WO2018192491A1

WO2018192491A1 - Information pushing method and device

Info

Publication number: WO2018192491A1
Application number: PCT/CN2018/083378
Authority: WO
Inventors: 赵夕炜; 徐夙龙; 胡景贺; 江雪
Original assignee: 北京京东尚科信息技术有限公司; 北京京东世纪贸易有限公司
Priority date: 2017-04-20
Filing date: 2018-04-17
Publication date: 2018-10-25
Also published as: CN107105031A

Abstract

Disclosed in the present application is an information pushing method and device. An embodiment of the method comprises: receiving a product information query request containing a search word, sent by a client; extracting multiple pieces of product information and multiple pieces of candidate information to be pushed, said information matching the search word; inputting the pieces of product information and the pieces of candidate information to be pushed into a previously trained order-placing rate prediction model, so as to obtain order-placing rates corresponding to the pieces of product information and the pieces of candidate information to be pushed, and inputting the pieces of candidate information to be pushed into a previously trained click-through rate prediction model, so as to obtain click-through rates corresponding to the pieces of candidate information to be pushed; determining, on the basis of the obtained order-placing rates and click-through rates, target information to be pushed among the pieces of candidate information to be pushed; and pushing to the client the target information to be pushed and the multiple pieces of product information. The embodiment of the application enables specific information pushing.

Description

Information push method and device

Cross-reference to related applications

This patent application claims to be submitted on April 20, 2017, with the application number of 201710260799.1, the applicant is Beijing Jingdong Shangke Information Technology Co., Ltd. and Beijing Jingdong Century Trading Co., Ltd., and the invention is entitled "Information Push Method and Device". Priority of the application, the entire contents of which are incorporated herein by reference.

Technical field

The present application relates to the field of computer technologies, and in particular, to the field of Internet technologies, and in particular, to an information push method and apparatus.

Background technique

With the development of computer technology, the use of e-commerce platforms for product transactions is becoming more and more common. In the process of searching for information by the e-commerce platform, users usually need to push the information needed by the user to reduce the information overload and reduce the time spent by the user on the network.

However, the existing information push method usually loads various push information directly at a certain fixed position of the search result, and the push information is greatly different from the search result, so that there is a problem that the information push is not targeted.

Summary of the invention

The purpose of the embodiments of the present application is to provide an improved information pushing method and apparatus to solve the technical problems mentioned in the above background art.

In a first aspect, an embodiment of the present application provides an information pushing method, which includes: receiving a product information query request that includes a search term sent by a client; and extracting multiple product information and multiple candidate pushes that match the search term. Information; inputting each product information and each candidate push information into a pre-trained order rate prediction model, obtaining an order rate corresponding to each product information and each candidate push information, and inputting each candidate push information to the pre-trained click The rate prediction model obtains a click rate corresponding to each candidate push information, wherein the order rate prediction model is used to represent the correspondence between the product information or the candidate push information and the order rate, and the click rate prediction model is used to represent the candidate push information and Corresponding relationship of click rates; determining target push information in the plurality of candidate push information based on the obtained order rate and click rate; pushing the target push information and the plurality of product information to the client.

In some embodiments, determining target push information in the plurality of candidate push information based on the obtained order rate and click rate, including: for each of the plurality of product information, corresponding to the product information The order rate is used as the order rate threshold, and the candidate push information of the plurality of candidate push information and the order rate is greater than the order rate threshold is selected, and the candidate push information set matching the product information is generated; for each generated a candidate push information set, determining a first expected value of each candidate push information in the candidate push information set, wherein a first expected value of each candidate push information is a click rate corresponding to the candidate push information and a preset a product of the charge value corresponding to the candidate push information; and based on the obtained first expected value, the target push information in the plurality of candidate push information is determined.

In some embodiments, each of the plurality of product information has a presentation order identifier for indicating a presentation order of the product information, and each candidate push information in the candidate push information set corresponding to the product information is presented Order identification.

In some embodiments, determining the target push information in the plurality of candidate push information based on the obtained first expected value comprises: using the candidate push information having the largest first expected value among the respective candidate push information sets as the target candidate push information. Generating a target candidate push information set; for each target candidate push information in the target candidate push information set, acquiring a preset order coefficient corresponding to the presentation order indicated by the presentation order identifier carried by the target candidate push information And determining a second expected value of the target candidate push information, wherein the second expected value is a product of a first expected value of the target candidate push information and a sequence coefficient corresponding to the target candidate push information; and a target candidate that maximizes the second expected value The push information is determined to be the target push information.

In some embodiments, the method further comprises: determining a presentation order indicated by the presentation order identifier carried by the target push information, and determining the determined presentation order as the target presentation order; determining each of the plurality of product information The presentation order indicated by the information indicates the presentation order indicated, and for each product information whose presentation order is not less than the target presentation order, the presentation order of the product information is increased by a first preset value;

In some embodiments, pushing the target push information and the plurality of product information to the client comprises: sorting the plurality of product information and the target product information in order of presentation from small to large; generating the plurality of products including the sorted Web page for information and target product information; send the web page to the client.

In some embodiments, before sorting the plurality of product information and the target product information in an order of presentation from small to large, the method further comprises: determining that a presentation order of the plurality of product information is less than a preset presentation order threshold The name of the category of the category to which the product information belongs, and the name of the category of the category to which the target push information belongs; the category name of the category to which the determined product information belongs is matched with the name of the category of the category to which the target push information belongs. Determining the number of product information whose category name matches the category name of the category to which the target push information belongs, and determining the total number of pieces of product information; in response to the ratio of the determined quantity to the total quantity being less than the preset ratio, the target is The presentation order of the push information is increased by a second preset value.

In some embodiments, each of the plurality of product information includes a product name, the target product information includes a target product name, and before the plurality of product information and the target product information are sorted in order of presentation from small to large, The method further includes: determining, for each product information of the plurality of product information that the presentation order is less than the preset presentation order threshold, the similarity between the product name in the product information and the target product name in the target push information; The determined similarity is less than a preset similarity threshold, and the product information is determined as the difference product information; and in response to determining that the quantity of the difference product information is greater than the preset number threshold, the display order of the target push information is increased by a third preset value. .

In some embodiments, before receiving the information query request sent by the client, the method further includes: extracting first feature information from the preset first training sample, wherein the first training sample includes The order identification of the order situation corresponding to the training sample; using the machine learning algorithm, based on the first feature information and the order identification, the training obtains the order rate prediction model.

In some embodiments, before receiving the information query request sent by the client, the method further includes: extracting second feature information from the preset second training sample, wherein the second training sample includes The click identifier of the click condition corresponding to the training sample; using the machine learning algorithm, the click rate prediction model is trained based on the second feature information and the click mark.

