a kind of for determining the search engine of the corresponding relevance of searches classification of target keyword, wherein, this search engine comprise as aforementioned according to a further aspect of the present invention for determining that the classification of the corresponding relevance of searches classification of target keyword determines equipment.

According to a further aspect of the invention, also provide a kind of for determining the search engine plug-in unit of the corresponding relevance of searches classification of target keyword, wherein, this search engine plug-in unit comprise as aforementioned according to a further aspect of the present invention for determining that the classification of the corresponding relevance of searches classification of target keyword determines equipment.

Compared with prior art, the present invention is by determining the target critical term clustering under target keyword, and then the corresponding relevance of searches classification of described target keyword, for subsequent treatment, effectively determine the corresponding relevance of searches classification of keyword thereby realized, and automatic test to batch keyword data, not only for the sequence of Optimizing Search engine search provides reference, and improve the testing efficiency to search engine relevance.And the present invention also can determine the corresponding preferred searching order routing information of target keyword, to adjust the searching order routing information of described target keyword, thereby further realize the sequence of Optimizing Search engine search, improve user profile and obtain efficiency.Further, the present invention also can determine keyword set to be optimized, determine the Optimizing Search sequence routing information of described one or more keywords to be optimized, for adjusting the searching order routing information of described one or more keywords to be optimized, thereby realize further the sequence of Optimizing Search engine search, improved user profile and obtain efficiency.

Accompanying drawing explanation

By reading the detailed description that non-limiting example is done of doing with reference to the following drawings, it is more obvious that other features, objects and advantages of the present invention will become:

Fig. 1 illustrates the equipment schematic diagram for definite corresponding relevance of searches classification of target keyword according to one aspect of the invention;

Fig. 2 illustrates the equipment schematic diagram for definite corresponding relevance of searches classification of target keyword in accordance with a preferred embodiment of the present invention;

Fig. 3 illustrates the method flow diagram for definite corresponding relevance of searches classification of target keyword according to a further aspect of the present invention;

Fig. 4 illustrates the method flow diagram for definite corresponding relevance of searches classification of target keyword in accordance with a preferred embodiment of the present invention.

In accompanying drawing, same or analogous Reference numeral represents same or analogous parts.

Embodiment

Below in conjunction with accompanying drawing, the present invention is described in further detail.

Fig. 1 illustrate according to one aspect of the invention for determining that the classification of the corresponding relevance of searches classification of target keyword determines equipment 1, wherein, classification determines that equipment 1 comprises cluster determining device 11 and classification determining device 12.Particularly, cluster determining device 11 is according to the searching order routing information of target keyword, determines the target critical term clustering under described target keyword from one or more keyword clustering; Classification determining device 12, according to described target critical term clustering, is determined the corresponding relevance of searches classification of described target keyword, for subsequent treatment.At this, the implication of described relevance of searches refers to the matching degree of keyword and Search Results.At this, classification determine equipment 1 include but not limited to the network equipment, subscriber equipment or the network equipment with subscriber equipment by the mutually integrated equipment forming of network.Wherein, the described network equipment includes but not limited to the cloud that network host, single network server, multiple webserver collection or multiple server form.At this, cloud is made up of a large amount of main frames based on cloud computing (Cloud Computing) or the webserver, and wherein, cloud computing is the one of Distributed Calculation, the super virtual machine being made up of the loosely-coupled computing machine collection of a group.It includes but not limited to any electronic product that can carry out with user man-machine interaction by keyboard, telepilot, touch pad or voice-operated device, such as computing machine, smart mobile phone, PDA or IPTV etc. described subscriber equipment.Described network includes but not limited to internet, wide area network, Metropolitan Area Network (MAN), LAN (Local Area Network), VPN network, wireless self-organization network (Ad Hoc network) etc.Those skilled in the art will be understood that above-mentioned classification determines that equipment 1 is only for for example; other network equipments existing or that may occur from now on or subscriber equipment are as applicable to the present invention; also should be included in protection domain of the present invention, and be contained in this at this with way of reference.

Particularly, cluster determining device 11, first according to pre-defined rule, is carried out clustering processing to multiple sample keywords, to obtain one or more keyword clustering; Again according to the searching order routing information of target keyword, from one or more keyword clustering, determine the target critical term clustering under described target keyword.At this, the implication of described searching order refers to that understanding and demand that search engine is inputted keyword to user analyze, use certain algorithm, in the predetermined web database extracting, pick out with user and input the webpage that keyword matches, and provide it to user, it includes but not limited to choose sequence as theme matching degree result, good result is proposed power sequence, cheating is clicked and is suppressed, general rise of prices of the stocks and other securities selected ci poem is got sequence, exercise question/summary assembling sequence etc., wherein, described result is proposed power sequence can comprise many son sequences, as: web sites authority is put forward power, power is put forward in official website, page richness is put forward power, power etc. is put forward in click.At this, described search engine includes but not limited to as baidu search engine of the Google search engine of Google company, company of Baidu etc., and searches the search engine plug-in units such as the MSN ToolBar of despot, Microsoft as the Baidu of the Google ToolBar of Google company, company of Baidu.At this, described searching order routing information for show search engine determine candidate search sort result information process the sort algorithm code path information of process, can use searching order ID of trace route path (Strategy Identifier, and branch mark (Branch Identifier SID), BID) represent, wherein, branch's mark is subordinated to searching order ID of trace route path, in the time that searching order is carried out to fine-grained mark, can be used.Those skilled in the art will be understood that above-mentioned searching order, search engine and searching order routing information are only for giving an example; other searching orders existing or that may occur from now on or search engine or searching order routing information are as applicable to the present invention; also should be included in protection domain of the present invention, and be contained in this at this with way of reference.

Particularly, cluster determining device 11, first according to pre-defined rule, is carried out clustering processing to multiple sample keywords, as adopted the unsupervised learning methods such as k-means, ISODATA, chain method, to obtain described one or more keyword clustering.Wherein, described pre-defined rule includes but not limited to following at least any one:

-according to the each self-corresponding searching order routing information of described multiple sample keywords, described multiple sample keywords are carried out to clustering processing, to obtain described one or more keyword clustering;

-according to the historical search recorded information of the each self-corresponding search subscriber of described multiple sample keyword, described multiple sample keywords are carried out to clustering processing, to obtain described one or more keyword clustering;

-according to described multiple sample keywords, each leisure meets the statistical information in the content of pages information of predetermined quality degree threshold value, described multiple sample keywords is carried out to clustering processing, to obtain described one or more keyword clustering.

For example, when described pre-defined rule comprises according to the each self-corresponding searching order routing information of described multiple sample keywords, described multiple sample keywords are carried out to clustering processing, when obtaining described one or more keyword clustering, suppose that multiple sample keywords are as sample keyword I to VI, its each self-corresponding searching order routing information is as shown in table 1 below, and wherein, S_* represents searching order ID of trace route path or branch's mark in the searching order path of keyword search request processing process:

Sample keyword	Searching order routing information
		I	S_A→S_C→S_D→S_B→S_E→S_G
II	S_A→S_E→S_G→S_C→S_D→S_B
		III	S_A→S_F→S_C→S_D→S_E→S_G
IV	S_A→S_C→S_D→S_F→S_E→S_G
		V	S_A→S_E→S_G→S_C→S_B→S_D
VI	S_M→S_N→S_C→S_B→S_G→S_D

Table 1

Cluster determining device 11 is according to the each self-corresponding searching order routing information of sample keyword I to VI, adopt the unsupervised learning methods such as k-means, ISODATA, chain method, to sample keyword, I to VI carries out clustering processing, obtain one or more keyword clustering as: 1.. the first keyword clustering cluster1, as sample keyword I, III and VI are classified as to a class, 2.. the second keyword clustering cluster2, as keyword II and V are classified as to a class, 3.. the 3rd keyword clustering cluster3, as sample keyword VI is classified as to a class, for another example, when comprising according to described multiple sample keywords each leisure, described pre-defined rule meets the statistical information in the content of pages information of predetermined quality degree threshold value, described multiple sample keywords are carried out to clustering processing, when obtaining described one or more keyword clustering, according to sample keyword I to VI, each leisure meets the statistical information in the content of pages information of predetermined quality degree threshold value to cluster determining device 11, the content of pages information that each leisure meets predetermined quality degree threshold value as sample keyword I to VI is as belonged to the height trusted site page as the frequency information occurring in http://www.sina.com.cn/, adopt k-means, ISODATA, the unsupervised learning methods such as chain method, to sample keyword, I to VI carries out clustering processing, obtain one or more keyword clustering as: 1.. the first keyword clustering cluster1, as by sample keyword I, II and III are classified as a class, 2.. the second keyword clustering cluster2, as sample keyword IV and VI are classified as to a class, 3.. the 3rd keyword clustering cluster3, as sample keyword V is classified as to a class.

