CN113793164A

CN113793164A - Advertisement putting method, device, equipment and storage medium

Info

Publication number: CN113793164A
Application number: CN202011387745.XA
Authority: CN
Inventors: 宋金波; 李勇; 彭长平; 包勇军; 颜伟鹏
Original assignee: Beijing Wodong Tianjun Information Technology Co Ltd
Current assignee: Beijing Wodong Tianjun Information Technology Co Ltd
Priority date: 2020-11-27
Filing date: 2020-12-01
Publication date: 2021-12-14

Abstract

The disclosure provides an advertisement putting method, an advertisement putting device, advertisement putting equipment and a storage medium, and relates to the technical field of electronic commerce. The method comprises the following steps: receiving an advertisement playing request; acquiring a candidate advertisement commodity set according to the advertisement playing request; acquiring advertisement price attribute information of a first candidate advertisement commodity in a candidate advertisement commodity set; acquiring a second candidate advertisement commodity based on the advertisement price attribute information of the first candidate advertisement commodity; acquiring the estimated click rate of a first candidate advertisement commodity and the estimated click rate of a second candidate advertisement commodity; obtaining a replacement reference lower limit value according to the estimated click rate of the first candidate advertisement commodity; and when the estimated click rate of the second candidate advertisement commodity is higher than the replacement reference lower limit value, replacing the first candidate advertisement commodity with the second candidate advertisement commodity to obtain an updated candidate advertisement commodity set for advertisement delivery. The method improves the rationality of advertisement putting to a certain extent.

Description

Advertisement putting method, device, equipment and storage medium

Technical Field

The present disclosure relates to the field of computer technologies, and in particular, to an advertisement delivery method, an advertisement delivery apparatus, an advertisement delivery device, and a readable storage medium.

Background

The rapid development of the internet has changed the world, and a myriad of online services have emerged to facilitate people's lives, from web portals to search engines, and from social networks to electronic commerce. The online services provided by internet companies (such as web portals, e-commerce platforms, short video applications, etc.) are mostly free for users, relying on selling advertising spots to advertisers to obtain revenue. An internet company provides valuable online products to users, accumulates a certain amount of users and reaches a certain user activity to generate traffic, and then advertisers put advertisements through advertisement slots of the online products and pay according to advertisement effects.

The number of goods that can be presented at an ad slot is limited and the number of goods to be presented at an ad slot may be numerous. In the related art, when candidate recommended commodities are determined from a large number of commodities to be displayed according to various factors, the flow is difficult to be reasonably distributed according to the advertising bid of the commodities, so that the rationality of advertising is low.

As described above, how to improve the rationality of advertisement placement is an urgent problem to be solved.

The above information disclosed in this background section is only for enhancement of understanding of the background of the disclosure and therefore it may contain information that does not constitute prior art that is already known to a person of ordinary skill in the art.

Disclosure of Invention

The invention aims to provide an advertisement putting method, an advertisement putting device, an advertisement putting equipment and a readable storage medium, which improve the rationality of advertisement putting at least to a certain extent.

Additional features and advantages of the disclosure will be set forth in the detailed description which follows, or in part will be obvious from the description, or may be learned by practice of the disclosure.

According to an aspect of the present disclosure, there is provided an advertisement delivery method including: receiving an advertisement playing request; acquiring a candidate advertisement commodity set according to the advertisement playing request; acquiring advertisement price attribute information of a first candidate advertisement commodity in the candidate advertisement commodity set; acquiring a second candidate advertisement commodity based on the advertisement price attribute information of the first candidate advertisement commodity; acquiring the estimated click rate of the first candidate advertisement commodity and the estimated click rate of the second candidate advertisement commodity; obtaining a replacement reference lower limit value according to the estimated click rate of the first candidate advertisement commodity; and when the estimated click rate of the second candidate advertisement commodity is higher than the replacement reference lower limit value, replacing the first candidate advertisement commodity with the second candidate advertisement commodity to obtain an updated candidate advertisement commodity set for advertisement delivery.

According to an embodiment of the present disclosure, the updated set of candidate advertising items includes an un-replaced candidate advertising item and a replacement candidate advertising item; the replacing the first candidate advertised item with the second candidate advertised item includes: replacing the first candidate advertisement commodity with the second candidate advertisement commodity to obtain a replacement candidate advertisement commodity; when the estimated click rate of the first candidate advertisement commodity is higher than the estimated click rate of the second candidate advertisement commodity, obtaining the estimated click rate of the replacement candidate advertisement commodity as the estimated click rate of the first candidate advertisement commodity; the obtaining an updated set of candidate advertising items for advertising comprises: and sequencing the non-replaced candidate advertisement commodities and the replaced candidate advertisement commodities according to the advertisement price attribute information and the estimated click rate of the non-replaced candidate advertisement commodities and the advertisement price attribute information and the estimated click rate of the replaced candidate advertisement commodities so as to carry out advertisement putting in sequence.

According to an embodiment of the present disclosure, the sorting the non-replaced candidate advertisement commodities and the replacement candidate advertisement commodities according to the advertisement price attribute information and the estimated click rate of the non-replaced candidate advertisement commodities and the advertisement price attribute information and the estimated click rate of the replacement candidate advertisement commodities includes: obtaining the ranking index of the non-replaced candidate advertisement commodity according to the advertisement price attribute information and the estimated click rate of the non-replaced candidate advertisement commodity; obtaining the ranking index of the replacement candidate advertisement commodity according to the advertisement price attribute information and the estimated click rate of the replacement candidate advertisement commodity; sequencing the non-replaced candidate advertisement commodities and the replaced candidate advertisement commodities according to the sequence of the sequencing indexes from large to small to obtain an advertisement commodity sequence to be delivered; the sequential advertisement delivery comprises: and carrying out advertisement putting on the sequence of the to-be-put advertisement commodities from front to back.

According to an embodiment of the present disclosure, the advertisement playing request includes a user identifier; the obtaining of the candidate advertisement commodity set according to the advertisement playing request includes: acquiring the candidate advertisement commodity set through a collaborative filtering algorithm based on the user identification, wherein first candidate advertisement commodities in the candidate advertisement commodity set are arranged from high to low according to an estimated satisfaction index; the obtaining a second candidate advertisement commodity based on the advertisement price attribute information of the first candidate advertisement commodity comprises: selecting a preset number of first candidate advertisement commodities in the first candidate advertisement commodities arranged according to the estimated satisfaction index; acquiring the second candidate advertisement commodity with advertisement price attribute information higher than the advertisement price attribute information of the first candidate advertisement commodity with the preset quantity in the front row.

According to an embodiment of the present disclosure, the method further comprises: obtaining historical advertising bids for goods belonging to a predetermined category; determining a reserved advertising bid for the predetermined category of goods according to the historical advertising bid; obtaining real-time advertising bids for goods belonging to the predetermined category; selecting commodities with real-time advertising bids higher than the reserved advertising bids, and obtaining a high-bid commodity set of the predetermined category; the obtaining a second candidate advertisement commodity based on the advertisement price attribute information of the first candidate advertisement commodity comprises: obtaining the predetermined category to which the first candidate advertisement commodity belongs; obtaining the second candidate advertising item from the predetermined category of high-bid item sets with a real-time advertising bid higher than the advertising bid of the first candidate advertising item.

According to an embodiment of the present disclosure, the determining a reserve advertising bid for the predetermined category of goods according to the historical advertising bid comprises: obtaining a price distribution of the historical advertising bids; and selecting the price of the preset quantile point of the price distribution as the reserved advertisement bid.

According to an embodiment of the present disclosure, the obtaining a replacement reference lower limit value according to the estimated click rate of the first candidate advertisement product includes: and multiplying the estimated click rate of the first candidate advertisement commodity by a preset percentage to obtain the replacement reference lower limit value.