In a second aspect, an embodiment of the present application provides an information pushing apparatus, where the apparatus includes: a receiving unit configured to receive a product information query request that includes a search term sent by a client; and a first extracting unit configured to extract and Searching for a plurality of product information and a plurality of candidate push information; the input unit is configured to input each product information and each candidate push information into a pre-trained order rate prediction model, and obtain various product information and each candidate Pushing the corresponding order rate of the information, and inputting each candidate push information to the pre-trained click rate prediction model to obtain a click rate corresponding to each candidate push information, wherein the order rate prediction model is used to represent the product information or the candidate push The correspondence between the information and the order rate, the click rate prediction model is used to represent the correspondence between the candidate push information and the click rate; the first determining unit is configured to determine a plurality of candidate pushes based on the obtained order rate and the click rate. Target push information in the message; push unit configured to push the target information and multiple Product information to the clients.

In some embodiments, the determining unit includes: a first generating module configured to select, according to each of the plurality of product information, a billing rate corresponding to the product information as a billing rate threshold, and select a plurality of candidates The candidate push information in the push information that has a higher order rate than the order rate threshold, generates a candidate push information set that matches the product information; and the first determining module is configured to use, for each candidate push information set generated, Determining a first expected value of each candidate push information in the candidate push information set, wherein a first expected value of each candidate push information is a click rate corresponding to the candidate push information and a preset corresponding to the candidate push information a product of the billing value; the second determining module is configured to determine target push information in the plurality of candidate push information based on the obtained first expected value.

In some embodiments, the second determining module includes: a generating submodule configured to use the candidate push information with the first expected value being the largest among the candidate push information sets as the target candidate push information, to generate the target candidate push information set; a determining submodule configured to: for each target candidate push information in the target candidate push information set, obtain a preset order coefficient corresponding to the presentation order indicated by the presentation order identifier carried by the target candidate push information And determining a second expected value of the target candidate push information, wherein the second expected value is a product of a first expected value of the target candidate push information and a sequence coefficient corresponding to the target candidate push information; and the second determining submodule is configured The target candidate push information that maximizes the second expected value is determined as the target push information.

In some embodiments, the apparatus further includes: a second determining unit configured to determine a presentation order indicated by the presentation order identifier carried by the target push information, and determine the determined presentation order as the target presentation order; An adding unit configured to determine a presentation order indicated by a presentation order identifier carried by each of the plurality of product information, and to display the product information for each product information whose presentation order is not less than the target presentation order The order is increased by a first preset value;

In some embodiments, the pushing unit includes: a sorting module configured to sort the plurality of product information and the target product information in an order of presentation from small to large; and the second generating module is configured to generate the content including the sorted a web page of product information and target product information; a sending module configured to send a web page to the client.

In some embodiments, the apparatus further includes: a third determining unit configured to determine a category name of the category of the product information in which the presentation order is less than the preset presentation order threshold, and determine the target push The name of the category of the category to which the information belongs; the matching unit is configured to match the determined category name of the category of each product information with the category name of the category to which the target push information belongs; the fourth determining unit is configured Determining the number of product information whose category name matches the category name of the category to which the target push information belongs, and determining the total number of the plurality of product information; and the second adding unit configured to respond to the determined quantity and the total quantity The ratio is less than the preset ratio, and the presentation order of the target push information is increased by a second preset value.

In some embodiments, each of the plurality of product information includes a product name, the target product information includes a target product name; and the apparatus further includes: a fifth determining unit configured to display the plurality of product information Determining a similarity between the product name in the product information and the target product name in the target push information; and determining that the similarity is less than the preset similarity threshold, The product information is determined as the difference product information; the third adding unit is configured to increase the display order of the target push information by a third preset value in response to determining that the quantity of the difference product information is greater than the preset number threshold.

In some embodiments, the apparatus further includes: a second extracting unit configured to extract first feature information from the preset first training sample, wherein the first training sample includes instructions for indicating corresponding to the first training sample The ordering identifier of the ordering situation; the first training unit is configured to utilize the machine learning algorithm to train the order rate prediction model based on the first feature information and the order identification.

In some embodiments, the apparatus further includes: a third extracting unit configured to extract second feature information from the preset second training sample, wherein the second training sample includes an indication to correspond to the second training sample The click identification of the click condition; the second training unit is configured to utilize the machine learning algorithm to train the click rate prediction model based on the second feature information and the click identifier.

The information pushing method and apparatus provided by the embodiment of the present application, by extracting a plurality of product information and a plurality of candidate push information that match the received search term, and then determining each product information and each candidate push information based on the order rate prediction model. Corresponding order rate, and determining a click rate corresponding to each candidate push information based on the click rate prediction model, and then determining target push information based on the obtained order rate and click rate, and finally pushing the target push information to the client, thereby realizing richness Targeted information push.

DRAWINGS

Other features, objects, and advantages of the present application will become more apparent from the detailed description of the accompanying drawings.

1 is an exemplary system architecture diagram to which the present application can be applied;

2 is a flow chart of one embodiment of an information push method according to the present application;

3 is a schematic diagram of an application scenario of an information pushing method according to the present application;

4 is a flow chart of still another embodiment of an information push method according to the present application;

FIG. 5 is a schematic structural diagram of an embodiment of an information pushing apparatus according to the present application; FIG.

6 is a block diagram showing the structure of a computer system suitable for implementing the server of the embodiment of the present application.

detailed description

The present application will be further described in detail below with reference to the accompanying drawings and embodiments. It is understood that the specific embodiments described herein are merely illustrative of the invention, rather than the invention. It is also to be noted that, for the convenience of description, only the parts related to the related invention are shown in the drawings.

It should be noted that the embodiments in the present application and the features in the embodiments may be combined with each other without conflict. The present application will be described in detail below with reference to the accompanying drawings.

FIG. 1 illustrates an exemplary system architecture 100 in which an information push method or information push device of the present application may be applied.

As shown in FIG. 1, system architecture 100 can include

terminal devices

101, 102, 103, network 104, and server 105. The network 104 is used to provide a medium for communication links between the

terminal devices

101, 102, 103 and the server 105. Network 104 may include various types of connections, such as wired, wireless communication links, fiber optic cables, and the like.

The user can interact with the server 105 over the network 104 using the

terminal devices

101, 102, 103 to receive or transmit messages and the like. Various communication client applications, such as a shopping application, a web browser application, a search application, an instant communication tool, a mailbox client, a social platform software, and the like, may be installed on the

terminal devices

101, 102, and 103.

The

terminal devices

101, 102, 103 may be various electronic devices having a display screen and supporting web browsing, including but not limited to smartphones, tablets, laptop portable computers, desktop computers, and the like.

The server 105 may be a server that provides various services, such as a background management server that provides support to the shopping websites that the user browses with the

terminal devices

101, 102, and 103. The background management server may analyze and process data such as the received product information query request, and feed back the processing result (for example, target push information and product information) to the terminal device.

