CN112785390A

CN112785390A - Recommendation processing method and device, terminal device and storage medium

Info

Publication number: CN112785390A
Application number: CN202110143085.9A
Authority: CN
Inventors: 张懿; 甘露; 吴伟佳; 洪庚伟; 李羽
Original assignee: Weimin Insurance Agency Co Ltd
Current assignee: Weimin Insurance Agency Co Ltd
Priority date: 2021-02-02
Filing date: 2021-02-02
Publication date: 2021-05-11
Anticipated expiration: 2041-02-02
Also published as: CN112785390B

Abstract

The embodiment of the application discloses a recommendation processing method, a recommendation processing device, terminal equipment and a storage medium, wherein the method comprises the following steps: acquiring fusion characteristic data of each element to be recommended in the element set to be recommended; performing sequencing prediction on each element to be recommended in the element set to be recommended according to the fusion characteristic data of each element to be recommended to obtain a predicted sequencing value of each element to be recommended; performing sequencing optimization processing based on historical characteristic data of each element to be recommended in the element set to be recommended to obtain a reordering factor sequence, wherein the reordering factor sequence comprises a reordering factor corresponding to each element to be recommended; and determining the recommended ordering value of each element to be recommended according to the predicted ordering value of each element to be recommended and the reordering factor of each element to be recommended, so as to determine the recommended ordering sequence of each element to be recommended in the element set to be recommended according to the recommended ordering value. By adopting the embodiment of the application, the recommendation accuracy can be improved, and the applicability is high.

Description

Recommendation processing method and device, terminal device and storage medium

Technical Field

The present application relates to the field of artificial intelligence technologies, and in particular, to a recommendation processing method and apparatus, a terminal device, and a storage medium.

Background

With the continuous development of online shopping platforms, recommendation systems have become irreplaceable important components in e-commerce. The recommendation system can learn the hidden preference information in the user historical behaviors, so that the shopping behaviors of the user are further predicted, customers are helped to select satisfied articles, and the income of an e-commerce platform is promoted to be improved. At present, many recommendation systems recommend articles according to different people, however, even different individuals of the same people may have different preferences, so that the recommendation accuracy is low.

Disclosure of Invention

The embodiment of the application provides a recommendation processing method, a recommendation processing device, a terminal device and a storage medium, which can improve recommendation accuracy and are high in applicability.

The embodiment of the application provides a recommendation processing method, which comprises the following steps:

acquiring fusion characteristic data of each element to be recommended in the element set to be recommended, wherein the fusion characteristic data comprises user characteristic data of a target recommendation user;

performing sequencing prediction on each element to be recommended in the element set to be recommended according to the fusion characteristic data of each element to be recommended to obtain a predicted sequencing value of each element to be recommended;

performing sequencing optimization processing based on historical characteristic data of each element to be recommended in the element set to be recommended to obtain a reordering factor sequence, wherein the reordering factor sequence comprises a reordering factor corresponding to each element to be recommended, and the historical characteristic data comprises user characteristic data of a sample user;

and determining the recommended ordering value of each element to be recommended according to the predicted ordering value of each element to be recommended and the reordering factor of each element to be recommended, so as to determine the recommended ordering sequence of each element to be recommended in the element set to be recommended according to the recommended ordering value.

An embodiment of the present application provides a recommendation processing apparatus, including:

the fusion characteristic data determining module is used for acquiring fusion characteristic data of each element to be recommended in the element set to be recommended, and the fusion characteristic data comprises user characteristic data of a target recommendation user;

the predicted ranking value determining module is used for performing ranking prediction on each element to be recommended in the element set to be recommended according to the fusion characteristic data of each element to be recommended to obtain a predicted ranking value of each element to be recommended;

the reordering factor acquisition module is used for performing ordering optimization processing based on historical characteristic data of each element to be recommended in the element set to be recommended to obtain a reordering factor sequence, wherein the reordering factor sequence comprises a reordering factor corresponding to each element to be recommended, and the historical characteristic data comprises user characteristic data of a sample user;

and the recommendation ranking value determining module is used for determining the recommendation ranking value of each element to be recommended according to the prediction ranking value of each element to be recommended and the reordering factor of each element to be recommended, so as to determine the recommendation ranking sequence of each element to be recommended in the element set to be recommended according to the recommendation ranking value.

The embodiment of the application provides terminal equipment, which comprises a processor and a memory, wherein the processor and the memory are connected with each other. The memory is configured to store a computer program that supports the terminal device to execute the method provided by the first aspect and/or any one of the possible implementation manners of the first aspect, where the computer program includes program instructions, and the processor is configured to call the program instructions to execute the provided method.

Embodiments of the present application provide a computer-readable storage medium storing a computer program, the computer program comprising program instructions, which, when executed by a processor, cause the processor to perform the method provided above.

In the embodiment of the application, fusion characteristic data of each element to be recommended in an element set to be recommended is obtained; performing sequencing prediction on each element to be recommended in the element set to be recommended according to the fusion characteristic data of each element to be recommended to obtain a predicted sequencing value of each element to be recommended; performing sequencing optimization processing based on historical characteristic data of each element to be recommended in the element set to be recommended to obtain a reordering factor sequence, wherein the reordering factor sequence comprises a reordering factor corresponding to each element to be recommended; and determining the recommended ordering value of each element to be recommended according to the predicted ordering value of each element to be recommended and the reordering factor of each element to be recommended, so as to determine the recommended ordering sequence of each element to be recommended in the element set to be recommended according to the recommended ordering value. By adopting the embodiment of the application, the recommendation accuracy can be improved, and the applicability is high.

Drawings

In order to more clearly illustrate the technical solutions in the embodiments of the present application, the drawings needed to be used in the description of the embodiments are briefly introduced below, and it is obvious that the drawings in the following description are only some embodiments of the present application, and it is obvious for those skilled in the art to obtain other drawings based on these drawings without creative efforts.

FIG. 1 is a system architecture diagram according to an embodiment of the present application;

fig. 2 is a schematic flowchart of a recommendation processing method provided in an embodiment of the present application;

FIG. 3 is a schematic diagram of an algorithm flow provided by an embodiment of the present application;

FIG. 4 is a schematic diagram illustrating an offline verification result of a reordering algorithm according to an embodiment of the present application;

FIG. 5 is a schematic diagram of the verification result of the AB test provided in the embodiments of the present application;

fig. 6 is a schematic structural diagram of a recommendation processing apparatus according to an embodiment of the present application;

fig. 7 is another schematic structural diagram of a recommendation processing apparatus according to an embodiment of the present application;

fig. 8 is a schematic structural diagram of a terminal device according to an embodiment of the present application.

Detailed Description

The technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the drawings in the embodiments of the present application, and it is obvious that the described embodiments are only a part of the embodiments of the present application, and not all of the embodiments. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present application.

