CN106408320A

CN106408320A - Advertisement index construction method and apparatus and advertisement retrieval method and system

Info

Publication number: CN106408320A
Application number: CN201510464622.4A
Authority: CN
Inventors: 王佳宾; 孙鹏
Original assignee: Beijing Qihoo Technology Co Ltd; Qizhi Software Beijing Co Ltd
Current assignee: Beijing Qihoo Technology Co Ltd; Qizhi Software Beijing Co Ltd
Priority date: 2015-07-31
Filing date: 2015-07-31
Publication date: 2017-02-15

Abstract

The invention discloses an advertisement index construction method and apparatus, an advertisement retrieval method and system, and relates to the technical field of advertisement indexing. The advertisement index construction method comprises steps of: acquiring the advertisement group object of each advertisement source object, wherein the advertisement group object includes at least one piece of advertisement content data and corresponds to at least one retrieval word; for each retrieval word, generating a first-level advertisement index for the advertisement group object according to a relation between the retrieval words and the advertisement group objects; based on the advertisement group object, selecting N pieces of advertisement content data each piece of the advertisement content data of the advertisement group object to generate a second-level advertisement index for the advertisement group object. The advertisement index construction method and apparatus achieve the effect of reducing the length of an index list under each retrieval word so as to reduce the traversal time, improve retrieval efficiency, and improve the display speed of the advertisement content data for the subsequent retrieval requests.

Description

Keyed advertising construction method and device, advertisement retrieval method and system

Technical field

The present invention relates to keyed advertising technical field is and in particular to a kind of keyed advertising construction method and dress Put, a kind of advertisement retrieval method and system.

Background technology

For advertising platform, each advertisement putting side can register advertising accounts in this advertising platform, so This businessman can log in advertising platform, each bar ad content data being editted in its client afterwards It is uploaded in this advertising platform, and select one or more term for each ad content data. So advertising platform can be directed to the ad content data structure keyed advertising of these advertising accounts.

Formerly in technology, keyed advertising is for each term, by letter in the index list of this term Single each ad content data is arranged, its keyed advertising structure as shown in Figure 1A, this advertisement rope In drawing, term 1 corresponds to ad content data a, ad content data b, ad content data c etc. Ad content data, term 2 corresponds to the ad contents such as ad content data m, ad content data n Data.But this kind of keyed advertising, for each retrieval request on line, it inquires this retrieval request In term after, can travel through analyze this term under all ad content data, due to each advertisement The ad content data amount of account is very big, and its traversal analysis time is long.And for advertising platform, extensively The displaying accusing content-data needs in time, accurately, and the keyed advertising of formerly technology, and its retrieval time is long, The displaying of ad content data is slower.

Content of the invention

In view of the above problems it is proposed that the present invention is to provide one kind to overcome the problems referred to above or at least partly The keyed advertising construction device that solves the above problems and corresponding keyed advertising construction method, and advertisement Retrieval device and advertisement retrieval method.

According to one aspect of the present invention, disclose a kind of keyed advertising construction method, including：

Obtain the advertisement group objects of each advertisement source object；Described advertisement group objects includes at least one advertisement Content-data；Described advertisement group objects corresponds at least one term；

For each term, based on the corresponding relation of term and advertisement group objects, generate and be directed to advertisement The rank advertisement index of group objects；

Based on described advertisement group objects, from each ad content data of described advertisement group objects, select N Individual ad content data, to generate the secondary advertisement index for ad content data.

Preferably, described each ad content data from described advertisement group objects, select N number of advertisement Content-data, to generate the secondary advertisement index for ad content data, including：

From each ad content data of described advertisement group objects, poll selects N number of ad content number According to generate the secondary advertisement index for ad content data.

Preferably, described each ad content data from described advertisement group objects, poll selects N number of Ad content data, to generate the secondary advertisement index for ad content data, including：

Obtain the access behavioral data of each ad content data of record in described advertisement group objects；

According to the access behavioral data of each ad content data, poll selects N number of ad content number According to generate the secondary advertisement index for ad content data.

Preferably, described for each term, based on the corresponding relation of term and advertisement group objects, Generate the rank advertisement index for advertisement group objects, including：

In each advertisement group objects corresponding with described term from advertisement source object, poll selects M Individual advertisement group objects, to generate the rank advertisement index for advertisement group objects.

Preferably, in each advertisement group objects corresponding with described term the described source object from advertisement, Poll selects the advertisement group objects of M, is included with the rank advertisement index generating for advertisement group objects：

According to the access behavioral data of each ad content data of record in advertisement group objects, obtain each wide Accuse the access behavioral data of group objects；

According to the access behavioral data of each advertisement group objects, poll selects the advertisement group objects of M, To generate the rank advertisement index for advertisement group objects.

According to another aspect of the present invention, also disclose a kind of advertisement retrieval method, including：

Receive the retrieval request for ad content；Described retrieval request includes term；

Based on rank advertisement index, search each advertisement group objects corresponding with described term；

For each the advertisement group objects finding, based on secondary advertisement index, from described advertisement group objects The lower ad content data selecting to match with described term；

Wherein, described rank advertisement index and secondary advertisement index are built by following steps：

Preferably, for each the advertisement group objects finding, indexed based on secondary advertisement, from described After selecting the ad content data matching with described term under advertisement group objects, also include：

In each advertisement group objects, record the access row of each ad content data of this advertisement group objects For data；

Further, described each ad content data from described advertisement group objects, poll selects N number of Ad content data, with generate for ad content data secondary advertisement index include：

For each ad content data in each advertisement group objects, obtain record in advertisement group objects The access behavioral data of each ad content data；

From each ad content data of described advertisement group objects, according to the access of each ad content data Behavioral data, poll selects N number of ad content data, to generate two grades for ad content data Keyed advertising.

Preferably, in each advertisement group objects corresponding with described term from advertisement source object, poll Select the advertisement group objects of M, included with the rank advertisement index generating for advertisement group objects：

In each advertisement group objects corresponding with described term from advertisement source object, according to each advertisement The access behavioral data of group objects, poll selects the advertisement group objects of M, is directed to advertisement group to generate The rank advertisement index of object.

According to the another one invention of the present invention, also disclose a kind of keyed advertising construction device, including：

Advertisement group objects acquisition module, is suitable to obtain the advertisement group objects of each advertisement source object；Described wide Accuse group objects and include at least one ad content data；At least one retrieval corresponding of described advertisement group objects Word；

One-level index construct module, is suitable to for each term, based on term and advertisement group objects Corresponding relation, generates the rank advertisement index for advertisement group objects；

Secondary index builds module, is suitable to based on described advertisement group objects, each from described advertisement group objects In ad content data, select N number of ad content data, to generate two for ad content data Level keyed advertising.