According to still another aspect of the present disclosure, there is provided an advertisement delivery apparatus including: the request receiving module is used for receiving an advertisement playing request; the commodity filtering module is used for acquiring a candidate advertisement commodity set according to the advertisement playing request; the information acquisition module is used for acquiring advertisement price attribute information of a first candidate advertisement commodity in the candidate advertisement commodity set; the commodity complement module is used for acquiring a second candidate advertisement commodity based on the advertisement price attribute information of the first candidate advertisement commodity; the click rate estimation module is used for acquiring the estimated click rate of the first candidate advertisement commodity and the estimated click rate of the second candidate advertisement commodity; the replacement preparation module is used for obtaining a replacement reference lower limit value according to the estimated click rate of the first candidate advertisement commodity; and the commodity recommending module is used for replacing the first candidate advertisement commodity with the second candidate advertisement commodity when the estimated click rate of the second candidate advertisement commodity is higher than the replacement reference lower limit value, and obtaining an updated candidate advertisement commodity set for advertisement putting.

According to an embodiment of the present disclosure, the updated set of candidate advertising items includes an un-replaced candidate advertising item and a replacement candidate advertising item; the commodity recommending module comprises: a replacement execution module, configured to replace the candidate advertisement commodity with the second candidate advertisement commodity, so as to obtain a replacement candidate advertisement commodity; the replacement execution module is further configured to obtain the estimated click rate of the replacement candidate advertisement commodity as the estimated click rate of the candidate advertisement commodity when the estimated click rate of the candidate advertisement commodity is higher than the estimated click rate of the second candidate advertisement commodity; and the commodity sequencing module is used for sequencing the non-replaced candidate advertisement commodities and the replacement candidate advertisement commodities according to the advertisement price attribute information and the estimated click rate of the non-replaced candidate advertisement commodities and the advertisement price attribute information and the estimated click rate of the replacement candidate advertisement commodities so as to launch the advertisements in sequence.

According to an embodiment of the present disclosure, the commodity ordering module includes: the user recommendation degree calculation module is used for obtaining the ranking index of the non-replaced candidate advertisement commodity according to the advertisement price attribute information and the estimated click rate of the non-replaced candidate advertisement commodity; the user recommendation degree calculation module is further used for obtaining the ranking index of the replacement candidate advertisement commodity according to the advertisement price attribute information and the estimated click rate of the replacement candidate advertisement commodity; the to-be-recommended commodity sequence obtaining module is used for sequencing the non-replaced candidate advertisement commodities and the replacement candidate advertisement commodities according to the sequence of the sequencing indexes from large to small to obtain a to-be-launched advertisement commodity sequence; and the commodity recommending module is also used for advertising the to-be-advertised commodity sequence from front to back.

According to an embodiment of the present disclosure, the advertisement playing request includes a user identifier; the commodity filtering module includes: the pre-algorithm module is used for acquiring the candidate advertisement commodity set through a collaborative filtering algorithm based on the user identification, and first candidate advertisement commodities in the candidate advertisement commodity set are arranged from high to low according to the estimated satisfaction index; the commodity complement module is further used for selecting a preset number of first candidate advertisement commodities in the first candidate advertisement commodities arranged according to the estimated satisfaction index; acquiring the second candidate advertisement commodity with advertisement price attribute information higher than the advertisement price attribute information of the first candidate advertisement commodity with the preset quantity in the front row.

According to an embodiment of the present disclosure, the apparatus further comprises: the historical data acquisition module is used for acquiring historical advertising bids of commodities belonging to a predetermined class; a reserve price determining module for determining a reserve advertisement bid for the predetermined category of goods according to the historical advertisement bid; the real-time data acquisition module is used for acquiring real-time advertising bids of commodities belonging to the preset category; the high bid commodity set obtaining module is used for selecting commodities with real-time advertising bids higher than the reserved advertising bids and obtaining a high bid commodity set of the predetermined category; the item replenishment module includes: the category query module is used for acquiring the preset category to which the first candidate advertisement commodity belongs; a supplemental merchandise acquisition module for acquiring the second candidate advertising merchandise from the predetermined category of high-bid merchandise set with a real-time advertising bid higher than the advertising bid of the first candidate advertising merchandise.

According to an embodiment of the present disclosure, the reserve price determination module is further configured to obtain a price distribution of the historical advertising bids; and selecting the price of the preset quantile point of the price distribution as the reserved advertisement bid.

According to an embodiment of the disclosure, the click rate estimation module is further configured to multiply the estimated click rate of the first candidate advertisement product by a predetermined percentage to obtain the replacement reference lower limit value.

According to yet another aspect of the present disclosure, there is provided an apparatus comprising: a memory, a processor and executable instructions stored in the memory and executable in the processor, the processor implementing any of the methods described above when executing the executable instructions.

According to yet another aspect of the present disclosure, there is provided a computer-readable storage medium having stored thereon computer-executable instructions that, when executed by a processor, implement any of the methods described above.

According to the commodity recommendation method provided by the embodiment of the disclosure, the candidate advertisement commodity set is obtained according to the received advertisement playing request, the second candidate advertisement commodity is obtained based on the advertisement price attribute information of the first candidate advertisement commodity in the candidate advertisement commodity set, the replacement reference lower limit value is obtained according to the estimated click rate of the first candidate advertisement commodity, then when the estimated click rate of the second candidate advertisement commodity is higher than the replacement reference lower limit value, the first candidate advertisement commodity is replaced by the second candidate advertisement commodity, the updated candidate advertisement commodity set is obtained for advertisement putting, so that the first candidate advertisement commodity in the candidate advertisement commodity set is replaced by the second candidate advertisement commodity obtained based on the advertisement bidding price, and the rationality of advertisement putting is improved to a certain extent.

It is to be understood that both the foregoing general description and the following detailed description are exemplary and explanatory only and are not restrictive of the disclosure.

Drawings

The above and other objects, features and advantages of the present disclosure will become more apparent by describing in detail exemplary embodiments thereof with reference to the attached drawings.

Fig. 1 shows a schematic diagram of a system architecture in an embodiment of the disclosure.

Fig. 2 shows a flowchart of an advertisement delivery method in an embodiment of the present disclosure.

Fig. 3 is a schematic diagram illustrating a processing procedure of step S212 shown in fig. 2 in an embodiment.

Fig. 4A is a flow diagram illustrating another method of advertisement delivery, according to an example embodiment.

FIG. 4B is a diagram illustrating a second candidate advertisement item selection process.

FIG. 4C illustrates an alternative strategy diagram for a second candidate advertised item.

Fig. 5 is a schematic view of an advertisement delivery process according to fig. 2 to 4.

FIG. 6 is a flow chart illustrating a method for obtaining a set of high priced items according to an exemplary embodiment.

Fig. 7 is a schematic diagram of a high-priced article collection obtaining process according to fig. 6.

Fig. 8 is a block diagram illustrating an advertising device in accordance with an exemplary embodiment.

Fig. 9 is a block diagram illustrating an advertising device in accordance with an exemplary embodiment.

Fig. 10 shows a schematic structural diagram of an electronic device in an embodiment of the present disclosure.

Detailed Description

Example embodiments will now be described more fully with reference to the accompanying drawings. Example embodiments may, however, be embodied in many different forms and should not be construed as limited to the examples set forth herein; rather, these embodiments are provided so that this disclosure will be thorough and complete, and will fully convey the concept of example embodiments to those skilled in the art. The drawings are merely schematic illustrations of the present disclosure and are not necessarily drawn to scale. The same reference numerals in the drawings denote the same or similar parts, and thus their repetitive description will be omitted.

Furthermore, the described features, structures, or characteristics may be combined in any suitable manner in one or more embodiments. In the following description, numerous specific details are provided to give a thorough understanding of embodiments of the disclosure. One skilled in the relevant art will recognize, however, that the subject matter of the present disclosure can be practiced without one or more of the specific details, or with other methods, apparatus, steps, etc. In other instances, well-known structures, methods, devices, implementations, or operations are not shown or described in detail to avoid obscuring aspects of the disclosure.

In the description of the present disclosure, "a plurality" means at least two, e.g., two, three, etc., unless explicitly specifically limited otherwise. The symbol "/" generally indicates that the former and latter associated objects are in an "or" relationship.

In the present disclosure, unless otherwise expressly specified or limited, the terms "connected" and the like are to be construed broadly, e.g., as meaning electrically connected or in communication with each other; may be directly connected or indirectly connected through an intermediate. The specific meaning of the above terms in the present disclosure can be understood by those of ordinary skill in the art as appropriate.