The recommendation processing method provided by the embodiment of the application can be widely applied to various online recommendation systems. For example, the online recommendation system may be a commodity recommendation system, a service recommendation system, a song recommendation system, a video recommendation system, and the like, and is determined according to an actual application scenario, which is not limited herein. The service recommendation system may be a system for recommending insurance services, and the like, which is not limited herein. For convenience of description, the object to be recommended is referred to as an element to be recommended, or an item to be recommended (item), and the like. The item to be recommended is a broad-sense item, and is a general term of all recommendable objects.

Referring to fig. 1, fig. 1 is a schematic diagram of a system architecture according to an embodiment of the present disclosure. As shown in fig. 1, the server 100a, the server 100b, and the terminal apparatus 101 may be connected to each other via a network. The terminal device shown in fig. 1 may include a mobile phone, a tablet computer, a notebook computer, a palm computer, a Mobile Internet Device (MID), a wearable device (e.g., a smart watch, a smart bracelet, etc.), and the like, which are not limited herein. It should be understood that the server 100a may perform data transmission with the server 100b, and the server 100a and the server 100b may also perform data transmission with the terminal device 101, respectively. For convenience of description, the application programs with various functionalities loaded on the terminal device may be exemplified by the first application. Optionally, the first application may be a video application, a music playing application, a game application, a shopping application, or the like, or the first application may also be an applet, a web page, or the like, which is not limited herein.

It is understood that the server 100a and the server 100b may be a local server of the first application, a remote server (e.g., a cloud server), and the like, and are not limited herein. Among other things, the server 100a may be used to store various types of data (e.g., user characteristic data of a registered user of the first application, element characteristic data of each element to be recommended, etc.) and information related to the first application. The server 100b may be configured to acquire various feature data required for recommending a task from the server 100a, and perform processing based on the respective feature data to generate a recommendation list to be transmitted to the terminal device for display. That is, when the first application in the terminal device sends a data acquisition request to the server 100b, the server 100b may return a data condition (e.g., a recommendation list as in fig. 1) to the first application in the terminal device based on the received data acquisition request to update the interface display condition of the first application in the terminal device.

The method in the embodiment of the application obtains fusion characteristic data of each element to be recommended in the element set to be recommended; performing sequencing prediction on each element to be recommended in the element set to be recommended according to the fusion characteristic data of each element to be recommended to obtain a predicted sequencing value of each element to be recommended; performing sequencing optimization processing based on historical characteristic data of each element to be recommended in the element set to be recommended to obtain a reordering factor sequence, wherein the reordering factor sequence comprises a reordering factor corresponding to each element to be recommended; and determining the recommended ordering value of each element to be recommended according to the predicted ordering value of each element to be recommended and the reordering factor of each element to be recommended, so as to determine the recommended ordering sequence of each element to be recommended in the element set to be recommended according to the recommended ordering value. By adopting the embodiment of the application, the recommendation accuracy can be improved.

The method and the related apparatus provided by the embodiments of the present application will be described in detail with reference to fig. 2 to 7, respectively.

Referring to fig. 2, fig. 2 is a schematic flow chart of a recommendation processing method according to an embodiment of the present application. The method provided by the embodiment of the application can comprise the following steps S201 to S204:

s201, obtaining fusion characteristic data of each element to be recommended in the element set to be recommended.

In some feasible embodiments, by obtaining the user characteristic data of the target recommending user and the element characteristic data of each element to be recommended in the element set to be recommended, fusion characteristic data corresponding to each element to be recommended can be generated according to the user characteristic data of the target recommending user and the element characteristic data of each element to be recommended. The target recommending user is a directional object for the element recommendation/article recommendation. For example, if the item recommendation is directed to the user1, the user1 is the target recommendation user, and if the item recommendation is directed to the user2, the user2 is the target recommendation user. For convenience of description, the following embodiments of the present application are all described by taking a target recommending user as an example. It is to be understood that the target recommending user may be a registered user in the recommending system, wherein the registered user may be a newly registered user, such as a user who has not performed purchasing behavior or made an order record, or may be an old registered user, i.e. a user who has performed purchasing behavior or made an order record, etc., without limitation. Wherein the user characteristic data of the target recommendation user comprises at least one of user gender, user age, user browsing history, user purchasing history and the like. It can be understood that at least two elements to be recommended in the element set to be recommended in the application are elements in the same recommendation task scene. For example, a recommended task scenario may be a special sale page scenario, a payment completion page scenario, a policy list page scenario, a child product page scenario, or the like, without limitation. The element to be recommended may also be referred to as an article to be recommended, and the element feature data of the element to be recommended may be referred to as article feature data of the article to be recommended. It should be understood that the item to be recommended of any item to be recommended includes at least one of an item identification, an item type, an item price, an item manufacturer, and the like.

It can be understood that the feature data of each dimension included in the user feature data and the article feature data can be converted into the feature data in LIBSVM format and then processed. That is, the present application may measure user characteristic data or element characteristic data of each dimension in the LIBSVM manner. For example, for gender in the user profile data, gender "male" may be translated into (gender _ male: 1) such a form that for the browsing history, assuming that the browsed products are item a, item b, and the unviewed products are item c, the browsing history profile may be structured as (browsing history _ item a: 1, browsing history _ item b: 1, browsing history _ item c: 0), without limitation.

In some feasible implementation manners, the element feature data of one element to be recommended in the element set to be recommended is spliced with the user feature data of the target recommending user to obtain the fusion feature data of the one element to be recommended, and further, the fusion feature data corresponding to each element to be recommended in the element set to be recommended can be generated. The splicing mode of the user characteristic data and the element characteristic data in the fused characteristic data can be determined according to an actual application scene, and is not limited herein. For example, assuming that the target recommending user is a user1, the user feature data corresponding to the user1 is feature1, and the element feature data of any element to be recommended a is feature2, the fusion feature data Ra corresponding to the element to be recommended is [ feature1, feature2], or the fusion feature data Ra corresponding to the element to be recommended is [ feature2, feature1], which is not limited herein.

S202, carrying out sequencing prediction on each element to be recommended in the element set to be recommended according to the fusion characteristic data of each element to be recommended to obtain a predicted sequencing value of each element to be recommended.

In some feasible embodiments, each element to be recommended in the element set to be recommended is subjected to sequencing prediction according to the fusion feature data of each element to be recommended, and a predicted sequencing value of each element to be recommended can be obtained. Specifically, the fusion characteristic data of each element to be recommended can be input into the ranking value prediction model, and ranking prediction is performed on each element to be recommended in the element set to be recommended through the ranking value prediction model, so that a predicted ranking value of each element to be recommended is obtained. The ranking value prediction model referred to in this application may be a Conversion Rate (CVR) prediction model, or the ranking value prediction model may also be a Click-Through-Rate (CTR) prediction model, and the like, which is not limited herein. When the ranking value prediction model is the CVR prediction model, the predicted ranking value is the predicted conversion rate, and correspondingly, when the ranking value prediction model is the CTR prediction model, the predicted ranking value is the predicted click rate.