Preferably, described secondary index builds module and includes：

Secondary index poll builds module, is suitable to from each ad content data of described advertisement group objects, Poll selects N number of ad content data, to generate the secondary advertisement index for ad content data.

Preferably, described secondary index poll builds module and includes：

Ad content accesses behavior acquisition module, and each being suitable to obtain record in described advertisement group objects is wide Accuse the access behavioral data of content-data；

Ad content poll builds module, is suitable to the access behavioral data according to each ad content data, Poll selects N number of ad content data, to generate the secondary advertisement index for ad content data.

Preferably, described one-level index construct module includes：

One-level index poll build module, be suitable to from advertisement source object corresponding with described term each In advertisement group objects, poll selects the advertisement group objects of M, to generate for advertisement group objects Level keyed advertising.

Preferably, described one-level index poll builds module and includes：

Advertisement group behavior acquisition module, is suitable to each ad content data according to record in advertisement group objects Access behavioral data, obtain each advertisement group objects access behavioral data；

Advertisement group polling builds module, is suitable to the access behavioral data according to each advertisement group objects, poll Select the advertisement group objects of M, to generate the rank advertisement index for advertisement group objects.

According to another aspect of the present invention, also disclose a kind of advertisement retrieval system, including：

Advertisement retrieval module and keyed advertising build module；

Described advertisement retrieval module includes：

Retrieval request receiver module, is suitable to receive the retrieval request for ad content；Described retrieval request Include term；

Advertisement group retrieves module, is suitable to index based on rank advertisement, searches corresponding with described term each Individual advertisement group objects；

Advertising content retrieval module, is suitable to for each the advertisement group objects finding, based on secondary advertisement Index, selects the ad content data matching with described term from described advertisement group objects；

Described keyed advertising builds module and includes：

Secondary index builds module, is suitable to based on described advertisement group objects, each from described advertisement group objects In ad content data, select N number of ad content data, generate two grades for ad content data Keyed advertising.

Preferably, described secondary index builds module and includes：

Preferably, also include in described advertisement retrieval module：

Content behavior record module, is suitable in each advertisement group objects, records each of this advertisement group objects The access behavioral data of individual ad content data；

Further, described secondary index poll builds module and includes：

Preferably, described one-level index construct module includes：

Preferably, described one-level index poll builds module and includes：

Advertisement group polling builds module, is suitable to the access behavioral data according to each advertisement group objects, and poll selects The advertisement group objects of M, to generate the rank advertisement index for advertisement group objects.

Keyed advertising construction method according to the present invention can be to each advertisement source object, this advertisement source object Can be understood as advertising accounts, ad content data will be recorded in the advertisement group objects of this advertisement source object, Then it is directed to advertisement group objects and builds rank advertisement index, build ad content data under advertisement group objects Secondary advertisement index, under each advertisement group objects as rank advertisement index, only select N number of wide Accuse content-data to index as secondary advertisement, so, in an advertisement group of an advertisement source object Each ad content data, without fully entering keyed advertising, greatly reduce the rope of each term Draw the length of list.During thus solving solution later retrieval, for each term, due to straight Connect the feelings needing to travel through all ad content data related to this term under all advertisement source objects Condition, and lead to retrieval time long, make the displaying time delay of ad content data higher, achieve and reduce often The length of the index list under individual term, hence for follow-up retrieval request, when can reduce traversal Between, improve recall precision, improve the beneficial effect of the displaying speed of ad content data.

Described above is only the general introduction of technical solution of the present invention, in order to better understand the skill of the present invention Art means, and being practiced according to the content of specification, and in order to allow, the present invention's is above and other Objects, features and advantages can become apparent, below especially exemplified by the specific embodiment of the present invention.

Brief description

By reading the detailed description of hereafter preferred embodiment, various other advantages and benefit are for this Field those of ordinary skill will be clear from understanding.Accompanying drawing is only used for illustrating the purpose of preferred embodiment, And it is not considered as limitation of the present invention.And in whole accompanying drawing, be denoted by the same reference numerals Identical part.In the accompanying drawings：

Fig. 1 shows that a kind of flow process of keyed advertising construction method according to an embodiment of the invention is illustrated Figure；

Figure 1A shows the keyed advertising example of formerly technology；

Figure 1B shows the keyed advertising example of the embodiment of the present invention；

Fig. 2 shows that a kind of flow process of keyed advertising construction method according to an embodiment of the invention is illustrated Figure；

Fig. 3 shows a kind of schematic flow sheet of advertisement retrieval method according to an embodiment of the invention；

Fig. 4 shows the schematic flow sheet of another kind advertisement retrieval method according to an embodiment of the invention

Fig. 5 shows a kind of structural representation of keyed advertising construction device according to an embodiment of the invention Figure；

Fig. 6 shows that the structure of another kind keyed advertising construction device according to an embodiment of the invention is shown It is intended to；

Fig. 7 shows a kind of structural representation of advertisement retrieval system according to an embodiment of the invention； And

Fig. 8 shows the structural representation of another kind advertisement retrieval system according to an embodiment of the invention Figure.

Specific embodiment

It is more fully described the exemplary embodiment of the disclosure below with reference to accompanying drawings.Although showing in accompanying drawing The exemplary embodiment of the disclosure it being understood, however, that may be realized in various forms the disclosure and not Should be limited by embodiments set forth here.On the contrary, these embodiments are provided to be able to more thoroughly Understand the disclosure, and complete for the scope of the present disclosure can be conveyed to those skilled in the art.

For the ease of understanding the embodiment of the present invention, below first introduce the present embodiments relate to term：

Advertisement source object：Can be understood as advertising accounts of merchant registration in advertising platform.Businessman with After this advertising accounts logs in advertising platform, can be in this advertising accounts, in the various advertisements that designed Hold data and be uploaded to advertising platform, these ad content data are then built keyed advertising by advertising platform.When So same businessman is likely to register multiple advertising accounts, and each advertising accounts is just an advertisement source pair As.

Advertisement group objects：In advertising accounts, businessman may arrange multiple packets, has in each group Multiple ad content data, this packet can be understood as advertisement group objects.In advertising platform, businessman can Think that each advertisement packet buys one or more word of bidding, this word of bidding can be in this advertisement packet All ad content data use, the wide of this advertisement packet can be finally associated with by this word of bidding Accuse on content-data.Wherein, word of bidding can be understood as the term of the present invention.

