CN104899462B - Determine the method and device of the action value of information releasing position - Google Patents

Determine the method and device of the action value of information releasing position Download PDF

Info

Publication number
CN104899462B
CN104899462B CN201510338063.2A CN201510338063A CN104899462B CN 104899462 B CN104899462 B CN 104899462B CN 201510338063 A CN201510338063 A CN 201510338063A CN 104899462 B CN104899462 B CN 104899462B
Authority
CN
China
Prior art keywords
releasing position
information releasing
publisher
described information
actual bid
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Active
Application number
CN201510338063.2A
Other languages
Chinese (zh)
Other versions
CN104899462A (en
Inventor
张涛
吴尧
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Beijing Hongxiang Technical Service Co Ltd
Original Assignee
Beijing Qihoo Technology Co Ltd
Qizhi Software Beijing Co Ltd
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Beijing Qihoo Technology Co Ltd, Qizhi Software Beijing Co Ltd filed Critical Beijing Qihoo Technology Co Ltd
Priority to CN201510338063.2A priority Critical patent/CN104899462B/en
Publication of CN104899462A publication Critical patent/CN104899462A/en
Application granted granted Critical
Publication of CN104899462B publication Critical patent/CN104899462B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Abstract

The invention provides a kind of method for the action value for determining information releasing position, including:For a keyword, the actual bid of multiple publishers on an information releasing position in scheduled duration is obtained;According to the actual bid of multiple publishers, expectancy of each publisher to information releasing position is determined;Based on each publisher to the expectancy of information releasing position and the quality score released news for the keyword corresponding with each publisher, it is determined that the sample action value of information releasing position corresponding with each publisher;Sample action value progress statistical disposition pair information releasing position corresponding with multiple publishers difference, determines the minimum value degree of information releasing position.In embodiments of the invention, be it is follow-up actually the deducting fees there is provided strong guarantee of the publisher is determined based on expectancy, and then can effectively and rapidly solve to sell problem cheap caused by the value of information releasing position is underestimated;And improve the ageing of information issuing system.