Generally, the conversion rate of a certain item is the ratio of the conversion amount to the click rate of the item, and the click rate of the item is the ratio of the click rate to the display amount of the item. For ease of understanding, the concepts of click-through rate and conversion rate are illustrated by taking an offline brick and mortar store as an example. For example, suppose that in a certain promotion, 10000 leaflets are printed by a merchant and sent out, 10000 people also see the leaflet, and the number of people actually entering the store is 1000. Then, 10000 people who see the display amount and 1000 people who enter the store are the click amount, and therefore the click rate is 1000/1000 × 100% — 10%. Further, assuming that 50 persons purchased the purchase after the 1000 persons entered the store, the conversion rate was 50/1000 × 100% — 5%.

Understandably, if the training sample used for training the initial prediction model is the user characteristic data of the sample user, the element characteristic data of the element to be recommended and the historical actual conversion rate corresponding to the element to be recommended, the ranking value prediction model is a conversion rate prediction model. Correspondingly, if the training sample used for training the initial prediction model is the user characteristic data of the sample user, the element characteristic data of the element to be recommended and the historical actual click rate corresponding to the element to be recommended, the ranking value prediction model is a click rate prediction model. The sample user is an old user in the recommendation system, which may or may not include a target recommendation user, and is determined according to an actual application scenario, which is not limited herein. For example, if the target recommending user is a newly registered user, the sample user may not include the target recommending user, and if the target recommending user is an old user, the sample user may include the target recommending user.

For convenience of understanding, the training process of the conversion rate prediction model is described by taking the ranking value prediction model as the conversion rate prediction model as an example. Before model training, a training sample set is obtained, and the training sample set may include a plurality of training samples. The training sample comprises user characteristic data of a sample user, element characteristic data of an element to be recommended and historical actual conversion rate corresponding to the element to be recommended. When model training is carried out, user characteristic data and element characteristic data in each training sample are input into the initial prediction model, and the prediction conversion rate of each training sample output by the initial prediction model can be obtained. Further, the difference (i.e., loss function) between the predicted conversion rate of each training sample output by the initial prediction model and the historical actual conversion rate included in each training sample is calculated, so that model parameters of the initial prediction model are continuously adjusted according to the difference between the predicted conversion rate of each training sample and the historical actual conversion rate, and the adjusted initial prediction model is determined as the conversion rate prediction model until the adjusted initial prediction model converges. It can be understood that the initial prediction model may be a Logistic Regression (LR) model, an XGBOOST model, a neural network model, or the like, and is determined according to an actual application scenario, which is not limited herein. Specifically, the cross entropy may be adopted as a loss function in model training in the embodiment of the present application, that is, the loss function in the embodiment of the present application satisfies:

wherein, y_iRepresenting the historical actual conversion rate, p, corresponding to the element to be recommended included in any training sample_iAnd the predicted conversion rate corresponding to the element to be recommended output by the conversion rate prediction model is represented. Can clean upIf the sum of the calculated cross entropies is less than or equal to the preset cross entropy, the conversion rate prediction model obtained by training meets the construction requirement, otherwise, the conversion rate prediction model obtained by training does not meet the construction requirement, and the training of the conversion rate prediction model is continued until the conversion rate prediction model meets the requirement.

S203, performing sequencing optimization processing based on the historical characteristic data of each element to be recommended in the element set to be recommended to obtain a reordering factor sequence.

In some possible embodiments, the sequence of reordering factors can be obtained by performing sorting optimization processing based on the historical feature data of each element to be recommended in the set of elements to be recommended. Specifically, historical characteristic data of each element to be recommended in the set of elements to be recommended is input into the ranking value prediction model, and a historical ranking value set corresponding to each element to be recommended can be obtained, wherein the historical ranking value set corresponding to any element to be recommended comprises: and the any element to be recommended has N historical sorting values for N sample users, wherein N is an integer greater than 0. Therefore, reordering optimization processing is carried out based on the historical ordering value set corresponding to each element to be recommended, and a reordering factor sequence can be obtained. Understandably, the historical feature data comprises element feature data of an element to be recommended and user feature data of a sample user. That is to say, the historical feature data of an element to be recommended is obtained by splicing the element feature data of the element to be recommended and the user feature data of a sample user. For example, if a sample user is a user2, the user feature data corresponding to the user2 is feature0, and the element feature data of the element to be recommended a is feature2, the historical feature data Ra 'corresponding to the element to be recommended is [ feature0, feature2], or the historical feature data Ra' corresponding to the element to be recommended is [ feature2, feature0], which is not limited herein. And the data format of the fusion characteristic data and the data format of the historical characteristic data are kept consistent. That is, if the order of splicing the user feature data and the element feature data is defined as [ user feature data, element feature data ], the data format of the historical feature data is [ user feature data of a sample user, element feature data of an element to be recommended ], and the data format of the fused feature data is [ user feature data of a target recommendation user, element feature data of an element to be recommended ]. If the splicing sequence of the specified user characteristic data and the specified element characteristic data is [ element characteristic data, user characteristic data ], the data format of the historical characteristic data is [ element characteristic data of an element to be recommended, user characteristic data of a sample user ], and the data format of the fused characteristic data is [ element characteristic data of an element to be recommended, user characteristic data of a target recommendation user ].

It is understood that, if there are N sample users in the present application, where N is an integer greater than 0 (i.e., there is at least one sample user in the present application), then N historical feature data corresponding to the element to be recommended (e.g., element to be recommended item a) may be obtained by concatenating the user feature data of each sample user of the N sample users with the element feature data of the element to be recommended (e.g., element to be recommended item a). Therefore, N historical characteristic data corresponding to the element to be recommended are input into the ranking value prediction model obtained through pre-training, and N historical ranking values corresponding to the element to be recommended can be obtained through the ranking value prediction model. Furthermore, for each element to be recommended in the set of elements to be recommended, N pieces of historical feature data corresponding to each element to be recommended, that is, a historical ranking value set corresponding to each element to be recommended, may be obtained.