Those skilled in the art will be understood that the above-mentioned mode that described multiple sample keywords are carried out to clustering processing is only for giving an example; other existing or modes that described multiple sample keywords are carried out to clustering processing that may occur are from now on as applicable to the present invention; also should be included in protection domain of the present invention, and be contained in this at this with way of reference.

Then, cluster determining device 11 is again according to the searching order routing information of target keyword, determines the target critical term clustering under described target keyword from one or more keyword clustering.Particularly, cluster determining device 11 first by such as search engine, browser, the application programming interfaces (API) of the third party devices such as target keyword equipment are provided, obtain target keyword, or, by the dynamic web page technique such as ASP, JSP, obtain the target keyword that user inputs by subscriber equipment; Then, cluster determining device 11 is again according to the searching order routing information of target keyword, determines the target critical term clustering under described target keyword from one or more keyword clustering.

For example, suppose that test man A is in assessment search engine relevance test process, at test platform keyword input field input target keyword goal-query, cluster determining device 11, by dynamic web page techniques such as ASP, JSP, just can get the target keyword goal-query that test man A inputs by subscriber equipment.

Those skilled in the art will be understood that the above-mentioned mode of obtaining target keyword is only for giving an example; other existing or modes of obtaining target keyword that may occur are from now on as applicable to the present invention; also should be included in protection domain of the present invention, and be contained in this at this with way of reference.

Finally, cluster determining device 11 is again according to the searching order routing information of target keyword, determines the target critical term clustering under described target keyword from one or more keyword clustering.At this, cluster determining device 11 determines that the method for described target critical term clustering includes but not limited to following at least any one:

1) the class searching order routing information of the searching order routing information of described target keyword and described keyword clustering is compared, to determine the target critical term clustering under described target keyword.For example, suppose that cluster determining device 11 carries out after clustering processing sample keyword I to VI as shown in table 1, each keyword clustering obtaining and the class searching order routing information that characterizes described keyword clustering are as shown in table 2 below:

Keyword clustering	Class searching order routing information
		cluster1	S_A→S_C→S_D→S_E→S_G→S_F
cluster2	S_A→S_E→S_G→S_C→S_D→S_B
		cluster3	S_M→S_N→S_C→S_B→S_G→S_D

Table 2

The searching order routing information of supposing the target keyword goal-query that cluster determining device 11 obtains is S_A → S_C → S_D → S_B → S_E → S_G, all identical routing information is maximum with order and searching order ID of trace route path in the class searching order routing information of the first keyword clustering cluster1 for they, and cluster determining device 11 determines that the target critical term clustering under target keyword goal-query is the first keyword clustering cluster1.

2) the searching order routing information of the reference keyword searching order routing information of described target keyword and described keyword clustering being comprised compares, to determine the target critical term clustering under described target keyword.For example, the searching order routing information of supposing the target keyword goal-query that cluster determining device 11 obtains is S_A → S_C → S_D → S_B → S_E → S_G, identical with the searching order routing information of the keyword I in the first keyword clustering cluster1, cluster determining device 11 determines that the target critical term clustering under target keyword goal-query is the first keyword clustering cluster1.

Those skilled in the art will be understood that the above-mentioned mode of determining the affiliated target critical term clustering of described target keyword is only for giving an example; the mode of the target critical term clustering under other definite described target keyword existing or that may occur is from now on as applicable to the present invention; also should be included in protection domain of the present invention, and be contained in this at this with way of reference.

First classification determining device 12 can add up included each of described target critical term clustering with reference to the corresponding relevance of searches descriptor of keyword, determines the corresponding relevance of searches classification of target critical term clustering; , then according to the corresponding relevance of searches classification of described target critical term clustering, determine the corresponding relevance of searches classification of described target keyword, for subsequent treatment then.At this, described relevance of searches classification includes but not limited to as high correlation classification, compared with low correlation classification, uncorrelated classification, cheating keyword classification etc.At this, described subsequent treatment include but not limited to as: 1) target keyword is carried out to Screening Treatment, as whether as test data etc.; 2) the searching order information of optimization aim keyword.Those skilled in the art will be understood that above-mentioned relevance of searches classification and subsequent treatment mode are only for giving an example; other relevance of searches classifications existing or that may occur from now on or subsequent treatment mode are as applicable to the present invention; also should be included in protection domain of the present invention, and be contained in this at this with way of reference.

For example, suppose that the target critical term clustering under the definite target keyword goal-query of cluster determining device 11 is the first keyword clustering cluster1, and the included sample keyword I of keyword clustering cluster1, it is high that the corresponding relevance of searches descriptor of II and III is respectively correlativity, correlativity is high, correlativity is low, due to the high correlativity descriptor of correlativity account for correlativity descriptor total quantity ratio meet be greater than threshold value as 0.65, classification determining device 12 determines that the corresponding relevance of searches classification of target keyword goal-query is the high classification of correlativity.For another example, suppose that the target critical term clustering under the definite target keyword goal-query of cluster determining device 11 is the first keyword clustering cluster2, and the corresponding relevance of searches descriptor of the included sample keyword IV of keyword clustering cluster2 and VI is respectively, correlativity is low, correlativity is low, the ratio that accounts for correlativity descriptor total quantity due to the low correlativity descriptor of correlativity meets and is greater than threshold value as 0.65, and classification determining device 12 determines that the corresponding relevance of searches classification of target keyword goal-query is the low classification of correlativity.

Those skilled in the art will be understood that the above-mentioned mode of determining the corresponding relevance of searches classification of described target keyword is only for giving an example; the mode of the corresponding relevance of searches classification of other definite described target keyword existing or that may occur is from now on as applicable to the present invention; also should be included in protection domain of the present invention, and be contained in this at this with way of reference.

Classification determines between each device of equipment 1 it is constant work.Particularly, cluster determining device 11 continues according to the searching order routing information of target keyword, the target critical term clustering from one or more keyword clustering under definite described target keyword; Classification determining device 12 continues according to described target critical term clustering, determines the corresponding relevance of searches classification of described target keyword, for subsequent treatment.At this, those skilled in the art be to be understood that " continuing " refer to classification determine equipment 1 each device constantly carry out respectively target critical term clustering determine and relevance of searches classification determine, until classification determine equipment 1 stop in a long time target critical term clustering determine.

Preferably, described keyword clustering comprises the class searching order routing information for characterizing described keyword clustering, and cluster determining device 11 comprises comparing unit (not shown), similarity determining unit (not shown) and cluster determining unit (not shown).Below with reference to Fig. 1, the preferred embodiment is described: comparing unit compares the searching order routing information of described target keyword and the corresponding described class searching order routing information of described one or more keyword clustering, to determine the smallest edit distance of searching order routing information and described class searching order routing information of described target keyword; Similarity determining unit, according to described smallest edit distance, is determined the searching order routing information of described target keyword and the sequence similarity of paths of described class searching order routing information; Cluster determining unit, according to described sequence similarity of paths, is determined described target critical term clustering.

Particularly, comparing unit is first according to pre-defined rule, multiple sample keywords are carried out to clustering processing, as adopted the unsupervised learning methods such as k-means, ISODATA, chain method, determine described one or more keyword clustering to obtain described one or more keyword clustering.At this, comparing unit obtains the mode of described one or more keyword clustering and cluster determining device 11, and to obtain the mode of described one or more keyword clustering same or similar, for simplicity's sake, thus do not repeat them here, and comprise by reference therewith.

Then, comparing unit compares the searching order routing information of described target keyword and the corresponding described class searching order routing information of described one or more keyword clustering, to determine the smallest edit distance of searching order routing information and described class searching order routing information of described target keyword.For example, the searching order routing information of supposing the target keyword goal-query that comparing unit obtains is S_A → S_C → S_D → S_B → S_E → S_G, and comparing unit is determined described one or more keyword clustering as shown in Table 2 above, comparing unit carries out serializing by searching order routing information S_A → S_C → S_D → S_B → S_E → S_G of target keyword goal-query and obtains character string goal-string=" ACDBEG ", equally corresponding keyword clustering cluster1 to cluster3 described class searching order routing information is carried out to serializing and obtain corresponding character string as cluster1-string=" ACDEGF ", cluster2-string=" AEGSDB ", cluster3-string=" MNCBGD ", then, comparing unit passes through such as dynamic programming, the smallest edit distance algorithms such as matrix method, the corresponding character string goal-string=of searching order routing information " ACDBEG " that calculates respectively target keyword goal-query with the corresponding character string of the each self-corresponding described class searching order routing information of keyword clustering cluster1 to cluster3 as cluster1-string=" ACDEGF ", cluster2-string=" AEGSDB ", the smallest edit distance of cluster3-string=" MNCBGD ", as obtain target keyword goal-query and the right smallest edit distance of keyword clustering cluster1 to cluster3 is respectively: 2,6 and 6.

Those skilled in the art will be understood that the above-mentioned mode of determining described smallest edit distance is only for giving an example; the mode of other definite described smallest edit distance existing or that may occur is from now on as applicable to the present invention; also should be included in protection domain of the present invention, and be contained in this at this with way of reference.