As described above, Internet companies may receive revenue through Internet advertising. The internet advertisement is divided according to the transaction mode between the internet company and the advertiser, and mainly comprises the following advertisement forms.

Contract advertising: contract internet advertisements are developed from traditional advertisement forms, and advertisers can use an internet platform as media like televisions, newspapers and the like to put banner advertisements on web pages. The transaction mode is similar to the traditional method for putting the advertisement, the internet company and the advertiser negotiate to determine the time for the advertiser to monopolize a certain position on the page and the price for the advertiser to pay the internet company.

Guarantee type targeted advertisement: with the development of internet advertisements, the way of exclusive page gradually develops into the way of allocation according to needs, for example: and putting advertisements of the shaver for male users and advertisements of the mask for female users. When the advertisement is put in, the flow can be labeled, such as gender, region, hobbies and interests, and the advertiser purchases the corresponding advertisement position according to the label. On one hand, the flow of the media realizes the refined selling, and on the other hand, the advertiser can also buy the flow aiming at the target crowd of the self product. At the moment, the price mechanism of the advertisement can be negotiated pricing, an advertiser purchases flow by selecting the label, the flow of different labels is different in size, the advertiser can put forward the flow requirement, and if the flow requirement is not met, the internet company pays the advertiser according to the negotiated price.

Real-Time Bid advertisement (RTB): for guaranteed advertisements, internet companies need to solve the problems of traffic estimation, distribution and the like, the workload is heavy, and the real-time bidding advertisements reduce the burden of the internet companies. The advertiser sets the targeted traffic and sets a price on the targeted traffic. In a certain directional traffic, an advertiser meeting a predetermined condition makes a bid, and an internet company determines the attribution of the directional traffic according to factors such as the bid of the advertiser. The selling mode of the RTB advertisement, namely the bidding categories of the advertisers, mainly comprises: thousands of exposure Charges (CPM), Per Click charges (CPC), and Per Action Charges (CPA), among others, where an "Action" in each Action charge may be understood as an advertiser product deal, an order placement, an application (App) installation, an App activation, and so on, which may also be referred to as a conversion.

When selecting recommended goods from a large number of advertised goods, the internet company may consider an advertising bidding factor of an advertiser, and may also consider an index factor representing the experience of the user with the advertised goods in order to maintain a stable flow rate. In some cases, advertising bids of advertisers may change, such as raising advertising bids to increase exposure to merchandise when new products are on shelf, or lowering advertising bids to save advertising costs when advertising budgets are reduced, etc. In the related art, when candidate recommended goods are determined according to the advertisement bids and the user experience indexes, the change of the advertisement bids may have a small influence on the determination of the candidate recommended goods. Accordingly, the present disclosure provides an advertisement delivery method for obtaining a candidate advertisement commodity set according to a received advertisement playing request, obtaining a second candidate advertisement commodity based on advertisement price attribute information of a first candidate advertisement commodity in the candidate advertisement commodity set, obtaining a replacement reference lower limit value according to an estimated click rate of the first candidate advertisement commodity, then when the estimated click rate of the second candidate advertisement commodity is higher than the replacement reference lower limit value, replacing the first candidate advertisement commodity with the second candidate advertisement commodity to obtain an updated candidate advertisement commodity set for advertisement delivery, therefore, the first candidate advertisement commodity in the candidate advertisement commodity set can be replaced by the second candidate advertisement commodity obtained based on the advertisement bid, the influence of the advertisement bid on the determination of the candidate advertisement putting set is improved to a certain extent, and the advertisement putting rationality is improved.

Fig. 1 illustrates an exemplary system architecture 10 to which the advertisement delivery method or advertisement delivery apparatus of the present disclosure may be applied.

As shown in fig. 1, system architecture 10 may include a terminal device 102, a network 104, and a server 106. Terminal device 102 may be a variety of electronic devices having a display screen and supporting web browsing, including but not limited to smart phones, tablets, laptop portable computers, desktop computers, and the like. Network 104 is the medium used to provide communication links between terminal device 102 and server 106. Network 104 may include various connection types, such as wired, wireless communication links, or fiber optic cables, to name a few. The server 106 may be a server, a cluster of servers, etc. that provide various services.

A user may use terminal device 102 to interact with server 106 via network 104 to receive or send messages and the like. Various communication client applications, such as shopping applications, social applications, etc., may be installed on the terminal device 102. For example, when a user uses a shopping application on the terminal device 102, the shopping application may obtain historical shopping information of the user from the server 106 through the network 104, and then perform advertisement placement according to the historical shopping information of the user.

The server 106 may be, for example, a background management server that provides support for shopping websites and the like browsed by the user using the terminal device 102. The background management server may analyze and process data received via the network 104, such as data of a user browsing a product on a website of the terminal device 102, clicking an advertisement, and the like, and feed back advertisement delivery information and the like to the terminal device.

It should be understood that the number of terminal devices, networks, and servers in fig. 1 is merely illustrative. There may be any number of terminal devices, networks, and servers, as desired for implementation.

Fig. 2 is a flow chart illustrating a method of advertisement delivery, according to an example embodiment. The method shown in fig. 2 may be applied to, for example, a server side of the system, and may also be applied to a terminal device of the system.

Referring to fig. 2, a method 20 provided by an embodiment of the present disclosure may include the following steps.

In step S202, an advertisement play request is received.

In step S204, a candidate advertisement product set is obtained according to the advertisement playing request. When candidate advertisement commodities are obtained, candidate advertisement commodities to be recommended to a user can be screened from massive commodities in the database according to a recommendation object (also called a user, such as a user of an online product of an internet company).

In some embodiments, for example, the set of candidate advertisement items may be obtained by the advertisement delivery server from the database server via a pre-recommendation algorithm based on the user's historical behavior. The pre-recommendation algorithm can be adopted to mine the commodities which are possibly interested by the user from the billion commodity sets based on the collaborative filtering method according to the historical clicking, purchase, purchasing, query and other behaviors of the user.

In other embodiments, the set of candidate advertisement items may also be obtained from the database server by the advertisement delivery client through a pre-recommendation algorithm based on the user's historical behavior, for example.

In other embodiments, for example, a deep learning model may also be used for personalized recommendation, and a neural network is trained through behaviors of a large number of users such as historical click, purchase, query, and the like, so that the neural network may output commodities that may be of interest to the user according to input historical behavior characteristics of the user.

In step S206, advertisement price attribute information of a first candidate advertisement item in the candidate advertisement item set is acquired. When an advertiser puts an advertisement of a candidate advertisement commodity in an advertisement putting system of an internet company, price attribute information such as an advertisement bid of the commodity is set, and the advertisement bid is stored in the advertisement putting system; the advertiser can also modify the advertisement bids according to needs, and the advertisement bids of the candidate advertisement commodities can be obtained through the advertisement putting system.

In step S208, a second candidate advertisement item is acquired based on the advertisement price attribute information of the first candidate advertisement item. The candidate advertisement commodity set comprises candidate advertisement commodities obtained by preliminary screening from massive commodities. A second candidate advertisement item (hereinafter referred to as "supplemental item") that is higher or lower than the advertisement bid for the candidate advertisement item may be selected from the advertisement item database according to actual needs.

In some embodiments, for example, if a supplemental item with an advertising bid higher than the candidate advertising item is desired, a set of high bid items may be generated at predetermined physical intervals by the offline system, and a supplemental item with an advertising bid higher than the candidate advertising item is selected from the set of high bid items to reduce the computational load of the search and improve system performance. The method for obtaining the high price commodity set can refer to fig. 6, and the architecture of the offline system for generating the high price commodity set can refer to fig. 7, which is not described in detail herein.