It should be noted that the webpage generating method provided by the embodiment of the present application is generally executed by the server 105. Accordingly, the webpage generating apparatus is generally disposed in the server 105.

It should be understood that the number of terminal devices, networks, and servers in Figure 1 is merely illustrative. Depending on the implementation needs, there can be any number of terminal devices, networks, and servers.

With continued reference to FIG. 2, a flow 200 of one embodiment of an information push method in accordance with the present application is illustrated. The information pushing method includes the following steps:

Step 201: Receive a product information query request that is sent by a client and includes a search term.

In this embodiment, the electronic device (for example, the server shown in FIG. 1) on which the information pushing method runs can use the wired connection method or the wireless connection manner to use the terminal for querying the product information from the user (for example, as shown in FIG. 1 ). The

terminal device

101, 102, 103) receives the product information query request, wherein the product information query request may include a search term. It should be noted that the above wireless connection manner may include but is not limited to 3G/4G connection, WiFi connection, Bluetooth connection, WiMAX connection, Zigbee connection, UWB (ultra wideband) connection, and other wireless connection methods that are now known or developed in the future. .

Generally, the user can edit the search term in the interface presented by the shopping application installed by the client, and click the search button in the interface to send a product information query request. In addition, the user can browse the shopping website by using a browser, edit the search term at the product search location of the webpage, and click the search button of the webpage page to send a product information query request. The above webpage may include html format, xhtml format, asp format, php format, jsp format, shtml format, nsp format, xml format webpage or other webpages to be developed in the future (as long as the webpage file of this format can be used by the browser) Open and browse the forms it contains).

Step 202: Extract a plurality of product information and a plurality of candidate push information that match the search term.

In this embodiment, a large amount of product information and candidate push information may be stored in the memory of the electronic device itself. At this time, the electronic device may directly extract a plurality of product information and multiple matches that match the search term locally. Candidate push information. In addition, the product information may also be stored in a remote server connected to the electronic device, and the electronic device may extract a plurality of product information matching the search term from the remote server. Further, the candidate push information may be stored in another server (for example, an advertisement server) connected to the electronic device, and the electronic device may extract a plurality of candidate push information matching the search term from the other server. It should be noted that the product information of each product may include various information related to the product, for example, may include but not limited to product name, product image, product introduction, product link, product price, product monthly sales, and the like. The candidate push information may include various information related to the candidate push product, and may include, but is not limited to, a candidate push product name, a candidate push product picture, a candidate push product profile, a candidate push product link, a candidate push product price, a candidate push product, and a candidate push product. Monthly sales, etc. The electronic device may extract a plurality of product information and a plurality of candidate push information that match the search term in various manners.

In some optional implementation manners of the embodiment, the electronic device may first retrieve a product name in the product information that includes the search term; and then determine product information including the retrieved product name as the search. The word matches the product information and extracts the determined product information.

In some optional implementation manners of the embodiment, the electronic device may first search for a product name in the product information that includes the search term or the search term (for example, an English translation, an abbreviation of the search term, etc.). Thereafter, the product information including the retrieved product name is determined as the product information matching the above search term, and the determined product information is extracted.

In some optional implementations of this embodiment, the product information of each product may match a preset keyword set. The above keyword set may include at least one keyword set in advance for describing the product. In practice, for each product, a set of keywords that match the product information of the product may be set by the user who sold the product and uploaded to the electronic device. After receiving the search term, the electronic device may match the search term with a keyword set corresponding to each product information. For each product information, if there is a keyword matching the search term in the keyword set corresponding to the product information, the electronic device may determine that the product information is product information that matches the search term, and extract the Determined product information.

In some optional implementation manners of the embodiment, the electronic device may extract candidate push information in the same manner as the product information is extracted. For example, the electronic device may first search for a candidate push product name of the candidate push information including the search term or the synonyms of the search term; and then, the candidate push information including the retrieved candidate push product name is determined as The above search words match the candidate push information, and extract the determined candidate push information. It should be noted that the manner of extracting the product information and the candidate push information may include, but is not limited to, the above enumeration, and details are not described herein again.

Step 203: Input each product information and each candidate push information into a pre-trained order rate prediction model, obtain an order rate corresponding to each product information and each candidate push information, and input each candidate push information to the pre-trained The click rate prediction model obtains a click rate corresponding to each candidate push information.

In this embodiment, the electronic device may input each product information and each candidate push information into a pre-trained order rate prediction model, and obtain an order rate corresponding to each product information and each candidate push information, and each candidate The push information is input to the pre-trained click rate prediction model to obtain a click rate corresponding to each candidate push information. The foregoing order rate prediction model may be used to represent the correspondence between the product information or the candidate push information and the order rate. The click rate prediction model may be used to represent the correspondence between the candidate push information and the click rate. The above-described order rate prediction model and the above-described click rate prediction model can be established in various ways. As an example, a deep learning method may be adopted, which is established based on a deep neural network, may also be established based on a deep neural network and a convolutional neural network, and may also be jointly established based on a deep neural network, a convolutional neural network, and a cyclic neural network. In practice, the order rate can refer to the click-through rate of the information, that is, the ratio of the actual number of clicks of the information to the amount of information displayed, which can be used to measure the effect of the information. The click rate can refer to the ratio of the number of times a content on a website page is clicked to the number of times it is displayed, and can be used to characterize the degree of attention of the content.

In some optional implementation manners of the embodiment, before executing step 201, the electronic device may pre-establish a ticket rate prediction model based on a learning ordering algorithm (for example, a Pairwise algorithm): first, the first training may be preset. The first feature information is extracted from the sample, where the first training sample includes a single order identifier for indicating an order situation corresponding to the first training sample, and the order identifier may be any character (number, letter, symbol) And so on) the composed string. It should be noted that the first training sample may be a sample selected based on a preset condition. As an example, a plurality of product information is displayed in a certain history page, and if one of the two adjacent product information is placed, if the product corresponding to one of the product information is placed, and the product corresponding to the other product information is not When placed, you can treat the two product information as a first training sample. Thereafter, the feature information in the first training sample may be extracted based on the depth neural network, the convolutional neural network, and the cyclic neural network, and the extracted feature information may be determined as the first feature information; and finally, the machine learning algorithm may be utilized, based on the foregoing The first feature information and the above-mentioned order identification are trained to obtain the above-mentioned order rate prediction model.

In some optional implementation manners of the embodiment, before performing step 201, the electronic device may pre-establish a click-rate prediction model based on a learning and sorting algorithm (for example, a Pairwise algorithm): first, a preset second training sample may be The first feature information is extracted, wherein the second training sample includes a click identifier for indicating a click situation corresponding to the second training sample, and the click identifier may be any character (number, letter, symbol, etc.) String. It should be noted that the second training sample may be a sample selected based on a preset condition. As an example, a plurality of product information is displayed in a certain history page, and if one of the two adjacent product information is clicked and another product information is not clicked, the two can be The product information is considered as a second training sample. Thereafter, the feature information in the second training sample may be extracted based on the deep neural network, the convolutional neural network, and the cyclic neural network, and the extracted feature information may be determined as the second feature information; finally, the machine learning algorithm may be utilized, based on the foregoing The second feature information and the click identifier are trained to obtain the above-mentioned click rate prediction model.