Similarity determining unit, according to described smallest edit distance, is determined the searching order routing information of described target keyword and the sequence similarity of paths of described class searching order routing information.For example, connect example, the described smallest edit distance that similarity determining unit is determined according to comparing unit, determine described sequence similarity of paths by following formula (1):

$r=\frac{1}{d+1}---\left(1\right)$

Wherein, d is smallest edit distance, and similarity determining unit determines that according to above-mentioned formula (1) the searching order routing information of target keyword goal-query is respectively with the sequence similarity of paths of keyword clustering cluster1 to cluster3 described class searching order routing information separately respectively: 1/3,1/7 and 1/7.

Preferably, the described smallest edit distance that similarity determining unit is determined according to comparing unit, determine described sequence similarity of paths by following formula (2):

$r=\frac{\mathrm{\α}\×\stackrel{\‾}{l}}{d+1}---\left(2\right)$

Wherein, α is normalization coefficient,

by the average string length of the corresponding character string of class searching order routing information, d is smallest edit distance, and wherein, normalization coefficient α can calculate by following formula (3):

$\mathrm{\α}=\frac{{x-x}_{\min }}{x_{\max }-x_{\min }}---\left(3\right)$

Wherein, x representation class searching order routing information the statistical length of corresponding character string in test process, if α=0.5, and the each self-corresponding class searching order routing information of keyword clustering cluster1 to cluster3 the average string length of corresponding character string be 6, similarity determining unit can determine that according to above-mentioned formula (2) the searching order routing information of target keyword goal-query is respectively with the sequence similarity of paths of keyword clustering cluster1 to cluster3 described class searching order routing information separately respectively: 1,3/7 and 3/7.

Those skilled in the art will be understood that the above-mentioned mode of determining described sequence similarity of paths is only for giving an example; the mode of other existing or definite described sequence similarity of paths that may occur is from now on as applicable to the present invention; also should be included in protection domain of the present invention, and be contained in this at this with way of reference.

Cluster determining unit, according to described sequence similarity of paths, is determined described target critical term clustering, as described in determining under target keyword as described in target critical term clustering be as described in sequence similarity of paths meet predetermined threshold as 0.8 corresponding keyword clustering.For example, connect example, similarity determining unit determines that the searching order routing information of target keyword goal-query is respectively with the sequence similarity of paths of keyword clustering cluster1 to cluster3 described class searching order routing information separately respectively: 1,3/7 and 3/7, and cluster determining unit determines that the described target critical term clustering under target keyword goal-query is cluster1.

Preferably, first cluster determining device 11 also can obtain pending keyword one or more to be measured, using as described target keyword; Then, according to the searching order routing information of described target keyword, the target critical term clustering from one or more keyword clustering under definite described target keyword; Classification determining device 12 also can, first according to described target critical term clustering, be determined the corresponding relevance of searches classification of described target keyword; Then,, according to described relevance of searches classification, described target keyword is carried out to Screening Treatment.

Particularly, cluster determining device 11 also can be first by such as search engine, browser, the application programming interfaces (API) of the third party devices such as keyword equipment to be measured are provided, obtain pending keyword one or more to be measured, using as described target keyword; Then, according to the searching order routing information of described target keyword, the target critical term clustering from one or more keyword clustering under definite described target keyword.At this, the mode of the target critical term clustering under the mode of the target critical term clustering under the definite described target keyword of cluster determining device 11 and the fixed described target keyword of aforementioned cluster determining device 11 is same or similar, for simplicity's sake, therefore do not repeat them here, and comprise by reference therewith.

Then, classification determining device 12 also can, first according to described target critical term clustering, be determined the corresponding relevance of searches classification of described target keyword.At this, the mode of classification determining device 12 definite corresponding relevance of searches classifications of described target keyword and aforementioned classification determining device 12 determine that the mode of the corresponding relevance of searches classifications of described target keyword is same or similar, for simplicity's sake, therefore do not repeat them here, and comprise by reference therewith.

Then, classification determining device 12, according to described relevance of searches classification, is carried out Screening Treatment to described target keyword.For example, suppose that the described target keyword that cluster determining device 11 is obtained comprises as query1, query2, query3 and query4, and classification determining device 12 is determined this target keyword query1, query2, it is high that the described relevance of searches classification that query3 and query4 are corresponding is respectively correlativity, in correlativity, correlativity is low, correlativity is high, classification determining device 12 is according to target keyword query1, query2, the each self-corresponding described relevance of searches classification of query3 and query4, it is carried out to Screening Treatment, as the keyword query3 that belongs to the low classification of correlativity is screened from keyword set to be measured, it is carried out to later stage searching order information optimization.

Preferably, classification determines that equipment 1 also comprises set determining device (not shown) and path optimizing determining device (not shown).Particularly, set determining device is determined the corresponding keyword set to be optimized of described keyword clustering; The corresponding common search sequence of the keyword one or more to be optimized routing information that path optimizing determining device comprises according to described keyword set to be optimized, determine the Optimizing Search sequence routing information of described one or more keywords to be optimized, for adjusting the searching order routing information of described one or more keywords to be optimized.

Particularly, the relevance of searches descriptor of all keywords that set determining device can comprise according to keyword clustering, as high in relevance of searches, relevance of searches is low etc., finds out the classification that relevance of searches is low, using as the corresponding keyword set to be optimized of described keyword clustering.At this, described keyword set to be optimized is corresponding to the keyword that belongs to the low classification of relevance of searches.For example, the keyword of supposing to belong in keyword clustering cluster1 the low classification of relevance of searches is sample keyword III, the keyword that belongs to the low classification of relevance of searches in keyword clustering cluster2 is sample keyword IV and VI, in keyword clustering cluster3, do not belong to the keyword of the low classification of relevance of searches, gather all keywords that belong to the low classification of relevance of searches that determining device can comprise keyword clustering cluster1 to cluster3 as described keyword set to be optimized, as comprise sample keyword III, IV and VI.

Preferably, set determining device also can compare the actual search results relevant information of all keywords included described keyword clustering and system index information, determines described keyword set to be optimized from described all keywords.At this, described actual search results relevant information includes but not limited to as returned to Search Results quantity, obtain click volume, return website authority, return to the quality degree of the content of pages of website etc.At this, described system index information comprises as returned to Search Results quantity, returning to the authority of website etc.For example, the included all keywords of described keyword clustering cluster1 to cluster3 that set determining device can be determined cluster determining device 11 are that actual search results relevant information and the system index information of sample keyword I to VI compares, from described all keywords, determine described keyword set to be optimized, as the actual search results relevant information of sample keyword I to VI being discontented with to the keyword of pedal system indication information, as described keyword set to be optimized.

Those skilled in the art will be understood that the above-mentioned mode of determining described keyword set to be optimized is only for giving an example; the mode of other definite described keyword set to be optimized existing or that may occur is from now on as applicable to the present invention; also should be included in protection domain of the present invention, and be contained in this at this with way of reference.

Then, the corresponding common search sequence of the keyword one or more to be optimized routing information that path optimizing determining device comprises according to described keyword set to be optimized, determine the Optimizing Search sequence routing information of described one or more keywords to be optimized, for adjusting the searching order routing information of described one or more keywords to be optimized.For example, connect example, the keyword set described to be optimized that set determining device is determined comprises sample keyword III, IV and VI, path optimizing determining device can be according to sample keyword III, the corresponding common search sequence of IV and VI routing information is as S_C → S_D → S-G, as described Optimizing Search sequence routing information, for adjusting the searching order routing information of described one or more keywords to be optimized, as delete sample keyword III, this Optimizing Search sequence routing information that the searching order routing information of IV and VI comprises, or, by sample keyword III, this Optimizing Search sequence routing information that the searching order routing information of IV and VI comprises is replaced by other searching order routing informations common search sequence as corresponding in the keyword of the high classification of relevance of searches routing information.

Preferably, can determine equipment 1 for the classification of determining the corresponding relevance of searches classification of target keyword by above-mentioned, combine with existing search engine, form a kind of new search engine, existing search engine includes but not limited to as baidu search engine of the Google search engine of Google company, company of Baidu etc.

Preferably, can determine equipment 1 for the classification of determining the corresponding relevance of searches classification of target keyword by above-mentioned, combine with existing search engine plug-in unit, form a kind of new search engine plug-in unit, existing including but not limited to searched the search engine plug-in units such as the MSN ToolBar of despot, Microsoft as the Baidu of the Google ToolBar of Google company, company of Baidu.