In step S210, the estimated click rate of the first candidate advertisement product and the estimated click rate of the second candidate advertisement product are obtained. Whether the internet company charges by adopting a CPM (CPM), CPC (CPC) or CPA (CPA), the income of the final flow change can be uniformly measured by eCPM (effective Cost Per Mile, income is shown every thousand times). CPC charges become the mainstream in the current internet advertisement, and when adopting CPC charges, the user triggers the advertisement action of clicking, and the advertisement owner's expense just can be collected to the internet. Accordingly, the advertising revenue of the internet company is related not only to the advertiser bid (i.e., CPC) but also to the Click Through Rate (CTR) of the advertisement. The expected revenue per presentation E is:

E＝CTR×CPC (1)

in the formula, the CTR may be used to represent the user satisfaction, i.e. a satisfaction index, and the higher the CTR is, the more satisfied the user is, the better the user experience is. The Click Rate of the goods can be estimated Through the Click Through Rate (CTR) estimation algorithm of the advertisement goods, and the estimated Click Through Rate (PCTR) of each goods is obtained. The CTR estimation algorithm based on the deep learning model can be adopted, for example, user behaviors (such as user browsing, buying, purchasing, inquiring, collecting and the like), user portrait characteristics (such as age, gender, academic calendar, address and the like) and commodity information (category, brand, price and the like) are input into a deep neural network, and the network outputs the click rate of each user on each commodity to obtain the PCTR of each user on each commodity.

In step S212, a replacement reference lower limit value is obtained according to the estimated click rate of the first candidate advertisement product. In some embodiments, for example, the replacement reference lower limit value may be obtained by multiplying the estimated click-through rate of the candidate advertisement item by a predetermined percentage; as another example, the replacement reference lower limit value may be obtained by subtracting a predetermined value from the estimated click rate of the candidate advertisement item. And obtaining a replacement reference lower limit value of the click rate so as to select a second candidate advertisement commodity for replacement in the following process, so that the estimated click rate of the second candidate advertisement commodity is not too low, and poor user experience caused by recommending the second candidate advertisement commodity with too low estimated click rate is avoided.

In step S214, when the estimated click rate of the second candidate advertisement product is higher than the replacement reference lower limit value, the first candidate advertisement product is replaced with the second candidate advertisement product, and an updated candidate advertisement product set is obtained for advertisement delivery. The estimated click rate of the complementary commodities is higher than the lower limit value of the click rate of the candidate recommended commodities, the estimated click rate of the complementary commodities is not too low, the user acceptance of the complementary commodities is not too poor, and at the moment, the complementary commodities can be used for replacing corresponding candidate advertisement commodities in the candidate advertisement commodity set, so that the complementary commodities obtained according to the advertisement bids are added into the candidate advertisement commodities.

In some embodiments, for example, after obtaining the updated candidate advertisement commodity set, the advertisements may be sorted according to the estimated click-through rate of the commodities therein and the advertisement price attribute information, for example, the advertisements may be sorted according to the order of the PCTR × CPC results from large to small; and then selecting a preset number of commodities arranged in the front for recommendation according to the advertisement position breadth. When the advertisement is placed, the first few commodities can be displayed at the front of the prominent position or the search result according to the sequence, for example, the first commodity is displayed at the center of the advertisement area, the second and third commodities are respectively displayed above and below the first commodity, and the like.

According to the advertisement delivery method provided by the disclosed embodiment, by obtaining a second candidate advertisement commodity based on the advertisement price attribute information of a first candidate advertisement commodity in the obtained candidate advertisement commodity set, obtaining a replacement reference lower limit value according to the estimated click rate of the first candidate advertisement commodity, when the estimated click rate of the second candidate advertisement commodity is higher than the replacement reference lower limit value, replacing the first candidate advertisement commodity in the candidate advertisement commodity set with the second candidate advertisement commodity to obtain an updated candidate advertisement commodity set for advertisement delivery, therefore, the first candidate advertisement commodity in the candidate advertisement commodity set can be replaced by the second candidate advertisement commodity obtained based on the advertisement price attribute information, the influence degree of the advertisement price attribute information on the determination of the candidate advertisement commodity set is improved, and the advertisement putting accuracy is further improved to a certain extent.

Fig. 3 is a schematic diagram illustrating a processing procedure of step S214 shown in fig. 2 in an embodiment. As shown in fig. 3, in the embodiment of the present disclosure, the step S214 may further include the following steps.

In step S2141, when the estimated click rate of the second candidate advertisement product is not higher than the replacement reference lower limit value, an unreplaced candidate advertisement product is obtained as the candidate advertisement product. And selecting a plurality of candidate advertisement commodities from the candidate advertisement commodity set to perform second candidate advertisement, obtaining a plurality of corresponding second candidate advertisement commodities, and judging whether to perform replacement according to a replacement reference lower limit value. And when the estimated click rate of the second candidate advertisement commodity is not higher than the corresponding replacement reference lower limit value, the expected user satisfaction of the second candidate advertisement commodity is low, the second candidate advertisement commodity is not added into the candidate advertisement commodity set, and the corresponding first candidate advertisement commodity is reserved.

In step S2142, when the estimated click rate of the second candidate advertisement product is higher than the replacement reference lower limit value, the first candidate advertisement product is replaced with the second candidate advertisement product, and a replacement candidate advertisement product is obtained. When the estimated click rate of the complementary commodity is higher than the corresponding replacement reference lower limit value, the expected user satisfaction of the complementary commodity is close to that of the corresponding candidate advertisement commodity, and if the influence on the user satisfaction caused by recommending the complementary commodity is small, the corresponding candidate advertisement commodity can be replaced by the complementary commodity.

In step S2143, when the estimated click rate of the first candidate advertisement product is higher than the estimated click rate of the second candidate advertisement product, the estimated click rate of the replacement candidate advertisement product is obtained as the estimated click rate of the first candidate advertisement product. Since the replacement candidate advertisement commodity (complementary commodity) is obtained according to the original candidate advertisement commodity, the satisfaction degrees of the user on the replacement candidate advertisement commodity and the complementary commodity are similar (the complementary commodity is a new product and the like, and the actions of historical click, purchase and the like are less, so that the estimated click rate obtained by adopting the deep learning model is lower), the estimated click rate of the candidate advertisement commodity can be used as the estimated click rate of the replacement candidate advertisement commodity, so that the influence of the advertisement bid of the replacement candidate advertisement commodity on the ranking of the candidate advertisement commodity is improved when the candidate advertisement commodity is ranked in the follow-up process.

In step S2124, when the estimated click rate of the candidate advertisement product is not higher than the estimated click rate of the second candidate advertisement product, the estimated click rate of the replacement candidate advertisement product is obtained as the estimated click rate of the second candidate advertisement product. If the estimated click rate of the complementary goods is higher than the estimated click rate of the original candidate advertisement goods, the estimated click rate of the complementary goods can be adopted, so that the influence of the advertisement bid for replacing the candidate advertisement goods on the ranking is improved, and the satisfaction degree of the user is not influenced.

In step S2125, the non-replacement candidate advertisement commodities and the replacement candidate advertisement commodities are sorted for advertisement delivery in order according to the advertisement price attribute information and the estimated click rate of the non-replacement candidate advertisement commodities and the advertisement price attribute information and the estimated click rate of the replacement candidate advertisement commodities. The method comprises the steps of obtaining a ranking index of the non-replaced candidate advertisement commodities according to the advertisement price attribute information and the estimated click rate of the non-replaced candidate advertisement commodities, obtaining a ranking index of the replaced candidate advertisement commodities according to the advertisement price attribute information and the estimated click rate of the replaced candidate advertisement commodities, ranking the non-replaced candidate advertisement commodities and the replaced candidate advertisement commodities according to the sequence of the ranking indexes from large to small to obtain a commodity sequence to be recommended, and then recommending commodities of the commodity sequence to be recommended from front to back.

In some embodiments, for example, in CPC billing mode, the Internet platform may rely on PCTR when selecting advertisements to be presented to the user^tThe results for x CPC sort the advertisements in descending order, where t>1 and t is a positive integer to increase the weight of PCTR to balance the impact of user experience and advertising revenue on ranking.

Internet platform according to PCTR^tWhen the x CPC ranks the candidate advertisements, the PCTR will dominate the ranking result of the candidate advertisements. Although the sorting mechanism ensures the relevance of users and improves the user experience, the bidding effect of advertisers is weakened, and the enthusiasm of the advertisers for improving the advertising bidding is influenced. When the advertiser is in urgent need of obtaining the display amount, the advertiser can choose to improve the advertisement bid, but under the condition that the t value is larger than 1, the advertiser is insensitive to the advertisement bid during sorting, and even if the advertiser increases the advertisement bid, the delivered goods cannot obtain more display amount, so that the sorting mechanism is poor in the experience of the advertiser.