One of core concept of the embodiment of the present invention is：For advertisement field, in advertising platform, One businessman may upload a collection of ad content data in same advertising accounts, and in these advertisements Hold data and buy various words of bidding as term.Advertising platform then needs according to above-mentioned bid word and advertisement Content-data, builds keyed advertising, thus subsequently retrieval request can be directed to, goes to retrieve corresponding advertisement Content-data.Formerly as shown in Figure 1A, it is all corresponding with this term for the keyed advertising of technology Ad content, arranges in an index list, and its spread length is long, and travel time is long, thus leading to Retrieval time is long.For sequence businessman rearward, the time of its advertisement retrieval is very long, because And cannot in time by advertising display to user, probably due to the retrieval time delay height of advertisement is not shown in time, and User has switched the page, leads to reduction advertisement cannot effectively be delivered to user, advertisement putting efficiency is low.

The embodiment of the present invention is then according to the situation that advertisement field is special, you can to press under each advertising accounts Advertisement packet arranges ad content data, word that the setting of each advertisement packet is bidded.So the present invention is implemented Example, then can arrange advertisement packet object to advertisement packet, and the word of bidding of this advertisement packet is retrieved as it Word, for each term, builds rank advertisement index with advertisement packet object, then divides in this advertisement Under group, selected section ad content data builds secondary advertisement index.

By the way, the embodiment of the present invention employ advertisement packet object as intermediate layer transfer right As, and the ad content data in advertisement packet is same businessman, then, limit advertisement group objects The number of the interior ad content as secondary advertisement index, for each term, its index column Table shortens, and the time traveling through this index list also shortens, therefore, it is possible to reduce the displaying of ad content data Time delay.For sequence advertising accounts rearward, can be retrieved faster, such that it is able to add Fast advertising display speed, improves advertisement putting efficiency.

Embodiment one

With reference to Fig. 1, the flow process that it illustrates a kind of keyed advertising construction method of the embodiment of the present invention is illustrated Figure, specifically can include：

Step 110, obtains the advertisement group objects of each advertisement source object；Described advertisement group objects include to A few ad content data；Described advertisement group objects corresponds at least one term；

In embodiments of the present invention, businessman can be in each advertising accounts of advertising platform registration, setting Multiple advertisement packet, then can upload one or more ad content number in each advertisement packet According to, and businessman can buy one or more word of bidding for each advertisement packet, for retrieving this Ad content in advertisement packet.

Certainly, above-mentioned advertisement source object corresponds to each advertising accounts.The corresponding advertisement of above-mentioned advertisement group objects divides Group.Above-mentioned word of bidding is the term creating index.Above-mentioned ad content data can include text, Various types of data such as static images, dynamic picture, according to the demand setting of businessman.

It is appreciated that in the embodiment of the present invention, advertising platform, when building keyed advertising, can obtain first The advertisement group objects of each advertisement source object.

Such as, businessman 1 is provided with 2 advertisement group objects, i.e. advertisement group pair in advertisement source object P1 As A and advertisement group objects B；Term 1, term 2 are provided with to advertisement group objects A；To advertisement Group objects B is provided with term 1；Advertisement group objects A uploads ad content data a1, a2, A3, a4, a5, a6, upload in advertisement group objects B ad content data b1, b2, b3, b4, b5.

Businessman 2 is provided with 2 advertisement group objects in advertisement source object P2, that is, advertisement group objects C and Advertisement group objects M；Term 1 is provided with to advertisement group objects C；Advertisement group objects M is provided with Term 2；Upload ad content data c1, c2, c3, c4 in advertisement group objects C, in advertisement Ad content data m1, m2, m3 is uploaded in group objects M.

Businessman 3 is provided with 1 advertisement group objects, i.e. advertisement group objects N in advertisement source object P3； Term 2 is provided with to advertisement group objects N；Upload ad content data in advertisement group objects N n1、n2、n3、n4.

Step 120, for each term, based on the corresponding relation of term and advertisement group objects, raw Become the rank advertisement index for advertisement group objects；

Such as obtain the advertisement group objects of each advertisement source object above-mentioned.Due to term and advertisement group pair The corresponding relation of elephant, can be directed to each term, find all advertisement groups pair related to this term As.Such as term 1, it can find advertisement group objects A, advertisement group objects B, advertisement group pair As C.For term 2, it can find advertisement group objects A, advertisement group objects M, advertisement group objects N.

Then the rank advertisement index of the group objects mark for advertisement group objects can be built.As advertisement group Object A is group objects mark.As shown in Figure 1B.Figure 1B it illustrates the wide of the embodiment of the present invention Accuse index example, rank advertisement index in the index list of term 1 in have advertisement group objects A, B, C etc..There are advertisement group objects M, advertisement group objects N etc. in the index list of term 2.

Certainly, in the embodiment of the present invention, can also be for advertisement group objects setting retrieval qualifications, than As release time and/or input region and/or negative word etc., when traveling through advertisement group objects, obtain The current time of retrieval request, current zone and term.If current time is not within the release time, Then no longer travel through the ad content data in this advertisement group objects, conversely, then continuing subsequent process.If Current zone is not thrown within region, then no longer travel through the ad content data in this advertisement group objects, Conversely, then continuing subsequent process.If current term and the negative word of this advertisement group objects, no Travel through the ad content data in this advertisement group objects again, conversely, then continuing subsequent process.

Step 130, based on described advertisement group objects, from each ad content data of described advertisement group objects In, select N number of ad content data, generate the secondary advertisement index for ad content data.

The advertisement group objects indexing as rank advertisement for each, due to which includes in multiple advertisements Hold data, in the embodiment of the present invention, N number of ad content data of therefrom selected section, partly wide with this The content identification accusing content-data generates secondary advertisement index.

Certainly, in the embodiment of the present invention, above-mentioned N number of can arrange according to actual needs, the present invention implement Example is not any limitation as to it.

In such as previous example, include in advertisement group objects A ad content data a1, a2, a3, a4, A5, a6,6 ad content data, ad content data a1, a2, a3, a4, a5, a6 are corresponding Content identification.In Figure 1B, setting is N number of to be 3, then select from advertisement group objects A is wide Accuse content-data a1, a2, a3 and build secondary advertisement index.

Wherein, N is the integer more than 0.

Certainly, in the embodiment of the present invention, if the number of the ad content data in certain advertisement group objects Inherently it is less than N number of, then then select the number of reality to build secondary advertisement index.

