CN112131473B

CN112131473B - Information recommendation method, device, equipment and storage medium

Info

Publication number: CN112131473B
Application number: CN202011016907.9A
Authority: CN
Inventors: 陈品殿
Original assignee: Beijing Dajia Internet Information Technology Co Ltd
Current assignee: Beijing Dajia Internet Information Technology Co Ltd
Priority date: 2020-09-24
Filing date: 2020-09-24
Publication date: 2024-03-26
Anticipated expiration: 2040-09-24
Also published as: CN112131473A

Abstract

The disclosure provides an information recommendation method, device, equipment and storage medium, and belongs to the technical field of Internet. The method comprises the following steps: acquiring first operation data, wherein the first operation data represents the operation of a terminal on first recommendation information, and the first recommendation information is information sent to the terminal at a first time point; acquiring a first interval duration according to the first operation data, wherein the first interval duration represents an interval duration between a time point of next sending of recommended information to the terminal and the first time point; and when the first interval duration after the first time point is reached, sending second recommendation information to the terminal, wherein the terminal is used for displaying the second recommendation information. The method can meet the personalized recommendation requirement of the user.

Description

Information recommendation method, device, equipment and storage medium

Technical Field

The disclosure relates to the technical field of internet, and in particular relates to an information recommendation method, device, equipment and storage medium.

Background

The development of internet technology enables people to acquire various information rapidly and in real time, provides great convenience for life and work of people, and has become a common function for recommending information to users. In the related art, information is generally recommended to the user according to a fixed period, but the recommendation mode cannot meet the personalized requirements of the user.

Disclosure of Invention

The disclosure provides an information recommendation method, device, equipment and storage medium, which can meet personalized recommendation requirements of users. The technical scheme of the present disclosure is as follows:

according to a first aspect of embodiments of the present disclosure, there is provided an information recommendation method, the method including:

acquiring first operation data, wherein the first operation data represents the operation of a terminal on first recommendation information, and the first recommendation information is information sent to the terminal at a first time point;

acquiring a first interval duration according to the first operation data, wherein the first interval duration represents an interval duration between a time point of next sending of recommended information to the terminal and the first time point;

and when the first interval duration after the first time point is reached, sending second recommendation information to the terminal, wherein the terminal is used for displaying the second recommendation information.

In one possible implementation manner, the obtaining a first interval duration according to the first operation data includes:

acquiring a stored second interval duration, wherein the second interval duration is the interval duration between a time point of sending third recommended information to the terminal and the first time point, and the third recommended information is information sent to the terminal before the first recommended information;

And adjusting the second interval duration according to the first operation data to obtain the first interval duration.

In another possible implementation manner, the adjusting the second interval duration according to the first operation data to obtain the first interval duration includes:

carrying out emotion recognition on the first operation data to obtain an emotion type, wherein the emotion type is a positive emotion type or a negative emotion type;

if the emotion type is a forward emotion type, reducing the second interval duration to obtain the first interval duration; or,

and if the emotion type is a negative emotion type, increasing the second interval duration to obtain the first interval duration.

In another possible implementation manner, before the acquiring the stored second interval duration, the method further includes:

acquiring second operation data, wherein the second operation data represents the operation of the terminal on the third recommendation information;

and acquiring the second interval duration according to the second operation data, and storing the second interval duration.

And selecting the second interval duration from the range of reference interval durations, and storing the second interval duration.

counting the first operation data to obtain a first adjustment parameter corresponding to the first operation data;

and adjusting the second interval duration according to the first adjustment parameter to obtain the first interval duration.

In another possible implementation manner, the first adjustment parameter is an adjustment duration, and the counting the first operation data to obtain a first adjustment parameter corresponding to the first operation data includes:

counting the first operation data to obtain an adjustment proportion corresponding to the first operation data;

and adjusting the reference time length according to the adjustment proportion to obtain the adjustment time length.

In another possible implementation manner, the counting the first operation data to obtain a first adjustment parameter corresponding to the first operation data includes:

acquiring the weight of each operation in the first operation data;

And acquiring the first adjustment parameters according to the weight of each operation.

In another possible implementation manner, the acquiring the weight of each operation in the first operation data includes:

acquiring an emotion recognition model, wherein the emotion recognition model comprises at least one operation weight, and the weight represents the importance degree of the operation on an emotion recognition result;

and reading the weight of each operation from the emotion recognition model.

In another possible implementation manner, the obtaining the first adjustment parameter according to the weight of each operation includes:

acquiring the execution times of each operation according to the first operation data;

and weighting the execution times of each operation according to the weight of each operation to obtain the first adjustment parameter.

In another possible implementation manner, the training process of the emotion recognition model includes:

acquiring a first training sample, wherein the first training sample comprises execution times and sample emotion types, the execution times are operation times of a sample user on recommended first sample information, the sample emotion types represent emotion types of second sample information recommended by the sample user after the first sample information, and the sample emotion types are positive emotion types or negative emotion types;

Invoking the emotion recognition model to perform emotion recognition on the execution times to obtain a predicted emotion type;

training the emotion recognition model according to the sample emotion type and the predicted emotion type.

and determining the sum of the weights of each operation as a first adjustment parameter corresponding to the first operation data.

acquiring a second training sample, wherein the second training sample comprises a sample operation and a sample emotion type, the sample operation is an operation of a sample user on recommended first sample information, the sample emotion type represents an emotion type of the sample user on recommended second sample information after the first sample information, and the sample emotion type is a positive emotion type or a negative emotion type;

invoking the emotion recognition model, and performing emotion recognition on the sample operation to obtain a predicted emotion type;

acquiring the execution time length of each operation according to the first operation data;

and weighting the execution time of each operation according to the weight of each operation to obtain the first adjustment parameter.

acquiring a third training sample, wherein the third training sample comprises an execution time length and a sample emotion type, the execution time length is the operation time length of a sample user on recommended first sample information, the sample emotion type represents the emotion type of the sample user on recommended second sample information after the first sample information, and the sample emotion type is a positive emotion type or a negative emotion type;

invoking the emotion recognition model to perform emotion recognition on the execution duration to obtain a predicted emotion type;

In another possible implementation manner, the acquiring the first operation data includes:

Receiving original operation data sent by the terminal, wherein the original operation data represents the operation of the terminal on the first recommendation information;

and selecting first operation data corresponding to the target operation from the original operation data.

In another possible implementation manner, the obtaining a first interval duration according to the first operation data includes:

and determining the interval duration corresponding to the emotion type as the first interval duration.

In another possible implementation manner, the first recommendation information includes at least one first user account recommended according to the user account logged in by the terminal;

the sending the second recommendation information to the terminal includes:

determining at least one recommended second user account according to the user account logged in by the terminal;

and sending the second recommendation information to the terminal, wherein the second recommendation information comprises the at least one second user account.

According to a second aspect of embodiments of the present disclosure, there is provided another information recommendation method, the method including:

Displaying a first recommendation interface, wherein the first recommendation interface comprises first recommendation information sent by a server at a first time point;

acquiring first operation data according to the operation detected in the first recommendation interface;

the first operation data is sent to the server, and the server is used for obtaining a first interval duration according to the first operation data, wherein the first interval duration represents an interval duration between a time point of next sending of recommended information and the first time point.

According to a third aspect of embodiments of the present disclosure, there is provided an information recommendation apparatus, the apparatus including:

an operation data acquisition unit configured to perform acquisition of first operation data representing an operation of a terminal on first recommendation information, the first recommendation information being information transmitted to the terminal at a first time point;

a first time length obtaining unit configured to obtain a first interval time length according to the first operation data, where the first interval time length represents an interval time length between a time point when the recommendation information is sent to the terminal next time and the first time point;

and a recommendation information transmitting unit configured to transmit second recommendation information to the terminal for displaying the second recommendation information when the first interval duration after the first time point is reached.

In one possible implementation manner, the first time length obtaining unit includes:

an acquisition subunit configured to perform acquisition of a stored second interval duration, where the second interval duration is an interval duration between a time point at which third recommendation information is sent to the terminal and the first time point, and the third recommendation information is information sent to the terminal before the first recommendation information;

and the adjustment subunit is configured to execute adjustment of the second interval duration according to the first operation data to obtain the first interval duration.

In another possible implementation manner, the adjustment subunit is configured to perform emotion recognition on the first operation data to obtain an emotion type, where the emotion type is a positive emotion type or a negative emotion type; if the emotion type is a forward emotion type, reducing the second interval duration to obtain the first interval duration; or if the emotion type is a negative emotion type, increasing the second interval duration to obtain the first interval duration.

In another possible implementation, the apparatus further includes:

a second time length acquisition unit configured to perform acquisition of second operation data representing an operation of the terminal on the third recommendation information; and acquiring the second interval duration according to the second operation data, and storing the second interval duration.

In another possible implementation, the apparatus further includes:

a second time length obtaining unit configured to perform selection of the second interval time length from the reference interval time length range, and store the second interval time length.

In another possible implementation manner, the adjustment subunit is configured to perform statistics on the first operation data to obtain a first adjustment parameter corresponding to the first operation data; and adjusting the second interval duration according to the first adjustment parameter to obtain the first interval duration.

In another possible implementation manner, the first adjustment parameter is an adjustment duration, and the adjustment subunit is configured to perform statistics on the first operation data to obtain an adjustment proportion corresponding to the first operation data; and adjusting the reference time length according to the adjustment proportion to obtain the adjustment time length.

In another possible implementation, the adjustment subunit is configured to perform obtaining a weight of each operation in the first operation data; and acquiring the first adjustment parameters according to the weight of each operation.

In another possible implementation manner, the adjusting subunit is configured to perform obtaining an emotion recognition model, where the emotion recognition model includes a weight of at least one operation, and the weight represents an importance degree of the operation on an emotion recognition result; and reading the weight of each operation from the emotion recognition model.

In another possible implementation manner, the adjusting subunit is configured to perform obtaining the number of times of execution of each operation according to the first operation data; and weighting the execution times of each operation according to the weight of each operation to obtain the first adjustment parameter.

In another possible implementation manner, the adjustment subunit is configured to determine, as the first adjustment parameter corresponding to the first operation data, a sum of weights of each operation.

In another possible implementation manner, the adjusting subunit is configured to obtain an execution duration of each operation according to the first operation data; and weighting the execution time of each operation according to the weight of each operation to obtain the first adjustment parameter.

In another possible implementation manner, the operation data obtaining unit is configured to perform receiving original operation data sent by the terminal, where the original operation data represents an operation of the terminal on the first recommendation information; and selecting first operation data corresponding to the target operation from the original operation data.