According to the method provided by the embodiment of the disclosure, when the estimated click rate of the second candidate advertisement commodity is higher than the replacement reference lower limit value, the first candidate advertisement commodity is replaced by the second candidate advertisement commodity, and the larger value of the estimated click rate of the first candidate advertisement commodity and the estimated click rate of the second candidate advertisement commodity is selected as the estimated click rate of the replacement candidate advertisement commodity, and the estimated click rate of the replacement candidate advertisement commodity and the advertisement price attribute information thereof jointly determine the advertisement ranking, so that the sensitivity of the ranking to the advertisement price attribute information can be improved on the premise of not influencing the satisfaction index, and the enthusiasm of an advertiser for bidding is enhanced.

Fig. 4A is a flow diagram illustrating another method of advertisement delivery, according to an example embodiment. The method shown in fig. 4A may be applied to, for example, a server side of the system, and may also be applied to a terminal device of the system.

Referring to fig. 4A, a method 40 provided by an embodiment of the present disclosure may include the following steps.

In step S402, a candidate advertisement commodity set is obtained through a collaborative filtering algorithm based on the user identifier in the advertisement playing request, and first candidate advertisement commodities in the candidate advertisement commodity set are arranged from high to low according to the estimated satisfaction index. Candidate advertisement commodities (also called parent commodities) which are possibly interested by the user can be mined out from a billion commodity set by adopting a pre-recommendation algorithm according to historical clicking, shopping, purchasing, querying and other behaviors of the user based on a collaborative filtering method, for example, a recommendation value of each advertisement in the commodity set put to the user is calculated based on the collaborative filtering method, then commodities are sorted from high to low according to the recommendation value, and a preset number of commodities arranged at the top are selected as a parent commodity queue, wherein the preset number can be 800, 1000 or 1200 and the like.

In step S404, advertisement price attribute information of the candidate advertisement item in the candidate recommendation set is acquired. For a specific implementation, refer to step S206, which is not described herein again.

In step S406, a predetermined number of candidate advertisement items listed in the candidate advertisement item set are selected. Although the complementary goods of each parent goods can be obtained in the step, then the replacement strategy is executed on all the parent goods in the subsequent steps, the calculation amount is large, and only a small number of goods can be shown due to page limitation and the like when the complementary goods are finally recommended to the user, so that the complementary goods of the previous preset number of parent goods can be obtained to reduce the calculation amount, and the system performance is improved.

In step S408, a second candidate advertising item having an advertising bid higher than the advertising bid of the top predetermined number of candidate advertising items is obtained. In an advertising system, a plurality of categories of goods may be included, each belonging to a unique category, and each category may include a plurality of goods, for example, the categories may include cell phones, computers, keyboards, mice, etc. When the complementary commodity with higher advertisement bid is selected, the complementary commodity can be selected from the commodity set with the same category as the parent commodity, so that the estimated click rate of the complementary commodity and the parent commodity is more comparable, and the satisfaction degree of a user is not influenced after replacement. In addition, when the complementary goods are selected, the balance of the account of the advertiser of the selected complementary goods can be ensured at the same time.

In some embodiments, for example, when the goods complement is performed according to the advertising bid of the parent goods queue, the top preset number (such as 8, 10 or 12, etc.) of parent goods is used as the key, and for each of the parent goods, the goods with the same category and higher CPC is found in the category → high bid goods set mapping table as the complement goods. FIG. 4B shows a schematic diagram of a supplemental merchandise selection process. As shown in fig. 4B, for the parent product 1, the parent product 2, the parent product 3, and the parent product … … obtained in the data mining step, an algorithm association step is performed, a complementary product 1, a complementary product 2, a complementary product 3, … …, and a complementary product 10 that correspond to the same category and have a higher CPC are obtained in the category → high offered product set mapping table, and a complementary queue is placed to perform a complementary strategy. The complementary commodity with CPC higher than that of the father commodity can be randomly selected from the high-price commodity set by adopting a random algorithm, so that the phenomenon that the advertising expense is consumed too fast due to overhigh exposure of the commodity when the complementary commodity with the highest CPC is selected is avoided.

In some embodiments, for example, a predetermined category to which the candidate advertising item belongs may be obtained first, and a supplemental item having a real-time advertising bid higher than the advertising bid of the candidate advertising item may be obtained from a set of high-bid items for the predetermined category. An embodiment of obtaining a predetermined category of high bid commodity set may refer to fig. 6, and is not described in detail herein.

In step S410, an estimated click rate of the candidate advertisement item and an estimated click rate of a second candidate advertisement item are obtained based on the user identification. The user identification can be obtained according to the user request, and the click rate of the user on each commodity (including a father commodity and a complementary commodity) is estimated through a CTR estimation algorithm, so that the PCTR of each commodity is obtained. The CTR estimation algorithm based on the deep learning model can be adopted, for example, user behaviors and user portrait characteristics are obtained according to user identification, user behaviors (such as user browsing, purchase, inquiry and collection), user portrait characteristics (such as age, sex, academic calendar, address and the like) and commodity information (category, brand, price and the like) are input into a deep neural network, and the network outputs the click rate of each user on each commodity, so that the PCTR of each user on each commodity is obtained. At this point the system may obtain the latest CPC for each parent item and complementary item for use in subsequent sorting steps.

In step S412, the estimated click rate of the candidate advertisement product is multiplied by a predetermined percentage to obtain a lower limit value of the click rate. The predetermined percentage k of the PCTR of the candidate advertised items may be preset, e.g., k may be set to 5%, 10%, or 15%, etc.

In step S414, when the estimated click rate of the second candidate advertisement product is higher than the replacement reference lower limit value, the candidate advertisement product is replaced with the second candidate advertisement productAnd obtaining an updated candidate advertisement commodity set for advertisement putting. Comparing the PCTR value of the complementary item with the PCTR value of the corresponding parent item, if PCTR is_{Supplemental merchandise}>PCTR_{Parent commodity}X (1-k), replacing the parent commodity in the parent commodity queue with the corresponding complementary commodity, and updating the parent commodity queue; otherwise, the replacement is not performed. On the premise that the parent commodity and the complementary commodity belong to the same kind of purpose, the satisfaction degrees of the user on the parent commodity and the complementary commodity are similar (the complementary commodity is probably a new commodity, the actions of historical clicking, purchasing and the like are less, and the PCTR obtained by adopting a deep learning model is lower), so that the PCTR of the complementary commodity added into the parent commodity queue is lower_{Supplemental merchandise}Replacement by PCTR_{Parent commodity}Then all the commodities in the parent commodity queue at the moment are in accordance with PCTR^tThe values of x CPC are sorted from large to small, and finally a preset number (e.g., 5, 10, or 12) of top ranked ads are delivered to the user.

In some embodiments, for example, FIG. 4C illustrates an alternative strategy diagram for replenishment of merchandise. As shown in fig. 4C, when k is 10%, the PCTR of the complementary item 2 is 0.28> (0.3 × 90%) 0.27, and therefore the parent item 2 in the parent item queue is replaced with the complementary item 2, and the PCTR of the complementary item 2 after replacement is replaced with 3; PCTR of the complementary item 3 is 0.25> (0.2 × 90%) 0.18, and thus the parent item 3 in the parent item queue is replaced with the complementary item 3, and 0.25>0.2, and thus the PCTR of the complementary item 3 after replacement is retained is 0.25; the PCTR of the complementary item 10 is 0.1> (0.11 × 90%) 0.099, and therefore the parent item 10 in the parent item queue is replaced with the complementary item 10, and the PCTR of the complementary item 10 after replacement is replaced with 0.11.

According to the method provided by the embodiment of the disclosure, parent commodities excavated by a complementary commodity replacement algorithm with high bids under the same category are used, and on the premise that the PCTR of the complementary commodities is not too low to influence the satisfaction degree of a user, the complementary commodities with higher bids are used for replacing the parent commodities to show higher-priced commodities, so that the sensibility to the bids when the commodities are recommended by an advertising system is improved; and the complement commodities of the former predetermined number of father commodities are obtained to reduce the operation amount and improve the system performance.