Fig. 2 illustrates the equipment schematic diagram for definite corresponding relevance of searches classification of target keyword in accordance with a preferred embodiment of the present invention, wherein, classification determines that equipment 1 comprises cluster determining device 11 ', classification determining device 12 ', preferred path determining device 13 ', adjusting gear 14 ' and generator 15 '.Particularly, cluster determining device 11 ' is according to the searching order routing information of target keyword, determines the target critical term clustering under described target keyword from one or more keyword clustering; Classification determining device 12 ', according to described target critical term clustering, is determined the corresponding relevance of searches classification of described target keyword, for subsequent treatment; Preferred path determining device 13 ' is determined the corresponding preferred searching order routing information of described target keyword; Adjusting gear 14 ', according to described preferred searching order routing information, is adjusted the searching order routing information of described target keyword; If search sequence and described target keyword match, generator 15 ', according to the searching order routing information of the described target keyword after adjusting, offers the corresponding user of described search sequence by corresponding described search sequence Search Results.At this, cluster determining device 11 ', classification determining device 12 ' are same or similar with corresponding intrument shown in Fig. 1 respectively, so locate to repeat no more, and mode is by reference contained in this.

Particularly, preferred path determining device 13 ' is determined the corresponding preferred searching order routing information of described target keyword.At this, preferred path determining device 13 ' determines that the mode of described preferred searching order routing information includes but not limited to following at least any one:

1) by included described target critical term clustering each with reference to the corresponding common search sequence of keyword routing information, as described preferred searching order routing information.For example, suppose that cluster determining device 11 ' determines that the target critical term clustering under target keyword goal-query is keyword clustering cluster1, preferred path determining device 13 ' can be that the corresponding common search sequence of sample keyword I, III and VI routing information is as S_A → S_C → S_D → S_E → S_G, as described preferred searching order routing information with reference to keyword by included cluster1 each.

2) included each of described target critical term clustering carried out to statistical study processing with reference to the corresponding searching order routing information of keyword, to determine described preferred searching order routing information, as statistics draws the searching order path frequency information that keyword is passed by, the frequency is met to the path of predetermined threshold, or using high frequency searching order path as described preferred searching order routing information.For example, connect example, preferred path determining device 13 ' can be that sample keyword I, III and the corresponding searching order routing information of VI carry out statistical study processing with reference to keyword by included cluster1 each, the frequency is met to predetermined threshold if the searching order path of frequency of occurrence satisfied 2 is as S_A → S_C → S_D → S_E → S_G → S_F, as described preferred searching order routing information.

Those skilled in the art will be understood that the above-mentioned mode of determining described preferred searching order routing information is only for giving an example; the mode of other existing or definite described preferred searching order routing informations that may occur is from now on as applicable to the present invention; also should be included in protection domain of the present invention, and be contained in this at this with way of reference.

Adjusting gear 14 ', according to described preferred searching order routing information, is adjusted the searching order routing information of described target keyword.For example, suppose that the described preferred searching order routing information that preferred path determining device 13 ' is determined is S_A → S_C → S_D → S_E → S_G, adjusting gear 14 ' is according to this preferred searching order routing information, by such as adjustment algorithm or machine learning model as SVM model, the searching order routing information of adjustment aim keyword goal-query, as being adjusted into the searching order routing information of target keyword goal-query as described in preferential execution preferably searching order routing information S_A → S_C → S_D → S_E → S_G.

Those skilled in the art will be understood that the above-mentioned mode of determining described keyword set to be optimized is only for giving an example; the mode of other definite described keyword set to be optimized existing or that may occur is from now on as applicable to the present invention; also should be included in protection domain of the present invention, and be contained in this at this with way of reference.

If search sequence and described target keyword match, generator 15 ', according to the searching order routing information of the described target keyword after adjusting, offers the corresponding user of described search sequence by corresponding described search sequence Search Results.Particularly, generator 15 ' first obtains search sequence; Then, judge whether described search sequence and described target keyword match, if coupling, generator 15 ', according to the searching order routing information of the described target keyword after adjusting, offers the corresponding user of described search sequence by corresponding described search sequence Search Results.At this, the implication of described coupling comprises search sequence and described target keyword is in full accord, search sequence is contained in described target keyword.

Particularly, generator 15 ' is first by dynamic web page techniques such as ASP, JSP, or the application programming interfaces that provide by search engine (API), obtains the search sequence that user inputs by subscriber equipment.For example, if search subscriber B inputs keyword " fresh flower " by its PC equipment in search engine search column, press "enter" key", generator 15 ', by dynamic web page techniques such as ASP, JSP or PHP, just can get the keyword " fresh flower " of search subscriber B input.It will be understood by those skilled in the art that the above-mentioned mode of obtaining search sequence is only for giving an example; other existing or modes of obtaining search sequence that may occur are from now on as applicable to the present invention; also should be included in protection domain of the present invention, and be contained in this at this with way of reference.

Then, the described target keyword that generator 15 ' obtains according to cluster determining device 11 ', by the mode of text comparison, judges whether described search sequence and described target keyword match.

If coupling, generator 15 ', according to the searching order routing information of the described target keyword after adjusting, offers the corresponding user of described search sequence by corresponding described search sequence Search Results.For example, connect example, suppose that described target keyword that cluster determining device 11 ' obtains is as " fresh flower ", " fresh flower express delivery " etc., generator 15 ' judges that search sequence " fresh flower " and target keyword are as " fresh flower ", " fresh flower express delivery " matches, then, generator 15 ' by search engine according to the corresponding Search Results of searching order routing information of this target keyword after adjusting as " fresh flower Baidu mhkc ", " fresh flower picture materials sky is off line ", as the corresponding Search Results of search sequence " fresh flower ", and pass through such as ASP, the dynamic web page technique such as JSP or PHP, or the communication mode of other agreements, as the communication protocol such as http or https, offering the corresponding user of described search sequence is user B, as this user's subscriber equipment, browse for user.

Fig. 3 illustrates the method flow diagram for definite corresponding relevance of searches classification of target keyword according to a further aspect of the present invention.

Particularly, in step S1, classification determines that equipment 1 is according to the searching order routing information of target keyword, the target critical term clustering from one or more keyword clustering under definite described target keyword; In step S2, classification determines that equipment 1 is according to described target critical term clustering, determines the corresponding relevance of searches classification of described target keyword, for subsequent treatment.At this, the implication of described relevance of searches refers to the matching degree of keyword and Search Results.At this, classification determine equipment 1 include but not limited to the network equipment, subscriber equipment or the network equipment with subscriber equipment by the mutually integrated equipment forming of network.Wherein, the described network equipment includes but not limited to the cloud that network host, single network server, multiple webserver collection or multiple server form.At this, cloud is made up of a large amount of main frames based on cloud computing (Cloud Computing) or the webserver, and wherein, cloud computing is the one of Distributed Calculation, the super virtual machine being made up of the loosely-coupled computing machine collection of a group.It includes but not limited to any electronic product that can carry out with user man-machine interaction by keyboard, telepilot, touch pad or voice-operated device, such as computing machine, smart mobile phone, PDA or IPTV etc. described subscriber equipment.Described network includes but not limited to internet, wide area network, Metropolitan Area Network (MAN), LAN (Local Area Network), VPN network, wireless self-organization network (Ad Hoc network) etc.Those skilled in the art will be understood that above-mentioned classification determines that equipment 1 is only for for example; other network equipments existing or that may occur from now on or subscriber equipment are as applicable to the present invention; also should be included in protection domain of the present invention, and be contained in this at this with way of reference.

Particularly, in step S1, classification determines that equipment 1 is first according to pre-defined rule, and multiple sample keywords are carried out to clustering processing, to obtain one or more keyword clustering; Again according to the searching order routing information of target keyword, from one or more keyword clustering, determine the target critical term clustering under described target keyword.At this, the implication of described searching order refers to that understanding and demand that search engine is inputted keyword to user analyze, use certain algorithm, in the predetermined web database extracting, pick out with user and input the webpage that keyword matches, and provide it to user, it includes but not limited to choose sequence as theme matching degree result, good result is proposed power sequence, cheating is clicked and is suppressed, general rise of prices of the stocks and other securities selected ci poem is got sequence, exercise question/summary assembling sequence etc., wherein, described result is proposed power sequence can comprise many son sequences, as: web sites authority is put forward power, power is put forward in official website, page richness is put forward power, power etc. is put forward in click.At this, described search engine includes but not limited to as baidu search engine of the Google search engine of Google company, company of Baidu etc., and searches the search engine plug-in units such as the MSN ToolBar of despot, Microsoft as the Baidu of the Google ToolBar of Google company, company of Baidu.At this, described searching order routing information for show search engine determine candidate search sort result information process the sort algorithm code path information of process, can use searching order ID of trace route path (Strategy Identifier, and branch mark (Branch Identifier SID), BID) represent, wherein, branch's mark is subordinated to searching order ID of trace route path, in the time that searching order is carried out to fine-grained mark, can be used.Those skilled in the art will be understood that above-mentioned searching order, search engine and searching order routing information are only for giving an example; other searching orders existing or that may occur from now on or search engine or searching order routing information are as applicable to the present invention; also should be included in protection domain of the present invention, and be contained in this at this with way of reference.