So, that is, construct the keyed advertising of the embodiment of the present invention.

In the embodiment of the present invention, such as can throw in for ad content data setting retrieval qualifications Time and/or input region and/or negative word etc., when traveling through ad content data, obtain retrieval The current time of request, current zone and term.If current time is not within the release time, No longer relevant matches are carried out to ad content data, conversely, then carrying out relevant matches, if related Property be more than threshold value, then return client be shown.If current zone is not within input region, No longer relevant matches are carried out to ad content data, conversely, then carrying out relevant matches, if related Property be more than threshold value, then return client be shown.If current term and this advertisement group objects Negative word, then no longer carry out relevant matches to ad content data, conversely, then carrying out relevant matches, If correlation is more than threshold value, returns client and be shown.

Preferably, each ad content data from described advertisement group objects described in step 130, select N number of ad content data, to generate the secondary advertisement index for ad content data, including：

Sub-step 131, from each ad content data of described advertisement group objects, poll selects N number of Ad content data, to generate the secondary advertisement index for ad content data.

In embodiments of the present invention, in order to ensure each ad content data in an advertisement group objects all Can be retrieved, make some ad content data will not retrieve always less than.The embodiment of the present invention can be determined Ad content data in phase poll advertisement group objects, changes the secondary advertisement index under advertisement group objects.

In first time index building, can arbitrarily select the N number of ad content number in advertisement group objects According to then to the ad content data as secondary advertisement index wide in this secondary index, in advertisement group objects In its relevant information is recorded.Then, at set intervals, according to certain poll rule, choosing Select new N number of ad content data and build secondary advertisement index.

Wherein, described poll rule such as, the N number of ad content data few by building number of times, remove structure Build secondary advertisement index.In such as previous example, for the first time with ad content data in advertisement group objects A A1, a2, a3 build secondary advertisement index, after a period of time, be polled, find a4, a5, It is 0 that a6 builds number of times, then delete the secondary advertisement index of a1, a2, a3, a4, a5, a6 are built Secondary index.I.e. the index under the advertisement group objects A of secondary advertisement index in Figure 1B is changed in advertisement Hold data a1, ad content data a2, ad content data a3.

Preferably, each ad content data from described advertisement group objects described in step 131, poll Select N number of ad content data, to generate the secondary advertisement index for ad content data, including：

Sub-step 1311, obtains the access row of each ad content data of record in described advertisement group objects For data；

In embodiments of the present invention, in advertising platform in retrieving, can be according to user in advertisement The access behavioral data holding data is recorded, and this access behavioral data record is in advertisement group objects.

It is appreciated that in the data structure of advertisement group objects of the embodiment of the present invention, being provided with one and be used for Record the structure of the access behavioral data of each ad content data.

In actual applications, each ad content number can be directed to according to the access log statistics of advertising platform According to access behavioral data.The displaying number of times of such as ad content data, the click time of ad content data Number etc..

So, the embodiment of the present invention then can obtain advertisement group objects from the above-mentioned record of advertisement group objects In each ad content data access behavioral data.

Sub-step 1312, according to the access behavioral data of each ad content data, poll selects N number of Ad content data, with generate for ad content data secondary advertisement index.

In embodiments of the present invention, poll can be gone to select according to the access behavioral data of ad content data Select N number of ad content data.

Such as, after last time index building, every the displaying number of times of scheduled time poll ad content data, Number of times sequence N number of ad content data rearward will be shown, rebuild to should advertisement group objects two Level keyed advertising.

Again such as, after ad content data shows that number of times reaches certain frequency threshold value, judge click time Number and the accounting showing number of times, are less than the ad content data of accounting threshold value, do not select these for accounting Ad content data builds secondary advertisement index.And from accounting be higher than accounting threshold value ad content data In, select N number of ad content data to build secondary advertisement index.Certainly, if accounting is higher than accounting threshold The number of the ad content data of value is less than N, such as M, then the then ad content number of N-N1 According to then can selecting from remaining ad content data.

Certainly, in the embodiment of the present invention, can be combined with the number of times that ad content data is fabricated index, Go to select N number of ad content data.Such as threshold value, then prioritizing selection are less than for index building number of times Index building is less than the ad content data of threshold value；Threshold value is more than or equal to for index building number of times, then According to the access behavioral data of each ad content data, poll selects N number of ad content data, goes Generate secondary advertisement index.

In embodiments of the present invention, for each ad content data in an advertisement group objects, Ke Yigen According to the access behavioral data of ad content data, judge to select which N number of structure secondary advertisement index is replaced Secondary secondary advertisement index.Some advertisements can be avoided can not to be demonstrated always it is also possible to reduce clicking rate The number of times of low advertisement index building, improves the actual input efficiency of advertisement.Wide according to click charge Accuse under input pattern, advertisement putting income can be increased.

Preferably, described in step 120 for each term, based on term and advertisement group objects Corresponding relation, generates the rank advertisement index for advertisement group objects, including：

Sub-step 121, in each advertisement group objects corresponding with described term from advertisement source object, Poll selects the advertisement group objects of M, to generate the rank advertisement index for advertisement group objects.

In embodiments of the present invention, the total length of the index list of each term can be reduced further. I.e. in this step, can from source object selected section advertisement group objects, such as M advertisement group pair As.Wherein M is the integer more than 0.

When building rank advertisement index for the first time, can arbitrarily select M advertisement group objects.Then, At set intervals, other M advertisement group objects are selected to build rank advertisement index replacement last The rank advertisement index building.

Wherein, can be identical between a period of time at interval in sub-step 121 and sub-step 131, also may be used With difference.

In embodiments of the present invention, can also be for advertisement source object and build zero level index, that is, term is first Corresponding advertisement source object, then advertisement group objects corresponding with term in advertisement source object is built one-level rope Draw.So, the relevant information of advertisement group objects can write in a data structure of advertisement source object, The access behavioral data of this relevant information such as number of times of the index building of advertisement group objects, advertisement group objects Deng.

Preferably, corresponding with described term from advertisement source object described in sub-step 121 each is wide Accuse in group objects, poll selects the advertisement group objects of M, to generate the one-level for advertisement group objects Keyed advertising includes：

Sub-step 1211, according to the access behavior number of each ad content data of record in advertisement group objects According to the access behavioral data of each advertisement group objects of acquisition；

In embodiments of the present invention, have recorded the visit of its each ad content data in each advertisement group objects Ask behavioral data, such as show number of times, number of clicks.

