CN102446171A

CN102446171A - Method and apparatus for evaluating quality score of promotion key word based on weighted average click-through rate

Info

Publication number: CN102446171A
Application number: CN201010298677XA
Authority: CN
Inventors: 焦世斗; 张怀亭
Original assignee: Beijing Baidu Netcom Science and Technology Co Ltd
Current assignee: Beijing Baidu Netcom Science and Technology Co Ltd
Priority date: 2010-09-30
Filing date: 2010-09-30
Publication date: 2012-05-09
Anticipated expiration: 2030-09-30
Also published as: CN102446171B

Abstract

The invention aims to provide a method and an apparatus for evaluating the quality score of a promotion key word based on a weighted average click-through rate, wherein the method comprises the following steps of: obtaining a promotion key word input by a user by a network apparatus; obtaining an individual click-through rate of the promotion key word corresponding to the user according to the promotion key word; obtaining the weighted average click-through rate of the promotion key word according to the promotion key word; and determining the quality score of the promotion key word corresponding to the user based on the individual click-through rate and the weighted average click-through rate. Compared with the prior art, the method provided in the invention evaluates the quality score of the promotion key word in combination with the weighted average click-through rate; therefore, more objective and effective evaluation result is provided for the promotion user so that the promotion user is possible to further improve the network information; and the selection of unappealing promotion key words by the promotion user is avoided and the use experience of the network user is enhanced.

Description

Advance the method and apparatus that keyword quality degree is promoted in the rate evaluation and test based on the weighted mean point

Technical field

The present invention relates to internet information disseminate technology field, relate in particular to a kind of by the computer implemented technology of advancing rate evaluation and test popularization keyword quality degree based on the weighted mean point.

Background technology

In the prior art; When the quality of evaluation and test popularization keyword is spent; How the point of the popularization keyword through a user advances the quality degree that rate is confirmed this popularization keyword, at this, puts rate and representes number of times that the pairing advertising creative of a certain popularization keyword is clicked on the Website page and the ratio that is represented number of times; Be clicks/views, it is a number percent.

But; Only advance rate and confirm that the quality degree of this popularization keyword is objective inadequately through the point of promoting keyword; Influence this click volume of putting rate into and possibly cause unusual fluctuations because of the network user's malicious act, thereby it is low excessively to make invocation point advance the degree of confidence of rate with the amount of representing.For example fryer search; Represent a certain popularization keyword in a large number and but do not click this popularization keyword; Perhaps to the clicking maliciously in a large number of a certain popularization keyword, it is low excessively to make that all the point of this popularization keyword advances the degree of confidence of rate, further causes the evaluation and test effect of this popularization keyword is reduced.

Therefore, how to provide a kind of objective, effective evaluation and test to promote the method for the quality degree of keyword, become one of present urgent problem.

Summary of the invention

The purpose of this invention is to provide a kind of method and apparatus that advances rate evaluation and test popularization keyword quality degree based on the weighted mean point.

According to an aspect of the present invention, provide a kind of and advanced the method that keyword quality degree is promoted in the rate evaluation and test by computer implemented based on the weighted mean point, this method may further comprise the steps:

A obtains the popularization keyword of user's input;

B is according to said popularization keyword, and the individuality point that obtains with the corresponding said popularization keyword of this user advances rate;

C is according to said popularization keyword, and the weighted mean point that obtains said popularization keyword advances rate;

D advances rate based on said individual point and said weighted average point advances rate, confirms the quality degree with the corresponding said popularization keyword of this user.

According to another aspect of the present invention, also provide a kind of and advanced the equipment that keyword quality degree is promoted in the rate evaluation and test based on the weighted mean point, this equipment comprises:

First deriving means is used to obtain the popularization keyword of user's input;

Second deriving means is used for according to said popularization keyword, and the individuality point that obtains with the corresponding said popularization keyword of this user advances rate;

The 3rd deriving means is used for according to said popularization keyword, and the weighted mean point that obtains said popularization keyword advances rate;

Confirm device, be used for advancing rate and advance rate, confirm and the quality degree of the corresponding said popularization keyword of this user with said weighted mean point according to said individual point.

Compared with prior art; The present invention advances the quality degree that keyword is promoted in the rate evaluation and test through combining the weighted mean point; Be not merely the popularization user more objective, effective evaluation result is provided; Making it further improve network information popularization becomes possibility, and evades the popularization user and select unappealing popularization keyword, has promoted the network user's experience.

Description of drawings

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

Fig. 1 illustrates and advances method flow diagram that rate evaluation and test promote keyword quality degree by computer implemented based on the weighted mean point according to one aspect of the invention;

Fig. 2 illustrate according to a further aspect of the present invention advance the equipment synoptic diagram that keyword quality degree is promoted in rate evaluation and test based on the weighted mean point.

Same or analogous Reference numeral is represented same or analogous parts in the accompanying drawing.

Embodiment

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

Fig. 1 is the method flow diagram according to one aspect of the invention, illustrates by computer implemented to advance the process that keyword quality degree is promoted in the rate evaluation and test based on the weighted mean point.Wherein, This process is accomplished at network equipment end usually; At this, the network equipment includes but not limited to network host, single network server, a plurality of webserver collection or based on the set of computers of cloud computing, wherein; Cloud computing is a kind of of Distributed Calculation, a super virtual machine of being made up of the loosely-coupled computing machine collection of a group.Those skilled in the art will be understood that the above-mentioned network equipment is merely for example, and other network equipments existing or that possibly occur from now on also should be included in the protection domain of the present invention, and be contained in this at this with way of reference as applicable to the present invention.

Particularly, the network equipment obtains the popularization keyword of user's input; According to said popularization keyword, the individuality point that obtains with the corresponding said popularization keyword of this user advances rate; According to said popularization keyword, the weighted mean point that obtains said popularization keyword advances rate; Advance rate according to said individual point and advance rate, confirm and the quality degree of the corresponding said popularization keyword of this user with said weighted mean point.

More specifically, in step S1, the network equipment obtains the popularization keyword of user's input.Particularly; In step S1; The network equipment obtains the popularization keyword of user's input in several ways; For example; User's mutual by with subscriber equipment, as carrying out man-machine interaction by modes such as keyboard, mouse, remote controller, touch pad, touch-screen or hand-written equipment, and subscriber equipment is connected with the network equipment via network; Send the popularization keyword of this user's input to the network equipment, the network equipment obtains the popularization keyword of user's input in view of the above; The user promotes input popularization keyword in the account at this then by this user's in the subscriber equipment logging in network equipment popularization account, and the network equipment obtains the popularization keyword of this user's input from this popularization account; The user will promote keyword and import crucial dictionary, and the network equipment and this key dictionary connect, and from this key dictionary, read the popularization keyword of this user's input.Those skilled in the art will be understood that the mode that the above-mentioned network equipment obtains the popularization keyword of user input is merely for example; Other network equipments existing or that possibly occur from now on obtain the mode of popularization keyword of user input as applicable to the present invention; Also should be included in the protection domain of the present invention, and be contained in this with way of reference at this.At this, network includes but not limited to internet, wide area network, Metropolitan Area Network (MAN), LAN, VPN network, wireless self-organization network (Ad Hoc network), GSM, WCDMA, CDMA2000, TD-SCDMA, GSM, CDMA1x, WIFI, WAPI, WiMax etc.Those skilled in the art will be understood that above-mentioned network is merely for example, and other networks existing or that possibly occur from now on also should be included in the protection domain of the present invention, and be contained in this at this with way of reference as applicable to the present invention.

Subsequently, in step S2, the network equipment is according to said popularization keyword, and the individuality point that obtains with the corresponding said popularization keyword of this user advances rate.Particularly, in step S2, the network equipment is according to the popularization keyword of its user who in step S1, obtains input, and the individuality point that obtains this popularization keyword of this user advances rate.At this, the source that obtains that the individuality of this popularization keyword point advances rate includes but not limited to, the memory storage of the network equipment itself, the management server of promoting the affiliated extension system of keyword and other equipment that can communicate by letter with this network equipment.The obtain manner that the individuality point of this popularization keyword advances rate includes but not limited to; For example; The network equipment itself is the management server of promoting the affiliated extension system of keyword, and the network equipment obtains this popularization keyword from the historical record of this popularization keyword individuality point advances rate; And for example; The network equipment is through reading, call the communication mode of certain applications routine interface (API) or other agreements in database; The individuality point that obtains this popularization keyword of this user advances rate; For example; The network equipment advances the application programming interfaces (API) that the third party device of rate provides through the individuality point that calls this popularization keyword that stores this user, and the individuality point that sends this popularization keyword that obtains this user to this third party device advances the request of rate, and the individuality point that receives this popularization keyword of this user that this third party device returns according to this request advances rate.Those skilled in the art will be understood that above-mentioned individual point advance rate obtain the source and obtain manner is merely for example; What other individuality points existing or that possibly occur from now on advanced rate obtains source or obtain manner as applicable to the present invention; Also should be included in the protection domain of the present invention, and be contained in this with way of reference at this.