Particularly, in step S1, classification determines that equipment 1 is first according to pre-defined rule, and multiple sample keywords are carried out to clustering processing, as adopted the unsupervised learning methods such as k-means, ISODATA, chain method, to obtain described one or more keyword clustering.Wherein, described pre-defined rule includes but not limited to following at least any one:

-according to the each self-corresponding searching order routing information of described multiple sample keywords, described multiple sample keywords are carried out to clustering processing, to obtain described one or more keyword clustering;

-according to the historical search recorded information of the each self-corresponding search subscriber of described multiple sample keyword, described multiple sample keywords are carried out to clustering processing, to obtain described one or more keyword clustering;

-according to described multiple sample keywords, each leisure meets the statistical information in the content of pages information of predetermined quality degree threshold value, described multiple sample keywords is carried out to clustering processing, to obtain described one or more keyword clustering.

For example, when described pre-defined rule comprises according to the each self-corresponding searching order routing information of described multiple sample keywords, described multiple sample keywords are carried out to clustering processing, when obtaining described one or more keyword clustering, suppose that multiple sample keywords are as sample keyword I to VI, its each self-corresponding searching order routing information is as shown in table 3 below, and wherein, S_* represents searching order ID of trace route path or branch's mark in the searching order path of keyword search request processing process:

Sample keyword	Searching order routing information
		I	S_A→S_C→S_D→S_B→S_E→S_G
II	S_A→S_E→S_G→S_C→S_D→S_B
		III	S_A→S_F→S_C→S_D→S_E→S_G
IV	S_A→S_C→S_D→S_F→S_E→S_G
		V	S_A→S_E→S_G→S_C→S_B→S_D
VI	S_M→S_N→S_C→S_B→S_G→S_D

Table 3

In step S1, classification determines that equipment 1 is according to the each self-corresponding searching order routing information of sample keyword I to VI, adopt the unsupervised learning methods such as k-means, ISODATA, chain method, to sample keyword, I to VI carries out clustering processing, obtain one or more keyword clustering as: 1.. the first keyword clustering cluster1, as sample keyword I, III and VI are classified as to a class, 2.. the second keyword clustering cluster2, as keyword II and V are classified as to a class, 3.. the 3rd keyword clustering cluster3, as sample keyword VI is classified as to a class, for another example, when comprising according to described multiple sample keywords each leisure, described pre-defined rule meets the statistical information in the content of pages information of predetermined quality degree threshold value, described multiple sample keywords are carried out to clustering processing, when obtaining described one or more keyword clustering, in step S1, classification determines that each leisure meets the statistical information in the content of pages information of predetermined quality degree threshold value to equipment 1 according to sample keyword I to VI, the content of pages information that each leisure meets predetermined quality degree threshold value as sample keyword I to VI is as belonged to the height trusted site page as the frequency information occurring in http://www.sina.com.cn/, adopt k-means, ISODATA, the unsupervised learning methods such as chain method, to sample keyword, I to VI carries out clustering processing, obtain one or more keyword clustering as: 1.. the first keyword clustering cluster1, as by sample keyword I, II and III are classified as a class, 2.. the second keyword clustering cluster2, as sample keyword IV and VI are classified as to a class, 3.. the 3rd keyword clustering cluster3, as sample keyword V is classified as to a class.

Those skilled in the art will be understood that the above-mentioned mode that described multiple sample keywords are carried out to clustering processing is only for giving an example; other existing or modes that described multiple sample keywords are carried out to clustering processing that may occur are from now on as applicable to the present invention; also should be included in protection domain of the present invention, and be contained in this at this with way of reference.

Then, in step S1, classification determines that equipment 1 is again according to the searching order routing information of target keyword, the target critical term clustering from one or more keyword clustering under definite described target keyword.Particularly, in step S1, classification determine equipment 1 first by such as search engine, browser, the application programming interfaces (API) of the third party devices such as target keyword equipment are provided, obtain target keyword, or, by the dynamic web page technique such as ASP, JSP, obtain the target keyword that user inputs by subscriber equipment; Then, cluster determining device 11 is again according to the searching order routing information of target keyword, determines the target critical term clustering under described target keyword from one or more keyword clustering.

For example, suppose that test man A is in assessment search engine relevance test process, at test platform keyword input field input target keyword goal-query, in step S1, classification determines that equipment 1, by dynamic web page techniques such as ASP, JSP, just can get the target keyword goal-query that test man A inputs by subscriber equipment.

Those skilled in the art will be understood that the above-mentioned mode of obtaining target keyword is only for giving an example; other existing or modes of obtaining target keyword that may occur are from now on as applicable to the present invention; also should be included in protection domain of the present invention, and be contained in this at this with way of reference.

Finally, in step S1, classification determines that equipment 1 is again according to the searching order routing information of target keyword, the target critical term clustering from one or more keyword clustering under definite described target keyword.At this, in step S1, classification determines that the method for equipment 1 definite described target critical term clustering includes but not limited to following at least any one:

1) the class searching order routing information of the searching order routing information of described target keyword and described keyword clustering is compared, to determine the target critical term clustering under described target keyword.For example, suppose in step S1, classification determines that equipment 1 carries out after clustering processing sample keyword I to VI as shown in table 3, and each keyword clustering obtaining and the class searching order routing information that characterizes described keyword clustering are as shown in table 4 below:

Keyword clustering	Class searching order routing information
		cluster1	S_A→S_C→S_D→S_E→S_G→S_F
cluster2	S_A→S_E→S_G→S_C→S_D→S_B
		cluster3	S_M→S_N→S_C→S_B→S_G→S_D

Table 4

Suppose in step S1, classification determines that the searching order routing information of the target keyword goal-query that equipment 1 obtains is S_A → S_C → S_D → S_B → S_E → S_G, all identical routing information is maximum with order and searching order ID of trace route path in the class searching order routing information of the first keyword clustering cluster1 for they,, in step S1, classification determines that the target critical term clustering under the definite target keyword goal-query of equipment 1 is the first keyword clustering cluster1.

2) the searching order routing information of the reference keyword searching order routing information of described target keyword and described keyword clustering being comprised compares, to determine the target critical term clustering under described target keyword.For example, suppose in step S1, classification determines that the searching order routing information of the target keyword goal-query that equipment 1 obtains is S_A → S_C → S_D → S_B → S_E → S_G, identical with the searching order routing information of the keyword I in the first keyword clustering cluster1,, in step S1, classification determines that the target critical term clustering under the definite target keyword goal-query of equipment 1 is the first keyword clustering cluster1.

Those skilled in the art will be understood that the above-mentioned mode of determining the affiliated target critical term clustering of described target keyword is only for giving an example; the mode of the target critical term clustering under other definite described target keyword existing or that may occur is from now on as applicable to the present invention; also should be included in protection domain of the present invention, and be contained in this at this with way of reference.

In step S2, classification determines that first equipment 1 can add up included each of described target critical term clustering with reference to the corresponding relevance of searches descriptor of keyword, determines the corresponding relevance of searches classification of target critical term clustering; , then according to the corresponding relevance of searches classification of described target critical term clustering, determine the corresponding relevance of searches classification of described target keyword, for subsequent treatment then.At this, described relevance of searches classification includes but not limited to as high correlation classification, compared with low correlation classification, uncorrelated classification, cheating keyword classification etc.At this, described subsequent treatment include but not limited to as: 1) target keyword is carried out to Screening Treatment, as whether as test data etc.; 2) the searching order information of optimization aim keyword.Those skilled in the art will be understood that above-mentioned relevance of searches classification and subsequent treatment mode are only for giving an example; other relevance of searches classifications existing or that may occur from now on or subsequent treatment mode are as applicable to the present invention; also should be included in protection domain of the present invention, and be contained in this at this with way of reference.

For example, suppose in step S1, classification determines that the target critical term clustering under the definite target keyword goal-query of equipment 1 is the first keyword clustering cluster1, and the included sample keyword I of keyword clustering cluster1, it is high that the corresponding relevance of searches descriptor of II and III is respectively correlativity, correlativity is high, correlativity is low, due to the high correlativity descriptor of correlativity account for correlativity descriptor total quantity ratio meet be greater than threshold value as 0.65, in step S2, classification determines that the definite corresponding relevance of searches classification of target keyword goal-query of equipment 1 is the high classification of correlativity.For another example, suppose in step S1, classification determines that the target critical term clustering under the definite target keyword goal-query of equipment 1 is the first keyword clustering cluster2, and that the corresponding relevance of searches descriptor of the included sample keyword IV of keyword clustering cluster2 and VI is respectively correlativity is low, correlativity is low, due to the low correlativity descriptor of correlativity account for correlativity descriptor total quantity ratio meet be greater than threshold value as 0.65, in step S2, classification determines that the definite corresponding relevance of searches classification of target keyword goal-query of equipment 1 is the low classification of correlativity.

Those skilled in the art will be understood that the above-mentioned mode of determining the corresponding relevance of searches classification of described target keyword is only for giving an example; the mode of the corresponding relevance of searches classification of other definite described target keyword existing or that may occur is from now on as applicable to the present invention; also should be included in protection domain of the present invention, and be contained in this at this with way of reference.