In some possible embodiments, the reordering optimization process is performed based on the historical ordering value set corresponding to each element to be recommended, and the obtaining of the reordering factor sequence may be: determining an equivalent ranking value H (i) of each element to be recommended according to the historical ranking value set of each element to be recommended, obtaining the total exposure number I (i) of each element to be recommended to N sample users, and obtaining the value P (i) of each element to be recommended. And determining the total conversion value Re v corresponding to the element set to be recommended according to the equivalent ranking value H (i) of each element to be recommended, the total exposure number I (i) of each element to be recommended to N sample users and the price P (i) of each element to be recommended. Further, a sequence of reordering factors is determined based on the total conversion value Rev. Wherein the total conversion value Rev satisfies:

wherein, the I (i) represents the total exposure number of each element i to be recommended to N sample users, H (i) represents the equivalent ranking value of each element i to be recommended, and P (i) represents the price of each element i to be recommended.

Understandably, when determining the total exposure number I (I) of each element to be recommended to N sample users, the exposure number I (I) of each element to be recommended to each sample user u can be obtained_u(i) And further determining the total exposure number I (i) of each element to be recommended to all sample users according to the sum of the exposure numbers of each element to be recommended to all sample users. When the ranking number of any element to be recommended in any sample user is smaller than or equal to M, the exposure number of any element to be recommended in any sample user is equal to 1, and when the ranking number of any element to be recommended in any sample user is larger than M, the exposure number of any element to be recommended in any sample user is equal to 0, wherein M is determined according to the maximum exposure number. That is, the total exposure number I (I) of any element I to be recommended to N sample users can be determined according to the independent exposure function I of each sample user_u(i) Are summed, i.e.

Wherein, I_u(i) Satisfies the following conditions:

wherein, rank (CR)_u(i) For a certain sample user u), the value range of the sorting result of the historical sorting value of a certain element i to be recommended in the historical sorting values corresponding to all elements to be recommended is

natural numbers

1,2 and …. M represents the maximum number of exposures for the current scene. Therefore, the exposure function can be understood as that when the sorting result of the element i to be recommended is less than or equal to M, the element i is considered to be exposed out,i.e. I_u(i) If the sorting result of the element I to be recommended is larger than M, I is equal to 1, otherwise_u(i) 0. Wherein CR_u(i) The value of (c) can be derived from the ranking value prediction model described above.

For example, suppose there are 50 elements to be recommended in a shopping mall on a certain line, where M is equal to 30, taking the sample user as user1 as an example. For any element to be recommended in the 50 elements to be recommended (for example, any element to be recommended is taken as item a for explanation), by splicing the user feature data feature1 of the user1 with the element feature data feature2 of the element to be recommended, the fusion feature data Ra corresponding to the item a [ feature1, feature2] can be obtained]. Therefore, for 50 elements to be recommended, 50 fused feature data corresponding to the 50 elements to be recommended can be obtained. It can be understood that 50 historical ranking values corresponding to the 50 elements to be recommended can be obtained by inputting the 50 fused feature data corresponding to the 50 elements to be recommended into the predictive ranking value model respectively. Wherein, assuming that the 50 historical sorting values are sorted from big to small, and then the historical sorting value of item a is determined to be at the 20 th position (i.e. the sorting number of item a is 20), since 20 < M, for the user1, the exposure number I of item a is_user1(item a) ═ 1. By analogy, the exposure number of item a to each of the N sample users can be obtained. For example, taking N-3 as an example, it is assumed that the 3 sample users include, in addition to the user1, a user2 and a user 3, where the exposure number I of item a to the sample user2_user2(item a) ═ 0, exposure number I of item a to sample user 3_user2(item a) ═ 1, then for item a, the total exposure number of item a to the 3 sample users is equal to 2, I (item a) ═ I_user1(item a)+I_user2(item a)+I_user3(item a) ═ 2. By analogy, for each element to be recommended included in the element set to be recommended, the total exposure number of each element to be recommended to the N sample users can be obtained.

Understandably, when the equivalent ranking value h (i) of each element to be recommended is determined according to the historical ranking value set of each element to be recommended, the association information set of each element to be recommended can be obtained, wherein the association information set of any element to be recommended comprises: and the association information between any element to be recommended and each sample user in the N sample users. Obtaining an initial ranking factor sequence, and determining an equivalent ranking value H (i) of each element to be recommended according to the initial ranking factor of each element to be recommended in the initial ranking factor sequence, the historical ranking value set corresponding to each element to be recommended, and the association information set of each element to be recommended.

Specifically, the association information between the any element to be recommended and each sample user of the N sample users includes the number of exposures of the any element to be recommended to each sample user. Therefore, the equivalent ranking value h (i) of each element to be recommended satisfies:

wherein γ (i) represents an initial reordering factor, CR, corresponding to each element i to be recommended_u(i) Representing the historical ranking value of each element I to be recommended to each sample user, I_u(i) Represents the number of exposures, ρ (rank (γ (i) × CR), of each element to be recommended to each sample user_u(i) ) is used for representing the conversion deviation coefficient of each element i to be recommended. Wherein the conversion deviation coefficient of each element to be recommended in each sample user comprises: and determining a coefficient from at least one candidate conversion deviation coefficient corresponding to each element to be recommended according to the sorting sequence number of each element to be recommended in the element set to be recommended, wherein the sorting sequence number of any element to be recommended is determined according to the product of the initial reordering factor corresponding to any element to be recommended and the historical sorting value.

That is, the deviation coefficient ρ (rank (γ (i) × CR) is converted in the present application_u(i) ) is determined from at least one candidate conversion deviation coefficient corresponding to the element i to be recommended according to the product of the initial reordering factor and the historical ordering value. Wherein one exposure position corresponds to one candidate conversion bias coefficient. At least one candidate as described aboveAnd the conversion deviation coefficient is determined according to at least one historical sorting value corresponding to the element i to be recommended at least one exposure position. Specifically, the conversion deviation coefficient corresponding to the element i to be recommended at any exposure position in the at least one exposure position is the ratio of the historical sorting value corresponding to the element i to be recommended at any exposure position to the historical sorting value corresponding to the first exposure position in the at least one exposure position.

For example, assuming that the historical conversion rate of item a at the first exposure position of the user1 is 0.7, the historical conversion rate at the second exposure position of the user1 is 0.6, and the historical conversion rate at the third exposure position of the user1 is 0.5, then according to the fact that the conversion deviation coefficient corresponding to the element i to be recommended at any exposure position is the ratio of the historical sorting value corresponding to the element i to be recommended at any exposure position to the historical sorting value corresponding to the first exposure position of at least one exposure position, the conversion deviation coefficient of item a at the first exposure position is 0.7/0.7-1, the conversion deviation coefficient at the second exposure position is 0.6/0.7-6/7, and the conversion deviation coefficient at the third exposure position is 0.5/0.7-5/7.