Step 204: Determine target push information in the plurality of candidate push information based on the obtained order rate and click rate.

In this embodiment, the electronic device may determine the multiple candidate pushes by using various manners based on the obtained order rate corresponding to each product information and each candidate push information, and a click rate corresponding to each candidate push information. Target candidate push information in the message.

In some optional implementation manners of the embodiment, for each of the plurality of product information, the electronic device may first determine a billing rate corresponding to the product information as a target order rate; thereafter, Among the plurality of candidate push information, the candidate push information having a higher order rate than the target order rate is determined as the target candidate push information.

In some optional implementation manners of the embodiment, for each of the plurality of product information, the electronic device may first determine a billing rate corresponding to the product information as a target order rate; thereafter, And extracting candidate push information that is higher than the target order rate in the plurality of candidate push information; and finally, candidate push information in the extracted candidate push information that has a click rate greater than a preset click rate preset Determine to push information as a target candidate.

In some optional implementation manners of the embodiment, for each of the plurality of product information, the electronic device may first use the order rate corresponding to the product information as the order rate threshold, and select the multiple Among the candidate push information, the candidate push information having a higher order rate than the above-mentioned order rate threshold is generated, and a candidate push information set matching the product information is generated. Then, for each of the generated candidate push information sets, a first expected value of each candidate push information in the candidate push information set may be determined, where the first expected value of each candidate push information may be corresponding to the candidate push information. The click rate and the preset product of the billing value corresponding to the candidate push information. In practice, the charging value corresponding to each candidate push information may be the transaction price of the bid of the candidate push information. Finally, the electronic device may determine target push information in the plurality of candidate push information in various manners based on the obtained first expected value.

In some optional implementations of the embodiment, the preset electronic device may be pre-stored with a preset first expected value, and based on the obtained first expected value, determining that the target push information in the multiple candidate push information may be The electronic device may determine, as the target push information, the candidate push information in the set of candidate push information that the first expected value is greater than the preset first expected value.

In some optional implementation manners of the embodiment, the preset electronic device may be pre-stored with a preset first expected value corresponding to each product information, and the determining the multiple candidate push information based on the obtained first expected value. The target push information may be performed as follows: for each generated candidate push information set, the electronic device may first determine product information corresponding to the candidate push information set; and thereafter, obtain corresponding to the determined product information. And preset the first expected value; and finally, the candidate push information in the candidate push information set that the first expected value is greater than the preset first expected value is determined as the target push information.

In some optional implementation manners of the embodiment, each of the plurality of product information may be provided with a presentation order identifier for indicating a presentation order of the product information, and a candidate push information set corresponding to the product information. Each of the candidate push information in the pair has the above-described presentation order identifier. In practice, the above presentation order identifiers may be numbers, such as 1, 2, and the like. The determining, according to the obtained first expected value, the target push information in the plurality of candidate push information may be performed as follows: First, candidate push information having the largest first expected value among the candidate push information sets may be used as a target candidate Push information to generate a target candidate push information set. Then, for each target candidate push information in the target candidate push information set, the electronic device may acquire a preset order coefficient corresponding to the presentation order indicated by the presentation order identifier carried by the target candidate push information, And determining a second expected value of the target candidate push information, wherein the second expected value is a product of a first expected value of the target candidate push information and a sequence coefficient corresponding to the target candidate push information. In practice, the order coefficients corresponding to the presentation order indicated by the respective presentation order identifiers may be predetermined and stored by the technician in the electronic device based on a large number of statistical calculations. Finally, the electronic device may determine the target candidate push information having the second expected value as the target push information.

Step 205: Push the target push information and the plurality of product information to the client.

In this embodiment, the electronic device may push the plurality of product information extracted in step 202 and the target push information determined in step 204 to the client.

With continued reference to FIG. 3, FIG. 3 is a schematic diagram 300 of an application scenario of the information pushing method according to the embodiment. In the application scenario of FIG. 3, first, the client 301 sends a product information query request 303 containing a search term to the server 302. Then, after receiving the product information query request 303, the server 302 extracts a plurality of product information 304 and a plurality of candidate push information 305 that match the search term. Thereafter, the server 302 inputs the product information and the candidate push information to the order rate prediction model, and inputs the candidate push information to the click rate prediction model to obtain the order rate 306 and the click rate 307. Then, the server 302 determines the target push information 308 of the plurality of candidate push information 305 based on the order rate 306 and the click rate 307. Finally, the server 302 pushes the target push information 308 and the plurality of product information 304 to the client 301.

The foregoing embodiment of the present application provides a method for extracting a plurality of product information and a plurality of candidate push information that match a received search term, and then determining each product information and each candidate push information corresponding to the next order rate prediction model. Single rate, and based on the click rate prediction model, determine the click rate corresponding to each candidate push information, and then determine the target push information based on the obtained order rate and click rate, and finally push the target push information to the client, achieving a targeted Information push.

With further reference to FIG. 4, a flow 400 of yet another embodiment of an information push method is illustrated. The flow 400 of the information pushing method includes the following steps:

Step 401: Receive a product information query request that is sent by a client and includes a search term.

terminal device

101, 102, 103) receives the product information query request, wherein the product information query request may include a search term.

Step 402: Extract a plurality of product information and a plurality of candidate push information that match the search term.

In this embodiment, a large amount of product information and candidate push information may be stored in the memory of the electronic device itself. At this time, the electronic device may directly extract a plurality of product information and multiple matches that match the search term locally. Candidate push information.

It should be noted that each of the plurality of product information may have a presentation order identifier for indicating a presentation order of the product information, and the presentation order identifier may be a number, for example, 1, 2, or the like.

Step 403: Input each product information and each candidate push information into a pre-trained order rate prediction model, obtain an order rate corresponding to each product information and each candidate push information, and input each candidate push information to the pre-trained The click rate prediction model obtains a click rate corresponding to each candidate push information.

In this embodiment, the electronic device may input each product information and each candidate push information into a pre-trained order rate prediction model, and obtain an order rate corresponding to each product information and each candidate push information, and each candidate The push information is input to the pre-trained click rate prediction model to obtain a click rate corresponding to each candidate push information. The foregoing order rate prediction model may be used to represent the correspondence between the product information or the candidate push information and the order rate. The click rate prediction model may be used to represent the correspondence between the candidate push information and the click rate.

It should be noted that the operations of the foregoing steps 401-403 are substantially the same as the operations of the foregoing steps 201-203, and details are not described herein again.