In another possible implementation manner, the first time length obtaining unit is configured to perform emotion recognition on the first operation data to obtain an emotion type, where the emotion type is a positive emotion type or a negative emotion type; and determining the interval duration corresponding to the emotion type as the first interval duration.

the recommendation information sending unit is configured to execute at least one second user account which is recommended according to the user account logged in by the terminal; and sending the second recommendation information to the terminal, wherein the second recommendation information comprises the at least one second user account.

According to a fourth aspect of embodiments of the present disclosure, there is provided an information recommendation apparatus, the apparatus including:

the recommendation information display unit is configured to display a first recommendation interface, wherein the first recommendation interface comprises first recommendation information sent by a server at a first time point;

an operation data acquisition unit configured to perform an operation according to the detection in the first recommendation interface, to acquire first operation data;

an operation data transmitting unit configured to perform transmission of the first operation data to the server, the server being configured to acquire a first interval duration according to the first operation data, the first interval duration representing an interval duration between a time point at which recommended information is transmitted next and the first time point.

According to a fifth aspect of embodiments of the present disclosure, there is provided an electronic device including:

a processor;

a memory for storing the processor-executable instructions;

wherein the processor is configured to execute the instructions to implement the information recommendation method of the above aspect.

According to a sixth aspect of embodiments of the present disclosure, there is provided a storage medium, which when executed by a processor of an electronic device, enables the electronic device to perform the information recommendation method described in the above aspects.

According to a seventh aspect of embodiments of the present disclosure, there is provided a computer program product, which when executed by a processor of an electronic device, enables the electronic device to perform the information recommendation method of the above aspect.

The technical scheme provided by the embodiment of the disclosure at least brings the following beneficial effects:

the present disclosure provides a scheme for personalizing recommended information, which does not recommend information to a user according to a fixed period, but determines an interval duration between a time point and a time point when recommended information is transmitted to a terminal next time according to an operation of the terminal on the recommended information transmitted at the time point. The operation of the terminal on the recommendation information represents the feedback of the user on the recommendation information, and the emotion type of the user on the recommendation information can be objectively reflected, so that the determined interval duration is matched with the emotion type of the user, and the personalized recommendation requirement of the user can be met.

Drawings

The accompanying drawings, which are incorporated in and constitute a part of this specification, illustrate embodiments consistent with the disclosure and together with the description, serve to explain the principles of the disclosure and do not constitute an undue limitation on the disclosure.

FIG. 1 is a schematic illustration of one implementation environment provided by embodiments of the present disclosure;

FIG. 2 is a flowchart of an information recommendation method provided by an embodiment of the present disclosure;

FIG. 3 is a flowchart of an information recommendation method provided by an embodiment of the present disclosure;

FIG. 4 is a flowchart of an information recommendation method provided by an embodiment of the present disclosure;

FIG. 5 is a schematic illustration of a recommendation interface provided by an embodiment of the present disclosure;

FIG. 6 is a flowchart of an information recommendation method provided by an embodiment of the present disclosure;

FIG. 7 is a block diagram of an information recommendation device provided by an embodiment of the present disclosure;

FIG. 8 is a block diagram of an information recommendation device provided by an embodiment of the present disclosure;

fig. 9 is a block diagram of a terminal provided by an embodiment of the present disclosure;

fig. 10 is a block diagram of a server provided by an embodiment of the present disclosure.

Detailed Description

In order to enable those skilled in the art to better understand the technical solutions of the present disclosure, the technical solutions of the embodiments of the present disclosure will be clearly and completely described below with reference to the accompanying drawings.

It should be noted that the terms "first," "second," and the like in the description of the present disclosure and the claims and the foregoing figures are used for distinguishing between similar objects and not necessarily for describing a particular sequential or chronological order. It is to be understood that the data so used may be interchanged where appropriate such that the embodiments of the disclosure described herein may be capable of operation in sequences other than those illustrated or described herein. The implementations described in the following exemplary examples are not representative of all implementations consistent with the present disclosure. Rather, they are merely examples of apparatus and methods consistent with some aspects of the present disclosure as detailed in the accompanying claims.

Another point to be noted is that the user information (including but not limited to user device information, user personal information, etc.) referred to in this disclosure may be information authorized by the user or sufficiently authorized by the parties.

It should be noted that, as used in this disclosure, the terms "each", "a plurality", and "any" and the like, a plurality includes two or more, each means each of the corresponding plurality, and any means any of the corresponding plurality. For example, the plurality of operations includes 10 operations, and each operation refers to each of the 10 operations, and any operation refers to any one of the 10 operations.

FIG. 1 is a schematic illustration of an implementation environment provided by embodiments of the present disclosure. Referring to fig. 1, the implementation environment includes a terminal 101 and a server 102. The terminal 101 and the server 102 are connected by a wireless or wired network. Terminal 101 is illustratively a computer, cell phone, tablet computer, or other terminal. The server 102 is illustratively a background server for the target application or a cloud server that provides services such as cloud computing and cloud storage.

Illustratively, the terminal 101 installs thereon a target application served by the server 102, through which the terminal 101 can implement functions such as data transmission, message interaction, and the like. The target application is illustratively a target application in the operating system of the terminal 101 or a target application provided for a third party. The target application has a function of recommending information, and illustratively, the target application has a function of recommending friends, a function of recommending games, a function of recommending videos, a function of recommending items, and the like, and of course, the target application can also have a function of recommending other information, which is not limited in this disclosure. Illustratively, the target application is a short video application, a music application, a gaming application, a shopping application, a chat application, or other application, to which the present disclosure is not limited.

In the embodiment of the disclosure, the server 102 is configured to recommend information to the terminal 101, and the terminal 101 displays a recommendation interface based on the information recommended by the server 102, where the recommendation interface includes recommendation information, and a user can operate the recommendation information on the recommendation interface.

The information recommendation method provided by the disclosure can be applied to any information recommendation scene, such as friend recommendation, article recommendation, game recommendation, audio recommendation, video recommendation, information recommendation and the like, and is not limited.

Fig. 2 is a flowchart of an information recommendation method provided by an embodiment of the present disclosure, referring to fig. 2, an execution body of the embodiment of the method is an electronic device, for example, the electronic device is a server, and the method includes:

in step 201, first operation data is acquired, the first operation data representing an operation of the terminal on first recommendation information, the first recommendation information being information of the terminal at a first time point.

In step 202, a first interval duration is acquired according to the first operation data, where the first interval duration represents an interval duration between a time point when the recommendation information is next sent to the terminal and the first time point.

In step 203, when a first interval duration after the first time point is reached, second recommendation information is sent to the terminal, and the terminal is used for displaying the second recommendation information.

The present disclosure provides a scheme for personalizing recommended information, which does not recommend information to a user according to a fixed period, but determines an interval duration between a time point and a time point when recommended information is transmitted to a terminal next time according to an operation of the terminal on the recommended information transmitted at the time point. The operation of the terminal on the recommendation information represents the feedback of the user on the recommendation information, and the emotion type of the user on the recommendation information can be objectively reflected, so that the determined interval duration is matched with the emotion type of the user, the personalized recommendation requirement of the user can be met, and the user viscosity is improved.

In one possible implementation, according to the first operation data, obtaining the first interval duration includes:

In another possible implementation manner, adjusting the second interval duration according to the first operation data to obtain the first interval duration includes:

if the emotion type is a forward emotion type, reducing the second interval duration to obtain a first interval duration; or,

In another possible implementation, before acquiring the stored second interval duration, the method further includes:

and acquiring a second interval duration according to the second operation data, and storing the second interval duration.

and selecting a second interval duration from the range of reference interval durations, and storing the second interval duration.

In another possible implementation manner, the first adjustment parameter is an adjustment duration, and the statistics is performed on the first operation data to obtain the first adjustment parameter corresponding to the first operation data, where the first adjustment parameter includes:

acquiring the weight of each operation in the first operation data;

and acquiring a first adjustment parameter according to the weight of each operation.

In another possible implementation, acquiring the weight of each operation in the first operation data includes:

the weight of each operation is read from the emotion recognition model.

In another possible implementation, the obtaining the first adjustment parameter according to the weight of each operation includes:

And weighting the execution times of each operation according to the weight of each operation to obtain a first adjustment parameter.

In another possible implementation, the training process of the emotion recognition model includes:

acquiring a first training sample, wherein the first training sample comprises execution times and sample emotion types, the execution times are the operation times of a sample user on recommended first sample information, the sample emotion types represent the emotion types of recommended second sample information after the sample user on the first sample information, and the sample emotion types are positive emotion types or negative emotion types;

invoking an emotion recognition model, and performing emotion recognition on the execution times to obtain a predicted emotion type;

training an emotion recognition model according to the sample emotion type and the predicted emotion type.

acquiring a second training sample, wherein the second training sample comprises a sample operation and a sample emotion type, the sample operation is an operation of a sample user on recommended first sample information, the sample emotion type represents an emotion type of second sample information recommended by the sample user after the first sample information, and the sample emotion type is a positive emotion type or a negative emotion type;

Invoking an emotion recognition model, and carrying out emotion recognition on sample operation to obtain a predicted emotion type;

acquiring the execution time of each operation according to the first operation data;

and weighting the execution time of each operation according to the weight of each operation to obtain a first adjustment parameter.

acquiring a third training sample, wherein the third training sample comprises an execution time length and a sample emotion type, the execution time length is the operation time length of a sample user on recommended first sample information, the sample emotion type represents the emotion type of second sample information recommended after the sample user on the first sample information, and the sample emotion type is a positive emotion type or a negative emotion type;

invoking an emotion recognition model, and performing emotion recognition on the execution duration to obtain a predicted emotion type;

In another possible implementation, acquiring the first operation data includes:

receiving original operation data sent by a terminal, wherein the original operation data represents the operation of the terminal on first recommendation information;

In another possible implementation manner, according to the first operation data, acquiring the first interval duration includes:

carrying out emotion recognition according to the first operation data to obtain an emotion type, wherein the emotion type is a positive emotion type or a negative emotion type;

and determining the interval duration corresponding to the emotion type as a first interval duration.

In another possible implementation manner, the first recommendation information includes at least one first user account recommended according to a user account logged in by the terminal;

sending second recommendation information to the terminal, including:

and sending second recommendation information to the terminal, wherein the second recommendation information comprises at least one second user account.

Fig. 3 is a flowchart of an information recommendation method provided by an embodiment of the present disclosure, referring to fig. 3, an execution body of the embodiment of the method is an electronic device, for example, the electronic device is a terminal, and the method includes:

In step 301, a first recommendation interface is presented, the first recommendation interface including first recommendation information sent by a server at a first point in time.