Then, in step S3, the network equipment is according to said popularization keyword, and the weighted mean point that obtains said popularization keyword advances rate.Particularly, in step S3, the network equipment is according to its popularization keyword that in step S1, obtains, and the weighted mean point that obtains this popularization keyword advances rate.The obtain manner that this weighted mean point advances rate includes but not limited to: obtained by network device computes; Promptly advance rate corresponding to the point of different user or different popularization unit according to this popularization keyword; And give these and put the into different weight of rate; Again these are put into rate and carry out weighted mean, advance rate (when these put the weight of rate into when identical, the weighted mean point of this popularization keyword advances rate and is the arithmetic mean point and advances rate) with the weighted mean point that obtains this popularization keyword; Directly obtain by the memory storage or the third party device of the network equipment from himself; For example; The network equipment is through reading, call the communication mode of certain applications routine interface (API) or other agreements in database, the weighted mean point that obtains this popularization keyword advances rate.For example; The application programming interfaces (API) that the network equipment provides through the third party device that calls the weighted mean point that stores this popularization keyword and advance rate; Send the weighted mean point that obtains this popularization keyword to this third party device and advance the request of rate, and the weighted mean point that receives this popularization keyword that this third party device returns according to this request advances rate.Those skilled in the art will be understood that above-mentioned obtain manner is merely for example, and other obtain manners existing or that possibly occur from now on also should be included in the protection domain of the present invention, and be contained in this at this with way of reference as applicable to the present invention.At this; The Consideration that the point of promoting keyword advances the weight of rate include but not limited to promote dog-eat-dog degree, the netizen of keyword the retrieval amount,, or promote industry under the keyword etc.; Those skilled in the art will be understood that the Consideration that the point of above-mentioned popularization keyword advances the weight of rate is merely for example; The point of other popularization keywords existing or that possibly occur from now on advances the Consideration of weight of rate as applicable to the present invention; Also should be included in the protection domain of the present invention, and be contained in this with way of reference at this.

Need to prove; Those skilled in the art will be understood that; Step S2 and step S3 are not that strict order is carried out relation, and the network equipment can first execution in step S2, and the individuality point that obtains with corresponding this popularization keyword of this user advances rate; Execution in step S3 again, the weighted mean point that obtains this popularization keyword advances rate; Also can first execution in step S3, the weighted mean point that obtains this popularization keyword advances rate, execution in step S2 again, the individuality point that obtains with corresponding this popularization keyword of this user advances rate; Simultaneously execution in step S2 and step S3 obtain the weighted mean point that individuality point with corresponding this popularization keyword of this user advances rate and this popularization keyword and advance rate; Even can be in same step; Obtain the weighted mean point that individuality point with corresponding this popularization keyword of this user advances rate and this popularization keyword simultaneously and advance rate; For example the network equipment with third party device mutual in, in same request and response, receive the weighted mean point that individuality point with corresponding this popularization keyword of this user advances rate and this popularization keyword and advance rate.For for simplicity, we explain network equipment execution in step S2 earlier, and the individuality point that obtains with corresponding this popularization keyword of this user advances rate, explains network equipment execution in step S3 again, and the weighted mean point that obtains this popularization keyword advances rate.It is convenient that those skilled in the art will be understood that such statement is merely statement in proper order, should be as limitation of the present invention.

Subsequently, in step S4, the network equipment advances rate based on said individual point and said weighted average point advances rate, confirms the quality degree with the corresponding said popularization keyword of this user.Particularly; In step S4; The network equipment advances rate according to the weighted mean point that the individuality point of this popularization keyword of its this user who in step S2, obtains advances rate and its this popularization keyword that in step S3, obtains, through confirm the quality degree of this popularization keyword of this user such as the mode of ratio, difference.For example, the weighted average point that the network equipment advances rate and this popularization keyword based on the individuality point of this popularization keyword of this user advances the ratio of rate, confirms the quality degree of this popularization keyword of this user; It is 0.2% that the individuality point of supposing this popularization keyword of this user advances rate, and it is 0.15% that the weighted average point of this popularization keyword advances rate, and their ratio is 1.33 so, the quality degree of this popularization keyword that this ratio is exactly this user.And for example, the weighted average point that the network equipment advances rate and this popularization keyword based on the individuality point of this popularization keyword of this user advances the difference of rate, confirms the quality degree of this popularization keyword of this user; It is 0.2% that the individuality point of supposing this popularization keyword of this user advances rate, and it is 0.15% that the weighted average point of this popularization keyword advances rate, and their difference is 0.05% so, the quality degree of this popularization keyword that this difference is exactly this user.Those skilled in the art will be understood that definite mode of above-mentioned quality degree is merely for example; Definite mode of other quality degree existing or that possibly occur from now on is as applicable to the present invention; Also should be included in the protection domain of the present invention, and be contained in this with way of reference at this.

Preferably, step S1, step S2, step S3 and step S4 are constant work.Particularly, the network equipment obtains the popularization keyword of user's input in step S1; In step S2, according to said popularization keyword, the individuality point that obtains with the corresponding said popularization keyword of this user advances rate; In step S3, according to said popularization keyword, the weighted mean point that obtains said popularization keyword advances rate; In step S4, advance rate according to said individual point and advance rate with said weighted mean point, confirm and the quality degree of the corresponding said popularization keyword of this user.At this; It will be understood by those skilled in the art that the weighted mean point that individuality point that " continuing " is meant that the network equipment requires to promote this popularization keyword of the obtaining of keyword, this user according to mode of operation that set or adjustment in real time respectively at step S1, step S2, step S3 and step S4 advances the obtaining of rate, this popularization keyword advances obtaining of rate; Reach confirming of this quality degree of promoting keyword, stop to evaluate and test the quality degree of promoting keyword in a long time until this network equipment.

In a preferred embodiment, in step S3, the network equipment is according to said popularization keyword, obtains and uses the user of the corresponding said popularization keyword of each user of said popularization keyword to put into rate; Put into rate according to said user, the weighted mean point that obtains said popularization keyword advances rate.Particularly; In step S3; The network equipment is according to its popularization keyword that in step S1, obtains; Obtain other users that use these popularization keywords and put into rate corresponding to the user of this popularization keyword, put into rate according to this (a bit) user then, the weighted mean point that obtains this popularization keyword advances rate.For example, put into the pairing amount of representing of rate according to the user and give the user and put the into different weight of rate, the weighted mean point that obtains this popularization keyword according to this weight advances rate; Suppose to promote keyword and be " fresh flower ", use the user of " fresh flower " to be user's first and user's second under the identical linguistic context, user's first puts into corresponding to the user of " fresh flower " that rate is 0.2%; The amount of representing is 1000 times; User's second puts into corresponding to the user of " fresh flower " that rate is 0.15%, and the amount of representing is 10000 times, and the user who gives user's first puts into that the weight of rate is 0.1; The user who gives user's second puts into that the weight of rate is 0.9, is 0.155% thereby the weighted mean point that obtains " fresh flower " advances rate.And for example, the weight that the user who supposes all users' same popularization keyword puts rate into is the same, and the weighted mean point of this popularization keyword advanced rate and was the arithmetic mean point and advances rate this moment; As promote keyword for " fresh flower "; User's first is put into corresponding to the user of " fresh flower ", and rate is 0.2%; User's second is put into corresponding to the user of " fresh flower ", and rate is 0.15%; User third puts into corresponding to the user of " fresh flower " that rate is 0.24%, and to advance rate be 0.197% to the weighted mean of " fresh flower " point at this moment.Those skilled in the art will be understood that the mode that the above-mentioned weighted mean point that obtains the popularization keyword advances rate is merely for example; Other existing or obtaining of possibly occurring from now on modes that weighted mean points of promoting keyword advance rate are as applicable to the present invention; Also should be included in the protection domain of the present invention, and be contained in this with way of reference at this.