Step 404: For each product information of the plurality of product information, using a billing rate corresponding to the product information as a billing rate threshold, selecting a candidate of the plurality of candidate push information that the order rate is greater than a billing rate threshold Push information to generate a set of candidate push information that matches the product information.

In this embodiment, for each of the plurality of product information, the electronic device may use an order rate corresponding to the product information as a billing rate threshold, and select one of the plurality of candidate push information. The candidate push information having a single rate greater than the above-mentioned order rate threshold generates a candidate push information set that matches the product information.

It should be noted that, for each of the plurality of product information, each candidate push information in the candidate push information set corresponding to the product information may be the same as the presentation order identifier carried by the product information. Show order identification.

Step 405: Determine, for each generated candidate push information set, a first expected value of each candidate push information in the candidate push information set.

In this embodiment, for each candidate push information set generated in step 404, the electronic device may determine a first expected value of each candidate push information in the candidate push information set, where the first candidate of each candidate push information The expected value may be a product of a click rate corresponding to the candidate push information and a preset billing value corresponding to the candidate push information. In practice, the charging value corresponding to each candidate push information may be the transaction price of the bid of the candidate push information.

Step 406: The candidate push information having the largest first expected value among the candidate push information sets is used as the target candidate push information, and the target candidate push information set is generated.

In this embodiment, the electronic device may generate the target candidate push information set by using the candidate push information having the largest first expected value in each candidate push information set generated in step 405 as the target candidate push information.

Step 407: For each target candidate push information in the target candidate push information set, acquire a preset order coefficient corresponding to the presentation order indicated by the presentation order identifier carried by the target candidate push information, and determine the target. The second expected value of the candidate push information.

In this embodiment, for each target candidate push information in the target candidate push information set generated in step 407, the electronic device may acquire a preset indication, which is indicated by the presentation order identifier of the target candidate push information. The order coefficient corresponding to the order is presented, and a second expected value of the target candidate push information is determined. The second expected value may be a product of a first expected value of the target candidate push information and a sequence coefficient corresponding to the target candidate push information. In practice, the order coefficients corresponding to the presentation order indicated by the respective presentation order identifiers may be predetermined and stored by the technician in the electronic device based on a large number of statistical calculations. Generally, the lower the presentation order, the larger the order coefficient corresponding to the presentation order.

Step 408: Determine target candidate push information with the second expected value as the target push information.

In this embodiment, the electronic device may determine the target candidate push information having the second highest expected value among the target candidate push information sets as the target push information.

Step 409, determining a presentation order indicated by the presentation order identifier carried by the target push information, and determining the determined presentation order as the target presentation order.

In this embodiment, the electronic device may determine a presentation order indicated by the presentation order identifier carried by the target push information, and determine the determined presentation order as the target presentation order.

Step 410: Determine a presentation order indicated by a presentation order identifier carried by each product information in the plurality of product information, and increase a presentation order of the product information to each product information in which the presentation order is not less than the target presentation order. Preset value.

In this embodiment, the electronic device may determine a presentation order indicated by a presentation order identifier carried by each of the plurality of product information, and for each product information whose presentation order is not less than the target presentation order, The presentation order of the product information is increased by a first preset value (for example, 1). As an example, if the target presentation order is 4, the electronic device may add 1 to the presentation order of the product information whose presentation order is not less than 4 (for example, 4, 5, 6, etc.), that is, the product information whose original presentation order is 4. The new presentation order is 5.

In step 411, the plurality of product information and the target product information are sorted in order of presentation order from small to large.

In this embodiment, the electronic device may sort the plurality of product information and the target product information in an order of presentation from small to large.

In some optional implementation manners of the embodiment, before the foregoing multiple pieces of product information and the target product information are sorted in order of presentation order, the electronic device may further perform the following steps: First, it may be determined a category name (for example, "clothing", "electric appliance", etc.) of the category of the product information in which the presentation order is smaller than the preset presentation order threshold, and the category of the category to which the target push information belongs is determined in the plurality of product information. name. Thereafter, the electronic device may match the determined category name of the category to which the respective product information belongs to the category name of the category to which the target push information belongs. Then, the number of product information whose category name matches the category name of the category to which the target push information belongs may be determined, and the total number of the plurality of product information described above is determined. Finally, in response to determining that the ratio of the number to the total number is less than a preset ratio, the electronic device may increase the presentation order of the target push information by a second predetermined value (eg, 1, 2, etc.).

In some optional implementation manners of the embodiment, before the foregoing multiple pieces of product information and the target product information are sorted in order of presentation order, the electronic device may further perform the following steps: In the product information, each of the product information in which the presentation order is smaller than the preset presentation order threshold, the electronic device may determine the similarity between the product name in the product information and the target product name in the target push information. Then, in response to the determined similarity being less than the preset similarity threshold, the product information may be determined as the difference product information. Finally, in response to determining that the number of the difference product information is greater than the preset number threshold, the electronic device may increase the presentation order of the target push information by a third preset value (eg, 1, 2, etc.).

Step 412: Generate a webpage including the sorted product information and the target product information.

In this embodiment, the electronic device may generate a webpage including the sorted plurality of product information and the target product information.

Step 413, sending a webpage to the client.

In this embodiment, the electronic device may send the webpage to the client.

As can be seen from Fig. 4, the flow 400 of the information push method in the present embodiment highlights the step of determining the presentation order of the target push information and the plurality of product information as compared with the embodiment corresponding to Fig. 2. Therefore, the solution described in this embodiment can dynamically determine the presentation position of the target push information, so that the user can find the information of interest more quickly, and realize the information-pushing information while realizing the targeted information push. flexibility.

With reference to FIG. 5, as an implementation of the method shown in the above figures, the present application provides an embodiment of an information pushing apparatus, and the apparatus embodiment corresponds to the method embodiment shown in FIG. Used in a variety of electronic devices.

As shown in FIG. 5, the information pushing apparatus 500 of the present embodiment includes: a receiving unit 501 configured to receive a product information query request including a search term sent by a client; and a first extracting unit 502 configured to extract and The plurality of product information and the plurality of candidate push information matched by the search term; the input unit 503 is configured to input each product information and each candidate push information into a pre-trained order rate prediction model, and obtain the product information and a billing rate corresponding to each candidate push information, and inputting each candidate push information to a pre-trained click rate prediction model to obtain a click rate corresponding to each candidate push information, wherein the order rate prediction model is used to represent product information Or the corresponding relationship between the candidate push information and the order rate, the click rate prediction model is used to represent the correspondence between the candidate push information and the click rate; the first determining unit 504 is configured to use the obtained order rate and the click rate, Determining target push information in the plurality of candidate push information; pushing unit 505 configured to use the above target The push information and the plurality of product information are pushed to the client.