In step 302, first operation data is obtained according to the operation detected in the first recommendation interface.

In step 303, first operation data is sent to a server, where the server is configured to obtain a first interval duration according to the first operation data, where the first interval duration represents an interval duration between a time point of next sending of the recommended information and the first time point.

The present disclosure provides a scheme for personalized recommendation of information, which does not recommend information to a user according to a fixed period, but obtains operation data according to an operation detected in a recommendation interface after the recommendation information sent by a server at a certain time point is displayed in the recommendation interface, so that the server determines an interval duration between the time point and a time point for sending the recommendation information next time according to the operation data. Because the detected operation in the recommendation interface represents the feedback of the user to the recommendation interface, the emotion type of the user to the recommendation information in the recommendation interface can be objectively reflected, and therefore the determined interval duration is matched with the emotion type of the user, personalized recommendation requirements of the user can be met, and user viscosity is improved.

Fig. 5 is a flowchart of an information recommendation method provided in an embodiment of the present disclosure, where an interaction body in the embodiment of the method is a terminal and a server. Referring to fig. 5, the method includes:

in step 401, the terminal displays a first recommendation interface, where the first recommendation interface includes first recommendation information sent by the server at a first time point.

Illustratively, the recommendation information is item information, friend information, video information, audio information, information, or other recommendation information. The first recommendation information is any one or more of the recommendation information described above, which is not limited by the present disclosure.

For example, the first recommendation information is friend information, and accordingly, the first recommendation information includes at least one first user account recommended according to a user account logged in by the terminal. The user account is used for representing the identity of the user, and is an identity card number, a mobile phone number and the like. The first user account is any user account, which is not limited in this embodiment of the disclosure. The first time point is any time point, which is not limited by the embodiments of the present disclosure.

The first recommendation interface refers to a recommendation interface displayed at this time, and optionally, the first recommendation interface is a home page of the target application, or is other pages of the target application, such as an active page, etc., which is not limited in this disclosure. The first recommendation interface is an application interface, and can jump to other application interfaces based on the first recommendation interface. Alternatively, the first recommended interface is another interface displayed on the upper layer of the application interface, for example, a popup interface.

The first recommendation interface also includes various functional options, and the functional options match the first recommendation information in the first recommendation interface. For example, the first recommendation information is item information, and the function options include a skip option of a detail page of the item, a collection option of the item, an acquisition option of the item, and the like. For another example, if the first recommendation information is friend information, the function options include a focus option of a friend, a skip option of a homepage of the friend, and the like. For another example, if the first recommendation information is video information, the function options include a collection option of the video, a play option of the video, a praise option of the video, a share option of the video, and the like, which is not limited in the disclosure.

Referring to fig. 5, fig. 5 is a schematic diagram of a recommendation interface, where the recommendation interface includes a plurality of identifications of friends and a skip option of a main page of each friend, the skip option is displayed as an image of the friend, and when the image is triggered, the user can enter the main page of the corresponding friend. The recommendation interface also comprises a concerned option corresponding to each friend, and the concerned option is triggered, so that the dynamics of the corresponding friend can be obtained in time afterwards. The recommendation interface also includes a close option "x", and when the close option is triggered, the recommendation interface can be exited.

In step 402, the terminal obtains first operation data according to the operation detected in the first recommendation interface.

After the terminal displays the first recommendation interface, the user can execute some operations based on the first recommendation interface, taking the first recommendation information in the first recommendation interface as friend information as an example, the user can execute attention operation, access operation (i.e. entering a homepage of a friend) and the like based on the first recommendation interface, taking the first recommendation information in the first recommendation interface as video information as an example, the user can execute play operation, praise operation, collection operation, sharing operation and the like based on the first recommendation interface, and taking the first recommendation information in the first recommendation interface as article information as an example, and the user can execute collection operation, acquisition operation and the like based on the first recommendation interface. Of course, the user can also perform other operations based on the first recommendation interface, such as closing the operation, i.e., exiting the first recommendation interface, etc.

Correspondingly, the terminal detects the operation in the first recommended interface, and generates first operation data according to the detected operation. In one possible implementation manner, the first operation data includes operation content, a user identifier corresponding to the operation, a time when the operation starts, and a time when the operation ends, where the user identifier corresponding to the operation refers to a user identifier currently logged in by the terminal, where the user identifier is used to represent a user identity, and the user identifier is exemplified by a mobile phone number, an account number, and the embodiment of the disclosure does not limit this. In another possible implementation, the first operation data further includes a number of executions of the operation, for example, a number of praise operations, i.e., a number of praise operations detected in the first recommendation interface. In another possible implementation manner, the first operation data further includes an execution duration of the operation, for example, a display duration of the first recommendation interface, where the display duration of the first recommendation interface refers to a duration from displaying the first recommendation interface to closing the first recommendation interface. Of course, other information can also be included in the first operation data, which is not limited by the embodiments of the present disclosure.

It should be noted that, the first recommended interface may include one page or multiple pages, for example, the first recommended interface includes multiple pages, the multiple pages may jump from each other, when the first recommended interface starts to be displayed, only one of the pages may be displayed, and then the terminal may jump to another page according to the operation detected on the page, or return to the page from another page.

For example, the first recommendation interface includes an item display page and a detail page for each item. The terminal can jump from the item display page to the detail page of a certain item, detect the acquisition operation on the item in the detail page of the item, and then can return to the item display page or reenter the detail page of other items. The operation detected by the terminal in any page included in the first recommendation interface belongs to the operation detected in the first recommendation interface.

According to the display mode, the recommended information can be displayed on the pages of the multiple layers, interference to the user caused by displaying excessive recommended information on the same page is avoided, the user is inspired, the page can be quickly jumped through convenient operation, and the user can conveniently browse more recommended information.

Another point to be described is that after the terminal obtains the first operation data, the first operation data is stored, and the terminal stores the first operation data in a redis (a database) for convenience in subsequent access to the first operation data due to high query and calculation efficiency of the redis.

In step 403, the terminal transmits the first operation data to the server.

In step 404, the server receives first operational data.

In one possible implementation, the step 402 is: the terminal obtains the original operation data according to the operation detected in the first recommendation interface, and correspondingly, step 403 is: the terminal sends the original operation data to the server, and step 404 is: the method comprises the steps that a server receives original operation data sent by a terminal, wherein the original operation data represent the operation of the terminal on first recommendation information; and the server selects first operation data corresponding to the target operation from the original operation data. Thus, the operation data corresponding to the key operation is screened from the original operation data, and the first interval duration acquired according to the operation data is more matched with the emotion type of the user.

The original operation data is generated according to all operations detected in the first recommendation interface. And the first operation data is the operation data corresponding to the target operation detected in the first recommendation interface.

Illustratively, the target operation is an operation that matches the information type of the first recommendation information. Correspondingly, the method further comprises the steps of: the server determines the information type of the sent first recommendation information and acquires the target operation matched with the information type. In one possible implementation, the server stores therein target operations for each information type match, for example, target operations for friend information match include a praise operation, an access operation, a focus operation, a close operation, and the like. As another example, the target operations for video information matching include a play operation, a share operation, a collection operation, and the like. For another example, the object operation of matching the item information includes a collection operation, an acquisition operation, and the like.

The target operation of each information type matching may be obtained according to a feature engineering or correlation method, and may be determined according to other methods, which is not limited by the embodiment of the present disclosure. The feature engineering refers to determining a target operation matched with each information type through a data analysis model, for example, inputting the information type of recommended information and various operations on the recommended information into the data analysis model, and outputting the target operation matched with the recommended information by the data analysis model. The training sample of the data analysis model is the type of the recommended information, various operations of the recommended information by the user and emotion types of the recommended information by the user, and the data analysis model can learn the target operation of each type of the recommended information, which influences the emotion types of the recommended information of the user, through the training sample, so that the data analysis model has the capability of determining the target operation matched with the information type according to the information type of the recommended information and various operations of the recommended information. Wherein the emotion type is used for representing the attitudes of the user to the first recommended information, such as like or dislike.

Correlation analysis refers to analyzing two or more variable elements with correlation, so as to measure the correlation degree of variable factors. In the embodiment of the disclosure, for each type of recommended information, each operation on the type of recommended information is taken as an independent variable, and the emotion type of the type of recommended information is taken as a dependent variable, so that a key operation affecting the emotion type, namely, a target operation, is determined.

In step 405, the server performs emotion recognition on the first operation data to obtain an emotion type, where the emotion type is a positive emotion type or a negative emotion type.

Emotion types are used to represent the user's attitudes to the first recommendation information, and in one possible implementation, emotion types include positive emotion types and negative emotion types. In the embodiment of the present disclosure, the positive emotion type is taken as a like, the negative emotion type is taken as a dislike, and of course, the positive emotion type and the negative emotion type can also be other, for example, the positive emotion type is interesting, the negative emotion type is not interesting, and the embodiment of the present disclosure does not limit this.

In the embodiment of the disclosure, after the terminal displays the first recommendation interface, the user may like the first recommendation information in the first recommendation interface or may dislike the first recommendation information, and the operations performed based on the first recommendation interface are different in the two cases that the user likes the first recommendation information and dislikes the first recommendation information. For example, the user may perform a collection operation, a sharing operation, or the like in a case where the user likes the first recommended information, and may directly perform a closing operation in a case where the user does not like the first recommended information. Therefore, the operation detected by the terminal in the first recommendation interface represents the emotion type of the user to the first recommendation information, and correspondingly, the first operation data generated according to the operation represents the emotion type to the first recommendation information.

In one possible implementation, the server does not use emotion recognition models, and the implementation of this step includes the following three:

(1) The server determines whether first reference operation matched with the positive emotion type is included in the first operation data, if the first reference operation is included in the first operation data, the emotion type is determined to be the positive emotion type, and if the first reference operation is not included in the first operation data, the emotion type is determined to be the negative emotion type.

Alternatively, the step includes: the server determines whether the first operation data comprises a second reference operation matched with the negative emotion type, if the first operation data comprises the second reference operation, the emotion type is determined to be the negative emotion type, and if the first operation data does not comprise the second reference operation, the emotion type is determined to be the positive emotion type.

Illustratively, the positive emotion type is like. Illustratively, the first reference operation that is forward emotion type matching is stored in the server. Illustratively, the first reference operation of forward emotion type matching includes a collection operation, a focus operation, a sharing operation, an acquisition operation, and the like. Illustratively, the second reference operation of negative emotion type matching includes a closing operation performed within a reference time after the first recommendation interface is presented, or a closing operation performed directly after the first recommendation interface is presented, or the like.