According to these, the embodiment of the present invention then can show that number of times and number of clicks etc. access behavioral data, meter Calculate the access behavioral data of respective advertisement group objects.

Ad content data a1 of such as aforementioned advertisement group objects A { shows number of times：500, number of clicks 100 }, a2 { shows number of times：700, number of clicks 100 }, a3 { show number of times：300, number of clicks 50 }, a4 { shows number of times：50, number of clicks 20 }, a5 { show number of times：1000, number of clicks 100 }, A6 { shows number of times：500, number of clicks 200 }.Advertisement group objects A then can be counted { shows number of times： 3050, number of clicks 770 }.By that analogy, the access behavior obtaining each advertisement group objects can be analyzed Data.

Sub-step 1212, according to the access behavioral data of each advertisement group objects, poll selects M Advertisement group objects, to generate the rank advertisement index for advertisement group objects.

Then, for an advertisement source object, can according to the access behavioral data of each advertisement group objects, Go to select M advertisement group objects.

Such as, after last time builds rank advertisement index, every the displaying of scheduled time poll advertisement group objects Number of times, will show number of times sequence M advertisement group objects rearward, rebuild to should advertisement group pair Elephant rank advertisement index, replace last time rank advertisement index in should advertisement source object advertisement group pair The index of elephant.

Again such as, after advertisement group objects shows that number of times reaches certain frequency threshold value, judge number of clicks With the accounting showing number of times, accounting is less than to the advertisement group objects of accounting threshold value, does not select these advertisements Group objects builds rank advertisement index.And from the advertisement group objects that accounting is higher than accounting threshold value, select M Individual advertisement group objects builds rank advertisement index.Certainly, if accounting is higher than the advertisement group pair of accounting threshold value The number of elephant is less than M, such as M1, then then the advertisement group objects of M-M1 then can be from remaining Select in advertisement group objects.

Certainly, in the embodiment of the present invention, can be combined with the number of times that advertisement group objects is fabricated index, go Select M advertisement group objects.Such as index building number of times is less than to the advertisement group objects of threshold value, then Prioritizing selection index building is less than the advertisement group objects of threshold value；Threshold value is more than or equal to for index building number of times Advertisement group objects, then the access behavioral data according to each advertisement group objects, poll selects that M's is wide Accuse group objects, go to generate rank advertisement index.

In embodiments of the present invention, for each advertisement group objects in an advertisement group objects, can basis The access behavioral data of advertisement group objects, judges to select that last time replaced in which N number of rank advertisement index that builds Rank advertisement indexes.Some advertisement groups can be avoided can not to be demonstrated always it is also possible to reduce clicking rate low Advertisement index building number of times, improve advertisement actual input efficiency.According to the advertisement clicking on charge Under input pattern, advertisement putting income can be increased.

Advertisement packet arranges ad content data, word that the setting of each advertisement packet is bidded.So present invention Embodiment, then can arrange advertisement packet object to advertisement packet, using the word of bidding of this advertisement packet as it Term, for each term, builds rank advertisement index with advertisement packet object, then wide at this Accuse under packet, selected section ad content data builds secondary advertisement index.

Embodiment two

With reference to Fig. 2, the flow process that it illustrates a kind of keyed advertising construction method of the embodiment of the present invention is illustrated Figure, specifically can include：

Step 210, obtains the advertisement group objects of each advertisement source object；Described advertisement group objects include to A few ad content data；Described advertisement group objects corresponds at least one term；

Step 220, according to the access behavioral data of each ad content data of record in advertisement group objects, Obtain the access behavioral data of each advertisement group objects；

It is appreciated that during first time index building, the access behavioral data of advertisement group objects can Think sky.Then constantly update during follow-up use.

It is appreciated that in embodiments of the present invention, can also be for advertisement source object and build zero level index, I.e. term first corresponds to advertisement source object, then by advertisement group objects corresponding with term in advertisement source object Build one-level index.So, the relevant information of advertisement group objects can write a number of advertisement source object According in structure, this relevant information such as number of times of the index building of advertisement group objects, the access of advertisement group objects Behavioral data etc..

Step 240, according to the access behavioral data of each advertisement group objects, poll selects the advertisement of M Group objects, to generate the rank advertisement index for advertisement group objects；

Such as, after last time builds rank advertisement index, every the displaying of scheduled time poll advertisement group objects Number of times, will show number of times sequence M advertisement group objects rearward, rebuild to should advertisement group objects Rank advertisement index, replace last time rank advertisement index in should advertisement source object advertisement group objects Index.

In embodiments of the present invention, the time interval for the poll of advertisement group objects is more than in advertisement Hold the time interval of the poll of data.

Step 250, obtains each ad content data of record in the described advertisement group objects after selecting Access behavioral data；

In embodiments of the present invention, for the advertisement group objects determining as one-level index, for it In ad content data, when building first time, the access behavioral data of its ad content data is sky. During the use of index, constantly update this access behavioral data.

Step 260, according to the access behavioral data of each ad content data, poll selects N number of wide Accuse content-data, to generate the secondary advertisement index for ad content data.

The embodiment of the present invention is similar with embodiment one similar step principle, and here is no longer superfluous to be chatted.

In embodiments of the present invention, first：For each advertisement group objects in an advertisement group objects, can With the access behavioral data according to advertisement group objects, judge to select which N number of structure rank advertisement index is replaced The rank advertisement index of last time.Some advertisement groups can be avoided can not to be demonstrated always it is also possible to reduce point The number of times of the low advertisement index building of rate of hitting, improves the actual input efficiency of advertisement.According to click charge Advertisement putting pattern under, advertisement putting income can be increased.

Secondly, for each ad content data in an advertisement group objects, can be according to ad content number According to access behavioral data, judge select which N number of build secondary advertisement index replace last time secondary advertisement Index.Some advertisements can be avoided can not to be demonstrated always and to build it is also possible to reduce the low advertisement of clicking rate The number of times of index, improves the actual input efficiency of advertisement.Under the advertisement putting pattern according to click charge, Advertisement putting income can be increased.

Again, for each term, its index list shortens, and travels through the time of this index list Also shorten, therefore, it is possible to reduce the displaying time delay of ad content data.Particularly with sequence advertisement rearward For account, can be retrieved faster, such that it is able to accelerate advertising display speed, improve advertisement putting Efficiency.