Understandably, by will

Is substituted into

In (b), the total conversion value Rev can be obtained to satisfy:

the formula of the total conversion value Rev shows that the formula only contains one unknown number gamma (i), so that the corresponding relation between the gamma (i) and the Rev can be directly learned. That is, a sequence of reordering factors can be determined based on the total conversion value Rev: for example, it may be detected whether the sum of the average revenue change values of the respective elements to be recommended satisfies the stop condition. And if so, determining a new reordering factor sequence based on the reordering factor sequence updating rule and the initial reordering factor sequence, and performing sequencing optimization processing based on the new reordering factor sequence. The reordering factor updating rule is determined according to the dot division result of the first reordering factor sequence and the second reordering factor sequence, the learning rate alpha and the average income change vector delta R (t). In this application, the first sequence of reordering factors refers to: and the reordering factor sequence obtained by the last iteration corresponding to the current initial reordering factor sequence. The second reordering factor sequence is: and the reordering factor sequence obtained by the last iteration corresponds to the reordering factor sequence obtained by the last iteration.

In other words, the application can obtain a reordering factor sequence which maximizes the above total transformation value Rev based on the reinforcement learning method training of strategy gradient. This is because the embodiment of the present application cannot acquire the optimal reordering factor sequence at one time, but can acquire the corresponding Rev value by continuously trying different reordering factor sequences. This process can be abstracted as a continuous decision process, i.e. updating its own policy based on environmental feedback.

First, the present application may define three variables of the learning process:

and a state S: the present application uses triplets { i (i), h (i), γ (i) } to represent the states of the learning process, and each iteration can be abstracted as a transition between states.

Action a: the present application defines the change of γ as the motion, and the motion vector at any time t may be represented as a (t) { Δ γ (0, t),. > Δ γ (N, t) }, where N denotes the number of the element to be recommended, and Δ γ (N, t) denotes the change value of the reordering factor of item N at time t. According to the above definition, the reordering factor vector γ (t +1) at time t +1 can be expressed as: γ (t +1) ═ γ (t) + a (t).

Reward R: given that the optimization goal of the present application is to optimize the total conversion value (i.e., the total conversion value), then naturally, the present application defines the current total conversion value as the reward R. Considering that the conversion value of each user is equally important to the application at any moment, the application does not reduce the right of R of different users at the same moment t during calculation. According to the above definition, the reward function can be calculated by the following formula:

according to the definition, the learning algorithm based on the strategy gradient is designed. In the context of the present application, the present application uses the average revenue change as the present application update direction vector, i.e.:

ΔR(t)＝[ΔR(0,t),...,ΔR(N,t)]

where Δ R (k, t) represents the average revenue rise of the element k to be recommended in the time t, and Δ R (k, t) ═ R (k, t) - (k, t-1). At the same time, the application also decides to update the gradient by using the difference of gamma of the two iterations. Based on the above derivation, the final update formula of the present application is as follows:

wherein the content of the first and second substances,

a dot division of the gamma vector between two iterations is represented and alpha represents the learning rate. By moving γ (t) to the left of the formula, the application can obtain the update action of each iteration. Wherein the stop condition satisfies:

where ε is a predefined hyperparameter.

That is to say, the application can obtain the reordering factor sequence which maximizes the above total transformation value Rev based on the reinforcement learning method training of strategy gradient. The reordering factor sequence comprises a reordering factor corresponding to each element to be recommended.

S204, determining the recommended ranking value of each element to be recommended according to the predicted ranking value of each element to be recommended and the reordering factor of each element to be recommended, and determining the recommended ranking sequence of each element to be recommended in the element set to be recommended according to the recommended ranking value.

In some feasible embodiments, the recommended ranking value of each element to be recommended can be determined according to the predicted ranking value corresponding to each element to be recommended and the reordering factor corresponding to each element to be recommended, and then the recommended ranking order of each element to be recommended in the element set to be recommended can be determined according to each recommended ranking value. Specifically, in the embodiment of the present application, a product of a predicted ranking value corresponding to any element to be recommended and a reordering factor corresponding to the element to be recommended may be determined as a recommended ranking value of the element to be recommended, and so on, a recommended ranking value of each element to be recommended in an element set to be recommended may be obtained.

Referring to fig. 3, fig. 3 is a schematic diagram of an algorithm flow provided in the embodiment of the present application. As shown in fig. 3, a predicted ranking value determined based on the ranking value prediction model can be obtained by inputting the fusion feature data of each element to be recommended in the set of elements to be recommended into the ranking value prediction model, and a historical ranking value set corresponding to each element to be recommended can be obtained by inputting the historical feature data of each element to be recommended in the set of elements to be recommended into the ranking value prediction model, so that a reordering factor sequence can be obtained by performing reordering optimization processing based on the historical ranking value set corresponding to each element to be recommended. And finally, multiplying the predicted ranking value of each element to be recommended by the reordering factor to obtain the recommended ranking value of each element to be recommended.

For example, taking the recommendation value prediction model as the conversion rate prediction model as an example, assuming that the target recommendation user is user1, the element set to be recommended includes an element to be recommended item a, an element to be recommended item b, and an element to be recommended item c. The method comprises the steps of fusing element feature data of an element to be recommended, and element feature data of an element to be recommended, and user feature data of a target recommending user1 to obtain fused feature data 1 corresponding to the element to be recommended, fused feature data 2 corresponding to the element to be recommended, and fused feature data 3 corresponding to the element to be recommended, respectively. And respectively inputting the fusion characteristic data 1, the fusion characteristic data 2 and the fusion characteristic data 3 into a conversion rate prediction model, so that the prediction conversion rate of the element to be recommended item a is 0.7, the prediction conversion rate of the element to be recommended item b is 0.65, and the prediction conversion rate of the element to be recommended item c is 0.32. Assuming that the reordering factor corresponding to item a in the obtained reordering factor sequence is equal to 0.8, the reordering factor corresponding to item b is equal to 0.9, and the reordering factor corresponding to item a is equal to 0.5, the recommended ordering value of item a to be recommended is 0.7 × 0.8 — 0.56, the recommended ordering value of item b is 0.65 × 0.9 — 0.585, and the recommended ordering value of item c is 0.32 × 0.5 — 0.16.

In some possible embodiments, the recommendation ranking value may be used to determine a ranking order of each element to be recommended in the set of elements to be recommended. For example, the set of elements to be recommended may be sorted in the order from the largest recommended sorting value to the smallest recommended sorting value of the elements to be recommended, so as to obtain the item arrangement order of the set of elements to be recommended. And then, according to the displayable number n of the display interface of the terminal equipment, displaying the first n elements to be recommended in the article arrangement sequence in the display interface of the terminal equipment. The number m of the items of the element set to be recommended in the recommendation system is larger than or equal to the displayable number n of the display interface. For example, assuming that m is 100 and n is 10, the elements to be recommended, which are 10 before the item arrangement order, of the 100 elements to be recommended may be sequentially displayed in the display interface of the recommendation system of the terminal device.