In this embodiment, the receiving unit 501 can receive a product information query request from a terminal (for example, the

terminal device

101, 102, and 103 shown in FIG. 1) that the user uses to perform product information query by using a wired connection manner or a wireless connection manner. The product information query request may include a search term.

In this embodiment, the first extracting unit 502 may extract a plurality of product information and a plurality of candidate push information that match the search term in various manners.

In this embodiment, the input unit 503 may input each product information and each candidate push information into a pre-trained order rate prediction model, and obtain an order rate corresponding to each product information and each candidate push information, and each The candidate push information is input to the pre-trained click rate prediction model to obtain a click rate corresponding to each candidate push information.

In this embodiment, the first determining unit 504 may determine the foregoing by using various manners based on the obtained order rate corresponding to each product information and each candidate push information, and the click rate corresponding to each candidate push information. Target candidate push information in the candidate push information.

In some optional implementation manners of the embodiment, the determining unit 504 may include a first generating module, a first determining module, and a second determining module (not shown). The first generation module may be configured to use, for each product information of the plurality of product information, a billing rate corresponding to the product information as a billing rate threshold, and select one of the plurality of candidate push information. The candidate push information with the order rate greater than the above-mentioned order rate threshold is generated, and a candidate push information set matching the product information is generated. The first determining module may be configured to determine, for each generated candidate push information set, a first expected value of each candidate push information in the candidate push information set, where a first expected value of each candidate push information is The click rate corresponding to the candidate push information is a product of a preset billing value corresponding to the candidate push information. The second determining module may be configured to determine target push information in the plurality of candidate push information based on the obtained first expected value.

In some optional implementation manners of the embodiment, each of the plurality of product information may be provided with a presentation order identifier for indicating a presentation order of the product information, and a candidate push information set corresponding to the product information. Each of the candidate push information in the present may be tagged with the above-described presentation order.

In some optional implementation manners of the embodiment, the foregoing second determining module may include a generating submodule, a first determining submodule, and a second determining submodule (not shown). The generation sub-module may be configured to use the candidate push information having the largest first expected value among the candidate push information sets as the target candidate push information to generate the target candidate push information set. The foregoing first determining sub-module may be configured to: for each target candidate push information in the target candidate push information set, obtain a preset, corresponding to a display order indicated by the display order identifier of the target candidate push information. And a second expected value of the target candidate push information, wherein the second expected value is a product of a first expected value of the target candidate push information and a sequence coefficient corresponding to the target candidate push information. The second determining submodule may be configured to determine the target candidate push information with the second expected value as the target push information.

In some optional implementation manners of the embodiment, the information pushing apparatus 500 may further include a second determining unit and a first adding unit (not shown). The second determining unit may be configured to determine a presentation order indicated by the presentation order identifier carried by the target push information, and determine the determined presentation order as the target presentation order. The first adding unit may be configured to determine a presentation order indicated by a presentation order identifier carried by each of the plurality of product information, and for each product information whose presentation order is not less than the target presentation order, The presentation order of the product information is increased by the first preset value.

In some optional implementation manners of the embodiment, the information pushing apparatus 500 may further include a third determining unit, a matching unit, a fourth determining unit, and a second adding unit (not shown). The third determining unit may be configured to determine a category name of the category of the product information in which the presentation order is less than the preset presentation order threshold, and determine the category of the category to which the target push information belongs. The name of the item. The matching unit may be configured to match the determined category name of the category to which the respective product information belongs to the category name of the category to which the target push information belongs. The fourth determining unit may be configured to determine the number of product information whose category name matches the category name of the category to which the target push information belongs, and determine the total number of the plurality of product information. The second adding unit may be configured to increase the display order of the target pushing information by a second preset value in response to determining that the ratio of the quantity to the total quantity is less than a preset ratio.

In some optional implementations of the embodiment, each of the plurality of product information includes a product name, and the target product information includes a target product name; and the information pushing device 500 further includes a fifth determining unit and The third addition unit (not shown). The fifth determining unit may be configured to determine, for each product information of the plurality of product information that the presentation order is less than the preset presentation order threshold, the product name in the product information and the target in the target push information. The similarity of the product name; determining the product information as the difference product information in response to the determined similarity being less than the preset similarity threshold. The third adding unit may be configured to increase the display order of the target push information by a third preset value in response to determining that the quantity of the difference product information is greater than a preset number threshold.

In this embodiment, the pushing unit 505 can push the plurality of product information and the target push information to the client.

In some optional implementation manners of the embodiment, the pushing unit 505 may include a sorting module, a second generating module, and a sending module (not shown). The sorting module may be configured to sort the plurality of product information and the target product information in an order of presentation from small to large. The second generation module may be configured to generate a webpage including the sorted plurality of product information and the target product information. The sending module may be configured to send the webpage to the client.

In some optional implementation manners of the embodiment, the information pushing apparatus 500 may further include a second extracting unit and a first training unit (not shown). The second extraction unit may be configured to extract the first feature information from the preset first training sample, where the first training sample includes an order for indicating an order corresponding to the first training sample. Single logo. The first training unit may be configured to use the machine learning algorithm to train the foregoing order rate prediction model based on the first feature information and the order identifier.

In some optional implementation manners of the embodiment, the information pushing apparatus 500 may further include a third extracting unit and a second training unit (not shown). The third extraction unit may be configured to extract second feature information from the preset second training sample, where the second training sample includes a click identifier for indicating a click situation corresponding to the second training sample. . The second training unit may be configured to use the machine learning algorithm to train the click rate prediction model based on the second feature information and the click identifier.

The apparatus provided by the above embodiment of the present application extracts a plurality of product information and a plurality of candidate push information that match the search term received by the receiving unit 501 to the first extracting unit 502, and then the input unit 503 predicts the order rate based on the order rate. The model determines the order rate corresponding to each product information and each candidate push information, and determines the click rate corresponding to each candidate push information based on the click rate prediction model, and then the first determining unit 504 determines the target push based on the obtained order rate and the click rate. The information, the final push unit 505 pushes the target push information to the client, and implements targeted information push.

Referring now to Figure 6, a block diagram of a computer system 600 suitable for use in implementing a server of an embodiment of the present application is shown. The server shown in FIG. 6 is merely an example, and should not impose any limitation on the function and scope of use of the embodiments of the present application.

As shown in FIG. 6, computer system 600 includes a central processing unit (CPU) 601 that can be loaded into a program in random access memory (RAM) 603 according to a program stored in read only memory (ROM) 602 or from storage portion 608. And perform various appropriate actions and processes. In the RAM 603, various programs and data required for the operation of the system 600 are also stored. The CPU 601, the ROM 602, and the RAM 603 are connected to each other through a bus 604. An input/output (I/O) interface 605 is also coupled to bus 604.

The following components are connected to the I/O interface 605: an input portion 606 including a keyboard, a mouse, etc.; an output portion 607 including, for example, a cathode ray tube (CRT), a liquid crystal display (LCD), and the like, and a storage portion 608 including a hard disk or the like. And a communication portion 609 including a network interface card such as a LAN card, a modem, or the like. The communication section 609 performs communication processing via a network such as the Internet. Driver 610 is also coupled to I/O interface 605 as needed. A removable medium 611, such as a magnetic disk, an optical disk, a magneto-optical disk, a semiconductor memory or the like, is mounted on the drive 610 as needed so that a computer program read therefrom is installed into the storage portion 608 as needed.

In particular, the processes described above with reference to the flowcharts may be implemented as a computer software program in accordance with an embodiment of the present disclosure. For example, an embodiment of the present disclosure includes a computer program product comprising a computer program embodied on a computer readable medium, the computer program comprising program code for executing the method illustrated in the flowchart. In such an embodiment, the computer program can be downloaded and installed from the network via communication portion 609, and/or installed from removable media 611. When the computer program is executed by the central processing unit (CPU) 601, the above-described functions defined in the method of the present application are performed. It should be noted that the computer readable medium described herein may be a computer readable signal medium or a computer readable storage medium or any combination of the two. The computer readable storage medium can be, for example, but not limited to, an electronic, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any combination of the above. More specific examples of computer readable storage media may include, but are not limited to, electrical connections having one or more wires, portable computer disks, hard disks, random access memory (RAM), read only memory (ROM), erasable Programmable read only memory (EPROM or flash memory), optical fiber, portable compact disk read only memory (CD-ROM), optical storage device, magnetic storage device, or any suitable combination of the foregoing. In the present application, a computer readable storage medium may be any tangible medium that can contain or store a program, which can be used by or in connection with an instruction execution system, apparatus or device. In the present application, a computer readable signal medium may include a data signal that is propagated in the baseband or as part of a carrier, carrying computer readable program code. Such propagated data signals can take a variety of forms including, but not limited to, electromagnetic signals, optical signals, or any suitable combination of the foregoing. The computer readable signal medium can also be any computer readable medium other than a computer readable storage medium, which can transmit, propagate, or transport a program for use by or in connection with the instruction execution system, apparatus, or device. . Program code embodied on a computer readable medium can be transmitted by any suitable 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 in accordance with various embodiments of the present application. In this regard, each block of the flowchart or block diagram can represent a module, a program segment, or a portion of code that includes one or more of the logic functions for implementing the specified. Executable instructions. It should also be noted that in some alternative implementations, the functions noted in the blocks may also occur in a different order than that illustrated in the drawings. For example, two successively represented blocks may in fact be executed substantially in parallel, and they may sometimes be executed in the reverse order, depending upon the functionality involved. It is also noted that each block of the block diagrams and/or flowcharts, and combinations of blocks in the block diagrams and/or flowcharts, can be implemented in a dedicated hardware-based system that performs the specified function or operation. Or it can be implemented by a combination of dedicated hardware and computer instructions.

The units involved in the embodiments of the present application may be implemented by software or by hardware. The described unit may also be provided in the processor, for example, as a processor including a receiving unit, a first extracting unit, an input unit, a first determining unit, and a pushing unit. The names of these units do not constitute a limitation on the unit itself under certain circumstances. For example, the receiving unit may also be described as "a unit that receives a product information inquiry request".

In another aspect, the present application also provides a computer readable medium, which may be included in the apparatus described in the above embodiments, or may be separately present and not incorporated into the apparatus. The computer readable medium carries one or more programs, when the one or more programs are executed by the device, causing the device to: receive a product information query request sent by a client and including a search term; and extract the search term Matching a plurality of product information and a plurality of candidate push information; inputting each product information and each candidate push information into a pre-trained order rate prediction model, and obtaining an order rate corresponding to each product information and each candidate push information, and Inputting each candidate push information into a pre-trained click rate prediction model to obtain a click rate corresponding to each candidate push information; and determining target push information in the plurality of candidate push information based on the obtained order rate and click rate; Pushing the target push information and the plurality of product information to the client.

The above description is only a preferred embodiment of the present application and a description of the principles of the applied technology. It should be understood by those skilled in the art that the scope of the invention referred to in the present application is not limited to the specific combination of the above technical features, and should also be covered by the above technical features or without departing from the above inventive concept. Other technical solutions formed by arbitrarily combining the equivalent features. For example, the above features are combined with the technical features disclosed in the present application, but are not limited to the technical features having similar functions.

Claims

An information pushing method, the method comprising:

Receiving a product information query request sent by the client including the search term;

Extracting a plurality of product information and a plurality of candidate push information that match the search term;

The product information and each candidate push information are input to the pre-trained order rate prediction model, and the order rates corresponding to the respective product information and each candidate push information are obtained, and each candidate push information is input to the pre-trained click rate prediction. a model obtains a click rate corresponding to each candidate push information, wherein the order rate prediction model is used to represent a correspondence relationship between the product information or the candidate push information and the order rate, and the click rate prediction model is used to represent the candidate push Correspondence between information and click-through rate;

Determining target push information in the plurality of candidate push information based on the obtained order rate and click rate;

Pushing the target push information and the plurality of product information to the client.
The information push method according to claim 1, wherein the determining the target push information in the plurality of candidate push information based on the obtained order rate and the click rate comprises:

For each of the plurality of product information, the order rate corresponding to the product information is used as a billing rate threshold, and the order rate of the plurality of candidate push information is greater than the order rate. a candidate push information of the threshold, generating a candidate push information set that matches the product information;

Determining, for each of the generated candidate push information sets, a first expected value of each candidate push information in the candidate push information set, wherein a first expected value of each candidate push information is a click rate and a corresponding click rate corresponding to the candidate push information. a predetermined product of billing values corresponding to the candidate push information;

Determining target push information in the plurality of candidate push information based on the obtained first expected value.
The information push method according to claim 2, wherein each of the plurality of product information has a presentation order identifier for indicating a presentation order of the product information, and candidate push corresponding to the product information Each candidate push information in the set of information carries the presentation order identification.
The information push method according to claim 3, wherein the determining the target push information in the plurality of candidate push information based on the obtained first expected value comprises:

Generating, by using the candidate push information having the largest first expected value among the candidate push information sets as the target candidate push information, generating a target candidate push information set;

For each target candidate push information in the target candidate push information set, acquiring a preset order coefficient corresponding to the presentation order indicated by the presentation order identifier carried by the target candidate push information, and determining the target candidate a second expected value of the push information, wherein the second expected value is a product of a first expected value of the target candidate push information and a sequence coefficient corresponding to the target candidate push information;

The target candidate push information having the second expected value is determined as the target push information.
The information pushing method according to claim 4, wherein the method further comprises:

Determining a presentation order indicated by the presentation order identifier carried by the target push information, and determining the determined presentation order as a target presentation order;