Embodiment three

With reference to Fig. 3, it illustrates a kind of schematic flow sheet of advertisement retrieval method of the embodiment of the present invention, Specifically can include：

Step 310, receives the retrieval request for ad content；Described retrieval request includes term；

In embodiments of the present invention, initiate retrieval request can be client.Such as, user is in client Access advertising platform in end, then in search box, input term, can send after clicking on determination and carry The retrieval request of term.When user accesses certain page in the client, the embodiment of the present invention then can obtain The keyword of content of pages taking this page, as term, then initiates retrieval request to advertising platform.

Certainly initiate retrieval request can also be other-end or server, and the present invention is to it not in addition Limit.

Certainly, may also include the release time in described retrieval request and/or throw in region etc..

Step 320, based on rank advertisement index, searches each advertisement group pair corresponding with described term As；

After advertising platform receives retrieval request, therefrom extract the contents such as term, then in aforementioned reality Corresponding advertisement group objects is retrieved in the one-level index applying example.

In the embodiment of the present invention, can be for advertisement group objects setting retrieval qualifications, when such as throwing in Between and/or throw in region and/or negative word etc., travel through advertisement group objects when, obtain retrieval request Current time, current zone and term.If current time is not within the release time, no longer Travel through the ad content data in this advertisement group objects, conversely, then continuing subsequent process.If current position Domain is not thrown within region, then no longer travel through the ad content data in this advertisement group objects, conversely, Then continue subsequent process.If current term and the negative word of this advertisement group objects, no longer travel through Ad content data in this advertisement group objects, conversely, then continue subsequent process.

Step 330, for each the advertisement group objects finding, based on secondary advertisement index, from described The ad content data matching with described term is selected under advertisement group objects；

After finding an advertisement group objects, then can according in the secondary index of previous embodiment, Search ad content data, be then based on term and carry out relevant matches with ad content data.

Step 340, obtains the advertisement group objects of each advertisement source object；Described advertisement group objects include to A few ad content data；Described advertisement group objects corresponds at least one term；

Step 350, for each term, based on the corresponding relation of term and advertisement group objects, raw Become the rank advertisement index for advertisement group objects；

Step 360, based on described advertisement group objects, from each ad content data of described advertisement group objects In, select N number of ad content data, to generate the secondary advertisement index for ad content data.

Preferably, each ad content data from described advertisement group objects described in step 360, select N number of ad content data, to generate the secondary advertisement index for ad content data, including：

Sub-step 361, from each ad content data of described advertisement group objects, poll selects N number of Ad content data, to generate the secondary advertisement index for ad content data.

Step 331, in each advertisement group objects, records each ad content number of this advertisement group objects According to access behavioral data；

In embodiments of the present invention, after retrieving completes, user is shown in client Ad content data is shown or clicks on etc. and accesses the behavior of behavioral data and recorded.The present invention is real Apply example and can will access behavioral data cumulative record in advertisement group objects.

Further, each ad content data from described advertisement group objects described in sub-step 361, Poll selects N number of ad content data, to generate the secondary advertisement index bag for ad content data Include：

Sub-step 3611, for each ad content data in each advertisement group objects, obtains advertisement group The access behavioral data of each ad content data of record in object；

Sub-step 3612, from each ad content data of described advertisement group objects, according in each advertisement Hold the access behavioral data of data, poll selects N number of ad content data, to generate in advertisement Hold the secondary advertisement index of data.

Preferably, described in step 350 for each term, based on term and advertisement group objects Corresponding relation, generates the rank advertisement index for advertisement group objects, including：

Sub-step 351, in each advertisement group objects corresponding with described term from advertisement source object, Poll selects the advertisement group objects of M, to generate the rank advertisement index for advertisement group objects.

Preferably, corresponding with described term from advertisement source object described in sub-step 351 each is wide Accuse in group objects, poll selects the advertisement group objects of M, to generate the one-level for advertisement group objects Keyed advertising includes：

Sub-step 3511, according to the access behavior number of each ad content data of record in advertisement group objects According to the access behavioral data of each advertisement group objects of acquisition；

Sub-step 3512, in each advertisement group objects corresponding with described term from advertisement source object, According to the access behavioral data of each advertisement group objects, poll selects the advertisement group objects of M, with life Become the rank advertisement index for advertisement group objects.

In the embodiment of the present invention, rank advertisement index and secondary advertisement index are built by following steps and real Apply the example one step principle similar with embodiment two to be similar to, here is no longer superfluous to be chatted.

Example IV

With reference to Fig. 4, it illustrates a kind of schematic flow sheet of advertisement retrieval method of the embodiment of the present invention, Specifically can include：

Step 410, receives the retrieval request for ad content；Described retrieval request includes term；

Step 412, based on rank advertisement index, searches each advertisement group pair corresponding with described term As；

Step 414, for each the advertisement group objects finding, based on secondary advertisement index, from described The ad content data matching with described term is selected under advertisement group objects；

Step 416, in each advertisement group objects, records each ad content number of this advertisement group objects According to access behavioral data.

Step 418, for each advertisement group objects, according to each ad content number of this advertisement group objects According to access behavioral data, analyze the access behavioral data of this advertisement group objects and recorded.

In embodiments of the present invention, because the ad content data in each advertisement group objects has specific visit Ask behavioral data, the present invention then can access, according to these, the visit that behavioral data analyzes each advertisement group objects Ask behavioral data, and recorded.Such as record in advertisement group objects it is also possible to record is on other ground Side.

Step 420, obtains the advertisement group objects of each advertisement source object；Described advertisement group objects include to A few ad content data；Described advertisement group objects corresponds at least one term；

Step 422, according to the access behavioral data of each ad content data of record in advertisement group objects, Obtain the access behavioral data of each advertisement group objects；

Step 424, according to the access behavioral data of each advertisement group objects, poll selects the advertisement of M Group objects, to generate the rank advertisement index for advertisement group objects；

Step 426, obtains each ad content data of record in the described advertisement group objects after selecting Access behavioral data；

Step 428, according to the access behavioral data of each ad content data, poll selects N number of wide Accuse content-data, to generate the secondary advertisement index for ad content data.

The embodiment of the present invention is similar with embodiment three similar step principle, and here is no longer superfluous to be chatted.

Embodiment five

With reference to Fig. 5, it illustrates a kind of structural representation of keyed advertising construction device of the embodiment of the present invention Figure, specifically can include：

Advertisement group objects acquisition module 510, is suitable to obtain the advertisement group objects of each advertisement source object；Institute State advertisement group objects and include at least one ad content data；At least one inspection corresponding of described advertisement group objects Rope word；

One-level index construct module 520, is suitable to for each term, based on term and advertisement group pair The corresponding relation of elephant, generates the rank advertisement index for advertisement group objects；

Secondary index builds module, is suitable to based on described advertisement group objects, each from described advertisement group objects In ad content data, select N number of ad content data, to generate two for ad content data Level keyed advertising 530.