In a further advantageous embodiment, in step S3, the network equipment obtains the one or more related keyword relevant with said popularization keyword according to said popularization keyword; The point that obtains said one or more related keywords advances rate; The point of said one or more related keywords is advanced rate average, the weighted mean point that obtains said popularization keyword advances rate; Wherein, said one or more related keywords include but not limited to, below at least each: the 1) synonymous keyword of said popularization keyword; 2) the near adopted keyword of said popularization keyword; 3) belong to the keyword of identical industry with said popularization keyword; 4) belong to the keyword of identical region with said popularization keyword.Particularly; In step S3; The network equipment is according to its popularization keyword that in step S1, obtains; Obtain to promote the relevant one or more related keyword of keyword with this; For example, this is promoted the synonym of keyword, nearly adopted keyword, promotes the perhaps keyword etc. of identical region of the identical industry of keyword with this, and the point that obtains this (a bit) related keyword advances rate; Again this (a bit) is put rate and average, thereby the weighted average point that obtains this popularization keyword advances rate.For example; Suppose to promote keyword and be " point advances rate "; The network equipment obtains synonymous keyword " CTR " and " clicking rate " through meaning of a word analysis, then through calling the application programming interfaces (API) that third party device that the point of storing keyword advances rate provides; Send the point that obtains " CTR " and " clicking rate " to this third party device and advance the request of rate, and the point that receives " CTR " and " clicking rate " that this third party device returns according to this request advances rate; Then the point of " CTR " and " clicking rate " is advanced rate and average, the weighted mean point that obtains " point advances rate " advances rate, and the weighted mean point of this moment advances rate and also advances rate for the arithmetic mean point.And for example, still hypothesis popularization keyword is " point advances rate ", and the network equipment is through meaning of a word analysis; Obtain synonymous keyword " CTR " and " clicking rate "; Nearly adopted keyword " click volume " and " equalization point advances rate ", then through reading the data of crucial dictionary, the point that obtains " CTR ", " clicking rate ", " click volume " and " equalization point advances rate " respectively advances rate; The weight that the point of setting " CTR " and " clicking rate " advances rate is 0.3; The weight that the point of " click volume " and " equalization point advances rate " advances rate is 0.2, these is put into rate ask weighted mean value, and the weighted mean point that can obtain " point advances rate " advances rate.And for example, still hypothesis popularization keyword is " point advances rate ", and the network equipment is through meaning of a word analysis; Obtain synonymous keyword " CTR " and " clicking rate "; Again these " CTR " and " clicking rate " are carried out the region cluster analysis, the region is " East China ", is 0.2% thereby the point of " CTR " that acquisition is used in " East China " area advances rate; It is 0.15% that the point of " clicking rate " used in " East China " area advances rate; The weight that the point that " CTR " be set advances rate is 0.4, and the weight that the point of " clicking rate " advances rate is 0.6, and it is 0.17% that the weighted mean point that draws " point advances rate " advances rate.Those skilled in the art will be understood that the obtain manner that the weighted mean point of above-mentioned popularization keyword advances rate is merely for example with related keyword; Obtain manner or the related keyword that the weighted mean point of other popularization keywords existing or that possibly occur from now on advances rate is as applicable to the present invention; Also should be included in the protection domain of the present invention, and be contained in this with way of reference at this.

In another preferred embodiment; In step S3; The network equipment is according to said popularization keyword; And the statistics point that combines the extension system under the said popularization keyword advances rate (i.e. the ratio of the click volume of all keywords and the amount of representing in this extension system), and the weighted mean point that obtains said popularization keyword advances rate.Particularly, in step S3, the network equipment is according to its popularization keyword that in step S1, obtains, and advances rate in conjunction with the statistics point of the extension system under this popularization keyword, and the weighted mean point that obtains this popularization keyword advances rate.For example to advance the confidence level of rate too low when the weighted average point of this popularizations keyword, and the statistics point of this extension system of network equipment selection advances rate and advances rate as the weighted average point of this popularization keyword; It is 0.09% that the weighted average point of supposing to promote keyword advances rate; It is 0.1% that the statistics point of extension system advances rate; The weighted average point of promoting keyword this moment advances the statistics point that rate is lower than extension system and advances rate; The statistics point that the network equipment obtains extension system advances rate; Promptly 0.1%, advance rate as the weighted average point of this popularization keyword.Preferably; The statistics point that the network equipment advances rate and extension system for the weighted mean point of promoting keyword advances rate and gives different weights; This weighted mean point is advanced rate and this statistics point advance rate and ask weighted mean value, advance rate with the new weighted mean point that obtains this popularization keyword; It is 0.09% that the weighted mean point of supposing to promote keyword advances rate; It is 0.6 that this weighted mean point advances the corresponding weight of rate, and it is 0.1% that the statistics point of extension system advances rate, and it is 0.4 that this statistics point advances the corresponding weight of rate; So, to advance rate be 0.094% to the new weighted mean point of this popularization keyword.Those skilled in the art will be understood that the obtain manner that the weighted mean point of above-mentioned popularization keyword advances rate is merely for example; The obtain manner that the weighted mean point of other popularization keywords existing or that possibly occur from now on advances rate is as applicable to the present invention; Also should be included in the protection domain of the present invention, and be contained in this with way of reference at this.At this; The source that obtains that the statistics point of the extension system under the popularization keyword advances rate includes but not limited to; The memory storage of the network equipment itself, the management server of extension system, other equipment that can communicate by letter with this network equipment; Those skilled in the art will be understood that the source that obtains that the statistics point of the extension system under the above-mentioned popularization keyword advances rate is merely for example; The statistics point of the extension system under other popularization keywords existing or that possibly occur from now on advances obtaining of rate originates as applicable to the present invention, also should be included in the protection domain of the present invention, and be contained in this at this with way of reference.

Preferably, in step S2, the network equipment is according to said popularization keyword, and other points of promoting keyword that obtain this user advance rate; Other points of promoting keyword according to said this user advance rate, and weighting is confirmed to advance rate with the individuality point of the corresponding said popularization keyword of this user.Particularly, in step S2, the network equipment is according to its popularization keyword that in step S1, obtains; Other points of promoting keyword that obtain this user advance rate; For example, this user, advance rate with this popularizations keyword other points of promoting keywords in same popularization unit, put rate according to this (a bit); Through confirming that such as the mode of meaning of a word cluster this (a bit) point advances the weight of rate, thereby weighting is confirmed the individuality point of this popularization keyword of this user and is advanced rate.For example; Suppose this management server of the network equipment as extension system; Promoting keyword is " point advances rate "; Other popularization keywords that the network equipment obtains this user in view of the above are " CTR ", " clicking rate ", " click volume " and " equalization point advances rate ", and the point that from this user's record, reads " CTR ", " clicking rate ", " click volume " and " equalization point advances rate " then advances rate, and " CTR ", " clicking rate ", " click volume " and " equalization point advances rate " are carried out meaning of a word analysis; Obtain synonymous keyword " CTR " and " clicking rate " of " point advances rate "; It is 0.3 that their point advances the corresponding weight of rate, near adopted keyword " click volume " and " equalization point advances rate " of " point advances rate ", and it is 0.2 that their point advances the corresponding weight of rate; These are put into rate ask weighted mean value, the individuality point that obtains " point advances rate " advances rate.Those skilled in the art will be understood that the obtain manner that the individuality point of above-mentioned popularization keyword advances rate is merely for example; The obtain manner that the individuality point of other popularization keywords existing or that possibly occur from now on advances rate is as applicable to the present invention; Also should be included in the protection domain of the present invention, and be contained in this with way of reference at this.At this; Other of this user are promoted keyword and are included but not limited to; This user, promote keyword other in same popularization unit with this and promote keywords; Perhaps this user, promote in unit other with these other of the same classification in popularization unit of promoting keyword place and promote keywords, perhaps this user, all promote in unit other and promote keywords.Those skilled in the art will be understood that other popularization keywords of above-mentioned this user are merely for example; Other of other these users existing or that possibly occur are from now on promoted keywords as applicable to the present invention; Also should be included in the protection domain of the present invention, and be contained in this with way of reference at this.

In a preferred embodiment, in step S2, the network equipment is according to said popularization keyword, and in conjunction with this user's user related information, the individuality point that obtains with the corresponding said popularization keyword of this user advances rate.Particularly, in step S2, the network equipment is according to its popularization keyword that in step S1, obtains; In conjunction with this user's user related information, for example, the individuality point of this popularization keyword of this user advances the historical record of rate; Be used for smoothly the unusual fluctuations of should the individuality point advancing rate; The submission record of this user's said popularization keyword is used to discern this user's malicious registration behavior, and the individuality point that obtains this popularization keyword of this user advances rate.For example; The popularization keyword of supposing user's input is " fresh flower "; Advance the historical record of rate according to the individuality point of this user " fresh flower "; It is 2% that the individuality point of 3:00PM-4:00PM time period yesterday advances rate, and it is 0.02% that the individuality point of 3:00PM-4:00PM time period today advances rate, judges that the individuality point of 3:00PM-4:00PM time period today advances the influence that rate possibly receive the network user's malicious act; Level and smooth these unusual fluctuations of the network equipment, the individuality point that obtains " fresh flower " are advanced the history average of rate (advancing the mean value of rate like the individuality point of same period in January in past) and are advanced rate as the individuality point of 3:00PM-4:00PM time period today " fresh flower ".Those skilled in the art will be understood that the mode that the above-mentioned individuality point that obtains this popularization keyword of this user advances rate is merely for example; The mode that other individuality points existing or this popularization keyword that obtains this user that possibly occur from now on advance rate is as applicable to the present invention; Also should be included in the protection domain of the present invention, and be contained in this with way of reference at this.