Description

Determine the method and device of the action value of information releasing position
Technical field
The present invention relates to field of computer technology, specifically, the present invention relates to a kind of valency for determining information releasing position The method and device of value degree.
Background technology
Bid release news be one kind independently delivered, managed independently by publisher, according to information distribution effect pay it is new Type Web information issuance form.The function that can be completed by this form of information issue includes:Support publisher in real time to hair The bid of cloth information is modified, and changes the particular content released news, selects information releasing position.Ensureing the function of the above In the case of, it is specified that information issuing system needs consideration to distribute to corresponding information releasing position when each displaying releases news Release news, and publisher is deducted fees;Accordingly, it would be desirable to set up a set of effective and reasonable information releasing position distribution side Formula and the scheme of deducting fees to publisher.
At present, the conventional information releasing position method of salary distribution is:By the bid of the quality score released news and publisher Be multiplied, obtain ranking score score, score be ranked up in the way of from big to small, based on score to release news into The distribution of row information releasing position, maximum the releasing news of score is assigned to first information releasing position, by that analogy.Often The mode deducted fees to publisher is:When the satisfaction that releases news positioned at some information releasing position deducts fees condition, For example:Deducted fees according to number of clicks release news to click as user releases news, and is deducted fees according to number of times is showed Release news and be presented to user as that will release news;Specific calculation of deducting fees is:The issue of some information releasing position Information is deducted fees as the score divided by the quality score that releases news of next information releasing position.
There is following problem in existing the information releasing position method of salary distribution and scheme of deducting fees:(1) when based on a certain In publisher's short time of keyword during lazy weight, the competition between publisher can be caused not fierce enough, deduct fees relatively low, underestimate The real value of the keyword;(2) when the overbid of some publisher forcibly occupies first information releasing position, short Its uncontrollable behavior of time, causes system resource allocation unfair;(3) when pre- forward price of some publisher to some keyword When value is significantly larger than other publishers, the problem of collection of charges is relatively low can be produced, causes keyword value to be underestimated.
The content of the invention
To overcome above-mentioned technical problem or solving above-mentioned technical problem at least in part, spy proposes following technical scheme:
Embodiments of the invention propose a kind of method for the action value for determining information releasing position, including:
For a keyword, actually going out for the interior multiple publishers on an information releasing position of scheduled duration is obtained Valency;
According to the actual bid of multiple publishers, expectancy of each publisher to information releasing position is determined;
Based on each publisher to the expectancy of information releasing position and it is corresponding with each publisher be directed to the key The quality score released news of word, it is determined that the sample action value of information releasing position corresponding with each publisher;
Sample action value progress statistical disposition pair information releasing position corresponding with multiple publishers difference, determines information The minimum value degree of releasing position.
Embodiments of the invention also proposed a kind of device for the action value for determining information releasing position, including:
Actual bid acquisition module, for for a keyword, obtaining in scheduled duration in an information releasing position The actual bid of upper multiple publishers;
Expectancy determining module, for the actual bid according to multiple publishers, determines that each publisher is sent out information The expectancy of cloth position;
Sample action value determining module, for based on each publisher to the expectancy of information releasing position and with it is each Publisher is directed to the quality score released news of the keyword accordingly, it is determined that information corresponding with each publisher issues position The sample action value put;
Minimum value degree determining module, the sample for pair information releasing position corresponding with multiple publishers difference is worth Degree carries out statistical disposition, determines the minimum value degree of information releasing position.
In embodiments of the invention, each publisher reflects the publisher to the expectancy of an information releasing position To the psychological endurance scope of the information releasing position, by minimum value degree of each publisher to an information releasing position Accurately estimated, psychological endurance limiting value of the publisher to the information releasing position can be obtained, be subsequently to be based on pre- forward price Value determines actually deducting fees there is provided strong guarantee for the publisher, and then can effectively and rapidly solve information releasing position Value sells problem cheap caused by being underestimated;And the present invention will not increase the load pressure of system, improve information issuing system It is ageing.Meanwhile, pass through the computing of formula 5, it is ensured that minimum value degree be using maximize the income of information issuing system as Target is calculated, the system income of information issuing system is improved.
The additional aspect of the present invention and advantage will be set forth in part in the description, and these will become from the following description Obtain substantially, or recognized by the practice of the present invention.
Brief description of the drawings
Of the invention above-mentioned and/or additional aspect and advantage will become from the following description of the accompanying drawings of embodiments Substantially and be readily appreciated that, wherein:
Fig. 1 is the schematic flow sheet of method one embodiment of the action value of determination information releasing position in the present invention;
Fig. 2 is the structural representation of device one embodiment of the action value of determination information releasing position in the present invention.
Embodiment
Embodiments of the invention are described below in detail, the example of the embodiment is shown in the drawings, wherein from beginning to end Same or similar label represents same or similar element or the element with same or like function.Below with reference to attached The embodiment of figure description is exemplary, is only used for explaining the present invention, and is not construed as limiting the claims.
Those skilled in the art of the present technique are appreciated that unless expressly stated, singulative " one " used herein, " one It is individual ", " described " and "the" may also comprise plural form.It is to be further understood that what is used in the specification of the present invention arranges Diction " comprising " refer to there is the feature, integer, step, operation, element and/or component, but it is not excluded that in the presence of or addition Other one or more features, integer, step, operation, element, component and/or their group.It should be understood that when we claim member Part is " connected " or during " coupled " to another element, and it can be directly connected or coupled to other elements, or can also exist Intermediary element.In addition, " connection " used herein or " coupling " can include wireless connection or wireless coupling.It is used herein to arrange Taking leave "and/or" includes one or more associated wholes or any cell for listing item and all combines.
Those skilled in the art of the present technique are appreciated that unless otherwise defined, all terms used herein (including technology art Language and scientific terminology), with the general understanding identical meaning with the those of ordinary skill in art of the present invention.Should also Understand, those terms defined in such as general dictionary, it should be understood that with the context with prior art The consistent meaning of meaning, and unless by specific definitions as here, otherwise will not use idealization or excessively formal implication To explain.