For example, assume that the set of elements to be recommended includes 3 elements to be recommended, item a, item b, and item c. And if the recommended ordering value of the item a is 0.56, the recommended ordering value of the item b is 0.585 and the recommended ordering value of the item c is 0.16, the order of the elements to be recommended can be determined as item b, item a and item c according to the descending order of the recommended ordering values. Therefore, the display effect of the elements to be recommended in the display interface of the terminal device is that item b, item a and item c are sequentially displayed from top to bottom.

In the embodiment of the application, fusion feature data corresponding to each element to be recommended is generated according to the user feature data of the target recommending user and the element feature data of each element to be recommended in the element set to be recommended. And determining a predicted ranking value corresponding to each element to be recommended according to the ranking value prediction model and the fusion characteristic data corresponding to each element to be recommended. And acquiring a reordering factor sequence, and determining a recommended ordering value of each element to be recommended according to the predicted ordering value corresponding to each element to be recommended and the reordering factor corresponding to each element to be recommended in the reordering factor sequence, wherein the recommended ordering value is used for determining the ordering sequence of the element set to be recommended. By adopting the embodiment of the application, the recommendation accuracy can be improved, and the applicability is high.

The recommendation processing method provided by the application can improve the recommendation accuracy of the target recommendation user and help customers to select satisfactory articles. Meanwhile, based on the recommendation processing method provided by the application, the income improvement of the electronic commerce platform can be promoted. For the convenience of understanding, the application verifies the effect of the model by using actual data. The data set extracted in the application is derived from real-world internet insurance data sales data, comprises about 560 pieces of data, and mainly comprises the following data of four recommended task scenes:

special sale page (scenario 1): for the recommended scene of a user with certain cognition, the sample number is 390W.

Payment completion page (scienio 2): the sample number is 18.3W for the recommended scenario of the user who has completed the product purchase.

Policy list page (scienio 3): the recommended scenario of the user has not been purchased yet, sample number 41W.

Daughter product page (scienio 4): sample number 115W for the product recommendation scenario prepared for children/children.

In the method, a multi-task learning model (SAMN) is used as a recommended value ordering model, an LR model is used as a base line, and offline verification of a reordering algorithm is performed on data of the four recommended task scenes respectively. Referring to fig. 4, fig. 4 is a schematic diagram illustrating an offline verification result of a reordering algorithm according to an embodiment of the present application. As can be seen from the verification results in fig. 4, we designed the reordering algorithm to achieve revenue improvements of 11.3%, 6.1%, 7.8% and 3% in the four scenarios. The improvement was more significant than baseline, 14.7%, 8.5%, 10.5% and 4.5%, respectively.

Meanwhile, in order to verify the effectiveness of the reordering algorithm, the invention is deployed in the real world and subjected to an AB test for 7 days. Please refer to fig. 5, fig. 5 is a schematic diagram illustrating a verification result of the AB test provided in the embodiment of the present application. As can be seen from the online verification results in fig. 5, the reordering algorithm in the embodiment of the present application (based on SAMN) achieves revenue improvements of 16.2%, 7.2%, 7.15% and 2.7% in four scenarios, respectively, compared to the baseline.

Referring to fig. 6, fig. 6 is a schematic structural diagram of a recommendation processing device according to an embodiment of the present application. The recommendation processing apparatus provided in the embodiment of the present application includes:

the fusion feature data determining module 31 is configured to obtain fusion feature data of each element to be recommended in the element set to be recommended, where the fusion feature data includes user feature data of a target recommending user;

the predicted ranking value determining module 32 is configured to perform ranking prediction on each element to be recommended in the element set to be recommended according to the fusion feature data of each element to be recommended, so as to obtain a predicted ranking value of each element to be recommended;

a reordering factor obtaining module 33, configured to perform ordering optimization processing based on historical feature data of each element to be recommended in the set of elements to be recommended, to obtain a reordering factor sequence, where the reordering factor sequence includes a reordering factor corresponding to each element to be recommended, and the historical feature data includes user feature data of a sample user;

and the recommended ranking value determining module 34 is configured to determine the recommended ranking value of each element to be recommended according to the predicted ranking value of each element to be recommended and the reordering factor of each element to be recommended, so as to determine the recommended ranking order of each element to be recommended in the element set to be recommended according to the recommended ranking value.

Referring to fig. 7, fig. 7 is another schematic structural diagram of a recommendation processing device according to an embodiment of the present application. Wherein:

in some possible implementations, the predicted rank value determination module 32 is configured to:

inputting the fusion characteristic data of each element to be recommended into a ranking value prediction model, and performing ranking prediction on each element to be recommended in the element set to be recommended through the ranking value prediction model to obtain a predicted ranking value of each element to be recommended;

the reordering factor obtaining module 33 comprises:

a historical ranking value set obtaining unit 331, configured to input historical feature data of each element to be recommended in the element set to be recommended into the ranking value prediction model, to obtain a historical ranking value set corresponding to each element to be recommended, where a historical ranking value set corresponding to any element to be recommended includes: n historical sorting values of any element to be recommended to N sample users are obtained, wherein N is an integer larger than 0;

and the optimization processing unit 332 is configured to perform reordering optimization processing based on the historical ordering value set corresponding to each element to be recommended, so as to obtain a reordering factor sequence.

In some possible embodiments, the optimization processing unit 332 includes:

the equivalent sorting value determining subunit 3321 is configured to determine, according to the historical sorting value set of each to-be-recommended element, an equivalent sorting value h (i) of each to-be-recommended element;

the first processing subunit 3322 is configured to obtain a total exposure number i (i) of each element to be recommended to N sample users, and obtain a value p (i) of each element to be recommended;

the total conversion value determining subunit 3323 is configured to determine a total conversion value Rev corresponding to the set of elements to be recommended according to the equivalent ranking value h (i) of each element to be recommended, the total exposure number i (i) of each element to be recommended to the N sample users, and the price p (i) of each element to be recommended;

a reordering factor sequence determining subunit 3324 for determining a reordering factor sequence based on the total conversion value Rev.

In some possible embodiments, the total conversion value Rev satisfies:

In some possible embodiments, the first processing subunit 3322 is specifically configured to:

acquiring the exposure number I of each element to be recommended to each sample user u_u(i)；

Determining the total exposure number I (i) of each element to be recommended to all sample users by the sum of the exposure numbers of each element to be recommended to all sample users;

when the ranking number of any element to be recommended in any sample user is smaller than or equal to M, the exposure number of any element to be recommended in any sample user is equal to 1, and when the ranking number of any element to be recommended in any sample user is larger than M, the exposure number of any element to be recommended in any sample user is equal to 0, wherein M is determined according to the maximum exposure number.