Classification determines between each step of equipment 1 it is constant work.Particularly, in step S1, classification determines that equipment 1 continues according to the searching order routing information of target keyword, the target critical term clustering from one or more keyword clustering under definite described target keyword; In step S2, classification determines that equipment 1 continues according to described target critical term clustering, determines the corresponding relevance of searches classification of described target keyword, for subsequent treatment.At this, those skilled in the art be to be understood that " continuing " refer to classification determine each step of equipment 1 constantly carry out respectively target critical term clustering determine and relevance of searches classification determine, until classification determine equipment 1 stop in a long time target critical term clustering determine.

Preferably, described keyword clustering comprises the class searching order routing information for characterizing described keyword clustering, and step S1 comprises step S11 (not shown), step S12 (not shown) and step S13 (not shown).Below with reference to Fig. 3, the preferred embodiment is described: in step S11, classification determines that equipment 1 compares the searching order routing information of described target keyword and the corresponding described class searching order routing information of described one or more keyword clustering, to determine the smallest edit distance of searching order routing information and described class searching order routing information of described target keyword; In step S12, classification determines that equipment 1 is according to described smallest edit distance, determines the searching order routing information of described target keyword and the sequence similarity of paths of described class searching order routing information; In step S13, classification determines that equipment 1, according to described sequence similarity of paths, determines described target critical term clustering.

Particularly, in step S11, classification determines that equipment 1 is first according to pre-defined rule, multiple sample keywords are carried out to clustering processing, as adopt the unsupervised learning method such as k-means, ISODATA, chain method, determine described one or more keyword clustering to obtain described one or more keyword clustering.At this, in step S11, classification determines that equipment 1 obtains the mode of described one or more keyword clustering with in step S1, it is same or similar that classification determines that equipment 1 obtains the mode of described one or more keyword clustering, for simplicity's sake, therefore do not repeat them here, and comprise by reference therewith.

Then, in step S11, classification determines that equipment 1 compares the searching order routing information of described target keyword and the corresponding described class searching order routing information of described one or more keyword clustering, to determine the smallest edit distance of searching order routing information and described class searching order routing information of described target keyword.For example, suppose in step S11, classification determines that the searching order routing information of the target keyword goal-query that equipment 1 obtains is S_A → S_C → S_D → S_B → S_E → S_G, and in step S11, classification is determined the definite described one or more keyword clustering of equipment 1 as shown in Table 2 above, in step S11, classification determines that equipment 1 carries out serializing by searching order routing information S_A → S_C → S_D → S_B → S_E → S_G of target keyword goal-query and obtains character string goal-string=" ACDBEG ", equally corresponding keyword clustering cluster1 to cluster3 described class searching order routing information is carried out to serializing and obtain corresponding character string as cluster1-string=" ACDEGF ", cluster2-string=" AEGSDB ", cluster3-string=" MNCBGD ", then, in step S11, classification determines that equipment 1 passes through such as dynamic programming, the smallest edit distance algorithms such as matrix method, the corresponding character string goal-string=of searching order routing information " ACDBEG " that calculates respectively target keyword goal-query with the corresponding character string of the each self-corresponding described class searching order routing information of keyword clustering cluster1 to cluster3 as cluster1-string=" ACDEGF ", cluster2-string=" AEGSDB ", the smallest edit distance of cluster3-string=" MNCBGD ", as obtain target keyword goal-query and the right smallest edit distance of keyword clustering cluster1 to cluster3 is respectively: 2, 6 and 6.

Those skilled in the art will be understood that the above-mentioned mode of determining described smallest edit distance is only for giving an example; the mode of other definite described smallest edit distance existing or that may occur is from now on as applicable to the present invention; also should be included in protection domain of the present invention, and be contained in this at this with way of reference.

In step S12, classification determines that equipment 1 is according to described smallest edit distance, determines the searching order routing information of described target keyword and the sequence similarity of paths of described class searching order routing information.For example, connect example, in step S12, classification is determined the described smallest edit distance that equipment 1 is determined according to comparing unit, determines described sequence similarity of paths by following formula (4):

$r=\frac{1}{d+1}---\left(4\right)$

Wherein, d is smallest edit distance, and similarity determining unit determines that according to above-mentioned formula (4) the searching order routing information of target keyword goal-query is respectively with the sequence similarity of paths of keyword clustering cluster1 to cluster3 described class searching order routing information separately respectively: 1/3,1/7 and 1/7.

Preferably, in step S12, classification is determined the described smallest edit distance that equipment 1 is determined according to comparing unit, determines described sequence similarity of paths by following formula (5):

$r=\frac{\mathrm{\α}\×\stackrel{\‾}{l}}{d+1}---\left(5\right)$

Wherein, α is normalization coefficient, by the average string length of the corresponding character string of class searching order routing information, d is smallest edit distance, and wherein, normalization coefficient α can calculate by following formula (6):

$\mathrm{\α}=\frac{{x-x}_{\min }}{x_{\max }-x_{\min }}---\left(6\right)$

Wherein, x representation class searching order routing information the statistical length of corresponding character string in test process, if α=0.5, and the each self-corresponding class searching order routing information of keyword clustering cluster1 to cluster3 the average string length of corresponding character string be 6, in step S12, classification determines that equipment 1 can determine that according to above-mentioned formula (5) the searching order routing information of target keyword goal-query is respectively with the sequence similarity of paths of keyword clustering cluster1 to cluster3 described class searching order routing information separately respectively: 1, 3/7 and 3/7.

Those skilled in the art will be understood that the above-mentioned mode of determining described sequence similarity of paths is only for giving an example; the mode of other existing or definite described sequence similarity of paths that may occur is from now on as applicable to the present invention; also should be included in protection domain of the present invention, and be contained in this at this with way of reference.

In step S13, classification determines that equipment 1 is according to described sequence similarity of paths, determine described target critical term clustering, as described in determining under target keyword as described in target critical term clustering be as described in sequence similarity of paths meet predetermined threshold as 0.8 corresponding keyword clustering.For example, connect example, in step S12, classification determines that the searching order routing information of equipment 1 definite target keyword goal-query is respectively with the sequence similarity of paths of keyword clustering cluster1 to cluster3 described class searching order routing information separately respectively: 1,3/7 and 3/7,, in step S13, classification determines that the described target critical term clustering under the definite target keyword goal-query of equipment 1 is cluster1.

Preferably, in step S1, classification determines that first equipment 1 also can obtain pending keyword one or more to be measured, using as described target keyword; Then, according to the searching order routing information of described target keyword, the target critical term clustering from one or more keyword clustering under definite described target keyword; In step S2, classification determines that equipment 1 also can, first according to described target critical term clustering, determine the corresponding relevance of searches classification of described target keyword; Then,, according to described relevance of searches classification, described target keyword is carried out to Screening Treatment.

Particularly, in step S1, classification determine equipment 1 also can be first by such as search engine, browser, the application programming interfaces (API) of the third party devices such as keyword equipment to be measured are provided, obtain pending keyword one or more to be measured, using as described target keyword; Then, according to the searching order routing information of described target keyword, the target critical term clustering from one or more keyword clustering under definite described target keyword.At this, in step S1, classification determines that equipment 1 determines the mode of the target critical term clustering under described target keyword and aforementioned in step S1, classification determines that the mode of the target critical term clustering under the fixed described target keyword of equipment 1 is same or similar, for simplicity's sake, therefore do not repeat them here, and comprise by reference therewith.

Then,, in step S2, classification determines that equipment 1 also can, first according to described target critical term clustering, determine the corresponding relevance of searches classification of described target keyword.At this, in step S2, classification determines that equipment 1 determines the mode of the corresponding relevance of searches classification of described target keyword and aforementioned in step S2, classification determines that the mode of equipment 1 definite corresponding relevance of searches classification of described target keyword is same or similar, for simplicity's sake, therefore do not repeat them here, and comprise by reference therewith.

Then,, in step S2, classification determines that equipment 1, according to described relevance of searches classification, carries out Screening Treatment to described target keyword.For example, suppose in step S 1, classification determines that the described target keyword that equipment 1 obtains comprises as query1, query2, query3 and query4, and in step S2, classification is determined definite this target keyword query1 of equipment 1, query2, it is high that the described relevance of searches classification that query3 and query4 are corresponding is respectively correlativity, in correlativity, correlativity is low, correlativity is high, in step S2, classification determines that equipment 1 is according to target keyword query1, query2, the each self-corresponding described relevance of searches classification of query3 and query4, it is carried out to Screening Treatment, as the keyword query3 that belongs to the low classification of correlativity is screened from keyword set to be measured, it is carried out to later stage searching order information optimization.