Fig. 1 is the schematic flow sheet of method one embodiment of the action value of determination information releasing position in the present invention.
Step S110:For a keyword, the interior multiple publishers on an information releasing position of scheduled duration are obtained Actual bid;Step S120:According to the actual bid of multiple publishers, determine each publisher to described information releasing position Expectancy;Step S130:Based on each publisher to the expectancy of information releasing position and corresponding to each publisher The quality score released news for the keyword, it is determined that the sample valency of information releasing position corresponding with each publisher Value degree;Step S140:Sample action value progress statistical disposition pair information releasing position corresponding with multiple publishers difference, really Determine the minimum value degree of information releasing position.
In embodiments of the invention, each publisher reflects the publisher to the expectancy of an information releasing position To the psychological endurance value of the information releasing position, by minimum value degree of each publisher to an information releasing position Accurately estimated, the publisher can be obtained the psychological endurance of the information releasing position is worth, be subsequently to be based on expectancy To determine actually deducting fees there is provided strong guarantee for the publisher, and then it can effectively and rapidly solve the valency of information releasing position Value sells problem cheap caused by being underestimated;And embodiments of the present invention will not increase the load pressure of system, improve information Delivery system it is ageing;Meanwhile, pass through the computing of formula 5, it is ensured that minimum value degree is to maximize information issuing system Income for calculate target, improve information issuing system system income.
Step S110:For a keyword, the interior multiple publishers on an information releasing position of scheduled duration are obtained Actual bid.
For example, for keyword " Beijing fresh flower express delivery ", obtaining in scheduled duration and owning on an information releasing position Or part publisher is the actual bid paid of releasing news for " Beijing fresh flower express delivery ".
Step S120:According to the actual bid of multiple publishers, expection of each publisher to information releasing position is determined Value.
Specifically, according to the actual bid of multiple publishers, pre- forward price of each publisher to information releasing position is determined The mode of value includes but is not limited to:
1) by the actual bid of publisher directly as its expectancy to information releasing position;For example, for key Word " Beijing fresh flower express delivery ", actual bid of the publisher on a certain information releasing position is 5 yuan, then the publisher is to the information The expectancy of releasing position is 5 yuan.
2) step S120 also includes step S121 (not shown)s and step S122 (not shown)s;Step S121: Obtain the actual bid on the next information releasing position of scheduled duration internal information releasing position and release news accordingly Quality score;Step S122:According to the actual bid on information releasing position and the quality score released news accordingly, and With reference to the actual bid on next information releasing position and the quality score released news accordingly, counted by prediction model Calculate, determine expectancy of each publisher to information releasing position.
Specifically, there is n publisher to compete k information releasing position, and n >=k, information is represented using t ∈ { 1 ..., k } The numbering of releasing position, the numbering of publisher is represented with i ∈ { 1 ..., n }, defines viRepresent publisher i expectancy, biTable Show publisher i actual bid;Define xt∈ (0,1] represent releasing position parameter, ei∈ (0,1] represent the quality that releases news Score value, releasing position parameter xtRepresent that be in t-th of releasing news for information releasing position is presented to user's by search Probability, and assume that releasing position parameter is successively decreased with the increase of information releasing position, i.e. x1> x2> ... > xk;Issue The quality score e of informationiUser clicks on the probability released news when representing to be presented to user when releasing news.According to auction mechanism In separability condition, the clicking rate released news is equal to releasing position parameter and the product of quality score released news.
Original distribution and the scheme of deducting fees is all release news according to ei·biIt is descending to be ranked up, define ei· biRepresent publisher i original value degree, and each release news meet when deducting fees condition it is original deduct fees for:
In information issuing system, publisher can not be changed their bid by time restriction, can also be moved back at any time Go out or into queue of vying for purchase.When just proceeding by information issue, every publisher only knows the actual bid of oneself and pre- The private informations such as forward price value, it is not known that the information of other publishers, but go study by constantly adjusting the bid of oneself repeatedly With infer other publishers true value be possible.By constantly attempting, publisher can converge on a kind of stabilization The publisher of state, i.e., one can not exchange position come additional income by the publisher with other positions;Now this information is sent out It is covered with sufficient below equation:
(vi-qi)eixi≥(vi-qj)eixjI ≠ j formula 2
Wherein, qiei=bi+1ei+1, represent that publisher i is put in i-th bit and release news than the receipts that released news in other j positions It is beneficial big.
Determined by arranging above-mentioned formula with lower inequality:
Formula 3
According to formula 3, it is determined that
Formula 4
Wherein,
Obtain the actual bid b on next information releasing position of scheduled duration internal information releasing positioni+1And it is corresponding The quality score e released newsi+1;According to the actual bid b on information releasing positioniAnd the quality released news accordingly Score value ei, and combine the actual bid b on next information releasing positioni+1And the quality score e released news accordinglyi+1, lead to Cross formula 4 to be calculated, determine expectancy v of each publisher to information releasing positioni
Preferably, this method also includes step S150 (not shown)s and step S160 (not shown)s;Step S150:When the sample size of the actual bid of multiple publishers in scheduled duration is less than predetermined sample amount threshold, to key Word is extended processing, it is determined that the one or more expanded keywords related to keyword;Step S160:For one or more Expanded keyword, obtains the actual bid on scheduled duration internal information releasing position and the quality score released news accordingly And actual bid on next information releasing position of information releasing position and the quality score that releases news accordingly;Step Rapid S122:According to for actual bid of the keyword on information releasing position and the quality score accordingly released news, With the actual bid on next information releasing position and the quality score released news accordingly, and for one or more Actual bid on the information releasing position of expanded keyword and the quality score released news accordingly and in next information Actual bid on releasing position and the quality score released news accordingly, are calculated by prediction model, are determined each Expectancy of the publisher to information releasing position.
Specifically, when the sample size of the actual bid of multiple publishers in scheduled duration is less than predetermined sample amount threshold When, processing is extended to keyword, it is determined that the one or more expanded keywords related to keyword.For example, will be with key The similar keyword of word text regard the keyword similar to keywords semantics as expanded keyword as expanded keyword.Pin To one or more expanded keywords, the actual bid b on scheduled duration internal information releasing position is obtainediAnd corresponding issue The quality score e of informationiAnd the actual bid b on next information releasing position of information releasing positioni+1And corresponding hair The quality score e of cloth informationi+1;According to the actual bid b on information releasing position for keywordiAnd corresponding issue The quality score e of informationi, and the actual bid b on next information releasing positioni+1And the quality released news accordingly point Value ei+1, and for the actual bid b on the information releasing position of one or more expanded keywordsiAnd corresponding issue letter The quality score e of breathiAnd the actual bid b on next information releasing positioni+1And the quality score released news accordingly ei+1, calculated by formula 4, determine expectancy of each publisher to information releasing position.