In some possible embodiments, the equivalent rank value determining subunit 3321 is specifically configured to:

acquiring the associated information set of each element to be recommended, wherein the associated information set of any element to be recommended comprises: the correlation information between any element to be recommended and each sample user in the N sample users;

obtaining an initial ranking factor sequence, and determining an equivalent ranking value H (i) of each element to be recommended according to an initial ranking factor of each element to be recommended in the initial ranking factor sequence, a historical ranking value set corresponding to each element to be recommended, and an associated information set of each element to be recommended.

In some possible embodiments, the association information between any element to be recommended and each of the N sample users includes: the exposure number of any element to be recommended to each sample user;

the equivalent ranking value H (i) of each element to be recommended meets the following conditions:

wherein γ (i) represents an initial reordering factor, CR, corresponding to each element i to be recommended_u(i) Representing the historical ranking value of each element I to be recommended to each sample user, I_u(i) Represents the number of exposures of each element to be recommended to each sample user, rho (rank (γ (i) × CR)_u(i) ) represents a transformation deviation coefficient of each element i to be recommended;

the conversion deviation coefficient of each element to be recommended in each sample user comprises: determining a coefficient from at least one candidate conversion deviation coefficient corresponding to each element to be recommended according to the sorting serial number of each element to be recommended in the element set to be recommended; the sorting sequence number of any element to be recommended is determined according to the product of the initial reordering factor corresponding to any element to be recommended and the historical sorting value.

In some possible embodiments, the reordering factor sequence determination subunit 3324 is specifically configured to:

detecting whether the sum of the average income change values of all elements to be recommended meets a stopping condition;

and if so, determining a new reordering factor sequence based on the reordering factor sequence updating rule and the initial reordering factor sequence, and performing sequencing optimization processing based on the new reordering factor sequence.

In some possible embodiments, the reordering factor updating rule is determined according to a dot division result of the first reordering factor sequence and the second reordering factor sequence, a learning rate α, and an average revenue variation vector Δ r (t);

the first reordering factor sequence is: a reordering factor sequence obtained by the last iteration corresponding to the current initial reordering factor sequence;

the second reordering factor sequence is: and the reordering factor sequence obtained by the last iteration corresponds to the reordering factor sequence obtained by the last iteration.

In some possible embodiments, the recommended ranking value determining module 34 is specifically configured to:

and multiplying the predicted ranking value of each element to be recommended by the reordering factor to obtain the recommended ranking value of each element to be recommended.

Based on the same inventive concept, the principle and the advantageous effect of the problem solving of the recommendation processing apparatus provided in the embodiment of the present application are similar to the principle and the advantageous effect of the problem solving of the message processing method in the embodiment of the present application, and reference may be made to the principle and the advantageous effect of the implementation of the method, and further, the description of the related content in the foregoing embodiment may be referred to for the relationship between the steps executed by the related modules, which is not described herein again for brevity.

Please refer to fig. 8, fig. 8 is a schematic structural diagram of a terminal device according to an embodiment of the present application. As shown in fig. 8, the terminal device in this embodiment may include: one or more processors 401 and memory 402. The processor 401 and the memory 402 are connected by a bus 403. The memory 402 is used to store a computer program comprising program instructions, and the processor 401 is used to execute the program instructions stored in the memory 402 to perform the following operations:

acquiring fusion characteristic data of each element to be recommended in an element set to be recommended, wherein the fusion characteristic data comprises user characteristic data of a target recommendation user;

sequencing and predicting each element to be recommended in the element set to be recommended according to the fusion characteristic data of each element to be recommended to obtain a predicted sequencing value of each element to be recommended;

performing sequencing optimization processing on the basis of historical characteristic data of each element to be recommended in the element set to be recommended to obtain a reordering factor sequence, wherein the reordering factor sequence comprises a reordering factor corresponding to each element to be recommended, and the historical characteristic data comprises user characteristic data of a sample user;

In some possible embodiments, the processor 401 is configured to:

the processor 401 is further configured to:

inputting the historical characteristic data of each element to be recommended in the set of elements to be recommended into the ranking value prediction model to obtain a historical ranking value set corresponding to each element to be recommended, wherein the historical ranking value set corresponding to any element to be recommended comprises: n historical sorting values of any element to be recommended to N sample users are obtained, wherein N is an integer larger than 0;

and performing reordering optimization processing based on the historical ordering value set corresponding to each element to be recommended to obtain a reordering factor sequence.

In some possible embodiments, the processor 401 is configured to:

determining an equivalent ranking value H (i) of each element to be recommended according to the historical ranking value set of each element to be recommended;

acquiring the total exposure number I (i) of each element to be recommended to N sample users, and acquiring the value P (i) of each element to be recommended;

determining a total conversion value Rev corresponding to the element set to be recommended according to the equivalent ranking value H (i) of each element to be recommended, the total exposure number I (i) of each element to be recommended to N sample users, and the price P (i) of each element to be recommended;

determining a sequence of reordering factors based on the total conversion value Re v.

In some possible embodiments, the total conversion value Rev satisfies:

In some possible embodiments, the processor 401 is configured to:

In some possible embodiments, the processor 401 is further configured to:

It should be appreciated that in some possible implementations, the processor 401 may be a Central Processing Unit (CPU), and the processor may be other general purpose processors, Digital Signal Processors (DSPs), Application Specific Integrated Circuits (ASICs), Field Programmable Gate Arrays (FPGAs) or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, and the like. A general purpose processor may be a microprocessor or the processor may be any conventional processor or the like. The memory 402 may include both read-only memory and random access memory, and provides instructions and data to the processor 401. A portion of the memory 402 may also include non-volatile random access memory. For example, the memory 402 may also store device type information.

In a specific implementation, the terminal device may execute the implementation manners provided in the steps in fig. 2 to fig. 3 through the built-in functional modules, which may specifically refer to the implementation manners provided in the steps, and are not described herein again.

Based on the same inventive concept, the principle and the beneficial effect of solving the problem of the intelligent device provided in the embodiment of the present application are similar to the principle and the beneficial effect of solving the problem of the message processing method in the embodiment of the present application, and reference may be made to the principle and the beneficial effect of the implementation of the method, and the relationship between the steps executed by the processor 401 may also refer to the description of the related contents in the foregoing embodiment, which is not described herein again for brevity.

An embodiment of the present application further provides a computer-readable storage medium, where a computer program is stored in the computer-readable storage medium, where the computer program includes program instructions, and when the program instructions are executed by a processor, the recommendation processing method provided in each step in fig. 2 to 3 is implemented.

The computer-readable storage medium may be the recommendation processing apparatus provided in any of the foregoing embodiments or an internal storage unit of the terminal device, such as a hard disk or a memory of an electronic device. The computer readable storage medium may also be an external storage device of the electronic device, such as a plug-in hard disk, a Smart Memory Card (SMC), a Secure Digital (SD) card, a flash card (flash card), and the like, which are provided on the electronic device. Further, the computer readable storage medium may also include both an internal storage unit and an external storage device of the electronic device. The computer-readable storage medium is used for storing the computer program and other programs and data required by the electronic device. The computer readable storage medium may also be used to temporarily store data that has been output or is to be output.