Fig. 5 is a schematic view of an advertisement delivery process according to fig. 2 to 4. Advertising can be carried out through the online system (50), and the user request can be obtained in real time (S502); mining a parent commodity which is possibly interested by the user from the billion commodity set based on a collaborative filtering method by adopting a pre-recommendation algorithm according to the actions of historical clicking, purchase adding, purchasing, query and the like of the user (S504); using the parent commodities which are arranged in the front preset number (such as 8, 10 or 12) as key words, finding the commodities which are similar to the parent commodities and have higher CPC in the category → high-price commodity set mapping table (5002) as complementary commodities, and obtaining a candidate commodity queue (S506); then, predicting the click rate of each commodity (including a father commodity and a complementary commodity) through a CTR prediction algorithm to obtain the PCTR of each commodity, wherein the system can obtain the latest CPC of each father commodity and the complementary commodity (S508); then, a replacement policy is executed (S510): setting the lower limit k of the ratio of PCTR, e.g. setting k to 5%, 10% or 15%, etc., comparing the PCTR value of the complementary item with the PCTR value of the corresponding parent item, if PCTR is_{Supplemental merchandise}>PCTR_{Parent commodity}X (1-k), replacing the parent commodity in the parent commodity queue with the corresponding complementary commodity, updating the parent commodity queue, or not replacing; then all the commodities in the parent commodity queue at the moment are in accordance with PCTR^tThe values of x CPC are sorted from large to small, and finally, a preset number of top-ranked advertisements are delivered to the user (S512).

FIG. 6 is a flow chart illustrating a method for obtaining a set of high priced items according to an exemplary embodiment. The method shown in fig. 6 may be applied to, for example, a server side of the system, and may also be applied to a terminal device of the system.

Referring to fig. 6, a method 60 provided by embodiments of the present disclosure may include the following steps.

In step S602, historical advertisement price attribute information of a commodity belonging to a predetermined category is acquired. The mapping relation between the commodities → the categories can be obtained according to the commodity attribute information base, the historical bid list of the commodities is obtained from the historical advertisement transaction information base of the advertisement delivery system, the commodity category relation table and the commodity historical bid information table are associated, the historical bid information table of each category is obtained, and then the historical advertisement bids of all the commodities of the preset category are obtained. Wherein: when the mapping relation of commodities → categories is mined, effective commodities which are not placed on shelves are selected, the category to which each commodity belongs is extracted, each commodity belongs to a unique category, and each category can comprise a plurality of commodities, for example, the categories can comprise a mobile phone, a computer, a keyboard, a mouse and the like; the behavior that each commodity is displayed and clicked for a plurality of times can occur, the bids at each click are different, information aggregation is carried out according to the commodities, historical bid information of each commodity is mined, and a historical bid list of the commodity is obtained.

In step S604, reserved advertisement price attribute information for a predetermined category of goods is determined from the historical advertisement price attribute information. A price distribution of historical advertising bids can be obtained, with prices at predetermined quantiles of the price distribution being selected as reserve advertising bids, e.g., bids at 25%, 30%, or 35% quantiles can be selected from the historical bid information table for each category as reserve prices for that category. The purpose of setting reserve prices is to obtain a threshold bid value based on the category's historical bid level, preventing the advertiser from lowering the bid too low, while reducing the amount of items in the set of higher priced items to reduce the effort of looking up data in the set.

In step S606, real-time advertisement price attribute information of a commodity belonging to a predetermined category is acquired. The advertisement delivery system can set CPC of the commodity in real time by an advertiser, and can extract the CPC of the commodity in real time from the real-time advertisement information base. And associating the commodity category relation table with the commodity real-time bidding information table to obtain the corresponding commodity under each category and the real-time CPC information thereof.

In step S608, a product with real-time advertisement price attribute information higher than the reserved advertisement price attribute information is selected, and a high price attribute information product set of a predetermined category is obtained. And screening out the commodities with the CPC higher than the reserve price under each category according to the reserve price of each category, and finally obtaining a category → high price commodity set mapping table.

According to the method provided by the embodiment of the disclosure, based on the commodity attribute information, the historical advertisement transaction information and the real-time advertisement information, the commodities with high bids under each category are obtained through a data mining and association algorithm, so that the advertisement bids of recommended commodities are guaranteed, the workload of searching for the commodities with high bids is reduced, and the performance of an online system is improved.

Fig. 7 is a schematic diagram of a high-priced commodity set obtaining flow according to fig. 6, and illustrates an architecture of an offline system 70 for obtaining a high-priced commodity set. The offline system 70 may update the set of high-priced goods at a preset frequency, such as every 0.5 hour, or 1 hour, or 1.5 hours, etc. to regenerate the set of high-priced goods. The offline system 70 may include a data acquisition module 702, a data mining module 704, and an algorithm association module 706.

The data acquiring module 702 is configured to acquire original information data, where the original information data may include a product attribute information library 7004, a historical advertisement transaction information library 7002, a real-time advertisement information library 7006, and the like, where: the commodity attribute information base 7004 can include attribute information of the brand, category, shelf getting-on and getting-off time and the like of the commodity; the historical advertisement deal information repository 7002 may include the advertisement item clicked, the user clicking on the advertisement item, the CPC per click, and so on; the real-time advertising information repository 7006 may include the CPC set by the advertiser in real-time, the advertiser's real-time account balance, the advertiser's Identification (ID), and the like. The data obtaining module 702 may obtain real-time advertisement information from the advertisement delivery system, and the real-time advertisement information in the advertisement delivery system is transmitted to the data obtaining module 702 of the offline system with a short delay.

The data mining module 704 is configured to respectively mine a product → category mapping relationship (S714), a historical bid list (S712) of the product, and a real-time CPC (S716) of the product from the product attribute information base, the historical advertisement delivery information base, and the real-time advertisement information base; in the advertisement delivery system, the advertiser may set the CPC of the product in real time, and the data mining module 704 may extract the CPC of the product in real time from the real-time advertisement information base through the data obtaining module 702.

The algorithm association module 706 may further obtain a reserve price corresponding to each category (S722), goods under each category and their real-time CPC (S724) based on the data of the data mining module 704, and finally obtain a set of goods with high bid under each category; and (3) screening out commodities with CPC higher than the reserve price under each category through algorithm association according to the reserve price of each category, and finally obtaining a category → high price commodity set mapping table (5002).

Fig. 8 is a block diagram illustrating an advertising device in accordance with an exemplary embodiment. The apparatus shown in fig. 8 may be applied to, for example, a server side of the system, and may also be applied to a terminal device of the system.

Referring to fig. 8, the apparatus 80 provided in the embodiment of the present disclosure may include a request receiving module 801, a goods filtering module 802, an information obtaining module 804, a goods complementing module 806, a click rate estimating module 808, a replacement preparing module 810, and a goods recommending module 812.

The request receiving module 801 may be configured to receive an advertisement play request.

The goods filtering module 802 may be configured to obtain a candidate advertisement goods set according to the advertisement playing request.

The information obtaining module 804 may be configured to obtain advertisement price attribute information for a first candidate advertisement item in the set of candidate advertisement items.

The item replenishment module 806 is operable to obtain a second candidate advertising item based on the advertising price attribute information for the first candidate advertising item.

The click rate estimation module 808 may be configured to obtain an estimated click rate of the first candidate advertisement item and an estimated click rate of the second candidate advertisement item.

The replacement preparation module 810 may be configured to obtain a replacement reference lower limit value according to the estimated click rate of the first candidate advertisement item.

The product recommending module 812 may be configured to replace the first candidate advertisement product with the second candidate advertisement product when the estimated click rate of the second candidate advertisement product is higher than the replacement reference lower limit value, and obtain an updated candidate advertisement product set for advertisement delivery.

Fig. 9 is a block diagram illustrating an advertising device in accordance with an exemplary embodiment. The apparatus shown in fig. 9 may be applied to, for example, a server side of the system, and may also be applied to a terminal device of the system.