Step S130:Based on each publisher to the expectancy of information releasing position and pin corresponding with each publisher To the quality score released news of the keyword, it is determined that the sample value of information releasing position corresponding with each publisher Degree.
Specifically, the sample action value for defining information releasing position corresponding with publisher i is si=ei·vi;Based on every Expectancy v of one publisher to information releasing positioniAnd releasing news for the keyword corresponding with each publisher Quality score ei, pass through si=ei·viCalculate, it is determined that the sample action value of information releasing position corresponding with each publisher si
Step S140:The sample action value of pair information releasing position corresponding with multiple publishers difference carries out Statistics Division Reason, determines the minimum value degree of information releasing position.
Specifically, step S140 includes step S141 (not shown)s and step S142 (not shown)s.
Step S141:Set up the probability distribution of the sample action value of information releasing position corresponding with multiple publishers difference Function and probability density function.
Specifically, the sample action value of information releasing position corresponding with multiple publishers difference obeys lognormal point Cloth, probability-distribution function is logarithm normal distribution function, and probability density function is the probability density function of logarithm normal distribution;Example Such as, sample action value si=ei·viMeet probability-distribution function F (si) and probability density function f (si)。
Step S142:Based on probability-distribution function and probability density function, the minimum value degree of information releasing position is determined.
Specifically, use
Formula 5
Determine the minimum value degree score_limit of information releasing position;Specifically, withAccordingly Zero point on curve, the corresponding s at zero pointiValue is minimum value degree score_limit;Wherein, siFor with any publisher's phase The sample action value for the described information releasing position answered, and siMeet probability-distribution function F (si) and probability density function f (si)。
In embodiments of the invention, each publisher reflects the publisher to the expectancy of an information releasing position To the psychological endurance scope of the information releasing position, by minimum value degree of each publisher to an information releasing position Accurately estimated, psychological endurance limiting value of the publisher to the information releasing position can be obtained, be subsequently to be based on pre- forward price Value determines actually deducting fees there is provided strong guarantee for the publisher, and then can effectively and rapidly solve information releasing position Value sells problem cheap caused by being underestimated;And the present invention will not increase the load pressure of system, improve information issuing system It is ageing.Meanwhile, pass through the computing of formula 5, it is ensured that minimum value degree be using maximize the income of information issuing system as Target is calculated, the system income of information issuing system is improved.
In a preferred embodiment (reference picture 1), this method also includes step S170 (not shown)s, step S170: For a publisher, according to the minimum value degree of information releasing position, and combine corresponding with the publisher for the key The quality score released news of word, determines that minimum on described information releasing position of the publisher is deducted fees.
Specifically, for a publisher, its corresponding releasing news in information releasing position for keyword is defined On it is minimum deduct fees as the ratio of the minimum value degree score_limit quality scores released news with this, i.e., it is minimum to deduct fees:
Formula 6
In the preferred embodiment, the minimum quality score deducted fees and released news is inversely proportional, therefore same information is sent out Minimum deduct fees of cloth position may be constantly different, can if the quality score released news on the information releasing position is higher Calculating determine it is less it is minimum deduct fees, determine larger minimum deduct fees conversely, then calculating;This preferred embodiment can promote hair Cloth person updates the quality that it releases news, so as to lift information distribution effect so that be presented in the releasing news of user Including more abundant, accurate content information, user's viewing experience is improved.
Preferably (reference picture 1), this method also includes step S180 (not shown)s, step S180:According to the issue Minimum on information releasing position of person is deducted fees, and combines the actual bid of the publisher, determines that the publisher issues in information Actually deducting fees on position.
Specifically, it is determined that the mode actually deducted fees includes:
1) size of minimum deduct fees p and the actual bid of the publisher of the publisher on information releasing position is compared Bi, selection smaller value is actually to deduct fees;
2) minimum deduct fees p, the actual bid bi and root of the publisher of the publisher on information releasing position are compared The original p that deducts fees determined according to formula 1Original iSize, to determine actually to deduct fees;For example, the min that actually deducts fees (bi, max (p, pOriginal i)), that is, select the minimum p and original p that deducts fees that deducts feesOriginal iIn higher value, then actually going out in the higher value and publisher Smaller value is selected in valency bi as actually deducting fees.
In the preferred embodiment of this programme, because publisher reflects it to information to the expectancy of information releasing position The psychological endurance value of releasing position, determines the scheme of deducting fees, it can be ensured that in the button based on obtained expectancy is accurately estimated Publisher will not frequently change the bid of oneself under expense scheme, reduce the information processing pressure of system, the stability of lifting system.
Fig. 2 is the structural representation of device one embodiment of the action value of determination information releasing position in the present invention.
Actual bid acquisition module 210 is obtained in scheduled duration on an information releasing position for a keyword The actual bid of multiple publishers;Expectancy determining module 220 determines each issue according to the actual bid of multiple publishers Expectancy of the person to described information releasing position;Sample action value determining module 230 is based on each publisher and information is issued The expectancy of position and the quality score released news for the keyword corresponding with each publisher, it is determined that with it is each The sample action value of the corresponding information releasing position of publisher;240 pairs of minimum value degree determining module is distinguished with multiple publishers The sample action value of corresponding information releasing position carries out statistical disposition, determines the minimum value degree of information releasing position.
In embodiments of the invention, each publisher reflects the publisher to the expectancy of an information releasing position To the psychological endurance value of the information releasing position, by minimum value degree of each publisher to an information releasing position Accurately estimated, the publisher can be obtained the psychological endurance of the information releasing position is worth, be subsequently to be based on expectancy To determine actually deducting fees there is provided strong guarantee for the publisher, and then it can effectively and rapidly solve the valency of information releasing position Value sells problem cheap caused by being underestimated;And embodiments of the present invention will not increase the load pressure of system, improve information Delivery system it is ageing;Meanwhile, pass through the computing of formula 5, it is ensured that minimum value degree is to maximize information issuing system Income for calculate target, improve information issuing system system income.