Embodiments of the present application further provide a computer program product containing instructions, which when run on a computer, cause the computer to execute the recommendation processing method described in the above method embodiments.

Embodiments of the present application also provide a computer program product or a computer program comprising computer instructions stored in a computer-readable storage medium. The processor of the computer device reads the computer instructions from the computer-readable storage medium, and the processor executes the computer instructions to cause the computer device to execute the method of recommendation processing.

The steps in the method of the embodiment of the application can be sequentially adjusted, combined and deleted according to actual needs.

The modules in the device can be merged, divided and deleted according to actual needs.

The terms "first", "second", "third", "fourth", and the like in the claims and in the description and drawings of the present application are used for distinguishing between different objects and not for describing a particular order. Furthermore, the terms "include" and "have," as well as any variations thereof, are intended to cover non-exclusive inclusions. For example, a process, method, system, article, or apparatus that comprises a list of steps or elements is not limited to only those steps or elements listed, but may alternatively include other steps or elements not listed, or inherent to such process, method, article, or apparatus.

Reference herein to "an embodiment" means that a particular feature, structure, or characteristic described in connection with the embodiment can be included in at least one embodiment of the application. The appearances of the phrase in various places in the specification are not necessarily all referring to the same embodiment, nor are separate or alternative embodiments mutually exclusive of other embodiments. It is explicitly and implicitly understood by one skilled in the art that the embodiments described herein can be combined with other embodiments. The term "and/or" as used in this specification and the appended claims refers to and includes any and all possible combinations of one or more of the associated listed items. Those of ordinary skill in the art will appreciate that the elements and algorithm steps of the examples described in connection with the embodiments disclosed herein may be embodied in electronic hardware, computer software, or combinations of both, and that the components and steps of the examples have been described in a functional general in the foregoing description for the purpose of illustrating clearly the interchangeability of hardware and software. Whether such functionality is implemented as hardware or software depends upon the particular application and design constraints imposed on the implementation. Skilled artisans may implement the described functionality in varying ways for each particular application, but such implementation decisions should not be interpreted as causing a departure from the scope of the present application.

The method and the related apparatus provided by the embodiments of the present application are described with reference to the flowchart and/or the structural diagram of the method provided by the embodiments of the present application, and each flow and/or block of the flowchart and/or the structural diagram of the method, and the combination of the flow and/or block in the flowchart and/or the block diagram can be specifically implemented by computer program instructions. These computer program instructions may be provided to a processor of a general purpose computer, special purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, create means for implementing the functions specified in the flowchart flow or flows and/or block or blocks of the block diagram. These computer program instructions may also be stored in a computer-readable memory that can direct a computer or other programmable data processing apparatus to function in a particular manner, such that the instructions stored in the computer-readable memory produce an article of manufacture including instruction means which implement the function specified in the flowchart flow or flows and/or block or blocks of the block diagram. These computer program instructions may also be loaded onto a computer or other programmable data processing apparatus to cause a series of operational steps to be performed on the computer or other programmable apparatus to produce a computer implemented process such that the instructions which execute on the computer or other programmable apparatus provide steps for implementing the functions specified in the flowchart flow or flows and/or block or blocks.

Those skilled in the art will appreciate that all or part of the steps in the methods of the above embodiments may be implemented by hardware related to instructions of a program, which may be stored in a computer readable storage medium, which may include: flash disks, Read-Only memories (ROMs), Random Access Memories (RAMs), magnetic or optical disks, and the like.

While the invention has been described with reference to a preferred embodiment, it will be understood by those skilled in the art that various changes in form and detail may be made therein without departing from the spirit and scope of the invention as defined by the appended claims.

Claims

1. A recommendation processing method, characterized in that the method comprises:

2. The method of claim 1,

the method for performing sequencing prediction on each element to be recommended in the element set to be recommended according to the fusion characteristic data of each element to be recommended to obtain a predicted sequencing value of each element to be recommended includes:

the method for performing sorting optimization processing based on the historical characteristic data of each element to be recommended in the element set to be recommended to obtain a reordering factor sequence includes:

3. The method of claim 2, wherein performing reordering optimization based on the historical ordering value set corresponding to each element to be recommended to obtain a reordering factor sequence comprises:

determining a sequence of reordering factors based on the total conversion value Rev.

4. The method of claim 3, wherein the total conversion value Rev satisfies:

5. The method of claim 3, wherein the obtaining of the total exposure number I (i) of each element to be recommended to N sample users comprises:

6. The method according to any one of claims 3 to 5, wherein the determining an equivalent ranking value H (i) of each of the elements to be recommended according to the historical ranking value set of each of the elements to be recommended comprises:

7. The method of claim 6, wherein the association information between any element to be recommended and each of the N sample users comprises: the exposure number of any element to be recommended to each sample user;

8. The method of claim 7, wherein said determining a sequence of reordering factors based on said total conversion value, Rev, further comprises:

9. The method of claim 8, wherein the reordering factor update rule is determined according to a dot division of the first sequence of reordering factors and the second sequence of reordering factors, a learning rate α, and an average revenue variation vector Δ r (t);

10. The method of claim 1, wherein determining the recommended ranking value of each of the elements to be recommended according to the predicted ranking value of each of the elements to be recommended and the reordering factor of each of the elements to be recommended comprises:

11. A recommendation processing apparatus, characterized in that the apparatus comprises:

the fusion characteristic data determining module is used for acquiring fusion characteristic data of each element to be recommended in the element set to be recommended, wherein the fusion characteristic data comprises user characteristic data of a target recommendation user;

a reordering factor obtaining module, configured to perform ordering optimization processing based on historical feature data of each element to be recommended in the set of elements to be recommended, to obtain a reordering factor sequence, where the reordering factor sequence includes a reordering factor corresponding to each element to be recommended, and the historical feature data includes user feature data of a sample user;

and the recommendation ranking value determining module is used for determining the recommendation ranking value of each element to be recommended according to the predicted ranking value of each element to be recommended and the reordering factor of each element to be recommended, so as to determine the recommendation ranking sequence of each element to be recommended in the element set to be recommended according to the recommendation ranking value.

12. A terminal device, comprising a processor and a memory, said processor and said memory being interconnected;

the memory for storing a computer program comprising program instructions, the processor being configured to invoke the program instructions to perform the method of any of claims 1-10.

13. A computer-readable storage medium, characterized in that the computer-readable storage medium stores a computer program comprising program instructions that, when executed by a processor, cause the processor to carry out the method according to any one of claims 1-10.