Preferably, user related information includes but not limited to, below at least each: 1) the individuality point of this popularization keyword of this user advances the historical record of rate; 2) the submission record of this popularization keyword of this user.Wherein, The historical record that the individuality point of this popularization keyword of this user advances rate includes but not limited to that historical record and this user's the individuality point of this popularization keyword that the individuality point of this popularization keyword of this user advances the adjacent moment of rate advances the historical record of a period of time of rate; Can be used for smoothly the unusual fluctuations of should the individuality point advancing rate; For example, advance the adjacent moment of rate or the historical record of the past period according to the individuality point of this popularization keyword of this user, filtering is because of the network user's malicious act; Perhaps click maliciously the unusual fluctuations that the individuality point that causes advances rate in a large number like fryer search (represent but in a large number and do not click).For example; The popularization keyword of supposing user's input is " fresh flower "; Advance the historical record of rate according to the individuality point of this user " fresh flower ", yesterday this user the individuality point of " fresh flower " to advance rate be 2%, have today the netizen to use fryer search " fresh flower "; Making this user's today the individuality point of " fresh flower " advance rate is 0.02%; The difference that the individuality point of this user's of yesterday and today " fresh flower " advances rate is 1.98%, has surpassed the relative value scope (as 0.5%-0.8%) that the preset individuality point of the network equipment advances rate, and the network equipment obtains critical value (as 1.2%) that this individuality point advances rate and advances rate as the individuality point of this user of today " fresh flower ".The submission record of this popularization keyword of this user includes but not limited to that this user promotes the record of submitting this popularization keyword in the unit at one; Perhaps this user submits to this to promote the record of keyword at it in all popularization unit; Can be used to discern this user's malicious registration behavior; For example this user promotes at one and has registered the very big still very little popularization keyword of click volume of the amount of representing in the unit, makes that the quality degree of this popularization keyword is very low, and this user is in order to eliminate this submission record; The renewal of registration should be promoted keyword in other popularization unit; Hope to reach equally the very big still very little purpose of click volume of the amount of representing, after the network equipment identifies this user's malicious registration behavior, advance rate according to the individuality point in the former submission record of this popularization keyword; The individuality point of confirming this popularization keyword of new registration advances rate; It is very low that the individuality point of promptly original this popularization keyword advances rate, and it is still very low that the individuality point of this popularization keyword of new registration advances rate, thereby make that the quality degree of this popularization keyword of new registration is also very low.For example; Supposing to promote keyword is " fresh flower ", and this user promotes to register in the unit at " product " is somebody's turn to do " fresh flower ", should " fresh flower " amount of representing in this " product " popularization unit be 100000 times; Click volume is 200 times, is 0.2% thereby the individual point of acquisition advances rate; This user promotes to register once more in the unit in " service " is somebody's turn to do " fresh flower "; The network equipment identifies this malicious registration behavior of this user; Individuality point of promoting in the unit at " product " according to this " fresh flower " advances rate, confirms that it is 0.2% that this " fresh flower " individuality point in " service " popularization unit advances rate.Those skilled in the art will be understood that above-mentioned user related information is merely for example, and other user related informations existing or that possibly occur from now on also should be included in the protection domain of the present invention, and be contained in this at this with way of reference as applicable to the present invention.

More preferably, this process also comprises step S5 (not shown), in step S5, and network equipment record or upgrade said user related information.Particularly, the network equipment writes down or upgrades this user's user related information based on the variation of user's popularization operation and/or user's popularization data.For example, when in the network equipment also not during this user related information, in case this popularization keyword of this user is clicked in network or represented, the network equipment writes down this user related information in step S5; When the network equipment has had the record of user related information; When the user carries out advertisement registration etc. and promotes operation; Perhaps when the point of this popularization keyword of this user advanced rate and changes, this user related information was upgraded in the variation according to this user related information in step S5.For example; When keyword is promoted in one of user's new registration; Think in the network equipment that not the individuality point about this popularization keyword of this user advances the historical record of rate this moment; As netizen during in accessing network resources, this popularization keyword of this user is represented or is clicked, and the network equipment will write down this user related information (promptly the individuality point of this popularization keyword of this user advances the original records of rate).Based on this; The network equipment calls this user related information; The individuality point of judging this popularization keyword of this user advances rate and unusual fluctuations whether occur; When the network equipment judges that it is unusual fluctuations that the individuality point of this popularization keyword of this user this moment advances rate; And after confirming that according to preset critical the individuality point of this popularizations keyword of this user this moment advances rate, the network equipment upgrades this user related information (promptly the individuality point of this popularization keyword of this user advances the historical record of rate at this moment) according to this preset critical.Those skilled in the art will be understood that the mode of above-mentioned record and renewal user related information is merely for example; Other records existing or that possibly occur from now on or the mode of upgrading user related information are as applicable to the present invention; Also should be included in the protection domain of the present invention, and be contained in this with way of reference at this.

In a further advantageous embodiment, this process also comprises step S6 (not shown), and in step S6, the network equipment is according to the variation of said quality degree, and the evaluation and test frequency of said popularization keyword is evaluated and tested in adjustment; According to adjusted evaluation and test frequency, evaluate and test the quality degree of said popularization keyword.Particularly, in step S6, the network equipment is based on the quality degree of its popularization keyword that in step S4, obtains, the variation of obtaining this quality degree, and change based on this, this promotes the evaluation and test frequency of keyword the adjustment evaluation and test.For example, if the variation of (as continuous two days) is less in the unit interval, this promotes the evaluation and test frequency of keyword then to reduce evaluation and test; If the variation of (as continuous two days) is bigger in the unit interval; This promotes the evaluation and test frequency of keyword then to improve evaluation and test; Then according to adjusted evaluation and test frequency; Carry out the evaluation and test of the quality degree of this follow-up popularization keyword, the adaptive in view of the above adjustment of the network equipment is promoted the evaluation and test frequency of the quality degree of keyword for this.Those skilled in the art will be understood that the mode of above-mentioned adjustment evaluation and test frequency is merely for example; The mode of other adjustment existing or that possibly occur from now on evaluation and test frequency is as applicable to the present invention; Also should be included in the protection domain of the present invention, and be contained in this with way of reference at this.

Preferably, in step S6, when the variable quantity of said quality degree in double evaluation and test was lower than the frequency reducing threshold value, the network equipment reduced said evaluation and test frequency; When the variable quantity of said quality degree in double evaluation and test surpassed the raising frequency threshold value, the network equipment improved said evaluation and test frequency.Particularly, in step S6, the network equipment contrasts the variable quantity of the quality degree of double evaluation and test, when this variable quantity is lower than the frequency reducing threshold value, reduces the evaluation and test frequency of the quality degree of this popularization keyword; When this variable quantity surpasses the raising frequency threshold value, improve the evaluation and test frequency of the quality degree of this popularization keyword.The network equipment improves or reduces the evaluation and test frequency for the quality degree of popularization keyword based on the situation of change of quality degree, makes that rationality of the present invention is higher.For example, suppose that the frequency reducing threshold value is 0.2, the variable quantity of promoting the double quality degree of keyword is 0.1, then reduces the evaluation and test frequency of the quality degree of this popularization keyword, as being reduced to 1 time/day from present 2 times/day.And for example, suppose that the raising frequency threshold value is 0.8, the variable quantity of promoting the double quality degree of keyword is 1.0, then improves the evaluation and test frequency of the quality degree of this popularization keyword, as being reduced to 3 times/day from present 2 times/day.Those skilled in the art will be understood that above-mentioned reduction and improve the mode of evaluating and testing frequency and are merely for example; Other reductions existing or that possibly occur from now on or the mode that improves the evaluation and test frequency are as applicable to the present invention; Also should be included in the protection domain of the present invention, and be contained in this with way of reference at this.

Fig. 2 is an equipment synoptic diagram according to a further aspect of the present invention, illustrates based on the weighted mean point to advance the network equipment that keyword quality degree is promoted in the rate evaluation and test.At this; The network equipment 1 includes but not limited to network host, single network server, a plurality of webserver collection or based on the set of computers of cloud computing; Wherein, cloud computing is a kind of of Distributed Calculation, a super virtual machine of being made up of the loosely-coupled computing machine collection of a group.Those skilled in the art will be understood that the above-mentioned network equipment is merely for example, and other network equipments existing or that possibly occur from now on also should be included in the protection domain of the present invention, and be contained in this at this with way of reference as applicable to the present invention.