Preferably, classification determines that equipment 1 also comprises step S6 (not shown) and step S7 (not shown).Particularly, in step S6, classification is determined the definite corresponding keyword set to be optimized of described keyword clustering of equipment 1; In step S7, classification is determined the corresponding common search sequence of the keyword one or more to be optimized routing information that equipment 1 comprises according to described keyword set to be optimized, determine the Optimizing Search sequence routing information of described one or more keywords to be optimized, for adjusting the searching order routing information of described one or more keywords to be optimized.

Particularly, in step S6, classification is determined the relevance of searches descriptor of all keywords that equipment 1 can comprise according to keyword clustering, as high in relevance of searches, relevance of searches is low etc., find out the classification that relevance of searches is low, using as the corresponding keyword set to be optimized of described keyword clustering.At this, described keyword set to be optimized is corresponding to the keyword that belongs to the low classification of relevance of searches.For example, the keyword of supposing to belong in keyword clustering cluster1 the low classification of relevance of searches is sample keyword III, the keyword that belongs to the low classification of relevance of searches in keyword clustering cluster2 is sample keyword IV and VI, in keyword clustering cluster3, do not belong to the keyword of the low classification of relevance of searches, in step S6, classification determines that all keywords that belong to the low classification of relevance of searches that equipment 1 can comprise keyword clustering cluster1 to cluster3 are as described keyword set to be optimized, as comprise sample keyword III, IV and VI.

Preferably, in step S6, classification determines that equipment 1 also can compare the actual search results relevant information of all keywords included described keyword clustering and system index information, determines described keyword set to be optimized from described all keywords.At this, described actual search results relevant information includes but not limited to as returned to Search Results quantity, obtain click volume, return website authority, return to the quality degree of the content of pages of website etc.At this, described system index information comprises as returned to Search Results quantity, returning to the authority of website etc.For example, in step S6, classification determines that actual search results relevant information and system index information that equipment 1 can be sample keyword I to VI by its included all keywords of described keyword clustering cluster1 to cluster3 definite in step S1 compare, from described all keywords, determine described keyword set to be optimized, as the actual search results relevant information of sample keyword I to VI being discontented with to the keyword of pedal system indication information, as described keyword set to be optimized.

Those skilled in the art will be understood that the above-mentioned mode of determining described keyword set to be optimized is only for giving an example; the mode of other definite described keyword set to be optimized existing or that may occur is from now on as applicable to the present invention; also should be included in protection domain of the present invention, and be contained in this at this with way of reference.

Then, in step S7, classification is determined the corresponding common search sequence of the keyword one or more to be optimized routing information that equipment 1 comprises according to described keyword set to be optimized, determine the Optimizing Search sequence routing information of described one or more keywords to be optimized, for adjusting the searching order routing information of described one or more keywords to be optimized.For example, connect example, in step S6, classification determines that the keyword set described to be optimized that equipment 1 is determined comprises sample keyword III, IV and VI, in step S7, classification determines that equipment 1 can be according to sample keyword III, the corresponding common search sequence of IV and VI routing information is as S_C → S_D → S-G, as described Optimizing Search sequence routing information, for adjusting the searching order routing information of described one or more keywords to be optimized, as delete sample keyword III, this Optimizing Search sequence routing information that the searching order routing information of IV and VI comprises, or, by sample keyword III, this Optimizing Search sequence routing information that the searching order routing information of IV and VI comprises is replaced by other searching order routing informations common search sequence as corresponding in the keyword of the high classification of relevance of searches routing information.

Fig. 4 illustrates the method flow diagram for definite corresponding relevance of searches classification of target keyword in accordance with a preferred embodiment of the present invention.

Wherein, classification determines that equipment 1 comprises step S1 ', step S2 ', step S3 ', step S4 ' and step S5 '.Particularly, in step S1 ', classification determines that equipment 1 is according to the searching order routing information of target keyword, the target critical term clustering from one or more keyword clustering under definite described target keyword; In step S2 ', classification determines that equipment 1 is according to described target critical term clustering, determines the corresponding relevance of searches classification of described target keyword, for subsequent treatment; In step S3 ', classification is determined the definite corresponding preferred searching order routing information of described target keyword of equipment 1; In step S4 ', classification determines that equipment 1, according to described preferred searching order routing information, adjusts the searching order routing information of described target keyword; If search sequence and described target keyword match, in step S5 ', classification determines that equipment 1, according to the searching order routing information of the described target keyword after adjusting, offers the corresponding user of described search sequence by corresponding described search sequence Search Results.At this, step S1 ' and step S2 ' are same or similar with corresponding step shown in Fig. 3 respectively, so locate to repeat no more, and mode is by reference contained in this.

Particularly, in step S3 ', classification is determined the definite corresponding preferred searching order routing information of described target keyword of equipment 1.At this, in step S3 ', classification determines described in equipment 1 that preferably the mode of searching order routing information includes but not limited to following at least any one:

1) by included described target critical term clustering each with reference to the corresponding common search sequence of keyword routing information, as described preferred searching order routing information.For example, suppose in step S1 ', classification determines that the target critical term clustering under the definite target keyword goal-query of equipment 1 is keyword clustering cluster1, in step S3 ', classification determines that equipment 1 can be that the corresponding common search sequence of sample keyword I, III and VI routing information is as S_A → S_C → S_D → S_E → S_G, as described preferred searching order routing information with reference to keyword by included cluster1 each.

2) included each of described target critical term clustering carried out to statistical study processing with reference to the corresponding searching order routing information of keyword, to determine described preferred searching order routing information, as statistics draws the searching order path frequency information that keyword is passed by, the frequency is met to the searching order path of predetermined threshold, or using high frequency searching order path as described preferred searching order routing information.For example, connect example, in step S3 ', classification determines that equipment 1 can be that sample keyword I, III and the corresponding searching order routing information of VI carry out statistical study processing with reference to keyword by included cluster1 each, the frequency is met to predetermined threshold if the searching order path of frequency of occurrence satisfied 2 is as S_A → S_C → S_D → S_E → S_G → S_F, as described preferred searching order routing information.

Those skilled in the art will be understood that the above-mentioned mode of determining described preferred searching order routing information is only for giving an example; the mode of other existing or definite described preferred searching order routing informations that may occur is from now on as applicable to the present invention; also should be included in protection domain of the present invention, and be contained in this at this with way of reference.

In step S4 ', classification determines that equipment 1, according to described preferred searching order routing information, adjusts the searching order routing information of described target keyword.For example, suppose in step S3 ', classification determines that the described preferred searching order routing information that equipment 1 is determined is S_A → S_C → S_D → S_E → S_G, in step S4 ', classification determines that equipment 1 is according to this preferred searching order routing information, by such as adjustment algorithm or machine learning model as SVM model, the searching order routing information of adjustment aim keyword goal-query, as being adjusted into the searching order routing information of target keyword goal-query as described in preferential execution preferably searching order routing information S_A → S_C → S_D → S_E → S_G.

Those skilled in the art will be understood that the above-mentioned mode of determining described keyword set to be optimized is only for giving an example; the mode of other definite described keyword set to be optimized existing or that may occur is from now on as applicable to the present invention; also should be included in protection domain of the present invention, and be contained in this at this with way of reference.

If search sequence and described target keyword match, in step S5 ', classification determines that equipment 1, according to the searching order routing information of the described target keyword after adjusting, offers the corresponding user of described search sequence by corresponding described search sequence Search Results.Particularly, in step S5 ', classification determines that first equipment 1 obtain search sequence; Then, judge whether described search sequence and described target keyword match, if coupling, in step S5 ', classification determines that equipment 1, according to the searching order routing information of the described target keyword after adjusting, offers the corresponding user of described search sequence by corresponding described search sequence Search Results.At this, the implication of described coupling comprises search sequence and described target keyword is in full accord, search sequence is contained in described target keyword.

Particularly, in step S5 ', classification determines that equipment 1 is first by dynamic web page techniques such as ASP, JSP, or the application programming interfaces that provide by search engine (API), obtains the search sequence that user inputs by subscriber equipment.For example, if search subscriber B inputs keyword " fresh flower " by its PC equipment in search engine search column, by "enter" key", in step S5 ', classification determines that equipment 1, by dynamic web page techniques such as ASP, JSP or PHP, just can get the keyword " fresh flower " of search subscriber B input.It will be understood by those skilled in the art that the above-mentioned mode of obtaining search sequence is only for giving an example; other existing or modes of obtaining search sequence that may occur are from now on as applicable to the present invention; also should be included in protection domain of the present invention, and be contained in this at this with way of reference.

Then, in step S5 ', classification is determined the described target keyword that equipment 1 obtains in step S 1 ' according to it, by the mode of text comparison, judges whether described search sequence and described target keyword match.

If coupling, in step S5 ', classification determines that equipment 1, according to the searching order routing information of the described target keyword after adjusting, offers the corresponding user of described search sequence by corresponding described search sequence Search Results.For example, connect example, suppose in step S 1 ', classification determines that described target keyword that equipment 1 obtains is as " fresh flower ", " fresh flower express delivery " etc., in step S5 ', classification determines that equipment 1 judges that search sequence " fresh flower " and target keyword are as " fresh flower ", " fresh flower express delivery " matches, then, in step S5 ', classification determine equipment 1 by search engine according to the corresponding Search Results of searching order routing information of this target keyword after adjusting as " fresh flower Baidu mhkc ", " fresh flower picture materials sky is off line ", as the corresponding Search Results of search sequence " fresh flower ", and pass through such as ASP, the dynamic web page technique such as JSP or PHP, or the communication mode of other agreements, as the communication protocol such as http or https, offering the corresponding user of described search sequence is user B, as this user's subscriber equipment, browse for user.

It should be noted that the present invention can be implemented in the assembly of software and/or software and hardware, for example, can adopt special IC (ASIC), general object computing machine or any other similar hardware device to realize.In one embodiment, software program of the present invention can carry out to realize step mentioned above or function by processor.Similarly, software program of the present invention (comprising relevant data structure) can be stored in computer readable recording medium storing program for performing, for example, and RAM storer, magnetic or CD-ROM driver or flexible plastic disc and similar devices.In addition, steps more of the present invention or function can adopt hardware to realize, for example, thereby as coordinate the circuit of carrying out each step or function with processor.

In addition, a part of the present invention can be applied to computer program, and for example computer program instructions, in the time that it is carried out by computing machine, by the operation of this computing machine, can call or provide the method according to this invention and/or technical scheme.And call the programmed instruction of method of the present invention, may be stored in fixing or movably in recording medium, and/or be transmitted by the data stream in broadcast or other signal bearing medias, and/or be stored in according in the working storage of the computer equipment of described programmed instruction operation.At this, comprise according to one embodiment of present invention a device, this device comprises storer for storing computer program instructions and the processor for execution of program instructions, wherein, in the time that this computer program instructions is carried out by this processor, trigger this device and move based on aforementioned according to the method for multiple embodiment of the present invention and/or technical scheme.

To those skilled in the art, obviously the invention is not restricted to the details of above-mentioned one exemplary embodiment, and in the situation that not deviating from spirit of the present invention or essential characteristic, can realize the present invention with other concrete form.Therefore, no matter from which point, all should regard embodiment as exemplary, and be nonrestrictive, scope of the present invention is limited by claims rather than above-mentioned explanation, is therefore intended to all changes that drop in the implication and the scope that are equal to important document of claim to be included in the present invention.Any Reference numeral in claim should be considered as limiting related claim.In addition, obviously other unit or step do not got rid of in " comprising " word, and odd number is not got rid of plural number.Multiple unit of stating in device claim or device also can be realized by software or hardware by a unit or device.The first, the second word such as grade is used for representing title, and does not represent any specific order.

Claims (19)

1. for determining the corresponding relevance of searches class of a target keyword method for distinguishing, wherein, the method comprises the following steps:

A is according to the searching order routing information of target keyword, determines the target critical term clustering under described target keyword from one or more keyword clustering;

B, according to described target critical term clustering, determines the corresponding relevance of searches classification of described target keyword, for subsequent treatment.

2. method according to claim 1, wherein, described keyword clustering comprises the class searching order routing information for characterizing described keyword clustering;

Wherein, described step a comprises:

-the searching order routing information of described target keyword and the corresponding described class searching order routing information of described one or more keyword clustering are compared, to determine the smallest edit distance of searching order routing information and described class searching order routing information of described target keyword;

-according to described smallest edit distance, determine the searching order routing information of described target keyword and the sequence similarity of paths of described class searching order routing information;

-according to described sequence similarity of paths, determine described target critical term clustering.

3. method according to claim 1 and 2, wherein, the method also comprises:

X determines the corresponding preferred searching order routing information of described target keyword;

-according to described preferred searching order routing information, adjust the searching order routing information of described target keyword;

Wherein, the method also comprises:

-Ruo search sequence and described target keyword match, and according to the searching order routing information of the described target keyword after adjusting, corresponding described search sequence Search Results are offered to the corresponding user of described search sequence.

4. method according to claim 3, wherein, described step x comprises:

-according to included each of described target critical term clustering with reference to the corresponding common search sequence of keyword routing information, determine described preferred searching order routing information.

5. method according to claim 3, wherein, described step x comprises:

-included each of described target critical term clustering carried out to statistical study processing with reference to the corresponding searching order routing information of keyword, to determine described preferred searching order routing information.

6. according to the method described in any one in claim 1 to 5, wherein, the method also comprises:

Y determines the corresponding keyword set to be optimized of described keyword clustering;

The corresponding common search sequence of-keyword one or more to be optimized that comprises according to described keyword set to be optimized routing information, determine the Optimizing Search sequence routing information of described one or more keywords to be optimized, for adjusting the searching order routing information of described one or more keywords to be optimized.

7. method according to claim 6, wherein, described step y comprises:

-the actual search results relevant information of all keywords included described keyword clustering and system index information are compared, from described all keywords, determine described keyword set to be optimized.

8. according to the method described in any one in claim 1 to 7, wherein, described step a comprises:

-obtain pending keyword one or more to be measured, using as described target keyword;

-according to the searching order routing information of described target keyword, the target critical term clustering from one or more keyword clustering under definite described target keyword;

Wherein, described step b comprises:

-according to described target critical term clustering, determine the corresponding relevance of searches classification of described target keyword;

-according to described relevance of searches classification, described target keyword is carried out to Screening Treatment.

9. determine an equipment for the classification of determining the corresponding relevance of searches classification of target keyword, wherein, this classification determines that equipment comprises:

Cluster determining device for according to the searching order routing information of target keyword, is determined the target critical term clustering under described target keyword from one or more keyword clustering;

Classification determining device, for according to described target critical term clustering, determines the corresponding relevance of searches classification of described target keyword, for subsequent treatment.

10. classification according to claim 9 is determined equipment, and wherein, described keyword clustering comprises the class searching order routing information for characterizing described keyword clustering;

Wherein, described cluster determining device comprises:

Comparing unit, for the searching order routing information of described target keyword and the corresponding described class searching order routing information of described one or more keyword clustering are compared, to determine the smallest edit distance of searching order routing information and described class searching order routing information of described target keyword;

Similarity determining unit, for according to described smallest edit distance, determines the searching order routing information of described target keyword and the sequence similarity of paths of described class searching order routing information;

Cluster determining unit, for according to described sequence similarity of paths, determines described target critical term clustering.

11. determine equipment according to the classification described in claim 9 or 10, and wherein, this classification determines that equipment also comprises:

Preferred path determining device, for determining the corresponding preferred searching order routing information of described target keyword;

Adjusting gear, for according to described preferred searching order routing information, adjusts the searching order routing information of described target keyword;

Wherein, this classification determines that equipment also comprises:

Generator, if match for search sequence and described target keyword, according to the searching order routing information of the described target keyword after adjusting, offers the corresponding user of described search sequence by corresponding described search sequence Search Results.

12. classifications according to claim 11 are determined equipment, and wherein, described preferred path determining device is used for:

-according to included each of described target critical term clustering with reference to the corresponding common search sequence of keyword routing information, determine described preferred searching order routing information.

13. classifications according to claim 11 are determined equipment, and wherein, described preferred path determining device is used for:

-included each of described target critical term clustering carried out to statistical study processing with reference to the corresponding searching order routing information of keyword, to determine described preferred searching order routing information.

14. determine equipment according to the classification described in any one in claim 9 to 13, and wherein, this classification determines that equipment also comprises:

Set determining device, for determining the corresponding keyword set to be optimized of described keyword clustering;

Path optimizing determining device, for the corresponding common search sequence of the keyword one or more to be optimized routing information comprising according to described keyword set to be optimized, determine the Optimizing Search sequence routing information of described one or more keywords to be optimized, for adjusting the searching order routing information of described one or more keywords to be optimized.

15. classifications according to claim 14 are determined equipment, and wherein, described set determining device is used for:

-the actual search results relevant information of all keywords included described keyword clustering and system index information are compared, from described all keywords, determine described keyword set to be optimized.

16. determine equipment according to the classification described in any one in claim 9 to 15, and wherein, described cluster determining device is used for:

-obtain pending keyword one or more to be measured, using as described target keyword;

-according to the searching order routing information of described target keyword, the target critical term clustering from one or more keyword clustering under definite described target keyword;

Wherein, described classification determining device is used for:

-according to described target critical term clustering, determine the corresponding relevance of searches classification of described target keyword;

-according to described relevance of searches classification, described target keyword is carried out to Screening Treatment.

17. 1 kinds of computer equipments, comprise that the classification as described in any one in claim 9 to 16 is determined equipment.

18. 1 kinds for determining the search engines of the corresponding relevance of searches classification of target keyword, and wherein, this search engine comprises that the classification as described in any one in claim 9 to 16 determines equipment.

19. 1 kinds for determining the search engine plug-in units of the corresponding relevance of searches classification of target keyword, and wherein, this search engine plug-in unit comprises that the classification as described in any one in claim 9 to 16 determines equipment.

Images