Referring to fig. 9, the apparatus 90 provided in the embodiment of the present disclosure may include a request receiving module 901, a commodity filtering module 902, an information obtaining module 904, a historical data obtaining module 9052, a reserve price determining module 9054, a real-time data obtaining module 9056, a high-price commodity set obtaining module 9058, a commodity complement module 906, a click rate estimating module 908, a replacement preparation module 910, and a commodity recommending module 912, where the commodity filtering module 902 may include a pre-algorithm module 9022, the commodity complement module 906 may include a category querying module 9062 and a complement commodity obtaining module 9064, the commodity recommending module 912 may include a replacement executing module 9122 and a commodity sorting module 9124, and the commodity sorting module 9124 may include a user recommendation degree calculating module 91242 and a to-be recommended commodity sequence obtaining module 91244.

The request receiving module 901 may be configured to receive an advertisement playing request. The advertisement play request includes a user identification.

The goods filtering module 902 may be configured to obtain a candidate advertisement goods set according to the advertisement playing request.

The pre-algorithm module 9022 may be configured to obtain a candidate advertisement commodity set through a collaborative filtering algorithm based on the user identifier, where a first candidate advertisement commodity in the candidate advertisement commodity set is arranged from high to low according to the estimated satisfaction index.

The information obtaining module 904 may be configured to obtain advertisement price attribute information for a first candidate advertisement item in the set of candidate advertisement items.

The historical data acquisition module 9052 may be operable to acquire historical advertising bids for items belonging to a predetermined category.

Reserve price determination module 9054 may be configured to determine a reserve advertising bid for a predetermined category of goods based on the historical advertising bids.

The reserve price determination module 9054 may also be configured to obtain a price distribution of the historical advertising bids; and selecting the price of the preset quantile point of the price distribution as the reserved advertisement bid.

The real-time data acquisition module 9056 may be operable to acquire real-time advertising bids for items belonging to a predetermined category.

The high bid commodity set obtaining module 9058 may be configured to select a commodity with a real-time advertisement bid higher than the reserved advertisement bid to obtain a high bid commodity set of a predetermined category.

The item replenishment module 906 may be configured to obtain a second candidate advertised item based on the advertising price attribute information for the first candidate advertised item.

The goods complement module 906 is further configured to select a predetermined number of first candidate advertisement goods in the first candidate advertisement goods arranged according to the pre-estimated satisfaction index; a second candidate advertisement item having advertisement price attribute information higher than that of the first candidate advertisement item ranked in the preceding predetermined number is acquired.

Category query module 9062 may be configured to obtain a predetermined category to which the first candidate advertised item belongs.

The supplemental item obtainment module 9064 can be operable to obtain a second candidate advertising item from the set of high-bid items for the predetermined category, with the real-time advertising bid higher than the advertising bid for the first candidate advertising item.

The click rate estimation module 908 may be configured to obtain an estimated click rate of the first candidate advertisement item and an estimated click rate of the second candidate advertisement item.

The click-through rate estimation module 908 may also be configured to multiply the estimated click-through rate for the first candidate advertising item by a predetermined percentage to obtain a replacement reference lower limit value.

The replacement preparation module 910 may be configured to obtain a replacement reference lower limit value according to the estimated click rate of the first candidate advertisement item.

The product recommending module 912 may be configured to replace the first candidate advertisement product with the second candidate advertisement product when the estimated click rate of the second candidate advertisement product is higher than the replacement reference lower limit value, and obtain an updated candidate advertisement product set for advertisement delivery. The updated set of candidate advertising items includes the non-replacement candidate advertising items and the replacement candidate advertising items.

The replacement execution module 9122 may be configured to replace the candidate advertised item with a second candidate advertised item, obtaining a replacement candidate advertised item.

The replacement performing module 9122 may be further configured to obtain the estimated click rate of the replacement candidate advertisement product as the estimated click rate of the candidate advertisement product when the estimated click rate of the candidate advertisement product is higher than the estimated click rate of the second candidate advertisement product.

The goods ordering module 9124 may be configured to order the non-replacement candidate advertisement goods and the replacement candidate advertisement goods according to the advertisement price attribute information and the estimated click rate of the non-replacement candidate advertisement goods and the advertisement price attribute information and the estimated click rate of the replacement candidate advertisement goods, so as to deliver the advertisements in sequence.

The user recommendation calculation module 91242 may be configured to obtain a ranking indicator of the un-replaced candidate advertisement product according to the advertisement price attribute information and the estimated click rate of the un-replaced candidate advertisement product.

The user recommendation degree calculation module 91242 may be further configured to obtain a ranking index of the replacement candidate advertisement product according to the advertisement price attribute information and the estimated click rate of the replacement candidate advertisement product.

The to-be-recommended commodity sequence obtaining module 91244 may be configured to sequence the non-replaced candidate advertisement commodities and the replaced candidate advertisement commodities according to a descending order of the ranking index, so as to obtain a to-be-advertised commodity sequence.

The product recommendation module 912 can also be configured to advertise the sequence of products to be advertised from front to back.

The specific implementation of each module in the apparatus provided in the embodiment of the present disclosure may refer to the content in the foregoing method, and is not described herein again.

Fig. 10 shows a schematic structural diagram of an electronic device in an embodiment of the present disclosure. It should be noted that the apparatus shown in fig. 10 is only an example of a computer system, and should not bring any limitation to the function and the scope of the application of the embodiments of the present disclosure.

As shown in fig. 10, the apparatus 1000 includes a Central Processing Unit (CPU)1001 that can perform various appropriate actions and processes in accordance with a program stored in a Read Only Memory (ROM)1002 or a program loaded from a storage section 1008 into a Random Access Memory (RAM) 1003. In the RAM 1003, various programs and data necessary for the operation of the apparatus 1000 are also stored. The CPU1001, ROM 1002, and RAM 1003 are connected to each other via a bus 1004. An input/output (I/O) interface 1005 is also connected to bus 1004.

The following components are connected to the I/O interface 1005: an input section 1006 including a keyboard, a mouse, and the like; an output section 1007 including a display such as a Cathode Ray Tube (CRT), a Liquid Crystal Display (LCD), and the like, and a speaker; a storage portion 1008 including a hard disk and the like; and a communication section 1009 including a network interface card such as a LAN card, a modem, or the like. The communication section 1009 performs communication processing via a network such as the internet. The driver 1010 is also connected to the I/O interface 1005 as necessary. A removable medium 1011 such as a magnetic disk, an optical disk, a magneto-optical disk, a semiconductor memory, or the like is mounted on the drive 1010 as necessary, so that a computer program read out therefrom is mounted into the storage section 1008 as necessary.

In particular, according to an embodiment of the present disclosure, the processes described above with reference to the flowcharts may be implemented as computer software programs. For example, embodiments of the present disclosure include a computer program product comprising a computer program embodied on a computer readable medium, the computer program comprising program code for performing the method illustrated in the flow chart. In such an embodiment, the computer program may be downloaded and installed from a network through the communication part 1009 and/or installed from the removable medium 1011. The above-described functions defined in the system of the present disclosure are executed when the computer program is executed by a Central Processing Unit (CPU) 1001.

It should be noted that the computer readable media shown in the present disclosure may be computer readable signal media or computer readable storage media or any combination of the two. A computer readable storage medium may be, for example, but not limited to, an electronic, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any combination of the foregoing. More specific examples of the computer readable storage medium may include, but are not limited to: an electrical connection having one or more wires, a portable computer diskette, a hard disk, a Random Access Memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disc read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the foregoing. In the present disclosure, a computer readable storage medium may be any tangible medium that can contain, or store a program for use by or in connection with an instruction execution system, apparatus, or device. In contrast, in the present disclosure, a computer-readable signal medium may include a propagated data signal with computer-readable program code embodied therein, for example, in baseband or as part of a carrier wave. Such a propagated data signal may take many forms, including, but not limited to, electro-magnetic, optical, or any suitable combination thereof. A computer readable signal medium may also be any computer readable medium that is not a computer readable storage medium and that can communicate, propagate, or transport a program for use by or in connection with an instruction execution system, apparatus, or device. Program code embodied on a computer readable medium may be transmitted using any appropriate medium, including but not limited to: wireless, wire, fiber optic cable, RF, etc., or any suitable combination of the foregoing.

The flowchart and block diagrams in the figures illustrate the architecture, functionality, and operation of possible implementations of systems, methods and computer program products according to various embodiments of the present disclosure. In this regard, each block in the flowchart or block diagrams may represent a module, segment, or portion of code, which comprises one or more executable instructions for implementing the specified logical function(s). It should also be noted that, in some alternative implementations, the functions noted in the block may occur out of the order noted in the figures. For example, two blocks shown in succession may, in fact, be executed substantially concurrently, or the blocks may sometimes be executed in the reverse order, depending upon the functionality involved. It will also be noted that each block of the block diagrams or flowchart illustration, and combinations of blocks in the block diagrams or flowchart illustration, can be implemented by special purpose hardware-based systems which perform the specified functions or acts, or combinations of special purpose hardware and computer instructions.

The modules described in the embodiments of the present disclosure may be implemented by software or hardware. The described modules may also be provided in a processor, which may be described as: a processor comprises a commodity filtering module, an information acquisition module, a commodity complement module, a click rate estimation module, a replacement preparation module and a commodity recommendation module. Where the names of these modules do not in some cases constitute a limitation on the module itself, for example, the item filtering module may also be described as a "module for obtaining candidate advertised items based on a coordinated filtering method".

As another aspect, the present disclosure also provides a computer-readable medium, which may be contained in the apparatus described in the above embodiments; or may be separate and not incorporated into the device. The computer readable medium carries one or more programs which, when executed by a device, cause the device to comprise: receiving an advertisement playing request; acquiring a candidate advertisement commodity set according to the advertisement playing request; acquiring advertisement price attribute information of a first candidate advertisement commodity in a candidate advertisement commodity set; acquiring a second candidate advertisement commodity based on the advertisement price attribute information of the first candidate advertisement commodity; acquiring the estimated click rate of a first candidate advertisement commodity and the estimated click rate of a second candidate advertisement commodity; obtaining a replacement reference lower limit value according to the estimated click rate of the first candidate advertisement commodity; and when the estimated click rate of the second candidate advertisement commodity is higher than the replacement reference lower limit value, replacing the first candidate advertisement commodity with the second candidate advertisement commodity to obtain an updated candidate advertisement commodity set for advertisement delivery.

Exemplary embodiments of the present disclosure are specifically illustrated and described above. It is to be understood that the present disclosure is not limited to the precise arrangements, instrumentalities, or instrumentalities described herein; on the contrary, the disclosure is intended to cover various modifications and equivalent arrangements included within the spirit and scope of the appended claims.

Claims

1. An advertisement delivery method, comprising:

receiving an advertisement playing request;

acquiring a candidate advertisement commodity set according to the advertisement playing request;

acquiring advertisement price attribute information of a first candidate advertisement commodity in the candidate advertisement commodity set;

acquiring a second candidate advertisement commodity based on the advertisement price attribute information of the first candidate advertisement commodity;

acquiring the estimated click rate of the first candidate advertisement commodity and the estimated click rate of the second candidate advertisement commodity;

obtaining a replacement reference lower limit value according to the estimated click rate of the first candidate advertisement commodity;

and when the estimated click rate of the second candidate advertisement commodity is higher than the replacement reference lower limit value, replacing the first candidate advertisement commodity with the second candidate advertisement commodity to obtain an updated candidate advertisement commodity set for advertisement delivery.

2. The method of claim 1 wherein the updated set of candidate advertised items includes an un-replaced candidate advertised item and a replaced candidate advertised item;

the replacing the first candidate advertised item with the second candidate advertised item includes:

replacing the first candidate advertisement commodity with the second candidate advertisement commodity to obtain a replacement candidate advertisement commodity;

when the estimated click rate of the first candidate advertisement commodity is higher than the estimated click rate of the second candidate advertisement commodity, obtaining the estimated click rate of the replacement candidate advertisement commodity as the estimated click rate of the first candidate advertisement commodity;

the obtaining an updated set of candidate advertising items for advertising comprises:

and sequencing the non-replaced candidate advertisement commodities and the replaced candidate advertisement commodities according to the advertisement price attribute information and the estimated click rate of the non-replaced candidate advertisement commodities and the advertisement price attribute information and the estimated click rate of the replaced candidate advertisement commodities so as to carry out advertisement putting in sequence.

3. The method of claim 2, wherein said ranking the non-replacement candidate advertisement item and the replacement candidate advertisement item according to the advertisement price attribute information and the estimated click rate of the non-replacement candidate advertisement item and the advertisement price attribute information and the estimated click rate of the replacement candidate advertisement item comprises:

obtaining the ranking index of the non-replaced candidate advertisement commodity according to the advertisement price attribute information and the estimated click rate of the non-replaced candidate advertisement commodity;

obtaining the ranking index of the replacement candidate advertisement commodity according to the advertisement price attribute information and the estimated click rate of the replacement candidate advertisement commodity;

sequencing the non-replaced candidate advertisement commodities and the replaced candidate advertisement commodities according to the sequence of the sequencing indexes from large to small to obtain an advertisement commodity sequence to be delivered;

the sequential advertisement delivery comprises:

and carrying out advertisement putting on the sequence of the to-be-put advertisement commodities from front to back.

4. The method of claim 1, wherein the advertisement play request includes a user identification;

the obtaining of the candidate advertisement commodity set according to the advertisement playing request includes:

acquiring the candidate advertisement commodity set through a collaborative filtering algorithm based on the user identification, wherein first candidate advertisement commodities in the candidate advertisement commodity set are arranged from high to low according to an estimated satisfaction index;

the obtaining a second candidate advertisement commodity based on the advertisement price attribute information of the first candidate advertisement commodity comprises:

selecting a preset number of first candidate advertisement commodities in the first candidate advertisement commodities arranged according to the estimated satisfaction index;

acquiring the second candidate advertisement commodity with advertisement price attribute information higher than the advertisement price attribute information of the first candidate advertisement commodity with the preset quantity in the front row.

5. The method of claim 1, further comprising:

obtaining historical advertising bids for goods belonging to a predetermined category;

determining a reserved advertising bid for the predetermined category of goods according to the historical advertising bid;

obtaining real-time advertising bids for goods belonging to the predetermined category;

selecting commodities with real-time advertising bids higher than the reserved advertising bids, and obtaining a high-bid commodity set of the predetermined category;

obtaining the predetermined category to which the first candidate advertisement commodity belongs;

obtaining the second candidate advertising item from the predetermined category of high-bid item sets with a real-time advertising bid higher than the advertising bid of the first candidate advertising item.

6. The method of claim 5, wherein determining a reserve advertising bid for the predetermined class of items based on the historical advertising bids comprises:

obtaining a price distribution of the historical advertising bids;

and selecting the price of the preset quantile point of the price distribution as the reserved advertisement bid.

7. The method of claim 1, wherein obtaining a replacement reference lower bound based on the estimated click rate of the first candidate advertised item comprises:

and multiplying the estimated click rate of the first candidate advertisement commodity by a preset percentage to obtain the replacement reference lower limit value.

8. An advertisement delivery device, comprising:

the request receiving module is used for receiving an advertisement playing request;

the commodity filtering module is used for acquiring a candidate advertisement commodity set according to the advertisement playing request;

the information acquisition module is used for acquiring advertisement price attribute information of a first candidate advertisement commodity in the candidate advertisement commodity set;

the commodity complement module is used for acquiring a second candidate advertisement commodity based on the advertisement price attribute information of the first candidate advertisement commodity;

the click rate estimation module is used for acquiring the estimated click rate of the first candidate advertisement commodity and the estimated click rate of the second candidate advertisement commodity;

the replacement preparation module is used for obtaining a replacement reference lower limit value according to the estimated click rate of the first candidate advertisement commodity;

and the commodity recommending module is used for replacing the first candidate advertisement commodity with the second candidate advertisement commodity when the estimated click rate of the second candidate advertisement commodity is higher than the replacement reference lower limit value, and obtaining an updated candidate advertisement commodity set for advertisement putting.

9. An apparatus, comprising: memory, processor and executable instructions stored in the memory and executable in the processor, characterized in that the processor implements the method according to any of claims 1-7 when executing the executable instructions.

10. A computer-readable storage medium having stored thereon computer-executable instructions, which when executed by a processor, implement the method of any one of claims 1-7.