Particularly, the network equipment 1 comprises first deriving means 11, second deriving means 12, the 3rd deriving means 13 and definite device 14.Wherein, first deriving means 11 obtains the popularization keyword of user's input; Second deriving means 12 is according to said popularization keyword, and the individuality point that obtains with the corresponding said popularization keyword of this user advances rate; The 3rd deriving means 13 is according to said popularization keyword, and the weighted mean point that obtains said popularization keyword advances rate; Confirm that device 14 advances rate according to said individual point and said weighted mean point advances rate, confirm quality degree with the corresponding said popularization keyword of this user.

More specifically, first deriving means 11 obtains the popularization keyword of user's input.Particularly; First deriving means 11 obtains the popularization keyword of user input in several ways, for example, and user's mutual through with subscriber equipment; As carrying out man-machine interaction through modes such as keyboard, mouse, telepilot, touch pad, touch-screen or hand-written equipment; And subscriber equipment is connected with the network equipment 1 via network, sends the popularization keyword of this user's input to the network equipment 1, and first deriving means 11 obtains the popularization keyword of user's input in view of the above; The user promotes input popularization keyword in the account at this then through this user's in the subscriber equipment logging in network equipment 1 popularization account, and first deriving means 11 obtains the popularization keyword of this user's input from this popularization account; The user will promote keyword and import crucial dictionary, and the network equipment 1 connects with this key dictionary, and first deriving means 11 reads the popularization keyword of this user's input from this key dictionary.Those skilled in the art will be understood that the mode that above-mentioned first deriving means obtains the popularization keyword of user input is merely for example; Other first deriving means existing or that possibly occur from now on obtain the mode of popularization keyword of user input as applicable to the present invention; Also should be included in the protection domain of the present invention, and be contained in this with way of reference at this.At this, network includes but not limited to internet, wide area network, Metropolitan Area Network (MAN), LAN, VPN network, wireless self-organization network (Ad Hoc network), GSM, WCDMA, CDMA2000, TD-SCDMA, GSM, CDMA1x, WIFI, WAPI, WiMax etc.Those skilled in the art will be understood that above-mentioned network is merely for example, and other networks existing or that possibly occur from now on also should be included in the protection domain of the present invention, and be contained in this at this with way of reference as applicable to the present invention.

Subsequently, second deriving means 12 is according to said popularization keyword, and the individuality point that obtains with the corresponding said popularization keyword of this user advances rate.Particularly, the popularization keyword that the user that second deriving means 12 obtains according to first deriving means 11 imports, the individuality point that obtains this popularization keyword of this user advances rate.At this, the source that obtains that the individuality of this popularization keyword point advances rate includes but not limited to, the memory storage of the network equipment 1 itself, the management server of promoting the extension system under the keyword and other equipment that can communicate by letter with this network equipment 1.The obtain manner that the individuality point of this popularization keyword advances rate includes but not limited to; For example; The network equipment 1 itself is the management server of promoting the affiliated extension system of keyword, and second deriving means 12 obtains this popularization keyword from the historical record of this popularization keyword individuality point advances rate; And for example; Second deriving means 12 is through reading, call the communication mode of certain applications routine interface (API) or other agreements in database; The individuality point that obtains this popularization keyword of this user advances rate; For example; Second deriving means 12 advances the application programming interfaces (API) that the third party device of rate provides through the individuality point that calls this popularization keyword that stores this user, and the individuality point that sends this popularization keyword that obtains this user to this third party device advances the request of rate, and the individuality point that receives this popularization keyword of this user that this third party device returns according to this request advances rate.Those skilled in the art will be understood that above-mentioned individual point advance rate obtain the source and obtain manner is merely for example; What other individuality points existing or that possibly occur from now on advanced rate obtains source or obtain manner as applicable to the present invention; Also should be included in the protection domain of the present invention, and be contained in this with way of reference at this.

Those skilled in the art will be understood that above-mentioned first deriving means and second deriving means are merely example, and in practice, they can be two independently modules, also can be integrated in the module.

Then, the 3rd deriving means 13 is according to said popularization keyword, and the weighted mean point that obtains said popularization keyword advances rate.Particularly, the 3rd deriving means 13 is according to the popularization keyword that first deriving means 11 obtains, and the weighted mean point that obtains this popularization keyword advances rate.The obtain manner that this weighted mean point advances rate includes but not limited to: calculate acquisition by the 3rd deriving means 13; Promptly advance rate corresponding to the point of different user or different popularization unit according to this popularization keyword; And give these and put the into different weight of rate; Again these are put into rate and carry out weighted mean, advance rate (when these put the weight of rate into when identical, the weighted mean point of this popularization keyword advances rate and is the arithmetic mean point and advances rate) with the weighted mean point that obtains this popularization keyword; Directly obtain by the memory storage or the third party device of the 3rd deriving means 13 from the network equipment 1 self; For example; The 3rd deriving means 13 is through reading, call the communication mode of certain applications routine interface (API) or other agreements in database, the weighted mean point that obtains this popularization keyword advances rate.For example; The application programming interfaces (API) that the 3rd deriving means 13 provides through the third party device that calls the weighted mean point that stores this popularization keyword and advance rate; Send the weighted mean point that obtains this popularization keyword to this third party device and advance the request of rate, and the weighted mean point that receives this popularization keyword that this third party device returns according to this request advances rate.Those skilled in the art will be understood that above-mentioned obtain manner is merely for example, and other obtain manners existing or that possibly occur from now on also should be included in the protection domain of the present invention, and be contained in this at this with way of reference as applicable to the present invention.At this; The Consideration that the point of promoting keyword advances the weight of rate include but not limited to promote dog-eat-dog degree, the netizen of keyword the retrieval amount,, or promote industry under the keyword etc.; Those skilled in the art will be understood that the Consideration that the point of above-mentioned popularization keyword advances the weight of rate is merely for example; The point of other popularization keywords existing or that possibly occur from now on advances the Consideration of weight of rate as applicable to the present invention; Also should be included in the protection domain of the present invention, and be contained in this with way of reference at this.

Need to prove; Those skilled in the art will be understood that; Second deriving means 12 and the 3rd deriving means 13 are not strict sequential working relations; The individuality point that second deriving means 12 can obtain earlier with corresponding this popularization keyword of this user advances rate, and the weighted mean point that the 3rd deriving means 13 obtains this popularization keyword again advances rate; The weighted mean point that also can the 3rd deriving means 13 obtains this popularization keyword earlier advances rate, and the individuality point that second deriving means 12 obtains with corresponding this popularization keyword of this user again advances rate; Second deriving means 12 and the 3rd deriving means 13 can also be worked simultaneously, obtain the weighted mean point that individuality point with corresponding this popularization keyword of this user advances rate and this popularization keyword respectively and advance rate; At this; Second deriving means 11 and the 3rd deriving means 12 are merely example; In practice, they can be two independently modules, also can be integrated in the module; For example the network equipment 1 with third party device mutual in, this integration module receives the weighted mean point that individuality point with corresponding this popularization keyword of this user advances rate and this popularization keyword and advances rate in same request and response.For for simplicity, we explain the individuality point that second deriving means 12 obtains with corresponding this popularization keyword of this user earlier and advance rate, explain the weighted mean point that the 3rd deriving means 13 obtains this popularization keyword again and advance rate.It is convenient that those skilled in the art will be understood that such statement is merely statement in proper order, should be as limitation of the present invention.

Those skilled in the art also will be understood that; Above-mentioned first deriving means, second deriving means and the 3rd deriving means are merely example, and in practice, they can be three independently modules; Perhaps any two devices are integrated in the module, also can all be integrated in the module.