Actual bid acquisition module 210 is obtained in scheduled duration on an information releasing position for a keyword The actual bid of multiple publishers.
For example, for keyword " Beijing fresh flower express delivery ", obtaining in scheduled duration and owning on an information releasing position Or part publisher is the actual bid paid of releasing news for " Beijing fresh flower express delivery ".
Expectancy determining module 220 determines that each publisher issues to information according to the actual bid of multiple publishers The expectancy of position.
Specifically, according to the actual bid of multiple publishers, pre- forward price of each publisher to information releasing position is determined The mode of value includes but is not limited to:
1) by the actual bid of publisher directly as its expectancy to information releasing position;For example, for key Word " Beijing fresh flower express delivery ", actual bid of the publisher on a certain information releasing position is 5 yuan, then the publisher is to the information The expectancy of releasing position is 5 yuan.
2) expectancy determining module 220 includes first acquisition unit (not shown) and computing unit (does not show in figure Go out);First acquisition unit obtain scheduled duration internal information releasing position next information releasing position on actual bid and The quality score released news accordingly;Then, computing unit is according to the actual bid on information releasing position and corresponding The quality score released news, and combine the actual bid on next information releasing position and the quality released news accordingly point Value, is calculated by prediction model, determines expectancy of each publisher to information releasing position.
Specifically, there is n publisher to compete k information releasing position, and n >=k, information is represented using t ∈ { 1 ..., k } The numbering of releasing position, the numbering of publisher is represented with i ∈ { 1 ..., n }, defines viRepresent publisher i expectancy, biTable Show publisher i actual bid;Define xt∈ (0,1] represent releasing position parameter, ei∈ (0,1] represent the quality that releases news Score value, releasing position parameter xtRepresent that be in t-th of releasing news for information releasing position is presented to user's by search Probability, and assume that releasing position parameter is successively decreased with the increase of information releasing position, i.e. x1> x2> ... > xk;Issue The quality score e of informationiUser clicks on the probability released news when representing to be presented to user when releasing news.According to auction mechanism In separability condition, the clicking rate released news is equal to releasing position parameter and the product of quality score released news.
Original distribution and the scheme of deducting fees is all release news according to ei·biIt is descending to be ranked up, define ei· biRepresent publisher i original value degree, and each release news meet when deducting fees condition it is original deduct fees for:
In information issuing system, publisher can not be changed their bid by time restriction, can also be moved back at any time Go out or into queue of vying for purchase.When just proceeding by information issue, every publisher only knows the actual bid of oneself and pre- The private informations such as forward price value, it is not known that the information of other publishers, but go study by constantly adjusting the bid of oneself repeatedly With infer other publishers true value be possible.By constantly attempting, publisher can converge on a kind of stabilization The publisher of state, i.e., one can not exchange position come additional income by the publisher with other positions;Now this information is sent out It is covered with sufficient below equation:
(vi-qi)eixi≥(vi-qj)eixjI ≠ j formula 2
Wherein, qiei=bi+1ei+1, represent that publisher i is put in i-th bit and release news than the receipts that released news in other j positions It is beneficial big.
Determined by arranging above-mentioned formula with lower inequality:
Formula 3
According to formula 3, it is determined that
Formula 4
Wherein,
Obtain the actual bid b on next information releasing position of scheduled duration internal information releasing positioni+1And it is corresponding The quality score e released newsi+1;According to the actual bid b on information releasing positioniAnd the quality released news accordingly Score value ei, and combine the actual bid b on next information releasing positioni+1And the quality score e released news accordinglyi+1, lead to Cross formula 4 to be calculated, determine expectancy v of each publisher to information releasing positioni
Preferably, the device also includes expanded keyword determining module (not shown) and relevant information acquisition module (not shown);When the sample size of the actual bid of multiple publishers in scheduled duration is less than predetermined sample amount threshold When, expanded keyword determining module is extended processing to keyword, it is determined that the one or more extensions related to keyword are closed Keyword;For one or more expanded keywords, relevant information acquisition module is obtained on scheduled duration internal information releasing position Actual bid and the quality score that releases news accordingly and on next information releasing position of information releasing position The actual quality score bid and released news accordingly;Computing unit according to for keyword on the information releasing position The actual quality score bid and released news accordingly, and the actual bid on next information releasing position and corresponding hair The quality score of cloth information, and for the actual bid on the information releasing position of one or more expanded keywords and accordingly The quality score released news and actual bid on next information releasing position and the quality that releases news accordingly Score value, is calculated by prediction model, determines expectancy of each publisher to information releasing position.
Specifically, when the sample size of the actual bid of multiple publishers in scheduled duration is less than predetermined sample amount threshold When, processing is extended to keyword, it is determined that the one or more expanded keywords related to keyword.For example, will be with key The similar keyword of word text regard the keyword similar to keywords semantics as expanded keyword as expanded keyword.Pin To one or more expanded keywords, the actual bid b on scheduled duration internal information releasing position is obtainediAnd corresponding issue The quality score e of informationiAnd the actual bid b on next information releasing position of information releasing positioni+1And corresponding hair The quality score e of cloth informationi+1;According to the actual bid b on information releasing position for keywordiAnd corresponding issue The quality score e of informationi, and the actual bid b on next information releasing positioni+1And the quality released news accordingly point Value ei+1, and for the actual bid b on the information releasing position of one or more expanded keywordsiAnd corresponding issue letter The quality score e of breathiAnd the actual bid b on next information releasing positioni+1And the quality score released news accordingly ei+1, calculated by formula 4, determine expectancy of each publisher to information releasing position.
Sample action value determining module 230 based on each publisher to the expectancy of information releasing position and with each hair Cloth person is directed to the quality score released news of the keyword accordingly, it is determined that information releasing position corresponding with each publisher Sample action value.
Specifically, the sample action value for defining information releasing position corresponding with publisher i is si=ei·vi;Based on every Expectancy v of one publisher to information releasing positioniAnd releasing news for the keyword corresponding with each publisher Quality score ei, pass through si=ei·viCalculate, it is determined that the sample action value of information releasing position corresponding with each publisher si
The sample action value of 240 pairs of minimum value degree determining module information releasing position corresponding with multiple publishers difference Statistical disposition is carried out, the minimum value degree of information releasing position is determined.