In the embodiment of the disclosure, the emotion type is determined by determining whether the first operation data comprises the reference operation matched with the emotion type, so that the method is simple and high in efficiency.

(2) The server determines whether the execution times of the reference operation which is included in the first operation data and is matched with the positive emotion type reach the corresponding reference execution times, if the corresponding reference execution times are reached, the emotion type is determined to be the positive emotion type, and if the corresponding reference execution times are not reached, the emotion type is determined to be the negative emotion type. Wherein, the reference execution times are set according to the needs, and the embodiment of the disclosure does not limit this.

For example, in the case that the first reference operation matched with the forward emotion type includes a collection operation, a concern operation and a sharing operation, the server determines whether the collection operation, the concern operation and the sharing operation respectively reach the corresponding reference execution times, if both the reference execution times are reached, determines that the emotion type is liked, and if any reference operation does not reach the corresponding reference execution times, determines that the emotion type is disliked.

In the embodiment of the disclosure, the emotion type is determined by determining whether the execution times of the reference operation matched with the emotion type in the first operation data reach the corresponding reference execution times, and the situation that the number of times of the reference operation matched with the forward emotion type is larger when the emotion type of the first recommended information is the forward emotion type is fully considered, for example, when the user likes friend information recommended in the first recommended interface, the number of times of attention executed in the first recommended interface is more, so that the determined emotion type is more accurate.

(3) The server determines whether the execution duration of the reference operation which is included in the first operation data and is matched with the positive emotion type reaches the corresponding reference execution duration, if so, the emotion type is determined to be the positive emotion type, and if not, the emotion type is determined to be the negative emotion type. Wherein, consult and carry out the duration setting as required, this is not limited to this in the embodiment of the disclosure.

For example, in the case that the first reference operation matched with the forward emotion type includes a play operation and an access operation, the server determines whether the play operation and the access operation respectively reach the corresponding reference execution time periods, if both reach the reference execution time periods, determines that the emotion type is favored, and if any reference operation does not reach the corresponding reference execution time periods, determines that the emotion type is disfavored.

In the embodiment of the disclosure, the emotion type is determined by determining whether the execution duration of the reference operation matched with the emotion type in the first operation data reaches the corresponding reference execution duration, and the situation that the duration of the reference operation matched with the forward emotion type is longer when the emotion type of the first recommended information is the forward emotion type is fully considered, for example, when the user likes friend information recommended in the first recommended interface, the access duration executed in the first recommended interface is longer, so that the determined emotion type is more accurate.

In another possible implementation, the server uses emotion recognition models, and the implementation of this step includes the following three kinds of:

(A) And the server calls an emotion recognition model to perform emotion recognition on the operation in the first operation data, so as to obtain an emotion type.

The training method of the emotion recognition model comprises the following steps: the server acquires a second training sample, wherein the second training sample comprises sample operation and sample emotion types, the sample operation is operation of a sample user on recommended first sample information, the sample emotion types represent emotion types of the sample user on the first sample information, and the sample emotion types are positive emotion types or negative emotion types; the server calls an emotion recognition model to perform emotion recognition on the sample operation to obtain a predicted emotion type; and the server trains an emotion recognition model according to the sample emotion type and the predicted emotion type.

Wherein the number of second training samples is any number, which is not limited by the embodiments of the present disclosure. The realization mode of training the emotion recognition model by the server according to the sample emotion type and the predicted emotion type is as follows: and the server acquires the error of the model according to the sample emotion type and the predicted emotion type, and iteratively trains the emotion recognition model according to the error.

Alternatively, the error may be MSE (Mean Squared Error, mean square error), RMSE (Root Mean Square Error ), MAPE (Mean Absolute Percentage Error, mean absolute percentage error), or other error, which is not limited by embodiments of the present disclosure.

After determining the error of the emotion recognition model, the server compares the error with a reference threshold, and if the error is greater than the reference threshold, the server continues to train the emotion recognition model in an iterative manner. For example, the server adjusts model parameters, or the server obtains more new training samples to train the emotion recognition model, which is not limited by the embodiments of the present disclosure. If the error is not greater than the reference threshold, the emotion recognition model training is completed. Wherein the reference threshold is set to any value as required, which is not limited by the embodiments of the present disclosure.

In the embodiment of the disclosure, since the second training sample includes the operation of the sample user on the recommended first sample information and the emotion type of the sample user on the first sample information, the emotion recognition model is trained through the second training sample, and the model can learn the relationship between the operation of the user on the first sample information and the emotion type of the user on the first sample information, thereby having the capability of determining the emotion type of the user on the recommended information according to the operation of the user on the recommended information, the emotion recognition is performed on the operation in the first operation data by calling the emotion recognition model to obtain the emotion type, and the emotion recognition efficiency is improved on the premise of ensuring the accuracy of the emotion recognition.

(B) The server acquires the execution times of each operation in the first operation data according to the first operation data, and invokes the emotion recognition model to perform emotion recognition on the execution times to obtain emotion types.

The training method of the emotion recognition model comprises the following steps: the method comprises the steps that a server obtains a first training sample, wherein the first training sample comprises execution times and sample emotion types, the execution times are the operation times of a sample user on recommended first sample information, the sample emotion types represent emotion types of the sample user on the first sample information, and the sample emotion types are positive emotion types or negative emotion types; the server calls an emotion recognition model to perform emotion recognition on the execution times to obtain a predicted emotion type; training an emotion recognition model according to the sample emotion type and the predicted emotion type. Wherein the sample operation is an operation matching the first sample information. The number of first training samples is any number, which is not limited by the embodiments of the present disclosure.

In the embodiment of the disclosure, since the first training sample includes the operation times of the first sample information recommended by the sample user and the emotion type of the first sample information by the user, the emotion recognition model is trained through the first training sample, and the model can learn the relationship between the operation times of the first sample information by the user and the emotion type of the first sample information recommended by the user, so that the emotion recognition method has the capability of determining the emotion type of the recommended information by the user according to the execution times, and therefore, the emotion recognition is carried out on the execution times by calling the emotion recognition model to obtain the emotion type, and the emotion recognition efficiency is improved on the premise of ensuring the emotion recognition accuracy.

(C) The server acquires the execution duration of each operation in the first operation data according to the first operation data, and invokes the emotion recognition model to perform emotion recognition on the execution duration to obtain emotion types.

The training method of the emotion recognition model comprises the following steps: the server acquires a third training sample, wherein the third training sample comprises an execution duration and a sample emotion type, the execution duration is the operation duration of a sample user on recommended first sample information, the sample emotion type represents the emotion type of the sample user on the first sample information, and the sample emotion type is a positive emotion type or a negative emotion type; the server calls an emotion recognition model to perform emotion recognition on the execution duration so as to obtain a predicted emotion type; training an emotion recognition model according to the sample emotion type and the predicted emotion type. The operation duration is the execution duration of the sample operation matched with the first sample information. The number of third training samples is any number, which is not limited by the embodiments of the present disclosure.

In the embodiment of the disclosure, the third training sample includes the operation duration of the first sample information by the sample user and the emotion type of the first sample information by the sample user, so that the emotion recognition model is trained by the first training sample, and the model can learn the relationship between the operation duration of the first sample information by the user and the emotion type of the first sample information by the user, thereby having the capability of determining the emotion type of the recommended information by the user according to the execution duration, and therefore, the emotion type is obtained by invoking the emotion recognition model to perform emotion recognition on the execution duration, and the emotion recognition efficiency is improved on the premise of ensuring the emotion recognition accuracy.

It should be noted that, the emotion recognition model may be trained by the server, or may be transmitted to the server after being trained by other terminals or servers, which is not limited in the embodiment of the present disclosure.

In step 406, the server determines an interval duration corresponding to the emotion type as a first interval duration.

In one possible implementation manner, the interval duration corresponding to the positive emotion type and the negative emotion type is stored in the server, and the interval duration corresponding to the positive emotion type is smaller than the interval duration corresponding to the negative emotion type. For example, the interval duration corresponding to the positive emotion type is 3 days, and the interval duration corresponding to the negative emotion type is 5 days. Of course, the interval duration corresponding to the positive emotion type and the interval duration corresponding to the negative emotion type may also be other interval durations, which are not limited in the embodiment of the present disclosure.

In the embodiment of the disclosure, the interval duration corresponding to the emotion type is determined to be a first interval duration, and the first interval duration is an interval duration matched with the emotion type of the user. And the interval duration corresponding to the positive emotion type is smaller than the interval duration corresponding to the negative emotion type, so that the determined first interval duration is shorter when the user likes the first recommended information, namely, the duration of displaying the recommended interface for the user next time is shorter, and the determined first interval duration is longer when the user dislikes the first recommended information, namely, the duration of displaying the recommended interface for the user next time is longer, thereby meeting the personalized recommended requirement of the user, improving the user experience and simultaneously guaranteeing the reasonable utilization of recommended resources.

In step 407, when a first interval duration after the first time point is reached, the server transmits second recommendation information to the terminal.

Illustratively, the recommendation information is item information, friend information, video information, audio information, information, or other recommendation information. The second recommendation information is any one of the recommendation information. And, the second recommendation information is the same as or different from the first recommendation information, which is not limited by the embodiments of the present disclosure.

The implementation manner of sending the second recommendation information to the terminal by the server is as follows: the server determines at least one recommended second user account according to the user account logged in by the terminal; and sending second recommendation information to the terminal, wherein the second recommendation information comprises at least one second user account.

Optionally, the server determines an interest tag of the user account according to the user account logged in by the terminal, and determines at least one second user account matched with the interest tag according to the interest tag, so that the user account interested by the user can be recommended to the user, and the recommendation effect is improved. The interest tags are used to represent interests of the user, for example, the interest tags are "fun", "pet", "travel", "history", and the like, which are not limited by the embodiments of the present disclosure.

In step 408, the terminal receives the second recommendation information, and displays a second recommendation interface, where the second recommendation interface includes the second recommendation information.

The point to be noted is that when the terminal displays the second recommendation interface, the terminal obtains third operation data according to the operation detected in the second recommendation interface, and sends the third operation data to the server; after receiving the third operation data, the server acquires a third interval duration according to the third operation data, wherein the third interval duration represents the interval duration between the time point when the server sends the recommended information to the terminal after the second recommended information and the time point when the server sends the second recommended information. The implementation is the same as the implementation of steps 401-406 described above and will not be described here again. Therefore, the interval duration of the time point distance for displaying the recommended information for the user is determined every time according to the operation of the user on the recommended information, so that the interval duration is always matched with the feedback operation executed by the user in the recommended interface, real-time change according to the real-time feedback operation of the user can be realized, and the personalized recommended requirement of the user is further met.