Determining a presentation order indicated by a presentation order identifier carried by each of the plurality of product information, and increasing the presentation order of the product information for each product information whose presentation order is not less than the target presentation order a preset value;
The information pushing method according to claim 5, wherein the pushing the target push information and the plurality of product information to the client comprises:

Sorting the plurality of product information and the target product information in an order of presentation from small to large;

Generating a webpage including the sorted product information and the target product information after sorting;

Sending the web page to the client.
The information push method according to claim 6, wherein the method further comprises: before the sorting the plurality of product information and the target product information in an order of presentation from small to large, the method further comprising:

Determining, in the plurality of product information, a category name of a category of the product information in which the presentation order is smaller than the preset presentation order threshold, and determining a category name of the category to which the target push information belongs;

Matching the determined category name of the category to which each product information belongs to the category name of the category to which the target push information belongs;

Determining the number of product information whose category name matches the category name of the category to which the target push information belongs, and determining the total number of the plurality of product information;

In response to determining that the ratio of the quantity to the total number is less than a preset ratio, the presentation order of the target push information is increased by a second predetermined value.
The information pushing method according to claim 6, wherein each of the plurality of product information includes a product name, and the target product information includes a target product name;

Before the sorting the plurality of product information and the target product information in an order of presentation from small to large, the method further includes:

Determining a similarity between the product name in the product information and the target product name in the target push information for each product information in the plurality of product information that exhibits a sequence less than a preset presentation order threshold; Determining the similarity is less than a preset similarity threshold, and determining the product information as the difference product information;

In response to determining that the number of difference product information is greater than a preset number threshold, the presentation order of the target push information is increased by a third predetermined value.
The information pushing method according to claim 1, wherein before the receiving the information query request sent by the client, the method further comprises:

Extracting, from the preset first training sample, the first feature information, where the first training sample includes an order identifier for indicating an order situation corresponding to the first training sample;

Using the machine learning algorithm, the order rate prediction model is trained based on the first feature information and the order identification.
The information pushing method according to claim 1, wherein before the receiving the information query request sent by the client, the method further comprises:

Extracting second feature information from the preset second training sample, wherein the second training sample includes a click identifier for indicating a click situation corresponding to the second training sample;

Using the machine learning algorithm, the click rate prediction model is trained based on the second feature information and the click identifier.
An information pushing device, characterized in that the device comprises:

a receiving unit, configured to receive a product information query request that is sent by the client and includes a search term;

a first extracting unit configured to extract a plurality of product information and a plurality of candidate push information that match the search term;

The input unit is configured to input each product information and each candidate push information into a pre-trained order rate prediction model, obtain an order rate corresponding to each product information and each candidate push information, and input each candidate push information to The pre-trained click rate prediction model obtains a click rate corresponding to each candidate push information, wherein the order rate prediction model is used to represent a product relationship or a correspondence between the candidate push information and the order rate, the click rate prediction The model is used to characterize the correspondence between the candidate push information and the click rate;

a first determining unit, configured to determine target push information in the plurality of candidate push information based on the obtained order rate and click rate;

a pushing unit configured to push the target push information and the plurality of product information to the client.
The information pushing apparatus according to claim 11, wherein the determining unit comprises:

a first generating module, configured to use, for each product information of the plurality of product information, a billing rate corresponding to the product information as a billing rate threshold, and select one of the plurality of candidate push information The candidate push information with a single rate greater than the order rate threshold is generated, and a candidate push information set matching the product information is generated;

a first determining module, configured to determine, for each generated candidate push information set, a first expected value of each candidate push information in the candidate push information set, where a first expected value of each candidate push information is a product of a click rate corresponding to the candidate push information and a preset billing value corresponding to the candidate push information;

The second determining module is configured to determine target push information in the plurality of candidate push information based on the obtained first expected value.
The information push apparatus according to claim 12, wherein each of said plurality of product information has a presentation order identifier for indicating a presentation order of said product information, and candidate push corresponding to said product information Each candidate push information in the set of information carries the presentation order identification.
The information pushing apparatus according to claim 13, wherein the second determining module comprises:

Generating a sub-module, configured to use candidate push information having the largest first expected value in each candidate push information set as target candidate push information, and generate a target candidate push information set;

a first determining submodule configured to: for each target candidate push information in the target candidate push information set, obtain a preset, corresponding to a presentation order indicated by the presentation order identifier of the target candidate push information a sequence coefficient, and determining a second expected value of the target candidate push information, wherein the second expected value is a product of a first expected value of the target candidate push information and a sequence coefficient corresponding to the target candidate push information;

And a second determining submodule configured to determine the target candidate push information with the second expected value as the target push information.
The information pushing apparatus according to claim 14, wherein the apparatus further comprises:

a second determining unit, configured to determine a presentation order indicated by the presentation order identifier carried by the target push information, and determine the determined presentation order as a target presentation order;

a first adding unit, configured to determine a presentation order indicated by a presentation order identifier carried by each of the plurality of product information, and for each product information whose presentation order is not less than the target presentation order, The order in which the product information is displayed is increased by a first preset value;
The information pushing apparatus according to claim 15, wherein the pushing unit comprises:

a sorting module, configured to sort the plurality of product information and the target product information in an order of presentation from small to large;

a second generating module, configured to generate a webpage including the sorted product information and the target product information;

And a sending module, configured to send the webpage to the client.
The information pushing apparatus according to claim 16, wherein the apparatus further comprises:

a third determining unit, configured to determine a category name of a category of the product information in which the presentation order is smaller than a preset presentation order threshold, and determine a category of the category to which the target push information belongs name;

a matching unit, configured to match the determined category name of the category to which each product information belongs to the category name of the category to which the target push information belongs;

a fourth determining unit, configured to determine a quantity of product information that matches a category name with a category name of the category to which the target push information belongs, and determine a total quantity of the plurality of product information;

And a second adding unit configured to increase the display order of the target push information by a second preset value in response to determining that the ratio of the quantity to the total number is less than a preset ratio.
The information pushing apparatus according to claim 16, wherein each of said plurality of product information includes a product name, and said target product information includes a target product name;

The device also includes:

a fifth determining unit, configured to determine, for each product information of the plurality of product information that the presentation order is less than a preset presentation order threshold, the product name in the product information and the target product in the target push information The similarity of the name; determining the product information as the difference product information in response to the determined similarity being less than the preset similarity threshold;

The third adding unit is configured to increase the display order of the target push information by a third preset value in response to determining that the quantity of the difference product information is greater than the preset number threshold.
A server that includes:

One or more processors;

a storage device for storing one or more programs,

The one or more programs are executed by the one or more processors such that the one or more processors implement the method of any of claims 1-10.
A computer readable storage medium having stored thereon a computer program, characterized in that the program, when executed by a processor, implements the method of any of claims 1-10.