Subsequently, confirm that device 14 advances rate based on said individual point and said weighted average point advances rate, confirm quality degree with the corresponding said popularization keyword of this user.Particularly; The weighted mean point that the individuality point of confirming this popularization keyword of this user that device 14 obtains according to second deriving means 12 advances this popularization keyword that rate and the 3rd deriving means 13 obtain advances rate, through confirm the quality degree of this popularization keyword of this user such as the mode of ratio, difference.For example, confirm that weighted average point that device 14 advances rate and this popularization keyword based on the individuality point of this popularization keyword of this user advances the ratio of rate, confirms the quality degree of this popularization keyword of this user; It is 0.2% that the individuality point of supposing this popularization keyword of this user advances rate, and it is 0.15% that the weighted average point of this popularization keyword advances rate, and their ratio is 1.33 so, the quality degree of this popularization keyword that this ratio is exactly this user.And for example, confirm that weighted average point that device 14 advances rate and this popularization keyword based on the individuality point of this popularization keyword of this user advances the difference of rate, confirms the quality degree of this popularization keyword of this user; It is 0.2% that the individuality point of supposing this popularization keyword of this user advances rate, and it is 0.15% that the weighted average point of this popularization keyword advances rate, and their difference is 0.05% so, the quality degree of this popularization keyword that this difference is exactly this user.Those skilled in the art will be understood that definite mode of above-mentioned quality degree is merely for example; Definite mode of other quality degree existing or that possibly occur from now on is as applicable to the present invention; Also should be included in the protection domain of the present invention, and be contained in this with way of reference at this.

Preferably, first deriving means 11, second deriving means 12, the 3rd deriving means 13 and definite device 14 are constant work.Particularly, first deriving means 11 obtains the popularization keyword of user's input; Second deriving means 12 is according to said popularization keyword, and the individuality point that obtains with the corresponding said popularization keyword of this user advances rate; The 3rd deriving means 13 is according to said popularization keyword, and the weighted mean point that obtains said popularization keyword advances rate; Confirm that device 14 advances rate according to said individual point and said weighted mean point advances rate, confirm quality degree with the corresponding said popularization keyword of this user.At this; It will be understood by those skilled in the art that the weighted mean point that individuality point that " continuing " is meant that first deriving means 11, second deriving means 12, the 3rd deriving means 13 and definite device 14 require to promote this popularization keyword of the obtaining of keyword, this user according to mode of operation that set or adjustment in real time respectively advances the obtaining of rate, this popularization keyword advances obtaining of rate; Reach confirming of this quality degree of promoting keyword, stop to evaluate and test the quality degree of promoting keyword in a long time until this network equipment.

In a preferred embodiment, the 3rd deriving means 13 is also according to said popularization keyword, obtains and uses the user of the corresponding said popularization keyword of each user of said popularization keyword to put into rate; Put into rate according to said user, the weighted mean point that obtains said popularization keyword advances rate.Particularly; The popularization keyword that the 3rd deriving means 13 obtains according to first deriving means 11; Obtain other users that use these popularization keywords and put into rate corresponding to the user of this popularization keyword, put into rate according to this (a bit) user then, the weighted mean point that obtains this popularization keyword advances rate.For example, the 3rd deriving means 13 is put into the pairing amount of representing of rate according to the user and is given the user and put the into different weight of rate, and the weighted mean point that obtains this popularization keyword according to this weight advances rate; Suppose to promote keyword and be " fresh flower ", use the user of " fresh flower " to be user's first and user's second under the identical linguistic context, user's first puts into corresponding to the user of " fresh flower " that rate is 0.2%; The amount of representing is 1000 times; User's second puts into corresponding to the user of " fresh flower " that rate is 0.15%, and the amount of representing is 10000 times, and the user who gives user's first puts into that the weight of rate is 0.1; The user who gives user's second puts into that the weight of rate is 0.9, is 0.155% thereby the weighted mean point that obtains " fresh flower " advances rate.And for example, the weight that the user who supposes all users' same popularization keyword puts rate into is the same, and the weighted mean point of this popularization keyword advanced rate and was the arithmetic mean point and advances rate this moment; As promote keyword for " fresh flower "; User's first is put into corresponding to the user of " fresh flower ", and rate is 0.2%; User's second is put into corresponding to the user of " fresh flower ", and rate is 0.15%; User third puts into corresponding to the user of " fresh flower " that rate is 0.24%, and to advance rate be 0.197% to the weighted mean of " fresh flower " point at this moment.Those skilled in the art will be understood that the mode that the above-mentioned weighted mean point that obtains the popularization keyword advances rate is merely for example; Other existing or obtaining of possibly occurring from now on modes that weighted mean points of promoting keyword advance rate are as applicable to the present invention; Also should be included in the protection domain of the present invention, and be contained in this with way of reference at this.

In a further advantageous embodiment, the 3rd deriving means 13 obtains the one or more related keyword relevant with said popularization keyword also according to said popularization keyword; The point that obtains said one or more related keywords advances rate; The point of said one or more related keywords is advanced rate average, the weighted mean point that obtains said popularization keyword advances rate; Wherein, said one or more related keywords include but not limited to, below at least each: the 1) synonymous keyword of said popularization keyword; 2) the near adopted keyword of said popularization keyword; 3) belong to the keyword of identical industry with said popularization keyword; 4) belong to the keyword of identical region with said popularization keyword.Particularly; The popularization keyword that the 3rd deriving means 13 obtains according to first deriving means 11 obtains to promote the relevant one or more related keyword of keyword with this, for example; This is promoted the synonym of keyword, nearly adopted keyword, promotes the perhaps keyword etc. of identical region of the identical industry of keyword with this; The point that obtains this (a bit) related keyword advances rate, again this (a bit) is put rate and average, thereby the weighted mean point that obtains this popularization keyword advances rate.For example; Suppose to promote keyword and be " point advances rate "; The 3rd deriving means 13 obtains synonymous keyword " CTR " and " clicking rate " through meaning of a word analyses, then through calling the application programming interfaces (API) that third party device that the point of storing keyword advances rate provides; Send the point that obtains " CTR " and " clicking rate " to this third party device and advance the request of rate, and the point that receives " CTR " and " clicking rate " that this third party device returns according to this request advances rate; Then the point of " CTR " and " clicking rate " is advanced rate and average, the weighted mean point that obtains " point advances rate " advances rate, and the weighted mean point of this moment advances rate and also advances rate for the arithmetic mean point.And for example, still hypothesis popularization keyword is " point advances rate ", and the 3rd deriving means 13 is through meaning of a word analyses; Obtain synonymous keyword " CTR " and " clicking rate "; Nearly adopted keyword " click volume " and " equalization point advances rate ", then through reading the data of crucial dictionary, the point that obtains " CTR ", " clicking rate ", " click volume " and " equalization point advances rate " respectively advances rate; The weight that the point of setting " CTR " and " clicking rate " advances rate is 0.3; The weight that the point of " click volume " and " equalization point advances rate " advances rate is 0.2, these is put into rate ask weighted mean value, and the weighted mean point that can obtain " point advances rate " advances rate.And for example, still hypothesis popularization keyword is " point advances rate ", and the 3rd deriving means 13 is through meaning of a word analyses; Obtain synonymous keyword " CTR " and " clicking rate "; Again these " CTR " and " clicking rate " are carried out the region cluster analysis, the region is " East China ", is 0.2% thereby the point of " CTR " that acquisition is used in " East China " area advances rate; It is 0.15% that the point of " clicking rate " used in " East China " area advances rate; The weight that the point that " CTR " be set advances rate is 0.4, and the weight that the point of " clicking rate " advances rate is 0.6, and it is 0.17% that the weighted mean point that draws " point advances rate " advances rate.Those skilled in the art will be understood that the obtain manner that the weighted mean point of above-mentioned popularization keyword advances rate is merely for example with related keyword; Obtain manner or the related keyword that the weighted mean point of other popularization keywords existing or that possibly occur from now on advances rate is as applicable to the present invention; Also should be included in the protection domain of the present invention, and be contained in this with way of reference at this.

In another preferred embodiment; The 3rd deriving means 13 is also according to said popularization keyword; And the statistics point that combines the extension system under the said popularization keyword advances rate (i.e. the ratio of the click volume of all keywords and the amount of representing in this extension system), and the weighted mean point that obtains said popularization keyword advances rate.Particularly, in step S3, the network equipment is according to its popularization keyword that in step S1, obtains, and advances rate in conjunction with the statistics point of the extension system under this popularization keyword, and the weighted mean point that obtains this popularization keyword advances rate.For example to advance the confidence level of rate too low when the weighted average point of this popularizations keyword, and the statistics point of this extension system of network equipment selection advances rate and advances rate as the weighted average point of this popularization keyword; It is 0.09% that the weighted average point of supposing to promote keyword advances rate; It is 0.1% that the statistics point of extension system advances rate; The weighted average point of promoting keyword this moment advances the statistics point that rate is lower than extension system and advances rate; The statistics point that the network equipment obtains extension system advances rate; Promptly 0.1%, advance rate as the weighted average point of this popularization keyword.Preferably; The statistics point that the network equipment advances rate and extension system for the weighted mean point of promoting keyword advances rate and gives different weights; This weighted mean point is advanced rate and this statistics point advance rate and ask weighted mean value, advance rate with the new weighted mean point that obtains this popularization keyword; It is 0.09% that the weighted mean point of supposing to promote keyword advances rate; It is 0.6 that this weighted mean point advances the corresponding weight of rate, and it is 0.1% that the statistics point of extension system advances rate, and it is 0.4 that this statistics point advances the corresponding weight of rate; So, to advance rate be 0.094% to the new weighted mean point of this popularization keyword.Those skilled in the art will be understood that the obtain manner that the weighted mean point of above-mentioned popularization keyword advances rate is merely for example; The obtain manner that the weighted mean point of other popularization keywords existing or that possibly occur from now on advances rate is as applicable to the present invention; Also should be included in the protection domain of the present invention, and be contained in this with way of reference at this.At this; The source that obtains that the statistics point of the extension system under the popularization keyword advances rate includes but not limited to; The memory storage of the network equipment itself, the management server of extension system, other equipment that can communicate by letter with this network equipment; Those skilled in the art will be understood that the source that obtains that the statistics point of the extension system under the above-mentioned popularization keyword advances rate is merely for example; The statistics point of the extension system under other popularization keywords existing or that possibly occur from now on advances obtaining of rate originates as applicable to the present invention, also should be included in the protection domain of the present invention, and be contained in this at this with way of reference.