Specifically, minimum value degree determining module 240 sets up unit (not shown) and minimum value degree including function Determining unit (not shown).
Function sets up the probability that unit sets up the sample action value of information releasing position corresponding with multiple publishers difference Distribution function and probability density function.
Specifically, the sample action value of information releasing position corresponding with multiple publishers difference obeys lognormal point Cloth, probability-distribution function is logarithm normal distribution function, and probability density function is the probability density function of logarithm normal distribution;Example Such as, sample action value si=ei·viMeet probability-distribution function F (si) and probability density function f (si)。
Minimum value degree determining unit is based on probability-distribution function and probability density function, determines information releasing position most Low value degree.
Specifically, use
Formula 5
Determine the minimum value degree score_limit of information releasing position;Specifically, withAccordingly Zero point on curve, the corresponding s at zero pointiValue is minimum value degree score_limit;Wherein, siFor with any publisher's phase The sample action value for the described information releasing position answered, and siMeet probability-distribution function F (si) and probability density function f (si)。
In embodiments of the invention, each publisher reflects the publisher to the expectancy of an information releasing position To the psychological endurance scope of the information releasing position, by minimum value degree of each publisher to an information releasing position Accurately estimated, psychological endurance limiting value of the publisher to the information releasing position can be obtained, be subsequently to be based on pre- forward price Value determines actually deducting fees there is provided strong guarantee for the publisher, and then can effectively and rapidly solve information releasing position Value sells problem cheap caused by being underestimated;And the present invention will not increase the load pressure of system, improve information issuing system It is ageing.Meanwhile, pass through the computing of formula 5, it is ensured that minimum value degree be using maximize the income of information issuing system as Target is calculated, the system income of information issuing system is improved.
In a preferred embodiment (reference picture 2), the device also includes minimum determining module (not shown) of deducting fees, most Low determining module of deducting fees according to the minimum value degree of information releasing position, and is combined and publisher's phase for a publisher The quality score released news for the keyword answered, determines minimum button of the publisher on described information releasing position Take.
Specifically, for a publisher, its corresponding releasing news in information releasing position for keyword is defined On it is minimum deduct fees as the ratio of the minimum value degree score_limit quality scores released news with this, i.e., it is minimum to deduct fees:
Formula 6
In the preferred embodiment, the minimum quality score deducted fees and released news is inversely proportional, therefore same information is sent out Minimum deduct fees of cloth position may be constantly different, can if the quality score released news on the information releasing position is higher Calculating determine it is less it is minimum deduct fees, determine larger minimum deduct fees conversely, then calculating;This preferred embodiment can promote hair Cloth person updates the quality that it releases news, so as to lift information distribution effect so that be presented in the releasing news of user Including more abundant, accurate content information, user's viewing experience is improved.
Preferably (reference picture 2), the device also includes actually deduct fees determining module (not shown), determination of actually deducting fees Module is deducted fees according to minimum on information releasing position of the publisher, and combines the actual bid of the publisher, determines the hair Cloth person actually deducting fees on information releasing position.
Specifically, it is determined that the mode actually deducted fees includes:
1) size of minimum deduct fees p and the actual bid of the publisher of the publisher on information releasing position is compared Bi, selection smaller value is actually to deduct fees;
2) minimum deduct fees p, the actual bid bi and root of the publisher of the publisher on information releasing position are compared The original p that deducts fees determined according to formula 1Original iSize, to determine actually to deduct fees;For example, the min that actually deducts fees (bi, max (p, pOriginal i)), that is, select the minimum p and original p that deducts fees that deducts feesOriginal iIn higher value, then actually going out in the higher value and publisher Smaller value is selected in valency bi as actually deducting fees.
In the preferred embodiment of this programme, because publisher reflects it to information to the expectancy of information releasing position The psychological endurance value of releasing position, determines the scheme of deducting fees, it can be ensured that in the button based on obtained expectancy is accurately estimated Publisher will not frequently change the bid of oneself under expense scheme, reduce the information processing pressure of system, the stability of lifting system.
Those skilled in the art of the present technique are appreciated that the present invention includes being related to for performing in operation described herein One or more of equipment.These equipment can be for needed for purpose and specially design and manufacture, or can also include general Known device in computer.These equipment have the computer program being stored in it, and these computer programs are optionally Activation is reconstructed.Such computer program can be stored in equipment (for example, computer) computer-readable recording medium or be stored in E-command and it is coupled to respectively in any kind of medium of bus suitable for storage, the computer-readable medium is included but not Be limited to any kind of disk (including floppy disk, hard disk, CD, CD-ROM and magneto-optic disk), ROM (Read-Only Memory, only Read memory), RAM (Random Access Memory, immediately memory), EPROM (Erasable Programmable Read-Only Memory, Erarable Programmable Read only Memory), EEPROM (Electrically Erasable Programmable Read-Only Memory, EEPROM), flash memory, magnetic card or light card Piece.It is, computer-readable recording medium includes storing or transmitting any Jie of information in the form of it can read by equipment (for example, computer) Matter.
Those skilled in the art of the present technique be appreciated that can be realized with computer program instructions these structure charts and/or The combination of each frame and these structure charts and/or the frame in block diagram and/or flow graph in block diagram and/or flow graph.This technology is led Field technique personnel be appreciated that these computer program instructions can be supplied to all-purpose computer, special purpose computer or other The processor of programmable data processing method is realized, so as to pass through the processing of computer or other programmable data processing methods The scheme that device is specified in the frame or multiple frames to perform structure chart disclosed by the invention and/or block diagram and/or flow graph.
Those skilled in the art of the present technique are appreciated that in the various operations discussed in the present invention, method, flow Step, measure, scheme can be replaced, changed, combined or deleted.Further, it is each with what is discussed in the present invention Kind operation, method, other steps in flow, measure, scheme can also be replaced, changed, reset, decomposed, combined or deleted. Further, it is of the prior art to have and the step in the various operations disclosed in the present invention, method, flow, measure, scheme It can also be replaced, changed, reset, decomposed, combined or deleted.
Described above is only some embodiments of the present invention, it is noted that for the ordinary skill people of the art For member, under the premise without departing from the principles of the invention, some improvements and modifications can also be made, these improvements and modifications also should It is considered as protection scope of the present invention.

Claims (16)

1. a kind of method for the action value for determining information releasing position, it is characterised in that including:
For a keyword, the actual bid of multiple publishers on an information releasing position in scheduled duration is obtained;
According to the actual bid of the multiple publisher, expectancy of each publisher to described information releasing position is determined;
Based on each publisher to the expectancy of described information releasing position and it is corresponding with each publisher be directed to the key The quality score released news of word, it is determined that the sample action value of described information releasing position corresponding with each publisher;
Sample action value progress statistical disposition pair described information releasing position corresponding with multiple publishers difference, it is determined that described The minimum value degree of information releasing position;
Wherein, the actual bid according to the multiple publisher, determines each publisher to described information releasing position Expectancy, is specifically included:
By the actual bid of each publisher directly as expectancy of the publisher to described information releasing position, and/or;
According to the actual bid on described information releasing position and the quality score released news accordingly, and with reference under described Actual bid on one information releasing position and the quality score released news accordingly, are calculated, really by prediction model Expectancy of the fixed each publisher to described information releasing position.
2. the method for the action value according to claim 1 for determining information releasing position, it is characterised in that according to described Actual bid on information releasing position and the quality score released news accordingly, and with reference to next information releasing position On actual bid and the quality score that releases news accordingly, calculated by prediction model, determine each publisher couple The expectancy of described information releasing position, is specifically included;
Obtain actual bid in the scheduled duration on next information releasing position of described information releasing position and corresponding The quality score released news;
According to the actual bid on described information releasing position and the quality score released news accordingly, and with reference under described Actual bid on one information releasing position and the quality score released news accordingly, are calculated, really by prediction model Expectancy of the fixed each publisher to described information releasing position.
3. the method for the action value according to claim 2 for determining information releasing position, it is characterised in that also include:
When the sample size of the actual bid of multiple publishers in the scheduled duration is less than predetermined sample amount threshold, to institute State keyword and be extended processing, it is determined that the one or more expanded keywords related to the keyword;
For one or more of expanded keywords, the reality on described information releasing position in the scheduled duration is obtained Bid and the quality score and the reality on next information releasing position of described information releasing position released news accordingly The quality score that border is bid and released news accordingly;
Wherein, according to the actual bid on described information releasing position and the quality score released news accordingly, and combine Actual bid on next information releasing position and the quality score released news accordingly, are counted by prediction model Calculate, determine expectancy of each publisher to described information releasing position, be specially:
According to for actual bid of the keyword on described information releasing position and the quality accordingly released news Score value, and actual bid on next information releasing position and the quality score that releases news accordingly, and be directed to Actual bid on the described information releasing position of one or more of expanded keywords and the quality released news accordingly Score value and actual bid on next information releasing position and the quality score released news accordingly, by estimating Model is calculated, and determines expectancy of each publisher to described information releasing position.
4. the method for the action value of the determination information releasing position according to claim any one of 1-3, wherein, pair with it is multiple The sample action value that publisher distinguishes corresponding described information releasing position carries out statistical disposition, determines described information releasing position Minimum value degree, specifically include:
Set up the probability-distribution function and generally of the sample action value of corresponding with multiple publishers difference described information releasing position Rate density function;
Based on the probability-distribution function and the probability density function, the minimum value degree of described information releasing position is determined.
5. the method for the action value according to claim 4 for determining information releasing position, wherein, based on the probability distribution Function and the probability density function, determine the minimum value degree of described information releasing position, specifically include:
Using formulaDetermine the minimum value degree of described information releasing position;
Wherein, siFor the sample action value of described information releasing position corresponding with any publisher, and siMeet probability distribution letter Number F (si) and probability density function f (si)。
6. the method for the action value according to claim 4 for determining information releasing position, wherein, described and multiple publishers The sample action value of corresponding described information releasing position obeys logarithm normal distribution respectively, and the probability-distribution function is logarithm Normal distyribution function, the probability density function is the probability density function of logarithm normal distribution.
7. the method for the action value of the determination information releasing position according to claim any one of 1-3, wherein, in addition to:
For a publisher, according to the minimum value degree of described information releasing position, and pin corresponding with the publisher is combined To the quality score released news of the keyword, determine that minimum on described information releasing position of the publisher is deducted fees.
8. the method for the action value according to claim 7 for determining information releasing position, wherein, in addition to:
Deducted fees according to minimum on described information releasing position of the publisher, and combine the actual bid of the publisher, it is determined that The publisher actually deducting fees on described information releasing position.
9. a kind of device for the action value for determining information releasing position, it is characterised in that including:
Actual bid acquisition module, for for a keyword, obtaining many on an information releasing position in scheduled duration The actual bid of individual publisher;
Expectancy determining module, for the actual bid according to the multiple publisher, determines each publisher to the letter Cease the expectancy of releasing position;
Sample action value determining module, for based on each publisher to the expectancy of described information releasing position and with it is each Publisher is directed to the quality score released news of the keyword accordingly, it is determined that described information corresponding with each publisher is sent out The sample action value of cloth position;
Minimum value degree determining module, the sample for pair described information releasing position corresponding with multiple publishers difference is worth Degree carries out statistical disposition, determines the minimum value degree of described information releasing position;
Wherein, the expectancy determining module, specifically for:
By the actual bid of each publisher directly as expectancy of the publisher to described information releasing position, and/or;
According to the actual bid on described information releasing position and the quality score released news accordingly, and with reference under described Actual bid on one information releasing position and the quality score released news accordingly, are calculated, really by prediction model Expectancy of the fixed each publisher to described information releasing position.
10. the device of the action value according to claim 9 for determining information releasing position, it is characterised in that the expection Value determining module is specifically included;
First acquisition unit, for obtaining next information releasing position in the scheduled duration in described information releasing position Actual bid and the quality score that releases news accordingly;
Computing unit, for according to the actual bid on described information releasing position and the quality released news accordingly point Value, and with reference to the actual bid on next information releasing position and the quality score released news accordingly, by estimating Model is calculated, and determines expectancy of each publisher to described information releasing position.
11. the device of the action value according to claim 10 for determining information releasing position, it is characterised in that also include:
Expanded keyword determining module, for being less than when the sample size of the actual bid of multiple publishers in the scheduled duration During predetermined sample amount threshold, processing is extended to the keyword, it is determined that related to the keyword is one or more Expanded keyword;
Relevant information acquisition module, for for one or more of expanded keywords, obtaining in the scheduled duration in institute State the actual bid on information releasing position and the quality score released news accordingly and in described information releasing position Actual bid on next information releasing position and the quality score released news accordingly;
Wherein, the computing unit is further used for according to the reality on described information releasing position for being directed to the keyword Bid and the quality score released news accordingly, and the actual bid on next information releasing position and corresponding hair The quality score of cloth information, and for actually going out on the described information releasing position of one or more of expanded keywords Valency and the quality score released news accordingly and the actual bid on next information releasing position and corresponding hair The quality score of cloth information, is calculated by prediction model, determines expection of each publisher to described information releasing position Value.
12. the device of the action value of the determination information releasing position according to claim any one of 9-11, wherein, it is described most Low value degree determining module is specifically included:
Function sets up unit, the sample action value for setting up corresponding with multiple publishers difference described information releasing position Probability-distribution function and probability density function;
Minimum value degree determining unit, for based on the probability-distribution function and the probability density function, determining the letter Cease the minimum value degree of releasing position.
13. the device of the action value according to claim 12 for determining information releasing position, wherein, the minimum value degree Determining unit is specifically for using formulaDetermine the minimum value degree of described information releasing position;
Wherein, siFor the sample action value of described information releasing position corresponding with any publisher, and siMeet probability distribution letter Number F (si) and probability density function f (si)。
14. the device of the action value according to claim 12 for determining information releasing position, wherein, described and multiple issues The sample action value that person distinguishes corresponding described information releasing position obeys logarithm normal distribution, the probability-distribution function for pair Number normal distyribution function, the probability density function is the probability density function of logarithm normal distribution.
15. the device of the action value of the determination information releasing position according to claim any one of 9-11, wherein, also wrap Include:
Minimum determining module of deducting fees, for for a publisher, according to the minimum value degree of described information releasing position, and tying The quality score released news for the keyword corresponding with the publisher is closed, determines that the publisher issues in described information Minimum on position is deducted fees.
16. the device of the action value according to claim 15 for determining information releasing position, wherein, in addition to:
Actually deduct fees determining module, for being deducted fees according to minimum on described information releasing position of the publisher, and combine should The actual bid of publisher, determines the publisher actually deducting fees on described information releasing position.
CN201510338063.2A 2015-06-17 2015-06-17 Determine the method and device of the action value of information releasing position Active CN104899462B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN201510338063.2A CN104899462B (en) 2015-06-17 2015-06-17 Determine the method and device of the action value of information releasing position

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN201510338063.2A CN104899462B (en) 2015-06-17 2015-06-17 Determine the method and device of the action value of information releasing position

Publications (2)

Publication Number Publication Date
CN104899462A CN104899462A (en) 2015-09-09
CN104899462B true CN104899462B (en) 2017-09-29

Family

ID=54032124

Family Applications (1)

Application Number Title Priority Date Filing Date
CN201510338063.2A Active CN104899462B (en) 2015-06-17 2015-06-17 Determine the method and device of the action value of information releasing position

Country Status (1)

Country Link
CN (1) CN104899462B (en)

Families Citing this family (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN106339897B (en) * 2016-08-18 2020-02-21 腾讯科技(深圳)有限公司 Method and device for determining release strategy
CN107169052B (en) * 2017-04-26 2019-03-05 北京小度信息科技有限公司 Recommended method and device
CN110648158B (en) * 2018-06-27 2022-07-19 阿里巴巴(中国)有限公司 Resource allocation method and device

Citations (8)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN1801210A (en) * 2004-12-04 2006-07-12 林赞松 Price competition ranking method for website search result list
CN101477664A (en) * 2008-01-04 2009-07-08 杨云国 Method for discovering rational price by network
CN101673279A (en) * 2009-03-19 2010-03-17 黄斌 Information release method and system for Internet search engine
CN102208077A (en) * 2010-03-29 2011-10-05 上海博泰悦臻电子设备制造有限公司 Network information issue method and network information service platform
CN102214182A (en) * 2010-04-12 2011-10-12 北京银万特科技有限公司 Accurate query searching method according to internet protocol (IP) address
WO2013131335A1 (en) * 2012-03-06 2013-09-12 百度在线网络技术(北京)有限公司 Method and device for issuing information in a network community
CN103425674A (en) * 2012-05-18 2013-12-04 同程网络科技股份有限公司 Method applicable to intelligent biding for new keyword releasing of search engine
CN103838819A (en) * 2013-12-20 2014-06-04 深圳指掌时代网络科技有限公司 Information publish method and system

Family Cites Families (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN103368812B (en) * 2012-03-26 2016-12-14 阿里巴巴集团控股有限公司 Information automatic delivery method and equipment
JP5814303B2 (en) * 2013-06-18 2015-11-17 ヤフー株式会社 Revenue index value generation system and revenue index value generation method

Patent Citations (8)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN1801210A (en) * 2004-12-04 2006-07-12 林赞松 Price competition ranking method for website search result list
CN101477664A (en) * 2008-01-04 2009-07-08 杨云国 Method for discovering rational price by network
CN101673279A (en) * 2009-03-19 2010-03-17 黄斌 Information release method and system for Internet search engine
CN102208077A (en) * 2010-03-29 2011-10-05 上海博泰悦臻电子设备制造有限公司 Network information issue method and network information service platform
CN102214182A (en) * 2010-04-12 2011-10-12 北京银万特科技有限公司 Accurate query searching method according to internet protocol (IP) address
WO2013131335A1 (en) * 2012-03-06 2013-09-12 百度在线网络技术(北京)有限公司 Method and device for issuing information in a network community
CN103425674A (en) * 2012-05-18 2013-12-04 同程网络科技股份有限公司 Method applicable to intelligent biding for new keyword releasing of search engine
CN103838819A (en) * 2013-12-20 2014-06-04 深圳指掌时代网络科技有限公司 Information publish method and system

Also Published As

Publication number Publication date
CN104899462A (en) 2015-09-09

Similar Documents

Publication Publication Date Title
CN108364190B (en) Mobile crowd sensing online excitation method combined with reputation updating
US20160303481A1 (en) Online game community with controlled cross-promotion
US20150310019A1 (en) Providing an energy target for high energy users
KR102111223B1 (en) Push information roughly select sorting method, device and computer storage media
US20120329542A1 (en) Method and device for fantasy sports auction recommendations
CA2695778A1 (en) Content item pricing
CN104899462B (en) Determine the method and device of the action value of information releasing position
CN108460618A (en) A kind of resource allocation method and device, electronic equipment
CN109003140A (en) Determination method, equipment and the storage medium that the dispensing of multimedia content is provided funds
JP5702422B2 (en) Information processing apparatus, information processing method, and information processing program
JP5426004B1 (en) Information processing apparatus, information processing method, and program
CN109492153A (en) A kind of Products Show method and apparatus
JP5827710B2 (en) Information processing apparatus, information processing method, information processing program, and information processing system
US20170084108A1 (en) System and method for sporting event wagering
CN107507023B (en) Information delivery method and device
US10632377B2 (en) Computer-readable recording medium and game system
JP6226095B1 (en) Information processing apparatus and program
JP5159592B2 (en) Katsuma voting ticket selection device and its selection program
WO2017043824A1 (en) Multilevel jackpot advertisement compensation service providing method
CN107451920A (en) Calculation server, dynamic income processing method and storage medium
KR101937849B1 (en) Method, search server and computer readable recording medium for advertising management based on exposure frequency of keyword
CN109376310A (en) User's recommended method, device, electronic equipment and computer readable storage medium
US20120303422A1 (en) Computer-Implemented Systems And Methods For Ranking Results Based On Voting And Filtering
JP7393716B2 (en) Information processing device, information processing method and program
CN111932315B (en) Method and device for data display, electronic equipment and computer readable storage medium

Legal Events

Date Code Title Description
C06 Publication
PB01 Publication
C10 Entry into substantive examination
SE01 Entry into force of request for substantive examination
GR01 Patent grant
GR01 Patent grant
TR01 Transfer of patent right
TR01 Transfer of patent right

Effective date of registration: 20220728

Address after: 300450 No. 9-3-401, No. 39, Gaoxin 6th Road, Binhai Science Park, Binhai New Area, Tianjin

Patentee after: 3600 Technology Group Co.,Ltd.

Address before: 100088 room 112, block D, 28 new street, new street, Xicheng District, Beijing (Desheng Park)

Patentee before: BEIJING QIHOO TECHNOLOGY Co.,Ltd.

Patentee before: Qizhi software (Beijing) Co.,Ltd.

TR01 Transfer of patent right
TR01 Transfer of patent right

Effective date of registration: 20230721

Address after: 1765, floor 17, floor 15, building 3, No. 10 Jiuxianqiao Road, Chaoyang District, Beijing 100015

Patentee after: Beijing Hongxiang Technical Service Co.,Ltd.

Address before: 300450 No. 9-3-401, No. 39, Gaoxin 6th Road, Binhai Science Park, Binhai New Area, Tianjin

Patentee before: 3600 Technology Group Co.,Ltd.