Preferably, described secondary index builds module and includes：

Preferably, described secondary index poll builds module and includes：

Preferably, described one-level index construct module includes：

Preferably, described one-level index poll builds module and includes：

Embodiment six

With reference to Fig. 6, it illustrates a kind of structural representation of keyed advertising construction device of the embodiment of the present invention Figure, specifically can include：

Advertisement group objects acquisition module 610, is suitable to obtain the advertisement group objects of each advertisement source object；Institute State advertisement group objects and include at least one ad content data；At least one inspection corresponding of described advertisement group objects Rope word；

One-level index construct module 620, including：

One-level index poll builds module 621, is suitable to, for each term, be suitable to from advertisement source In each advertisement group objects corresponding with described term in object, poll selects the advertisement group pair of M As to generate the rank advertisement index for advertisement group objects, specifically including：

Advertisement group behavior acquisition module 6211, is suitable to each according to record in advertisement group objects The access behavioral data of individual ad content data, obtains the access behavioral data of each advertisement group objects；

Advertisement group polling builds module 6212, is suitable to the access according to each advertisement group objects Behavioral data, poll selects the advertisement group objects of M, wide to generate the one-level being directed to advertisement group objects Accuse index.

Secondary index builds module 630, including:

Secondary index poll builds module 631, is suitable to based on described advertisement group objects, from described wide Accuse in each ad content data of group objects, poll selects N number of ad content data, be directed to generating The secondary advertisement index of ad content data, specifically includes:

Ad content accesses behavior acquisition module, and 6311 are suitable to obtain in described advertisement group objects The access behavioral data of each ad content data of record；

Ad content poll builds module 6312, is suitable to the visit according to each ad content data Ask behavioral data, poll selects N number of ad content data, to generate two for ad content data Level keyed advertising.

Embodiment seven

With reference to Fig. 7, it illustrates a kind of structural representation of advertisement retrieval system of the embodiment of the present invention, Specifically can include：

Advertisement retrieval module 710 and keyed advertising build module 750；

Described advertisement retrieval module 710 includes：

Retrieval request receiver module 711, is suitable to receive the retrieval request for ad content；Described retrieval Request includes term；

Advertisement group retrieves module 712, is suitable to index based on rank advertisement, searches corresponding with described term Each advertisement group objects；

Advertising content retrieval module 713, is suitable to for each the advertisement group objects finding, based on two grades Keyed advertising, selects the ad content data matching with described term from described advertisement group objects；

Described keyed advertising builds module 750 and includes：

Advertisement group objects acquisition module 751, is suitable to obtain the advertisement group objects of each advertisement source object；Institute State advertisement group objects and include at least one ad content data；At least one inspection corresponding of described advertisement group objects Rope word；

One-level index construct module 752, is suitable to for each term, based on term and advertisement group pair The corresponding relation of elephant, generates the rank advertisement index for advertisement group objects；

Secondary index builds module 753, is suitable to based on described advertisement group objects, from described advertisement group objects Each ad content data in, select N number of ad content data, generate for ad content data Secondary advertisement indexes.

Preferably, described secondary index builds module and includes：

Preferably, also include in described advertisement retrieval module：

Further, described secondary index poll builds module and includes：

Preferably, described one-level index construct module includes：

Preferably, described one-level index poll builds module and includes：

Embodiment eight

With reference to Fig. 8, it illustrates a kind of structural representation of advertisement retrieval system of the embodiment of the present invention, Specifically can include：

Advertisement retrieval module 810 and keyed advertising build module 850；

Described advertisement retrieval module 810 includes：

Retrieval request receiver module 811, is suitable to receive the retrieval request for ad content；Described retrieval Request includes term；

Advertisement group retrieves module 812, is suitable to index based on rank advertisement, searches corresponding with described term Each advertisement group objects；

Advertising content retrieval module 813, is suitable to for each the advertisement group objects finding, based on two grades Keyed advertising, selects the ad content data matching with described term from described advertisement group objects；

Content behavior record module 814, is suitable in each advertisement group objects, records this advertisement group objects Each ad content data access behavioral data；

Described keyed advertising builds module 850 and includes：

Advertisement group objects acquisition module 851, is suitable to obtain the advertisement group objects of each advertisement source object；Institute State advertisement group objects and include at least one ad content data；At least one inspection corresponding of described advertisement group objects Rope word；

One-level index construct module 852, including：

One-level index poll builds module 8521, is suitable to, for each term, be suitable to from advertisement source In each advertisement group objects corresponding with described term in object, poll selects the advertisement group pair of M As to generate the rank advertisement index for advertisement group objects, specifically including：

Advertisement group behavior acquisition module 85211, is suitable to according to record in advertisement group objects The access behavioral data of each ad content data, obtains the access behavioral data of each advertisement group objects；

Advertisement group polling builds module 85212, is suitable to the visit according to each advertisement group objects Ask behavioral data, poll selects the advertisement group objects of M, to generate the one-level for advertisement group objects Keyed advertising.

Secondary index builds module 853, including:

Secondary index poll builds module 8531, is suitable to based on described advertisement group objects, from described wide Accuse in each ad content data of group objects, poll selects N number of ad content data, be directed to generating The secondary advertisement index of ad content data, specifically includes:

Ad content accesses behavior acquisition module 85311, is suitable to obtain described advertisement group objects The access behavioral data of each ad content data of middle record；

Ad content poll builds module 85312, is suitable to the visit according to each ad content data Ask behavioral data, poll selects N number of ad content data, to generate two for ad content data Level keyed advertising.

Algorithm and display be not solid with any certain computer, virtual system or miscellaneous equipment provided herein There is correlation.Various general-purpose systems can also be used together with based on teaching in this.As described above, It is obvious for constructing the structure required by this kind of system.Additionally, the present invention be also not for any specific Programming language.It is understood that, it is possible to use various programming languages realize the content of invention described herein, And the description above language-specific done is the preferred forms in order to disclose the present invention.

In specification mentioned herein, illustrate a large amount of details.It is to be appreciated, however, that this Inventive embodiment can be put into practice in the case of not having these details.In some instances, not It is shown specifically known method, structure and technology, so as not to obscure the understanding of this description.