Preferably, second deriving means 12 is according to said popularization keyword, and other points of promoting keyword that obtain this user advance rate; Other points of promoting keyword according to said this user advance rate, and weighting is confirmed to advance rate with the individuality point of the corresponding said popularization keyword of this user.Particularly; Second deriving means 12 is according to the popularization keyword that first deriving means 11 obtains, and other points of promoting keyword that obtain this user advance rate, for example; This user, with this popularizations keyword in same popularization unit other the popularization keywords points advance rate; Advance rate according to this (a bit) point, put the weight of rate, thereby weighting is confirmed the individuality point of this popularization keyword of this user and advanced rate through confirm this (a bit) such as the mode of meaning of a word cluster.For example; Suppose 1 management server of the network equipment as extension system; Promoting keyword is " point advances rate "; Other popularization keywords that second deriving means 12 obtains this user in view of the above are " CTR ", " clicking rate ", " click volume " and " equalization point advances rate ", and the point that from this user's record, reads " CTR ", " clicking rate ", " click volume " and " equalization point advances rate " then advances rate, and " CTR ", " clicking rate ", " click volume " and " equalization point advances rate " are carried out meaning of a word analysis; Obtain synonymous keyword " CTR " and " clicking rate " of " point advances rate "; It is 0.3 that their point advances the corresponding weight of rate, near adopted keyword " click volume " and " equalization point advances rate " of " point advances rate ", and it is 0.2 that their point advances the corresponding weight of rate; These are put into rate ask weighted mean value, the individuality point that obtains " point advances rate " advances rate.Those skilled in the art will be understood that the obtain manner that the individuality point of above-mentioned popularization keyword advances rate is merely for example; The obtain manner that the individuality point of other popularization keywords existing or that possibly occur from now on advances rate is as applicable to the present invention; Also should be included in the protection domain of the present invention, and be contained in this with way of reference at this.At this; Other of this user are promoted keyword and are included but not limited to; This user, promote keyword other in same popularization unit with this and promote keywords; Perhaps this user, promote in unit other with these other of the same classification in popularization unit of promoting keyword place and promote keywords, perhaps this user, all promote in unit other and promote keywords.Those skilled in the art will be understood that other popularization keywords of above-mentioned this user are merely for example; Other of other these users existing or that possibly occur are from now on promoted keywords as applicable to the present invention; Also should be included in the protection domain of the present invention, and be contained in this with way of reference at this.

In a preferred embodiment, second deriving means 12 is also according to said popularization keyword, and in conjunction with this user's user related information, the individuality point that obtains with the corresponding said popularization keyword of this user advances rate.Particularly, the popularization keyword that second deriving means 12 obtains according to first deriving means 11 is in conjunction with this user's user related information; For example; The individuality point of this popularization keyword of this user advances the historical record of rate, is used for smoothly the unusual fluctuations of should the individuality point advancing rate, the submission record of this user's said popularization keyword; Be used to discern this user's malicious registration behavior, the individuality point that obtains this popularization keyword of this user advances rate.For example; The popularization keyword of supposing user's input is " fresh flower "; Advance the historical record of rate according to the individuality point of this user " fresh flower "; It is 2% that the individuality point of 3:00PM-4:00PM time period yesterday advances rate, and it is 0.02% that the individuality point of 3:00PM-4:00PM time period today advances rate, judges that the individuality point of 3:00PM-4:00PM time period today advances the influence that rate possibly receive the network user's malicious act; 12 level and smooth these unusual fluctuations of second deriving means, the individuality point that obtains " fresh flower " are advanced the history average of rate (advancing the mean value of rate like the individuality point of same period in January in past) and are advanced rate as the individuality point of 3:00PM-4:00PM time period today " fresh flower ".Those skilled in the art will be understood that the mode that the above-mentioned individuality point that obtains this popularization keyword of this user advances rate is merely for example; The mode that other individuality points existing or this popularization keyword that obtains this user that possibly occur from now on advance rate is as applicable to the present invention; Also should be included in the protection domain of the present invention, and be contained in this with way of reference at this.

Preferably, user related information includes but not limited to, below at least each: 1) the individuality point of this popularization keyword of this user advances the historical record of rate; 2) the submission record of this popularization keyword of this user.Wherein, The historical record that the individuality point of this popularization keyword of this user advances rate includes but not limited to that historical record and this user's the individuality point of this popularization keyword that the individuality point of this popularization keyword of this user advances the adjacent moment of rate advances the historical record of a period of time of rate; Can be used for smoothly the unusual fluctuations of should the individuality point advancing rate; For example, advance the adjacent moment of rate or the historical record of the past period according to the individuality point of this popularization keyword of this user, filtering is because of the network user's malicious act; Perhaps click maliciously the unusual fluctuations that the individuality point that causes advances rate in a large number like fryer search (represent but in a large number and do not click).For example; The popularization keyword of supposing user's input is " fresh flower "; Advance the historical record of rate according to the individuality point of this user " fresh flower ", yesterday this user the individuality point of " fresh flower " to advance rate be 2%, have today the netizen to use fryer search " fresh flower "; Making this user's today the individuality point of " fresh flower " advance rate is 0.02%; The difference that the individuality point of this user's of yesterday and today " fresh flower " advances rate is 1.98%, has surpassed the relative value scope (as 0.5%-0.8%) that preset individuality point advances rate, and second deriving means 12 obtains critical value (as 1.2%) that this individuality point advances rate and advances rate as the individuality point of this user of today " fresh flower ".The submission record of this popularization keyword of this user includes but not limited to that this user promotes the record of submitting this popularization keyword in the unit at one; Perhaps this user submits to this to promote the record of keyword at it in all popularization unit; Can be used to discern this user's malicious registration behavior; For example this user promotes at one and has registered the very big still very little popularization keyword of click volume of the amount of representing in the unit, makes that the quality degree of this popularization keyword is very low, and this user is in order to eliminate this submission record; The renewal of registration should be promoted keyword in other popularization unit; Hope to reach equally the very big still very little purpose of click volume of the amount of representing, after the malicious registration behavior that identifies this user, second deriving means 12 advances rate according to the individuality point in the former submission record of this popularization keyword; The individuality point of confirming this popularization keyword of new registration advances rate; It is very low that the individuality point of promptly original this popularization keyword advances rate, and it is still very low that the individuality point of this popularization keyword of new registration advances rate, thereby make that the quality degree of this popularization keyword of new registration is also very low.For example; Supposing to promote keyword is " fresh flower ", and this user promotes to register in the unit at " product " is somebody's turn to do " fresh flower ", should " fresh flower " amount of representing in this " product " popularization unit be 100000 times; Click volume is 200 times, is 0.2% thereby the individual point of acquisition advances rate; This user promotes to register once more in the unit in " service " is somebody's turn to do " fresh flower "; After this malicious registration behavior that identifies this user; Individuality point that second deriving means 12 is promoted in the unit at " product " based on this " fresh flower " advances rate, confirms that it is 0.2% that this " fresh flower " individuality point in " service " popularization unit advances rate.Those skilled in the art will be understood that above-mentioned user related information is merely for example, and other user related informations existing or that possibly occur from now on also should be included in the protection domain of the present invention, and be contained in this at this with way of reference as applicable to the present invention.