And the server selects first operation data corresponding to the target operation from the original operation data, so that operation data corresponding to the key operation is screened from the original operation data, and the first interval duration acquired according to the operation data is more matched with the emotion type of the user.

Fig. 6 is a flowchart of an information recommendation method provided in an embodiment of the present disclosure, where in the method embodiment, an interaction body is a terminal and a server. Referring to fig. 6, the method includes:

in step 601, the terminal displays a first recommendation interface, where the first recommendation interface includes first recommendation information sent by the server at a first time point.

In step 602, the terminal obtains first operation data according to the operation detected in the first recommendation interface.

In step 603, the terminal transmits the first operation data to the server.

In step 604, the server receives first operational data.

The implementation of steps 601-604 is the same as the implementation of steps 401-404 and will not be described here again.

In step 605, the server acquires a stored second interval duration, where the second interval duration is an interval duration between a time point when third recommendation information is sent to the terminal and the first time point, and the third recommendation information is information sent to the terminal before the first recommendation information.

Illustratively, the recommendation information is item information, friend information, video information, audio information, information, or other recommendation information. The third recommendation information is any one of the recommendation information. And, the third recommendation information is the same as or different from the first recommendation information, which is not limited by the embodiments of the present disclosure.

The second interval duration is an interval duration between a time point at which the third recommended information is transmitted to the terminal and the first time point, and the third recommended information is information that is transmitted to the terminal before the first recommended information, that is, the second interval duration is determined before the first recommended information is transmitted to the terminal, and the server stores the second interval duration after determining the second interval duration. The method for determining the second interval duration comprises the following two steps:

firstly, a server displays a third recommendation interface; the server acquires second operation data, wherein the second operation data represents the operation of the terminal on the third recommendation information; and the server acquires a second interval duration according to the second operation data. The implementation manner is the same as that of the server for obtaining the first interval duration according to the first operation data, and is not described herein again.

Second, the server selects a second interval duration from a range of reference interval durations.

The reference interval duration range is set as required, for example, is set to [1-10] days, and of course, can be set to other, which is not limited by the embodiment of the disclosure.

Illustratively, the server selects the second interval duration from the range of reference interval durations by: the server randomly selects a second interval duration from the range of reference interval durations. The server can change the recommended interval duration according to the feedback operation of the user on the recommended information every time, so that the server directly selects the second interval duration from the range of the reference interval duration as the initial interval duration, and the matching of the subsequent interval duration and the feedback operation of the user is not influenced. And the method for determining the second interval duration is simple and high in efficiency.

It should be noted that the execution sequence between step 605 and steps 601-604 is not limited.

In step 606, the server adjusts the second interval duration according to the first operation data, to obtain the first interval duration.

In one possible implementation manner, the server adjusts the second interval duration according to the first operation data to obtain the first interval duration, including: the server carries out emotion recognition on the first operation data to obtain emotion types; if the emotion type is a forward emotion type, reducing the second interval duration to obtain a first interval duration; and if the emotion type is a negative emotion type, increasing the second interval duration to obtain the first interval duration. Illustratively, the positive emotion type is like and the negative emotion type is dislike. Therefore, the interval time of the recommendation is shorter when the user likes the first recommendation information, and the interval time of the recommendation is longer when the user dislikes the first recommendation information, so that personalized recommendation requirements of the user are met, interference of the recommendation to the user is reduced, and reasonable utilization of recommendation resources can be guaranteed.

The implementation manner of the server for identifying the emotion of the first operation data to obtain the emotion type is described in step 405, which is not described herein.

The server determines a reduced duration corresponding to the positive emotion type and an increased duration corresponding to the negative emotion type, wherein the difference between the second interval duration and the reduced duration is used as a first interval duration when the emotion type is the positive emotion type, and the sum of the second interval duration and the increased duration is used as the first interval duration when the emotion type is the negative emotion type.

For example, the reduction duration corresponding to the positive emotion type is 2 days, the increase duration corresponding to the negative emotion type is 2 days, and correspondingly, if the second interval duration is 5 days, the first interval duration is 3 days when the obtained emotion type is the positive emotion type, and the second interval duration is 7 days when the obtained emotion type is the negative emotion type.

Of course, the reduced duration corresponding to the positive emotion type and the increased duration corresponding to the negative emotion type can be other durations, which are not limited by the embodiments of the present disclosure.

In another possible implementation manner, the server adjusts the second interval duration according to the first operation data to obtain the first interval duration, including: the server counts the first operation data to obtain a first adjustment parameter corresponding to the first operation data; and the server adjusts the second interval duration according to the first adjustment parameter to obtain the first interval duration.

In one possible implementation manner, the server performs statistics on the first operation data to obtain a first adjustment parameter corresponding to the first operation data, where the statistics includes: the server acquires the weight of each operation in the first operation data; and the server acquires the first adjustment parameters according to the weight of each operation.

In one possible implementation, a weight for each operation is set in the server, the weight being determined according to the law between the operation and the corresponding emotion type. Correspondingly, the server obtains the weight of each operation in the first operation data, including: the server queries the weight of each operation in the first operation data from the stored weights of each operation.

In another possible implementation, the weights of the operations are determined by emotion recognition models. Correspondingly, the server obtains the weight of each operation in the first operation data, including: the method comprises the steps that a server acquires an emotion recognition model, wherein the emotion recognition model comprises at least one operation weight, and the weight represents the importance degree of the operation on an emotion recognition result; the server reads the weight of each operation from the emotion recognition model.

In the embodiment of the present disclosure, the implementation manner of obtaining the first adjustment parameter by the server according to the weight of each operation includes the following three methods:

(1) The server acquires the execution times of each operation according to the first operation data; and the server performs weighting processing on the execution times of each operation according to the weight of each operation to obtain a first adjustment parameter.

Wherein the weighting process includes weighted summation, or weighted summation followed by multiplication by a reference coefficient, etc., which is not limited by embodiments of the present disclosure.

In the case of acquiring the first adjustment parameter in this way, the training method of the emotion recognition model includes:

the method comprises the steps that a server acquires a first training sample, wherein the first training sample comprises execution times and sample emotion types, the execution times are the operation times of a sample user on recommended first sample information, the sample emotion types represent emotion types of second sample information recommended by the sample user after the first sample information, and the sample emotion types are positive emotion types or negative emotion types; the server calls an emotion recognition model to perform emotion recognition on the execution times to obtain a predicted emotion type; training an emotion recognition model according to the sample emotion type and the predicted emotion type. Wherein the execution times are execution times of sample operations matched with the first sample information. The number of first training samples is any number, which is not limited by the embodiments of the present disclosure.

In the embodiment of the disclosure, since the first training sample includes the execution times and the sample emotion types, the emotion recognition model is trained according to the training sample, and the emotion recognition model can learn the relationship between the execution times and the sample emotion types, and the relationship between the execution times and the sample emotion types is represented by a plurality of model parameters, and the plurality of model parameters include the weight of each operation. Correspondingly, after the emotion recognition model is trained, the weight of each operation can be obtained from the model parameters of the emotion recognition model. Because the weight of each operation reflects the importance degree of the execution times of each operation on the emotion recognition result, the execution times of each operation are weighted according to the weight of each operation to obtain a first adjustment parameter, the first adjustment parameter can reflect the influence degree of the execution times of each operation on the emotion recognition result, and the obtained first interval duration is more matched with the emotion type of the user after the second interval duration is adjusted according to the first adjustment parameter, so that the personalized recommendation requirement of the user can be met.

(2) And the server determines the sum of the weights of each operation as a first adjustment parameter corresponding to the first operation data.

In the case of acquiring the first adjustment parameter in this way, the training method of the emotion recognition model includes: the server acquires a second training sample, wherein the second training sample comprises a sample operation and a sample emotion type, the sample operation is the operation of a sample user on recommended first sample information, the sample emotion type represents the emotion type of the recommended second sample information after the sample user on the first sample information, and the sample emotion type is a positive emotion type or a negative emotion type; the server calls an emotion recognition model to perform emotion recognition on the sample operation to obtain a predicted emotion type; and the server trains an emotion recognition model according to the sample emotion type and the predicted emotion type. Wherein the sample operation is an operation matching the first sample information. The number of second training samples is any number, which is not limited by the embodiments of the present disclosure.

In the embodiment of the disclosure, since the second training sample includes a sample operation and a sample emotion type, the emotion recognition model is trained according to the training sample, and the emotion recognition model can learn the relationship between the sample operation and the sample emotion type, and the relationship between the sample operation and the sample emotion type is represented by a plurality of model parameters, and the plurality of model parameters include weights of each operation. Correspondingly, after the emotion recognition model is trained, the weight of each operation can be obtained from the model parameters of the emotion recognition model. Because the weight of each operation reflects the importance degree of each operation on the emotion recognition result, the sum of the weights of each operation is determined to be a first adjustment parameter corresponding to the first operation data, the first adjustment parameter can reflect the influence degree of each operation on the emotion recognition result, and the obtained first interval duration is more matched with the emotion type of the user after the second interval duration is adjusted according to the first adjustment parameter, so that the personalized recommendation requirement of the user can be met.

(3) The server acquires the execution time of each operation according to the first operation data; and the server performs weighting processing on the execution duration of each operation according to the weight of each operation to obtain a first adjustment parameter.

In the case of acquiring the first adjustment parameter in this way, the training method of the emotion recognition model includes: the server acquires a third training sample, wherein the third training sample comprises an execution time length and a sample emotion type, the execution time length is the operation time length of a sample user on recommended first sample information, the sample emotion type represents the emotion type of second sample information recommended after the sample user on the first sample information, and the sample emotion type is a positive emotion type or a negative emotion type; the server calls an emotion recognition model to perform emotion recognition on the execution duration so as to obtain a predicted emotion type; and the server trains an emotion recognition model according to the sample emotion type and the predicted emotion type. Wherein the execution time length is the execution time length of the sample operation matched with the first sample information. The number of third training samples is any number, which is not limited by the embodiments of the present disclosure.

In the embodiment of the present disclosure, since the third training sample includes an execution duration and a sample emotion type, the emotion recognition model is trained according to the training sample, and the emotion recognition model can learn a relationship between the execution duration and the sample emotion type, where the relationship between the execution duration and the sample emotion type is represented by a plurality of model parameters, and the plurality of model parameters include weights of each operation. Correspondingly, after the emotion recognition model is trained, the weight of each operation can be obtained from the model parameters of the emotion recognition model. Because the weight of each operation reflects the importance degree of the execution time length of each operation on the emotion recognition result, the execution time length of each operation is weighted according to the weight of each operation to obtain a first adjustment parameter, the first adjustment parameter can reflect the influence degree of the execution time length of each operation on the emotion recognition result, and the obtained first interval time length is more matched with the emotion type of the user after the second interval time length is adjusted according to the first adjustment parameter, so that the personalized recommendation requirement of the user can be met.

It should be noted that, the three methods for obtaining the first adjustment parameter have different attention angles, and the first method obtains the first adjustment parameter from the point of view of the execution times of each operation, considering that the execution times of each operation are different for different emotion types of the user. The second method takes into account the fact that the user performs different operations under different emotion types, and obtains the first adjustment parameters from the viewpoint of the operation itself. The third method takes into consideration that the execution time length of each operation is different under different emotion types of the user, and obtains the first adjustment parameter from the aspect of the execution time length. It should be noted that these three methods can be combined in any manner to obtain the first adjustment parameter, which is not limited by the embodiments of the present disclosure.

In one possible implementation manner, the first adjustment parameter is an adjustment duration, and the server performs statistics on the first operation data to obtain the first adjustment parameter corresponding to the first operation data, where the method includes: the server counts the first operation data to obtain an adjustment proportion corresponding to the first operation data; and the server adjusts the reference time length according to the adjustment proportion to obtain the adjustment time length.

The method comprises the steps that a server counts first operation data to obtain an implementation mode of an adjustment proportion corresponding to the first operation data, the implementation mode of the first adjustment parameter corresponding to the first operation data is the same as that of the first operation data counted by the server in the steps (1) - (3), then the server adjusts a reference time length through the adjustment proportion to obtain an adjustment time length, and the adjustment time length is used as a first adjustment parameter to adjust a second interval time length.

The reference time length is set to be any time length according to needs, and the embodiment of the disclosure does not limit the reference time length. The server uses the product of the reference time length and the adjustment proportion as the adjustment time length, and uses the sum of the second interval time length and the adjustment time length as the first interval time length. For example, the second interval period is 5 days, the adjustment ratio is 0.8, and the reference period is 2 days, then the product of the reference period and the adjustment ratio is "1.6 days" as the adjustment period, and the sum of the second interval period and the first adjustment parameter is "6.6 days" as the first interval period.

In step 607, the server transmits second recommendation information to the terminal when a first interval duration after the first time point is reached.

In step 608, the terminal receives the second recommendation information, and displays a second recommendation interface, where the second recommendation interface includes the second recommendation information.

The implementation of steps 607-608 is the same as the implementation of steps 407-408 described above and will not be repeated here.

The proposal provided by the disclosure improves the recommended function of users by 12% under the premise of ensuring user experience in the AB test.

Another point to be described is that the above two embodiments are only two exemplary descriptions of acquiring the first interval duration according to the operation data, and the implementation manner of acquiring the first interval duration according to the operation data can also be other, which is not limited in this disclosure.

In the embodiment of the disclosure, the server adjusts the current recommendation interval duration according to the operation data of the recommendation information to obtain the interval duration between the time point of sending the recommendation information and the time point of sending the recommendation information to the terminal next time, so that the recommendation interval duration gradually changes according to the operation of the user on the recommendation information each time, the recommendation interval duration can always be matched with the emotion type of the user on the recommendation information, and the personalized recommendation requirement of the user is met.

In addition, the server can change the recommended interval duration according to the feedback operation of the user on the recommended information each time, so that the server directly selects the second interval duration from the range of the reference interval duration as the initial interval duration, and the matching of the subsequent interval duration and the feedback operation of the user is not influenced. And the method for determining the second interval duration is simple and high in efficiency.

And the server reduces the second interval duration to obtain the first interval duration under the condition that the emotion type is determined to be the positive emotion type, and increases the second interval duration to obtain the first interval duration under the condition that the emotion type is determined to be the negative emotion type, so that the recommended interval duration is ensured to be matched with the emotion type of the user, and the personalized recommendation requirement of the user is met. And the recommended interval duration is gradually adjusted according to the emotion type of the user, so that the recommended interval duration can be always consistent with the emotion type of the user, and the viscosity of the user is improved.

In the embodiment of the disclosure, since the weight of each operation reflects the importance degree of the execution times of each operation on the emotion recognition result, the execution times of each operation are weighted according to the weight of each operation, the obtained first adjustment parameter can reflect the influence degree of the execution times of each operation on the emotion recognition result, the obtained first interval time length is adjusted according to the first adjustment parameter to be more matched with the emotion type of the user, and the personalized recommendation requirement of the user can be met.

In the embodiment of the disclosure, since the weight of each operation reflects the importance degree of each operation on the emotion recognition result, the sum of the weights of each operation is determined as the first adjustment parameter corresponding to the first operation data, the first adjustment parameter can reflect the influence degree of each operation on the emotion recognition result, the second interval duration is adjusted according to the first adjustment parameter, the obtained first interval duration is more matched with the emotion type of the user, and the personalized recommendation requirement of the user can be met.

In the embodiment of the disclosure, since the weight of each operation reflects the importance degree of the execution duration of each operation on the emotion recognition result, the execution duration of each operation is weighted according to the weight of each operation, the obtained first adjustment parameter can reflect the influence degree of the execution duration of each operation on the emotion recognition result, and the obtained first interval duration is adjusted to be more matched with the emotion type of the user according to the first adjustment parameter, so that the personalized recommendation requirement of the user can be met.

Fig. 7 is a block diagram of an information recommendation apparatus provided in an embodiment of the present disclosure, referring to fig. 7, the apparatus includes:

An operation data acquisition unit 701 configured to perform acquisition of first operation data representing an operation of the terminal on first recommendation information, the first recommendation information being information of the terminal at a first time point;

a first time length obtaining unit 702 configured to obtain a first interval time length according to the first operation data, the first interval time length representing an interval time length between a time point when the recommendation information is next transmitted to the terminal and a first time point;

the recommendation information transmitting unit 703 is configured to transmit second recommendation information to the terminal when a first interval duration after the first time point is reached, the terminal being configured to display the second recommendation information.

In one possible implementation, the first time length obtaining unit 702 includes:

an acquisition subunit configured to perform acquisition of a stored second interval duration, the second interval duration being an interval duration between a time point at which third recommendation information is transmitted to the terminal and the first time point, the third recommendation information being information transmitted to the terminal before the first recommendation information;

and the adjusting subunit is configured to execute the adjustment of the second interval duration according to the first operation data to obtain the first interval duration.

In another possible implementation manner, the adjustment subunit is configured to perform emotion recognition on the first operation data to obtain an emotion type, where the emotion type is a positive emotion type or a negative emotion type; if the emotion type is a forward emotion type, reducing the second interval duration to obtain a first interval duration; or if the emotion type is a negative emotion type, increasing the second interval duration to obtain the first interval duration.

In another possible implementation, the apparatus further includes:

a second time length acquisition unit configured to perform acquisition of second operation data representing an operation of the terminal on the third recommendation information; and acquiring a second interval duration according to the second operation data, and storing the second interval duration.

In another possible implementation, the apparatus further includes:

a second time length acquisition unit configured to perform selection of a second interval time length from the reference interval time length range, and store the second interval time length.

In another possible implementation, the adjustment subunit is configured to perform obtaining a weight of each operation in the first operation data; and acquiring a first adjustment parameter according to the weight of each operation.

In another possible implementation manner, the adjusting subunit is configured to perform obtaining an emotion recognition model, where the emotion recognition model includes a weight of at least one operation, and the weight represents an importance degree of the operation on an emotion recognition result; the weight of each operation is read from the emotion recognition model.

In another possible implementation manner, the adjusting subunit is configured to perform obtaining the execution times of each operation according to the first operation data; and weighting the execution times of each operation according to the weight of each operation to obtain a first adjustment parameter.

In another possible implementation, the adjustment subunit is configured to perform determining the sum of the weights of each operation as the first adjustment parameter corresponding to the first operation data.

In another possible implementation manner, the adjusting subunit is configured to perform obtaining an execution duration of each operation according to the first operation data; and weighting the execution time of each operation according to the weight of each operation to obtain a first adjustment parameter.

In another possible implementation manner, the operation data obtaining unit 701 is configured to execute original operation data sent by the receiving terminal, where the original operation data represents an operation of the terminal on the first recommendation information; and selecting first operation data corresponding to the target operation from the original operation data.

In another possible implementation manner, the first time length obtaining unit 702 is configured to perform emotion recognition on the first operation data to obtain an emotion type, where the emotion type is a positive emotion type or a negative emotion type; and determining the interval duration corresponding to the emotion type as a first interval duration.

a recommendation information transmitting unit 703 configured to perform determining at least one recommended second user account according to the user account logged in by the terminal; and sending second recommendation information to the terminal, wherein the second recommendation information comprises at least one second user account.

Fig. 8 is a block diagram of an information recommendation apparatus provided in an embodiment of the present disclosure, referring to fig. 8, the apparatus includes:

a recommendation information presentation unit 801 configured to perform presentation of a first recommendation interface including first recommendation information sent by a server at a first point in time;

an operation data acquisition unit 802 configured to perform acquisition of first operation data according to the operation detected in the first recommendation interface;

an operation data transmitting unit 803 configured to perform transmission of first operation data to a server for acquiring a first interval duration indicating an interval duration between a time point at which the recommendation information is next transmitted and a first time point, based on the first operation data.

It should be noted that: in the information recommendation device provided in the above embodiment, only the division of the above functional modules is used for illustration, and in practical application, the above functional allocation may be performed by different functional modules according to needs, that is, the internal structure of the terminal is divided into different functional modules, so as to complete all or part of the functions described above. In addition, the information recommending apparatus and the information recommending method provided in the foregoing embodiments belong to the same concept, and specific implementation processes of the information recommending apparatus and the information recommending method are detailed in the method embodiments and are not described herein again.

Fig. 9 shows a block diagram of a terminal 900 according to an exemplary embodiment of the present application. The terminal 900 may be: a smart phone, a tablet computer, an MP3 player (Moving Picture Experts Group Audio Layer III, motion picture expert compression standard audio plane 3), an MP4 (Moving Picture Experts Group Audio Layer IV, motion picture expert compression standard audio plane 4) player, a notebook computer, or a desktop computer. Terminal 900 may also be referred to by other names of user devices, portable terminals, laptop terminals, desktop terminals, etc.

In general, the terminal 900 includes: a processor 901 and a memory 902.

Processor 901 may include one or more processing cores, such as a 4-core processor, an 8-core processor, and the like. The processor 901 may be implemented in at least one hardware form of DSP (Digital Signal Processing ), FPGA (Field-Programmable Gate Array, field programmable gate array), PLA (Programmable Logic Array ). The processor 901 may also include a main processor and a coprocessor, the main processor being a processor for processing data in an awake state, also referred to as a CPU (Central Processing Unit ); a coprocessor is a low-power processor for processing data in a standby state. In some embodiments, the processor 901 may integrate a GPU (Graphics Processing Unit, image processor) for rendering and drawing of content required to be displayed by the display screen. In some embodiments, the processor 901 may also include an AI (Artificial Intelligence ) processor for processing computing operations related to machine learning.

The memory 902 may include one or more computer-readable storage media, which may be non-transitory. The memory 902 may also include high-speed random access memory, as well as non-volatile memory, such as one or more magnetic disk storage devices, flash memory storage devices. In some embodiments, a non-transitory computer readable storage medium in memory 902 is used to store at least one program code for execution by processor 901 to implement the information recommendation methods provided by the method embodiments herein.

In some embodiments, the terminal 900 may further optionally include: a peripheral interface 903, and at least one peripheral. The processor 901, memory 902, and peripheral interface 903 may be connected by a bus or signal line. The individual peripheral devices may be connected to the peripheral device interface 903 via buses, signal lines, or circuit boards. Specifically, the peripheral device includes: at least one of radio frequency circuitry 904, a display 905, a camera assembly 906, audio circuitry 907, a positioning assembly 908, and a power source 909.

The peripheral interface 903 may be used to connect at least one peripheral device associated with an I/O (Input/Output) to the processor 901 and the memory 902. In some embodiments, the processor 901, memory 902, and peripheral interface 903 are integrated on the same chip or circuit board; in some other embodiments, either or both of the processor 901, the memory 902, and the peripheral interface 903 may be implemented on separate chips or circuit boards, which is not limited in this embodiment.

The Radio Frequency circuit 904 is configured to receive and transmit RF (Radio Frequency) signals, also known as electromagnetic signals. The radio frequency circuit 904 communicates with a communication network and other communication devices via electromagnetic signals. The radio frequency circuit 904 converts an electrical signal into an electromagnetic signal for transmission, or converts a received electromagnetic signal into an electrical signal. Optionally, the radio frequency circuit 904 includes: antenna systems, RF transceivers, one or more amplifiers, tuners, oscillators, digital signal processors, codec chipsets, subscriber identity module cards, and so forth. The radio frequency circuit 904 may communicate with other terminals via at least one wireless communication protocol. The wireless communication protocol includes, but is not limited to: metropolitan area networks, various generations of mobile communication networks (2G, 3G, 4G, and 5G), wireless local area networks, and/or WiFi (Wireless Fidelity ) networks. In some embodiments, the radio frequency circuit 904 may also include NFC (Near Field Communication ) related circuits, which are not limited in this application.

The display 905 is used to display a UI (User Interface). The UI may include graphics, text, icons, video, and any combination thereof. When the display 905 is a touch display, the display 905 also has the ability to capture touch signals at or above the surface of the display 905. The touch signal may be input as a control signal to the processor 901 for processing. At this time, the display 905 may also be used to provide virtual buttons and/or a virtual keyboard, also referred to as soft buttons and/or a soft keyboard. In some embodiments, the display 905 may be one, providing a front panel of the terminal 900; in other embodiments, the display 905 may be at least two, respectively disposed on different surfaces of the terminal 900 or in a folded design; in other embodiments, the display 905 may be a flexible display disposed on a curved surface or a folded surface of the terminal 900. Even more, the display 905 may be arranged in an irregular pattern other than rectangular, i.e., a shaped screen. The display 905 may be made of LCD (Liquid Crystal Display ), OLED (Organic Light-Emitting Diode) or other materials.

The camera assembly 906 is used to capture images or video. Optionally, the camera assembly 906 includes a front camera and a rear camera. Typically, the front camera is disposed on the front panel of the terminal and the rear camera is disposed on the rear surface of the terminal. In some embodiments, the at least two rear cameras are any one of a main camera, a depth camera, a wide-angle camera and a tele camera, so as to realize that the main camera and the depth camera are fused to realize a background blurring function, and the main camera and the wide-angle camera are fused to realize a panoramic shooting and Virtual Reality (VR) shooting function or other fusion shooting functions. In some embodiments, camera assembly 906 may also include a flash. The flash lamp can be a single-color temperature flash lamp or a double-color temperature flash lamp. The dual-color temperature flash lamp refers to a combination of a warm light flash lamp and a cold light flash lamp, and can be used for light compensation under different color temperatures.

The audio circuit 907 may include a microphone and a speaker. The microphone is used for collecting sound waves of users and the environment, converting the sound waves into electric signals, and inputting the electric signals to the processor 901 for processing, or inputting the electric signals to the radio frequency circuit 904 for voice communication. For purposes of stereo acquisition or noise reduction, the microphone may be plural and disposed at different portions of the terminal 900. The microphone may also be an array microphone or an omni-directional pickup microphone. The speaker is used to convert electrical signals from the processor 901 or the radio frequency circuit 904 into sound waves. The speaker may be a conventional thin film speaker or a piezoelectric ceramic speaker. When the speaker is a piezoelectric ceramic speaker, not only the electric signal can be converted into a sound wave audible to humans, but also the electric signal can be converted into a sound wave inaudible to humans for ranging and other purposes. In some embodiments, the audio circuit 907 may also include a headphone jack.

The location component 908 is used to locate the current geographic location of the terminal 900 to enable navigation or LBS (Location Based Service, location-based services). The positioning component 908 may be a positioning component based on the United states GPS (Global Positioning System ), the Beidou system of China, the Granati system of Russia, or the Galileo system of the European Union.

The power supply 909 is used to supply power to the various components in the terminal 900. The power supply 909 may be an alternating current, a direct current, a disposable battery, or a rechargeable battery. When the power supply 909 includes a rechargeable battery, the rechargeable battery can support wired or wireless charging. The rechargeable battery may also be used to support fast charge technology.

In some embodiments, terminal 900 can further include one or more sensors 910. The one or more sensors 910 include, but are not limited to: acceleration sensor 911, gyroscope sensor 912, pressure sensor 913, fingerprint sensor 914, optical sensor 915, and proximity sensor 916.

The acceleration sensor 911 can detect the magnitudes of accelerations on three coordinate axes of the coordinate system established with the terminal 900. For example, the acceleration sensor 911 may be used to detect components of gravitational acceleration in three coordinate axes. The processor 901 may control the display 905 to display the user interface in a landscape view or a portrait view according to the gravitational acceleration signal acquired by the acceleration sensor 911. The acceleration sensor 911 may also be used for the acquisition of motion data of a game or a user.

The gyro sensor 912 may detect a body direction and a rotation angle of the terminal 900, and the gyro sensor 912 may collect a 3D motion of the user on the terminal 900 in cooperation with the acceleration sensor 911. The processor 901 may implement the following functions according to the data collected by the gyro sensor 912: motion sensing (e.g., changing UI according to a tilting operation by a user), image stabilization at shooting, game control, and inertial navigation.

The pressure sensor 913 may be provided at a side frame of the terminal 900 and/or at a lower layer of the display 905. When the pressure sensor 913 is provided at a side frame of the terminal 900, a grip signal of the user to the terminal 900 may be detected, and the processor 901 performs left-right hand recognition or shortcut operation according to the grip signal collected by the pressure sensor 913. When the pressure sensor 913 is provided at the lower layer of the display 905, the processor 901 performs control of the operability control on the UI interface according to the pressure operation of the user on the display 905. The operability controls include at least one of a button control, a scroll bar control, an icon control, and a menu control.

The fingerprint sensor 914 is used for collecting the fingerprint of the user, and the processor 901 identifies the identity of the user according to the fingerprint collected by the fingerprint sensor 914, or the fingerprint sensor 914 identifies the identity of the user according to the collected fingerprint. Upon recognizing that the user's identity is a trusted identity, the processor 901 authorizes the user to perform relevant sensitive operations including unlocking the screen, viewing encrypted information, downloading software, paying for and changing settings, etc. The fingerprint sensor 914 may be provided on the front, back or side of the terminal 900. When a physical key or a vendor Logo is provided on the terminal 900, the fingerprint sensor 914 may be integrated with the physical key or the vendor Logo.

The optical sensor 915 is used to collect the intensity of ambient light. In one embodiment, the processor 901 may control the display brightness of the display panel 905 based on the intensity of ambient light collected by the optical sensor 915. Specifically, when the ambient light intensity is high, the display luminance of the display screen 905 is turned up; when the ambient light intensity is low, the display luminance of the display panel 905 is turned down. In another embodiment, the processor 901 may also dynamically adjust the shooting parameters of the camera assembly 906 based on the ambient light intensity collected by the optical sensor 915.

A proximity sensor 916, also referred to as a distance sensor, is typically provided on the front panel of the terminal 900. Proximity sensor 916 is used to collect the distance between the user and the front of terminal 900. In one embodiment, when the proximity sensor 916 detects that the distance between the user and the front face of the terminal 900 gradually decreases, the processor 901 controls the display 905 to switch from the bright screen state to the off screen state; when the proximity sensor 916 detects that the distance between the user and the front surface of the terminal 900 gradually increases, the processor 901 controls the display 905 to switch from the off-screen state to the on-screen state.

Those skilled in the art will appreciate that the structure shown in fig. 9 is not limiting and that more or fewer components than shown may be included or certain components may be combined or a different arrangement of components may be employed.

Fig. 10 is a schematic structural diagram of a server according to an embodiment of the present application, where the server 1000 may have a relatively large difference due to different configurations or performances, and may include one or more processors (central processing units, CPU) 1001 and one or more memories 1002, where at least one program code is stored in the memories 1002, and the at least one program code is loaded and executed by the processors 1001 to implement the information recommendation method provided in the foregoing method embodiments. Of course, the server may also have a wired or wireless network interface, a keyboard, an input/output interface, and other components for implementing the functions of the device, which are not described herein.

In an exemplary embodiment, a computer readable storage medium, such as a memory comprising instructions executable by a processor in a terminal to perform the information recommendation method in the above embodiment is also provided. Alternatively, the computer readable storage medium may be a non-transitory computer readable storage medium, for example, a ROM, a Random Access Memory (RAM), a CD-ROM, a magnetic tape, a floppy disk, an optical data storage device, and the like.

In an exemplary embodiment, a computer program product is also provided, which, when instructions in the computer program product are executed by a processor of an electronic device, enables the electronic device to perform the information recommendation method in the above-described embodiments.

Other embodiments of the disclosure will be apparent to those skilled in the art from consideration of the specification and practice of the disclosure disclosed herein. This application is intended to cover any adaptations, uses, or adaptations of the disclosure following, in general, the principles of the disclosure and including such departures from the present disclosure as come within known or customary practice within the art to which the disclosure pertains. It is intended that the specification and examples be considered as exemplary only, with a true scope and spirit of the disclosure being indicated by the following claims.

It is to be understood that the present disclosure is not limited to the precise arrangements and instrumentalities shown in the drawings, and that various modifications and changes may be effected without departing from the scope thereof. The scope of the present disclosure is limited only by the appended claims.

Claims

1. An information recommendation method, the method comprising:

acquiring an emotion recognition model, and reading the weight of each operation from the emotion recognition model, wherein the emotion recognition model comprises at least one weight of the operation, and the weight represents the importance degree of the operation on an emotion recognition result;

acquiring the execution times of each operation according to the first operation data, and carrying out weighting processing on the execution times of each operation according to the weight of each operation to obtain a first adjustment proportion; determining the sum of the weights of each operation as a second adjustment proportion; acquiring the execution time length of each operation according to the first operation data; weighting the execution duration of each operation according to the weight of each operation to obtain a third adjustment proportion;

according to the first adjustment proportion, the second adjustment proportion and the third adjustment proportion, adjusting the reference time length to obtain a first adjustment parameter of the first operation data;

According to the first adjustment parameters, the second interval duration is adjusted to obtain a first interval duration, wherein the first interval duration represents the interval duration between the time point of next sending of recommended information to the terminal and the first time point;

when the first interval duration after the first time point is reached, second recommendation information is sent to the terminal, and the terminal is used for displaying the second recommendation information;

receiving third operation data sent by the terminal, wherein the third operation data is acquired by the terminal according to the detected operation of an interface displaying the second recommendation information; and acquiring a third interval duration according to the third operation data, wherein the third interval duration represents the interval duration between the time point of sending the recommended information to the terminal after the second recommended information and the time point of sending the second recommended information.

2. The information recommendation method according to claim 1, wherein the adjusting the second interval duration to obtain the first interval duration includes:

3. The information recommendation method according to claim 1 or 2, wherein before said acquiring the stored second interval duration, the method further comprises:

4. The information recommendation method according to claim 1 or 2, wherein before said acquiring the stored second interval duration, the method further comprises:

5. The information recommendation method according to claim 1, wherein the method further comprises:

6. The information recommendation method according to claim 5, wherein the training process of the emotion recognition model includes:

7. The information recommendation method according to claim 1, wherein the method further comprises:

8. The information recommendation method according to claim 7, wherein the training process of the emotion recognition model includes:

9. The information recommendation method according to claim 1, wherein the method further comprises:

10. The information recommendation method according to claim 9, wherein the training process of the emotion recognition model includes:

11. The information recommendation method according to claim 1, wherein the acquiring the first operation data includes:

12. The information recommendation method according to claim 1, wherein the method further comprises:

13. The information recommendation method according to claim 1, wherein the first recommendation information includes at least one first user account recommended according to a user account registered by the terminal;

the sending the second recommendation information to the terminal includes:

14. An information recommendation method, the method comprising:

the first operation data is sent to the server, the server is used for acquiring stored second interval duration, the second interval duration is interval duration between a time point of sending third recommendation information to a terminal and the first time point, and the third recommendation information is information sent to the terminal before the first recommendation information; acquiring an emotion recognition model, and reading the weight of each operation from the emotion recognition model, wherein the emotion recognition model comprises at least one weight of the operation, and the weight represents the importance degree of the operation on an emotion recognition result; acquiring the execution times of each operation according to the first operation data, and carrying out weighting processing on the execution times of each operation according to the weight of each operation to obtain a first adjustment proportion; determining the sum of the weights of each operation as a second adjustment proportion; acquiring the execution time length of each operation according to the first operation data; weighting the execution duration of each operation according to the weight of each operation to obtain a third adjustment proportion; adjusting the reference time length according to the first adjustment proportion, the second adjustment proportion and the third adjustment proportion to obtain a first adjustment parameter of the first operation data; according to the first adjustment parameters, adjusting the second interval duration to obtain a first interval duration, wherein the first interval duration represents the interval duration between the time point of next sending of second recommended information and the first time point;

According to the detected operation on the interface displaying the second recommendation information, third operation data are obtained, and the third operation data are sent to the server; the server is further configured to obtain a third interval duration according to the third operation data after receiving the third operation data, where the third interval duration represents an interval duration between a time point when the server sends the recommended information to the terminal after sending the second recommended information and a time point when the server sends the second recommended information.

15. An information recommendation device, characterized in that the device comprises:

the first time length acquisition unit comprises an acquisition subunit and an adjustment subunit;

the acquisition subunit is configured to perform acquisition of a stored second interval duration, wherein the second interval duration is an interval duration between a time point of sending third recommended information to the terminal and the first time point, and the third recommended information is information sent to the terminal before the first recommended information;

The adjusting subunit is configured to acquire an emotion recognition model, and read the weight of each operation from the emotion recognition model, wherein the emotion recognition model comprises the weight of at least one operation, and the weight represents the importance degree of the operation on the emotion recognition result; acquiring the execution times of each operation according to the first operation data, and carrying out weighting processing on the execution times of each operation according to the weight of each operation to obtain a first adjustment proportion; determining the sum of the weights of each operation as a second adjustment proportion; acquiring the execution time length of each operation according to the first operation data; weighting the execution duration of each operation according to the weight of each operation to obtain a third adjustment proportion; adjusting the reference time length according to the first adjustment proportion, the second adjustment proportion and the third adjustment proportion to obtain a first adjustment parameter of the first operation data; according to the first adjustment parameters, the second interval duration is adjusted to obtain a first interval duration, wherein the first interval duration represents the interval duration between the time point of next sending of recommended information to the terminal and the first time point;

A recommendation information transmitting unit configured to transmit second recommendation information to the terminal for displaying the second recommendation information when the first interval duration after the first time point is reached;

a module for performing the steps of: receiving third operation data sent by the terminal, wherein the third operation data is acquired by the terminal according to the detected operation of an interface displaying the second recommendation information; and acquiring a third interval duration according to the third operation data, wherein the third interval duration represents the interval duration between the time point of sending the recommended information to the terminal after the second recommended information and the time point of sending the second recommended information.

16. The information recommendation device of claim 15, wherein,

the adjusting subunit is configured to perform emotion recognition on the first operation data to obtain an emotion type, wherein the emotion type is a positive emotion type or a negative emotion type; if the emotion type is a forward emotion type, reducing the second interval duration to obtain the first interval duration; or if the emotion type is a negative emotion type, increasing the second interval duration to obtain the first interval duration.

17. The information recommendation device according to claim 15 or 16, wherein said device further comprises:

18. The information recommendation device according to claim 15 or 16, wherein said device further comprises:

19. The information recommendation device of claim 15, wherein,

the adjusting subunit is configured to execute the first operation data to obtain the execution times of each operation; and weighting the execution times of each operation according to the weight of each operation to obtain the first adjustment parameter.

20. The information recommendation device of claim 19, wherein the training process of emotion recognition model comprises:

21. The information recommendation device of claim 15, wherein,

the adjusting subunit is configured to determine the sum of the weights of each operation as a first adjusting parameter corresponding to the first operation data.

22. The information recommendation device of claim 21, wherein the training process of emotion recognition model comprises:

23. The information recommendation device of claim 15, wherein,

the adjusting subunit is configured to execute the first operation data to obtain the execution duration of each operation; and weighting the execution time of each operation according to the weight of each operation to obtain the first adjustment parameter.

24. The information recommendation device of claim 23, wherein the training process of emotion recognition model comprises:

25. The information recommendation device of claim 15, wherein,

the operation data acquisition unit is configured to perform receiving original operation data sent by the terminal, wherein the original operation data represents the operation of the terminal on the first recommendation information; and selecting first operation data corresponding to the target operation from the original operation data.

26. The information recommendation device of claim 15, wherein,

the first time length obtaining unit is configured to perform emotion recognition on the first operation data to obtain an emotion type, wherein the emotion type is a positive emotion type or a negative emotion type; and determining the interval duration corresponding to the emotion type as the first interval duration.

27. The information recommendation device according to claim 15, wherein the first recommendation information includes at least one first user account recommended according to a user account registered by the terminal;

28. An information recommendation device, characterized in that the device comprises:

an operation data transmitting unit configured to perform transmission of the first operation data to the server, the server being configured to acquire a stored second interval duration, the second interval duration being an interval duration between a time point at which third recommendation information is transmitted to a terminal and the first time point, the third recommendation information being information transmitted to the terminal before the first recommendation information; acquiring an emotion recognition model, and reading the weight of each operation from the emotion recognition model, wherein the emotion recognition model comprises at least one weight of the operation, and the weight represents the importance degree of the operation on an emotion recognition result; acquiring the execution times of each operation according to the first operation data, and carrying out weighting processing on the execution times of each operation according to the weight of each operation to obtain a first adjustment proportion; determining the sum of the weights of each operation as a second adjustment proportion; acquiring the execution time length of each operation according to the first operation data; weighting the execution duration of each operation according to the weight of each operation to obtain a third adjustment proportion; adjusting the reference time length according to the first adjustment proportion, the second adjustment proportion and the third adjustment proportion to obtain a first adjustment parameter of the first operation data; according to the first adjustment parameters, adjusting the second interval duration to obtain a first interval duration, wherein the first interval duration represents the interval duration between the time point of next sending of second recommended information and the first time point;

A module for performing the steps of: according to the detected operation on the interface displaying the second recommendation information, third operation data are obtained, and the third operation data are sent to the server; the server is further configured to obtain a third interval duration according to the third operation data after receiving the third operation data, where the third interval duration represents an interval duration between a time point when the server sends the recommended information to the terminal after sending the second recommended information and a time point when the server sends the second recommended information.

29. An electronic device, comprising:

a processor;

a memory for storing the processor-executable instructions;

wherein the processor is configured to execute the instructions to implement the information recommendation method of any one of claims 1 to 14.

30. A storage medium, characterized in that instructions in the storage medium, when executed by a processor of an electronic device, enable the electronic device to perform the information recommendation method of any one of claims 1 to 14.