Similarly it will be appreciated that in order to simplify the disclosure and help understand one of each inventive aspect Or multiple, in the description to the exemplary embodiment of the present invention above, each feature of the present invention is sometimes It is grouped together in single embodiment, figure or descriptions thereof.However, should not be by the disclosure Method be construed to reflect following intention：I.e. the present invention for required protection requires ratio in each claim The more feature of middle feature be expressly recited.More precisely, as the following claims reflect As, inventive aspect is all features less than single embodiment disclosed above.Therefore, it then follows Claims of specific embodiment are thus expressly incorporated in this specific embodiment, wherein each right The separate embodiments all as the present invention for the requirement itself.

Those skilled in the art are appreciated that and the module in the equipment in embodiment can be carried out certainly Adaptively change and they are arranged in one or more equipment different from this embodiment.Permissible Module in embodiment or unit or assembly are combined into a module or unit or assembly, and in addition may be used To be divided into multiple submodule or subelement or sub-component.Except such feature and/or process or Outside at least some of unit excludes each other, using any combinations, (companion can be included to this specification With claim, summary and accompanying drawing) disclosed in all features and so disclosed any method or All processes of person's equipment or unit are combined.Unless expressly stated otherwise, this specification (includes companion With claim, summary and accompanying drawing) disclosed in each feature can be by providing identical, equivalent or phase Alternative features like purpose to replace.

Although additionally, it will be appreciated by those of skill in the art that some embodiments described herein include it Included some features rather than further feature in its embodiment, but the group of the feature of different embodiment Closing means to be within the scope of the present invention and formed different embodiments.For example, in following power In sharp claim, embodiment required for protection one of arbitrarily can in any combination mode making With.

The all parts embodiment of the present invention can be realized with hardware, or with one or more process The software module run on device is realized, or is realized with combinations thereof.Those skilled in the art should Understand, basis can be realized using microprocessor or digital signal processor (DSP) in practice The embodiment of the present inventionKeyed advertising buildsSome or all parts in equipment and advertisement retrieval equipment Some or all functions.The present invention be also implemented as execute method as described herein one Partly or completely equipment or program of device (for example, computer program and computer program). Such program realizing the present invention can store on a computer-readable medium, or can have one Or the form of multiple signals.Such signal can be downloaded from internet website and obtain, or carrying There is provided on body signal, or provided with any other form.

It should be noted that above-described embodiment the present invention will be described rather than limits the invention, And those skilled in the art can design replacement without departing from the scope of the appended claims Embodiment.In the claims, any reference symbol between bracket should not be configured to right The restriction requiring.Word "comprising" does not exclude the presence of element not listed in the claims or step.Position Word "a" or "an" before element does not exclude the presence of multiple such elements.The present invention can Realize with by means of the hardware including some different elements and by means of properly programmed computer. If in the unit claim listing equipment for drying, several in these devices can be by same Individual hardware branch is embodying.The use of word first, second, and third does not indicate that any order. These words can be construed to title.

The invention discloses A1, a kind of keyed advertising construction method, including：

A2, the method as described in A1, described each ad content data from described advertisement group objects, Select N number of ad content data, to generate the secondary advertisement index for ad content data, including：

A3, the method as described in A2, described each ad content data from described advertisement group objects, Poll selects N number of ad content data, is indexed with the secondary advertisement generating for ad content data, Including：

A4, the method as described in A1, described for each term, based on term and advertisement group pair The corresponding relation of elephant, generates the rank advertisement index for advertisement group objects, including：

A5, the method as described in A4, the described source object from advertisement corresponding with described term each In advertisement group objects, poll selects the advertisement group objects of M, to generate for advertisement group objects Level keyed advertising includes：

The invention also discloses B6, a kind of advertisement retrieval method, including：

B7, the method as described in B6, described each ad content data from described advertisement group objects, Select N number of ad content data, to generate the secondary advertisement index for ad content data, including：

B8, the method as described in B7 are for each the advertisement group objects finding, wide based on two grades Accuse index, select from described advertisement group objects the ad content data that matches with described term it Afterwards, also include：

B9, the method as described in B6, described for each term, based on term and advertisement group pair The corresponding relation of elephant, generates the rank advertisement index for advertisement group objects, including：

B10, the method as described in B9, corresponding with described term from advertisement source object each is wide Accuse in group objects, poll selects the advertisement group objects of M, to generate the one-level for advertisement group objects Keyed advertising includes：

The invention also discloses C11, a kind of keyed advertising construction device, including：

C12, the device as described in C11, described secondary index builds module and includes：

C13, the device as described in C12, described secondary index poll builds module and includes：

C14, the device as described in C11, described one-level index construct module includes：

C15, the device as described in C14, described one-level index poll builds module and includes：

The invention also discloses D16, a kind of advertisement retrieval system, including：

Advertisement retrieval module and keyed advertising build module；

Described advertisement retrieval module includes：

Described keyed advertising builds module and includes：

D17, the system as described in D16, described secondary index builds module and includes：

D18, the system as described in D17, also include in described advertisement retrieval module：

Further, described secondary index poll builds module and includes：

D19, the system as described in D16, described one-level index construct module includes：

D20, the system as described in D19, described one-level index poll builds module and includes：

Claims

1. a kind of keyed advertising construction method, including：

2. the method for claim 1 it is characterised in that described from described advertisement group objects In each ad content data, select N number of ad content data, to generate for ad content data Secondary advertisement indexes, including：

3. method as claimed in claim 2 it is characterised in that described from described advertisement group objects In each ad content data, poll selects N number of ad content data, is directed to ad content number to generate According to secondary advertisement index, including：

4. the method for claim 1 it is characterised in that described for each term, base In the corresponding relation of term and advertisement group objects, generate the rank advertisement for advertisement group objects and index, Including：

5. method as claimed in claim 4 it is characterised in that the described source object from advertisement with institute State in term each advertisement group objects corresponding, poll selects the advertisement group objects of M, to generate Rank advertisement index for advertisement group objects includes：

6. a kind of advertisement retrieval method, including：

7. method as claimed in claim 6 it is characterised in that described from described advertisement group objects In each ad content data, select N number of ad content data, to generate for ad content data Secondary advertisement indexes, including：

8. method as claimed in claim 7 is it is characterised in that for each advertisement finding Group objects, based on secondary advertisement index, selects to match with described term from described advertisement group objects Ad content data after, also include：

9. a kind of keyed advertising construction device, including：

10. a kind of advertisement retrieval system, including：

Advertisement retrieval module and keyed advertising build module；

Described advertisement retrieval module includes：

Described keyed advertising builds module and includes：