More preferably, the network equipment 1 also comprises the updating device (not shown), writes down or upgrades said user related information.Particularly, updating device writes down or upgrades this user's user related information based on the variation of user's popularization operation and/or user's popularization data.For example, when in the network equipment 1 also not during this user related information, in case this popularization keyword of this user is clicked in network or represented, this user related information of updating device record; When the network equipment 1 has had the record of user related information; When the user carries out advertisement registration etc. and promotes operation; Perhaps when the point of this popularization keyword of this user advanced rate and changes, updating device upgraded this user related information according to the variation of this user related information.For example; When keyword is promoted in one of user's new registration; Think in the network equipment 1 that not the individuality point about this popularization keyword of this user advances the historical record of rate this moment; As netizen during in accessing network resources, this popularization keyword of this user is represented or is clicked, and updating device will write down this user related information (promptly the individuality point of this popularization keyword of this user advances the original records of rate).Based on this; Second deriving means 12 calls this user related information; The individuality point of judging this popularization keyword of this user advances rate and unusual fluctuations whether occur; When second deriving means 12 judges that it is unusual fluctuations that the individuality point of this popularization keyword of this user this moment advances rate; And after confirming that according to preset critical the individuality point of this popularizations keyword of this user this moment advances rate, updating device upgrades this user related information (promptly the individuality point of this popularization keyword of this user advances the historical record of rate at this moment) according to this preset critical.Those skilled in the art will be understood that the mode of above-mentioned record and renewal user related information is merely for example; Other records existing or that possibly occur from now on or the mode of upgrading user related information are as applicable to the present invention; Also should be included in the protection domain of the present invention, and be contained in this with way of reference at this.

In a further advantageous embodiment, the network equipment 1 also comprises the adjusting gear (not shown), according to the variation of said quality degree, and the evaluation and test frequency of the said popularization keyword of adjustment evaluation and test; According to adjusted evaluation and test frequency, evaluate and test the quality degree of said popularization keyword.Particularly, the adjusting gear basis is confirmed the quality degree of the popularization keyword that device 14 obtains, the variation of obtaining this quality degree, and according to this variation, the evaluation and test frequency of this popularization keyword of adjustment evaluation and test.For example, if the variation of (as continuous two days) is less in the unit interval, adjusting gear then reduces the evaluation and test frequency of this popularization keyword of evaluation and test; If the variation of (as continuous two days) is bigger in the unit interval; Adjusting gear then improves the evaluation and test frequency of this popularization keyword of evaluation and test; Then according to adjusted evaluation and test frequency; Carry out the evaluation and test of the quality degree of this follow-up popularization keyword, the adaptive in view of the above adjustment of adjusting gear is promoted the evaluation and test frequency of the quality degree of keyword for this.Those skilled in the art will be understood that the mode of above-mentioned adjustment evaluation and test frequency is merely for example; The mode of other adjustment existing or that possibly occur from now on evaluation and test frequency is as applicable to the present invention; Also should be included in the protection domain of the present invention, and be contained in this with way of reference at this.

Preferably, when the variable quantity of said quality degree in double evaluation and test was lower than the frequency reducing threshold value, adjusting gear reduced said evaluation and test frequency; When the variable quantity of said quality degree in double evaluation and test surpassed the raising frequency threshold value, adjusting gear improved said evaluation and test frequency.Particularly, adjusting gear contrasts the variable quantity of the quality degree of double evaluation and test, when this variable quantity is lower than the frequency reducing threshold value, reduces the evaluation and test frequency of the quality degree of this popularization keyword; When this variable quantity surpasses the raising frequency threshold value, improve the evaluation and test frequency of the quality degree of this popularization keyword.Adjusting device improves or reduces the evaluation and test frequency for the quality degree of popularization keyword based on the situation of change of quality degree, makes that rationality of the present invention is higher.For example, suppose that the frequency reducing threshold value is 0.2, the variable quantity of promoting the double quality degree of keyword is 0.1, and adjusting gear then reduces the evaluation and test frequency of the quality degree of this popularization keyword, as being reduced to 1 time/day from present 2 times/day.And for example, suppose that the raising frequency threshold value is 0.8, the variable quantity of promoting the double quality degree of keyword is 1.0, and adjusting gear then improves the evaluation and test frequency of the quality degree of this popularization keyword, as being reduced to 3 times/day from present 2 times/day.Those skilled in the art will be understood that above-mentioned reduction and improve the mode of evaluating and testing frequency and are merely for example; Other reductions existing or that possibly occur from now on or the mode that improves the evaluation and test frequency are as applicable to the present invention; Also should be included in the protection domain of the present invention, and be contained in this with way of reference at this.

To those skilled in the art, obviously the invention is not restricted to the details of above-mentioned example embodiment, and under the situation that does not deviate 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 accompanying claims rather than above-mentioned explanation, therefore is intended to the implication of the equivalents that drops on claim and all changes in the scope are included in the present invention.Should any Reference numeral in the claim be regarded as limit related claim.In addition, obviously other unit or step do not got rid of in " comprising " speech, and odd number is not got rid of plural number.A plurality of unit of stating in system's claim or device also can be realized through 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

1. one kind is advanced the method that keyword quality degree is promoted in rate evaluation and test by computer implemented based on the weighted mean point, and this method may further comprise the steps:

A obtains the popularization keyword of user's input;

2. method according to claim 1, wherein, said step c also comprises:

-according to said popularization keyword, obtain and use the user of the corresponding said popularization keyword of each user of said popularization keyword to put into rate;

-put into rate according to said user, the weighted mean point that obtains said popularization keyword advances rate.

3. method according to claim 1, wherein, said step c also comprises:

-according to said popularization keyword, obtain the one or more related keyword relevant with said popularization keyword;

-the point that obtains said one or more related keywords advances rate;

-point of said one or more related keywords being advanced rate average, the weighted mean point that obtains said popularization keyword advances rate;

Wherein, said one or more related keyword comprise following at least each:

The synonymous keyword of-said popularization keyword;

The near adopted keyword of-said popularization keyword;

-belong to the keyword of identical industry with said popularization keyword;

-belong to the keyword of identical region with said popularization keyword.

4. method according to claim 1, wherein, said step c also comprises:

-based on said popularization keyword, and combine the statistics point of the extension system under the said popularization keyword to advance rate, the weighted average point that obtains said popularization keyword advances rate.

5. according to each described method in the claim 1 to 4, wherein, said step b also comprises:

-according to said popularization keyword, other points of promoting keyword that obtain this user advance rate;

-advance rate based on other points of promoting keyword of said this user, weighting is confirmed to advance rate with the individuality point of the corresponding said popularization keyword of this user.

6. according to each described method in the claim 1 to 4, wherein, said step b also comprises:

-based on said popularization keyword, in conjunction with this user's user related information, the individuality point that obtains with the corresponding said popularization keyword of this user advances rate.

7. method according to claim 6, wherein, said user related information comprise following at least each:

The individuality point of-this user's said popularization keyword advances the historical record of rate;

The submission record of-this user's said popularization keyword.

8. according to claim 6 or 7 described methods, wherein, this method also comprises:

E record or upgrade said user related information.

9. according to each described method in the claim 1 to 8, wherein, this method also comprises step f:

-based on the variation of said quality degree, the evaluation and test frequency of said popularization keyword is evaluated and tested in adjustment;

-based on adjusted evaluation and test frequency, evaluate and test the quality degree of said popularization keyword.

10. method according to claim 9, wherein, said step f also comprises:

-when the variable quantity of said quality degree in double evaluation and test is lower than the frequency reducing threshold value, reduce said evaluation and test frequency;

-when the variable quantity of said quality degree in double evaluation and test surpasses the raising frequency threshold value, improve said evaluation and test frequency.

11. one kind is advanced rate based on the weighted mean point and evaluates and tests the equipment of promoting keyword quality degree, this equipment comprises:

12. equipment according to claim 11, wherein, said the 3rd deriving means also is used for:

13. equipment according to claim 11, wherein, said the 3rd deriving means also is used for:

-the point that obtains said one or more related keywords advances rate;

Wherein, said one or more related keyword comprise following at least each:

The synonymous keyword of-said popularization keyword;

The near adopted keyword of-said popularization keyword;

-belong to the keyword of identical industry with said popularization keyword;

-belong to the keyword of identical region with said popularization keyword.

14. equipment according to claim 11, wherein, said the 3rd deriving means also is used for:

15. according to each described equipment in the claim 11 to 14, wherein, said second deriving means also is used for:

16. according to each described equipment in the claim 11 to 14, wherein, said second deriving means also is used for:

17. equipment according to claim 16, wherein, said user related information comprise following at least each:

The submission record of-this user's said popularization keyword.

18. according to claim 16 or 17 described equipment, wherein, this equipment also comprises:

Updating device is used for record or upgrades said user related information.

19. according to each described equipment in the claim 11 to 18, wherein, this equipment also comprises adjusting gear, said adjusting gear is used for:

20. equipment according to claim 19, wherein, said adjusting gear also is used for: