Video ringtone recommendation method and electronic device

Abstract

The present disclosure provides a video ringback tone recommendation method and an electronic device, and relates to the technical field of computers, which comprises the following steps: determining a target video ringback tone from a plurality of selectable video ringback tones according to ringback tone features of the plurality of selectable video ringback tones, user features of each friend user of a target user, and user features of the target user; and recommending the target video ringback tone to the target user and / or the friend user. The video processing scheme provided in the present disclosure can enrich the functions of the video ringback tone.

Description

Technical Field

[0001] This invention relates to the field of computer technology, and in particular to a video ringback tone recommendation method and electronic device. Background Technology

[0002] In mobile communication scenarios, when a calling user makes a call and requests a call, a video content will be displayed on the calling user's terminal before the called user answers the call. This video content can be called a video ringback tone.

[0003] In related technologies, the ringback tone server can recommend video ringback tones to the called user. The called user can choose a video ringback tone according to their own preferences so that the calling user can watch the video ringback tone they set after initiating a call request.

[0004] However, the video ringback tone recommendation solutions provided in related technologies are generally targeted at individual potential users, making recommendations based on the potential user's interests, resulting in a low conversion rate for video ringback tone settings. Summary of the Invention

[0005] This disclosure provides a video ringback tone recommendation method and an electronic device to at least solve the problem of limited video ringback tone functionality in related technologies. The technical solution of this disclosure is as follows:

[0006] According to a first aspect of this disclosure, a video ringback tone recommendation method is provided, comprising:

[0007] Based on the ringback tone characteristics of multiple optional video ringback tones, the user characteristics of each friend of the target user, and the user characteristics of the target user, the target video ringback tone is determined from the multiple optional video ringback tones;

[0008] The target video ringback tone will be recommended to the target user and / or the user's friends.

[0009] According to a second aspect of this disclosure, an electronic device is provided, comprising:

[0010] Processor; and

[0011] Stored program memory,

[0012] The program includes instructions that, when executed by the processor, cause the processor to perform the method as described in the first aspect.

[0013] The video ringback tone recommendation method and electronic device disclosed herein can, during the video ringback tone recommendation process, determine the target video ringback tones to be recommended to the target user and / or friends based on the user characteristics of the target user and the user characteristics of the target user's friends, thereby increasing the probability that the target user and / or friends will set the recommended target video ringback tone as their own video ringback tone and improving the conversion rate of the target user and / or friends' video ringback tone setting. Attached Figure Description

[0014] Further details, features, and advantages of this disclosure are disclosed in the following description of exemplary embodiments in conjunction with the accompanying drawings, in which:

[0015] Figure 1 A schematic diagram illustrating an implementation scenario of a video ringback tone recommendation method according to an exemplary embodiment of this disclosure is shown.

[0016] Figure 2 A flowchart illustrating an exemplary embodiment of the present disclosure of a video ringback tone recommendation method is shown;

[0017] Figure 3 A flowchart illustrating a method for determining a target video ringback tone according to an exemplary embodiment of this disclosure is shown.

[0018] Figure 4 A flowchart illustrating a method for determining user metrics of friend users, as exemplified in this disclosure, is shown.

[0019] Figure 5 A schematic diagram of the display interface of a user terminal according to an exemplary embodiment of the present disclosure is shown;

[0020] Figure 6 A schematic diagram of the display interface of another user terminal according to an exemplary embodiment of the present disclosure is shown;

[0021] Figure 7 A block diagram of a video ringback tone recommendation device according to an exemplary embodiment of the present disclosure is shown;

[0022] Figure 8 A schematic block diagram of a chip according to an exemplary embodiment of the present disclosure is shown;

[0023] Figure 9 A structural block diagram of an exemplary electronic device that can be used to implement embodiments of the present disclosure is shown. Detailed Implementation

[0024] Embodiments of this disclosure will now be described in more detail with reference to the accompanying drawings. While some embodiments of this disclosure are shown in the drawings, it should be understood that this disclosure can be implemented in various forms and should not be construed as limited to the embodiments set forth herein. Rather, these embodiments are provided to provide a more thorough and complete understanding of this disclosure. It should be understood that the accompanying drawings and embodiments of this disclosure are for illustrative purposes only and are not intended to limit the scope of protection of this disclosure.

[0025] It should be understood that the steps described in the method embodiments of this disclosure may be performed in different orders and / or in parallel. Furthermore, the method embodiments may include additional steps and / or omit the steps shown. The scope of this disclosure is not limited in this respect.

[0026] The term "comprising" and its variations as used herein are open-ended, meaning "including but not limited to". The term "based on" means "at least partially based on". The term "one embodiment" means "at least one embodiment"; the term "another embodiment" means "at least one additional embodiment"; the term "some embodiments" means "at least some embodiments". Definitions of other terms will be given in the description below. It should be noted that the concepts of "first", "second", etc., used in this disclosure are only used to distinguish different devices, modules, or units, and are not intended to limit the order of functions performed by these devices, modules, or units or their interdependencies.

[0027] It should be noted that the terms "a" and "a plurality of" used in this disclosure are illustrative rather than restrictive, and those skilled in the art should understand that, unless otherwise expressly indicated in the context, they should be understood as "one or more".

[0028] The names of messages or information exchanged between multiple devices in the embodiments of this disclosure are for illustrative purposes only and are not intended to limit the scope of such messages or information.

[0029] In related technologies, when a user sets a video ringback tone, the user terminal can respond to the acquisition of the video ringback tone setting operation by sending a ringback tone acquisition request to the ringtone server. The ringtone server can select a video ringback tone that the user is interested in based on the user's interest characteristics and send it to the user terminal for the user to set the video ringback tone.

[0030] However, the video ringback tone recommendation solutions provided in related technologies focus more on the interests of individual potential users who want to set up video ringback tones. Recommendations are made based on the interests of potential users, resulting in a low conversion rate for setting up video ringback tones.

[0031] To overcome the above problems, an exemplary embodiment of this disclosure provides a video ringback tone recommendation method. This method can determine a target video ringback tone from among multiple optional video ringback tones based on the ringback tone characteristics of multiple optional video ringback tones, the user characteristics of each friend of the target user, and the user characteristics of the target user. The target video ringback tone is then recommended to the target user and / or their friends. This method can determine the video ringback tones to be recommended to the target user and / or their friends based on the characteristics of the target user and their friends, thereby improving the conversion rate of video ringback tone settings.

[0032] Figure 1 A schematic diagram illustrating an implementation scenario of the video ringback tone recommendation method, an exemplary embodiment of this disclosure, is shown. For example... Figure 1 As shown, this implementation scenario 100 includes a user terminal 101 and a ringtone server 102. The user terminal 101 can be a terminal device used by a user, which allows the user to set video ringtones. This terminal device can include devices such as smartphones, personal computers, tablets, or wearable devices. The ringtone server 102 can be a server providing ringtone data services. It should be understood that... Figure 1 The implementation scenario shown is only an example. In actual scenarios, in addition to the calling user terminal and the server, there may be other devices.

[0033] like Figure 1 As shown, the calling user terminal 101 can establish a communication link with the ringtone server 102. For example, the calling user terminal 101 and the ringtone server 102 can be connected to the same wired or wireless network to generate video ringtones.

[0034] Figure 2 A flowchart illustrating an exemplary embodiment of the present disclosure of a video ringback tone recommendation method is shown, such as... Figure 2 As shown, the method in this embodiment of the disclosure may include:

[0035] Step S201: Based on the ringback tone characteristics of multiple optional video ringback tones, the user characteristics of each friend of the target user, and the user characteristics of the target user, determine the target video ringback tone from the multiple optional video ringback tones.

[0036] Step S202: Recommend the target video ringback tone to the target user and / or friend users;

[0037] In summary, the video ringback tone recommendation method provided in this disclosure can determine the target video ringback tone to be recommended to the target user and / or friends based on the user characteristics of the target user and the user characteristics of the target user's friends during the video ringback tone recommendation process. This increases the probability that the target user and / or friends will set the recommended target video ringback tone as their own video ringback tone, thereby increasing the conversion rate of video ringback tone setting for the target user and / or friends.

[0038] The following are Figure 2 The specific implementation methods of each step in the illustrated embodiment are described in detail below:

[0039] In step S201, the target video ringback tone is determined from the multiple optional video ringback tones based on the ringback tone characteristics of the multiple optional video ringback tones, the user characteristics of each friend of the target user, and the user characteristics of the target user.

[0040] In this embodiment of the disclosure, the target user is the user who sets a video ringback tone; the optional video ringback tones include those recommended to the user from the video ringback tone library.

[0041] It should be noted that the optional video ringback tone features may include video ringback tone type (such as cute pets, scenery, or animation), video ringback tone duration, video ringback tone exposure, user setting rate, and / or preview rate. The user features of the friend user include user attribute information and interaction information between the friend user and the video ringback tone. User attribute information may include user age, gender, and education level, while interaction information may include: the video types of video ringback tones set by the friend user, the frequency with which the friend user sets video ringback tones, and / or the storage duration of the video ringback tones set by the friend user. The user features of the target user are similar to those of the friend user, and will not be elaborated upon in this embodiment.

[0042] In one alternative implementation, such as Figure 3 As shown, the process by which the ringtone server determines the target video ringtone from multiple optional video ringtones based on the ringtone characteristics of multiple optional video ringtones, the user characteristics of each friend of the target user, and the user characteristics of the target user, may include:

[0043] Step S301: Based on the ringback tone characteristics of multiple optional video ringback tones, the user characteristics of each friend user, and the user characteristics of the target user, determine multiple candidate video ringback tones and the recommended value of each candidate video ringback tone.

[0044] In this embodiment of the disclosure, the multiple candidate video ringback tones are optional video ringback tones that are of interest to the friends, or optional video ringback tones that are of interest to both the target user and the friends. The recommendation value is used to characterize the degree of interest of multiple friends in the candidate video ringback tones.

[0045] In one optional implementation, if multiple candidate video ringback tones are optional video ringback tones that are of interest to both the target user and their friends, then the ringtone server determines the target video ringback tone from among the multiple optional video ringback tones based on the ringback tone features of the multiple optional video ringback tones, the user features of each friend of the target user, and the user features of the target user. This process may include: determining multiple candidate video ringback tones based on the user features of each friend, the user features of the target user, the ringback tone features of the multiple optional video ringback tones, and a pre-trained recall model; further, inputting the user features of each friend and the ringback tone features of the multiple candidate video ringback tones into a pre-trained interest value prediction model to determine the interest index value of each friend for each candidate video ringback tone; then, determining the recommendation value of each candidate video ringback tone based on the interest index value of each friend for each candidate video ringback tone and the user index of each friend, where the friend's user index is determined based on the friend's ringback tone setting willingness coefficient and the call probability value of the friend calling the target user. On the one hand, when there is a large number of video ringback tones to choose from, the system can quickly identify candidate video ringback tones that both the target user and their friends may be interested in based on their user characteristics, thus improving the efficiency and reliability of recommending video ringback tones to the target user. On the other hand, by considering the user metrics of friends and their interest in the candidate video ringback tones, the system can determine the recommendation value for each candidate video ringback tone, which can further increase the willingness of the target user and / or their friends to set video ringback tones, thereby improving the conversion rate of video ringback tone setting for the target user and / or their friends.

[0046] The process by which the ringtone server determines the recommendation value of each candidate video ringtone based on each friend's interest index value for each candidate video ringtone and each friend's user index may include: for each friend, determining the friend's interest index value for each candidate video ringtone and the product of the friend's user index to obtain the initial recommendation value of each candidate video ringtone for the target friend; determining the sum of the initial recommendation values ​​of each candidate video ringtone for multiple target friends to obtain the recommendation value of each candidate video ringtone.

[0047] For example, the ringtone server determines the recommendation value for each candidate video ringtone based on each friend's interest index value for each candidate video ringtone, as well as each friend's user index and a preset formula. The preset formula is as follows:

[0048]

[0049] Among them, STR′(vrbt k Score(U) represents the recommended value of the k-th candidate video ringback tone. j ,? vrbt k Value(U) represents the interest index value of the j-th friend user towards the k-th candidate video ringback tone. j Let represent the user metrics of the j-th friend user, where J is the number of multiple friend users.

[0050] In one optional implementation, if multiple candidate video ringback tones are optional video ringback tones that a friend user is interested in, then the ringtone server determines the target video ringback tone from the multiple optional video ringback tones based on the ringback tone features of the multiple optional video ringback tones, the user features of each friend user of the target user, and the user features of the target user. This process may include: determining multiple candidate video ringback tones based on the user features of each friend user, the ringback tone features of the multiple optional video ringback tones, and a pre-trained recall model; further, inputting the user features of each friend user and the ringback tone features of the multiple candidate video ringback tones into a pre-trained interest value prediction model to determine the interest index value of each friend user for each candidate video ringback tone; and then, determining the recommendation value of each candidate video ringback tone based on the interest index value of each friend user for each candidate video ringback tone and the user index of each friend user. On the one hand, when there is a large number of video ringback tones to choose from, the user characteristics of friends can be used to quickly identify candidate video ringback tones that friends may be interested in, thereby improving the efficiency and reliability of recommending video ringback tones to target users. On the other hand, by considering the user metrics of friends and their interest metrics in candidate video ringback tones, the recommendation value of each candidate video ringback tones can be determined, which can further increase the willingness of target users and / or friends to set video ringback tones, thereby improving the conversion rate of video ringback tones set by target users and / or friends.

[0051] In one optional implementation, the process of determining multiple candidate video ringback tones based on the user characteristics of each friend user, the ringback tone characteristics of multiple optional video ringback tones, and a pre-trained recall model may include: determining the number of candidate video ringback tones associated with each friend user and the number of candidate video ringback tones associated with the target user based on the user metrics of the target user, the user metrics of each friend user, and the number of multiple candidate video ringback tones; further, inputting the user characteristics of each friend user and the ringback tone characteristics of multiple optional video ringback tones into the pre-trained recall model to obtain at least one candidate video ringback tone specified by the number of candidate video ringback tones associated with each friend user; inputting the user characteristics of the target user and the ringback tone characteristics of multiple optional video ringback tones into the pre-trained recall model to obtain at least one candidate video ringback tone specified by the number of candidate video ringback tones associated with the target user; finally, combining the at least one candidate video ringback tone specified by the number of candidate video ringback tones associated with each friend user and the at least one candidate video ringback tone specified by the number of candidate video ringback tones associated with the target user to obtain multiple candidate video ringback tones. In this embodiment, the user metrics of the target user are preset parameters, which can be 1. The number of candidate video ringback tones can be determined based on actual needs, and this embodiment does not limit this. By combining the user metrics of each friend or target user, different numbers of candidate video ringback tones can be selected from multiple available options. Among these candidate video ringback tones, those associated with friend users or target users with high user metrics have a higher number of candidate video ringback tones, which can further stimulate the willingness of target users or friend users to set video ringback tones.

[0052] In one optional implementation, the process by which the ringtone server determines the number of candidate video ringtones associated with each target friend user and the number of candidate video ringtones associated with the target user, based on the target user's user metrics, the user metrics of each target friend user, and the number of multiple candidate video ringtones, may include: determining the target user's user metrics and the ratio of the user metrics of multiple target friend users to the total user metrics of the target user, to obtain a weight value associated with the friend user; determining the user metrics of each target friend user and the ratio of the user metrics of multiple target friend users to the total user metrics of the target user, to obtain a weight value associated with each target friend user; determining the number of multiple candidate video ringtones and the product of the weight values ​​associated with each target friend user, to obtain the number of candidate video ringtones associated with each target friend user; and determining the number of multiple candidate video ringtones and the product of the weight values ​​associated with the target user, to obtain the number of candidate video ringtones associated with the target user.

[0053] For example, if the number of candidate video ringback tones can be L, then the number of candidate video ringback tones associated with the target friend user can be:

[0054]

[0055] Among them, Recall(U j Value(U) represents the number of candidate video ringback tones associated with the j-th target friend user. j ) represents the user metrics of the j-th target friend user. This represents the total user metrics for multiple target friend users and target users.

[0056] In one optional implementation, the process by which the ringtone server determines the number of candidate video ringtones associated with each target friend user and the number of candidate video ringtones associated with the target user, based on the user metrics of the target user, the user metrics of each target friend user, and the number of multiple candidate video ringtones, may include: determining the user metrics of each target friend user and the ratio of the user metrics of multiple target friend users to the total user metrics of the target user, to obtain a weight value associated with each target friend user; determining the product of the multiple candidate video ringtones and the weight values ​​associated with each target friend user, to obtain the number of candidate video ringtones associated with each target friend user; and determining the difference between the multiple candidate video ringtones and the total number of candidate video ringtones associated with at least one target friend user, to obtain the number of candidate video ringtones associated with the target user.

[0057] It should be noted that, in this embodiment of the disclosure, before determining the target video ringback tone from multiple optional video ringback tones based on the ringback tone characteristics of multiple optional video ringback tones, the user characteristics of each friend of the target user, and the user characteristics of the target user, the ringback tone server may also pre-determine the user indicators of each friend of the target user, such as... Figure 4 As shown,

[0058] Step S401: Determine the ringback tone setting willingness coefficient for each friend user based on the historical ringback tone setting information of each friend user associated with the target user and the pre-trained ringback tone setting willingness prediction model.

[0059] In this embodiment of the disclosure, historical ringback tone setting information is used to characterize the ringback tone setting status within a historical time period. The historical time period may include multiple historical time units, and the duration of each time unit can be determined based on actual needs; this embodiment of the disclosure does not limit this. For example, the duration of a time unit may be 1 day, 2 days, or 1 week, etc. The historical ringback tone setting information may include the number of video ringback tone settings in each historical time unit. Optionally, the historical ringback tone setting information may also include historical time units where video ringback tone promotions exist, or historical time units that belong to holidays, etc.

[0060] It should be noted that, in the embodiments of this disclosure, the user's ringback tone setting intention is usually a time series. Optionally, the ringback tone setting intention prediction model trained in the embodiments of this disclosure may include a time series model that can achieve time series prediction. For example, the ringback tone setting intention prediction model may include a Long Short-Term Memory (LSTM) network model.

[0061] In one optional implementation, the process of the ringtone server training the ringback tone setting intention prediction model may include: obtaining sample ringback tone setting information of each friend user associated with the target user in multiple sample units of time; determining a first training dataset based on the sample ringback tone setting information of each friend user in multiple sample units of time; and training the ringback tone setting intention prediction model to be trained using the first training dataset until the ringback tone setting intention prediction model converges, thereby obtaining the trained ringback tone setting intention prediction model; wherein, the number of multiple sample units of time can be determined based on actual needs, and this embodiment of the disclosure does not limit this. For example, if the unit of time is 1 day, the multiple sample units of time may include 365 days.

[0062] It should be noted that, in this embodiment of the disclosure, multiple sample time units may include multiple time units prior to the time unit in which the ringback tone setting intention prediction model is trained. Optionally, the ringback tone setting information may change in each time unit. In order to train a ringback tone setting intention prediction model that can more accurately predict the ringback tone setting intention coefficient for each time unit, the training time for the ringback tone setting intention prediction model can be the beginning of each time unit. For example, if the time unit is 1 day, at the beginning of each day, sample ringback tone setting information of each friend user associated with the target user for multiple days prior to that day can be obtained, and the ringback tone setting intention prediction model can be trained to more accurately predict the ringback tone setting intention coefficient for that day.

[0063] In one optional implementation, the ringtone server trains the ringback tone setting intention prediction model at the start time of the current unit of time. The ringtone server can use multiple units of time preceding the current unit of time as multiple sample units of time to train the ringback tone setting intention prediction model associated with the current unit of time. The current unit of time is the unit of time in which the terminal device obtains the video ringback tone setting operation. During the process of the server training the ringback tone setting intention prediction model associated with the current unit of time, the process of determining the first training dataset based on the sample ringback tone setting information of each friend user in multiple sample units of time may include: for each friend user, based on the ringback tone setting... The correspondence between the number of sample ringback tone settings and the willingness coefficient is established. The willingness coefficient corresponding to the number of sample ringback tone settings in the sample ringback tone setting information for each sample unit of time is determined. The willingness coefficients of sample ringback tone settings for multiple consecutive sample units of time are determined as the first sample data. The willingness coefficient of sample ringback tone settings for the next sample unit of time after multiple consecutive sample units of time is determined as the first label data, thus obtaining a set of first training data. This process of determining a set of first training data is repeated to obtain multiple sets of first training data. Furthermore, multiple sets of first training data associated with each friend user can be combined to obtain a first training dataset for training the ringback tone setting willingness prediction model. The number of consecutive sample units of time can be determined based on actual needs, and this embodiment does not limit this, for example, seven consecutive days or four consecutive weeks. It is understood that the higher the willingness coefficient of ringback tone settings, the stronger the user's tendency to set a video ringback tone.

[0064] For example, if the unit of time is 1 day, and the ringtone server obtains sample ringtone setting information for each friend user associated with target user A in the 365 days prior to that day, then for each friend user, based on the correspondence between the number of ringtone settings and the ringtone setting willingness coefficient, the sample ringtone setting willingness coefficient corresponding to the number of sample ringtone settings each day in the 365 days prior to that day can be determined. Furthermore, starting from the first day of the 365 days prior to that day, the sample ringtone setting willingness coefficients for 5 consecutive days are determined as the first sample data, and the sample ringtone setting willingness coefficients for the next day after 5 consecutive days are determined as the first label data, thus obtaining multiple sets of first training data for each friend user, and combining the multiple sets of first training data for each friend user to obtain the first training dataset.

[0065] It should be noted that the correspondence between the number of times a ringback tone is set and the ringback tone setting willingness coefficient can be determined based on actual needs, and this embodiment does not limit this. For example, the correspondence between the number of times a ringback tone is set and the ringback tone setting willingness coefficient may include: if the number of times a ringback tone is set is 0, the corresponding ringback tone setting willingness coefficient is 0.5; if the number of times a ringback tone is set is greater than or equal to 1 and less than or equal to 2, the corresponding ringback tone setting willingness coefficient is 0.8; if the number of times a ringback tone is set is greater than 2, the corresponding ringback tone setting willingness coefficient may be 1.

[0066] For example, one set of first training data may include: first sample data of 1.0, 0.5, 0.8, 0.5, 0.5, with the corresponding first label data of 1.0; another set of first training data may include: first sample data of 0.5, 0.8, 0.5, 0.5, 1.0, with the corresponding first label data of 0.5; yet another set of first training data may include: first sample data of 0.8, 0.5, 0.5, 1.0, 0.5, with the corresponding first label data of 1.0.

[0067] Understandably, the process by which the ringtone server determines the ringtone setting intention coefficient for each friend user based on their historical ringtone settings and a pre-trained ringtone setting intention prediction model can include: inputting the ringtone settings information of each friend user from multiple historical time units prior to the current time unit into the ringtone setting intention prediction model associated with the current time unit to obtain the first ringtone setting intention coefficient for each friend user at the current time unit; and determining the ringtone setting intention coefficient for each friend user at the current time unit based on the first ringtone setting intention coefficient for each friend user at the current time unit. The number of multiple historical time units can be the same as the number of consecutive sample time units used when training the ringtone setting intention prediction model; for example, if the consecutive sample time units are 7 consecutive days, then the multiple historical time units are 7 consecutive days. The accuracy of determining the ringtone setting intention coefficient for each friend user at the current time unit can be improved by using a pre-trained ringtone setting intention prediction model associated with the current time unit.

[0068] In one optional implementation, the process of determining the ringback tone setting willingness coefficient of each friend user in the current unit time based on the first ringback tone setting willingness coefficient of each friend user in the current unit time may include: determining the ringback tone setting willingness coefficient of each friend user in the current unit time based on the first ringback tone setting willingness coefficient of each friend user in the current unit time; and determining the ringback tone setting willingness coefficient of each friend user in the current unit time based on a ringback tone setting willingness prediction model associated with the current unit time, thereby improving the efficiency of determining the ringback tone setting willingness coefficient of each friend user in the current unit time.

[0069] It should be noted that, in this embodiment of the disclosure, the probability of a user calling the target user in each unit of time will affect the probability of calling the target user in subsequent units of time. For example, after setting a video ringback tone in a certain unit of time, the willingness to set a video ringback tone in subsequent units of time will increase or decrease. Therefore, the ringback tone setting willingness coefficient of the user in a certain unit of time and the ringback tone setting willingness coefficient of multiple units of time after that unit of time can be comprehensively considered to determine the ringback tone setting willingness coefficient of the user in that unit of time, so as to improve the accuracy of the determined ringback tone setting willingness coefficient of the user in that unit of time. At the beginning of the current unit of time, the ringtone server can also determine multiple units of time before the current unit of time as multiple sample units of time and train a ringback tone setting willingness prediction model associated with the target unit of time.

[0070] It should be noted that when the ringtone server trains the ringtone setting intention prediction model associated with the target unit time, the method for determining the first training dataset in the process of determining the first training data based on the sample ringtone setting information of each friend user in multiple sample units of time can include: determining the sample ringtone setting intention coefficients of multiple consecutive sample units of time as the first sample data, and determining the sample ringtone setting intention coefficients of the nth unit of time after multiple consecutive sample units of time as the first label data, thus obtaining a set of first training data. Here, n is the number of unit time intervals between the target unit time and the current unit time.

[0071] For example, if the unit time is 1 day, the current unit time is day t, and the target unit time is day t+3, and the ringtone server obtains the sample ringtone settings information of each friend user associated with the target user A in the 365 days prior to the current day, then the method for determining the first training data may include: starting from the first day of the 365 days prior to the current day, determining the sample ringtone setting willingness coefficients for 5 consecutive days as the first sample data, and determining the sample ringtone setting willingness coefficients for the 8th day after the 5 consecutive days as the first label data, thus obtaining a set of first training data.

[0072] Understandably, before the ringtone server determines the ringtone setting willingness coefficient for each friend user based on the first ringtone setting willingness coefficient for each friend user at the current unit time, the ringtone server can input the number of ringtone settings for each friend user at multiple historical units of time before the current unit time into the ringtone setting willingness prediction model associated with multiple target units of time to obtain the second ringtone setting willingness coefficient for each friend user at each target unit time. Here, the target unit time is later than the current unit time, and the target unit time can be multiple consecutive units of time after the current unit time. The number of target units of time can be determined based on actual needs, and this embodiment does not limit this. For example, the number of target units of time can be 7 or 5.

[0073] Furthermore, the process by which the ringtone server determines each friend's ringtone setting intention coefficient for the current unit of time, based on the first ringtone setting intention coefficient for each friend in the current unit of time, can include: taking the maximum value of each friend's first ringtone setting intention coefficient for the current unit of time and the second ringtone setting intention coefficient for each friend in each target unit of time, to determine the ringtone setting intention coefficient for each friend in the current unit of time. The changing trend of users' willingness to set video ringtones can be considered, combining each friend's first ringtone setting intention coefficient for the current unit of time with the second ringtone setting intention coefficients associated with multiple units of time after the current unit of time to determine the ringtone setting intention coefficient for each friend in the current unit of time, thereby improving the reliability of the obtained ringtone setting intention coefficient for each friend in the current unit of time.

[0074] For example, if the current unit time is t, and the target unit time is T units after the current unit time t, then the ringback tone setting willingness coefficient of the i-th friend of the target user at the current unit time t is:

[0075]

[0076] in, This represents the willingness coefficient of the target user's i-th friend to set a custom ringback tone at the current unit of time t. This represents the willingness coefficient of the i-th friend of the target user to set the first ringback tone at the current unit of time t. This represents the willingness coefficient of the target user's i-th friend to set a second ringback tone at the target unit time t+1. The value represents the willingness coefficient of the i-th friend of the target user to set the second ringback tone at the target unit time t+T.

[0077] Step S402: Based on the sequence of friends who have called the target user and the pre-trained call probability prediction model, determine the call probability value of each friend who has called the target user.

[0078] In this embodiment of the disclosure, the sequence of friends who have called the target user is used to represent the information of friends who have called the target user. The friends in the sequence can be arranged from earliest to latest according to the time of calling the target user.

[0079] It should be noted that, in this embodiment of the disclosure, the sequence of friends who have called the target user is usually a time series. Optionally, the call probability prediction model trained in this embodiment of the disclosure may include a time series model that can achieve time series prediction. For example, the call probability prediction model may include a Long Short-Term Memory (LSTM) network model.

[0080] In one optional implementation, the process of the ringtone server training the call probability prediction model may include: obtaining a sequence of sample friend users who have called the target user; further, determining a second training dataset based on the sample friend user sequence; and using the second training dataset to train the call probability prediction model to be trained until the call probability prediction model converges, thereby obtaining the trained call probability prediction model; wherein, the sample friend user sequence includes a preset number of sample friend users who have called the target user, the preset number can be determined based on actual needs, and this embodiment of the disclosure does not limit this; for example, the preset number can be 400.

[0081] It should be noted that in this embodiment, the friends of the target user who will be calling will change over time. To train a call probability prediction model that can more accurately predict the call probability value for each unit of time, the training time for the call probability prediction model can be the beginning of each unit of time. For example, if the unit of time is one day, at the beginning of each day, a list of sample friends who have called the target user before that day can be obtained, and the call probability prediction model can be trained to more accurately predict the call probability for that day.

[0082] In one optional implementation, the ringtone server can train the call probability prediction model at the start of the current unit of time. The ringtone server can train the call probability prediction model associated with the current unit of time based on the sequence of sample friend users who have called the target user before the current unit of time. During the process of the ringtone server training the call probability prediction model associated with the current unit of time, the process of determining the second training dataset based on the sample friend user sequence can include: determining multiple consecutive sample friend users in the sample friend user sequence as second sample data, determining the next sample friend user after the multiple consecutive sample friend users as second label data, obtaining a set of second training data; and repeating the process of determining a set of second training data to obtain multiple sets of second training data. Further, the multiple sets of second training data are combined to obtain a second training dataset for training the call probability prediction model. The number of consecutive sample friend users can be determined based on actual needs, and this embodiment does not limit this; for example, five consecutive sample friend users, or three consecutive sample friend users, etc.

[0083] For example, the sample friend user sequence includes 400 sample friend users who have called the target user. Starting from the first sample friend user among the 400 sample friend users, the next 5 consecutive sample friend users can be determined as the second sample data. The next sample friend user after the next 5 consecutive sample friend users can be determined as the second label data, thus obtaining multiple sets of second training data. The multiple sets of second training data are combined to obtain the second training dataset.

[0084] Understandably, the process by which the ringtone server determines the call probability value of each friend calling the target user based on the sequence of friends who have called the target user and a pre-trained call probability prediction model can include: inputting the sequence of friends who called the target user before the current unit of time into the call probability prediction model associated with the current unit of time to obtain a first call probability value for each friend calling the target user at the current unit of time; further, determining the call probability value of each friend calling the target user at the current unit of time based on the first call probability value of each friend calling the target user at the current unit of time. The number of friends in the sequence of friends who called the target user before the current unit of time can be the same as the number of consecutive sample friends when training the call probability prediction model. For example, if the number of consecutive sample friends is 5, then the number of friends in the sequence of friends who called the target user before the current unit of time can be 5. The accuracy of determining the probability of each friend calling the target user at the current unit of time can be improved by using a pre-trained call probability prediction model associated with the current unit of time.

[0085] In one optional implementation, the call probability prediction model can be an LSTM model. The ringtone server will input the sequence of friends who called the target user before the current unit time into the call probability prediction model associated with the current unit time to obtain the first call probability value of each friend calling the target user in the current unit time. This process may include: inputting the sequence of friends who called the target user before the current unit time into the call probability prediction model associated with the current unit time for feature extraction, obtaining the output vector of the hidden layer in the call probability prediction model associated with the current unit time, and further processing the output vector through an activation function to obtain the first call probability value of each friend calling the target user in the current unit time.

[0086] In one optional implementation, the process by which the ringtone server determines the call probability value of each friend user calling the target user within the current unit of time, based on the first call probability value of each friend user calling the target user within the current unit of time, may include: determining the call probability value of each friend user calling the target user within the current unit of time from the first call probability value of each friend user calling the target user within the current unit of time. This can be achieved by using a call probability prediction model associated with the current unit of time, thereby improving the efficiency of determining the call probability value of each friend user calling the target user within the current unit of time.

[0087] It should be noted that the probability of a friend calling the target user in each unit of time will affect the probability of calling the target user in subsequent units of time. For example, after calling the target user in a certain unit of time, the probability of calling the target user in subsequent units of time will increase or decrease. Therefore, the probability of a friend calling the target user in a certain unit of time can be determined by comprehensively considering the probability of a friend calling the target user in that unit of time, as well as the probability of calling the target user in subsequent units of time. This improves the accuracy of the determined probability of a friend calling the target user in that unit of time. At the beginning of the current unit of time, the ringtone server can also use the sample sequence of friends who have called the target user before the current unit of time to train a call probability prediction model associated with the target unit of time.

[0088] It should be noted that, in the process of training the call probability prediction model associated with the target unit time, the method for determining the second training dataset based on the sample friend user sequence includes: determining multiple consecutive sample friend users in the sample friend user sequence as the second sample data; determining the sample unit time that the latest caller among the multiple consecutive sample friend users calls the target user; and determining the first sample friend user who calls the target user in the nth unit time after the latest caller's sample unit time as the second label data, thus obtaining a set of second training data.

[0089] For example, if the unit time is 1 day, the current unit time is day t, the target unit time is day t+3, and the sample friend user sequence includes 400 sample friend users who have called the target user, we can start from the first sample friend user among the 400 sample friend users, determine the 5 consecutive sample friend users as the second sample data, determine the sample unit time of the latest call to the target user among the 5 consecutive sample friend users, and determine the sample friend user who calls the target user earliest on the 3rd day after the latest call to the target user as the second label data, thus obtaining a set of second training data.

[0090] Understandably, before the ringtone server determines the call probability value of each friend user calling the target user in the current unit of time based on the first call probability value of each friend user calling the target user in the current unit of time, the ringtone server can input the sequence of friends users who have called the target user before the current unit of time into the call probability prediction model associated with multiple target unit times to obtain the second call probability value of each friend user calling the target user in each target unit of time.

[0091] Furthermore, the process by which the ringtone server determines the call probability value of each friend user calling the target user within the current unit of time, based on the first call probability value of each friend user calling the target user within the current unit of time, may include: determining the maximum value between the first call probability value of each friend user calling the target user within the current unit of time and the second call probability value of each friend user calling the target user within each target unit of time, to determine the call probability value of each friend user calling the target user within the current unit of time. The changing trend of friend users calling the target friend can be considered, combining the first call probability value of each friend user calling the target friend within the current unit of time with the second call probability values ​​of calling the target user in multiple units of time after the current unit of time to determine the call probability value of each friend user calling the target user within the current unit of time, thereby improving the reliability of the determined call probability value of each friend user calling the target user within the current unit of time.

[0092] For example, if the current unit time is t, and the target unit time is T units after the current unit time t, then the probability that the target user's i-th friend will call the target user in the current unit time t is:

[0093]

[0094] in, This represents the probability that the target user's i-th friend will call the target user in the current unit of time t. This represents the probability value of the target user's i-th friend calling the target user in the current unit time t. This represents the probability value of the second call from the target user's i-th friend at the target time t+1. This represents the probability value of the second call from the target user's i-th friend at the target unit time t+T.

[0095] Step S403: Determine the user metrics for each friend user based on the ringback tone setting willingness coefficient for each friend user and the call probability value for each friend user to call the target user.

[0096] In this embodiment of the disclosure, user metrics for each friend user are used to characterize the frequency of calls between each friend user and the target user, as well as the degree of willingness to set video ringback tones.

[0097] In one optional implementation, the process by which the ringtone server determines the user metrics for each friend user based on each friend user's ringtone setting willingness coefficient and the call probability value for each friend user to call the target user may include: determining the product of each friend user's ringtone setting willingness coefficient and the call probability value for calling the target user to obtain the user metrics for each friend user. It is understood that each friend user's ringtone setting willingness coefficient may include the ringtone setting willingness coefficient for each friend user in the current unit of time, and the call probability value for each friend user to call the target user may include the call probability value for each friend user to call the target user in the current unit of time; therefore, each friend user's user metrics may include the user metrics for each friend user in the current unit of time.

[0098] For example, let the current unit of time be t, and let the user metrics of the target user's i-th friend user in the current unit of time be... for:

[0099]

[0100] It is understood that, in this embodiment of the disclosure, after receiving a video ringback tone setting operation sent by the target user's user terminal, the ringtone server determines the target video ringback tone from the multiple optional video ringback tones based on the ringback tone characteristics of multiple optional video ringback tones, the user characteristics of each of the target user's friends, and the user characteristics of the target user. The target user can perform the video ringback tone setting operation on their user terminal, and the target user's user terminal can respond to the video ringback tone setting operation by sending a video ringback tone setting operation to the ringtone server.

[0101] In one optional implementation, the process by which the ringtone server determines the target video ringtone from among multiple optional video ringtones based on the ringtone characteristics of multiple optional video ringtones, the user characteristics of each of the target user's friends, and the target user's user characteristics may include: in response to obtaining a video ringtone setting operation from the target user, determining a list of the target user's friends; further, if a selection operation is obtained for at least one target friend in the friend list, then determining the target video ringtone from among the multiple optional video ringtones based on the ringtone characteristics of multiple optional video ringtones, the user characteristics of each target friend, and the target user's user characteristics. The target video ringtone can be determined by combining the target friend selected by the target user from the friend list, thereby improving the flexibility of video ringtone recommendations and enhancing the user's video ringtone setting experience.

[0102] It should be noted that, in the embodiments of this disclosure, the process by which the ringtone server determines the target video ringtone from among the multiple optional video ringtones based on the ringtone characteristics of multiple optional video ringtones, the user characteristics of each target friend user, and the user characteristics of the target user is similar to the process in the above embodiments by which the ringtone server determines the target video ringtone from among the multiple optional video ringtones based on the ringtone characteristics of multiple optional video ringtones, the user characteristics of each friend user, and the user characteristics of the target user. Therefore, this disclosure will not elaborate on this process.

[0103] It is understood that, in this embodiment of the disclosure, after the ringtone server determines the target user's friend list, it can send the target user's friend list to the target user's user terminal. If the user terminal obtains a selection operation for at least one target friend in the friend list, it can send the operation information to the ringtone server, so that the ringtone server can determine the target video ringtone from the multiple optional video ringtones based on the ringtone characteristics of multiple optional video ringtones, the user characteristics of each target friend, and the user characteristics of the target user.

[0104] In one optional implementation, the friend user list can be a list of displayable friend users, where the user metrics of the friends in the list are greater than or equal to a preset threshold. The preset threshold can be determined based on actual needs, and this embodiment does not limit its implementation. It is understood that in this embodiment, if a friend user's user metrics are greater than or equal to the preset threshold, it indicates that the friend user has a high frequency of calls with the target user and is more willing to set video ringback tones.

[0105] The process by which the ringtone server determines the list of friends that can be displayed to the target user may include: identifying at least one friend whose user metrics at the current unit of time are greater than or equal to a preset threshold as a displayable friend, thus obtaining the target user's list of displayable friends at the current unit of time. Based on the target user's friends' willingness to set ringtones and the probability of them calling the target user, friends with a high willingness to set ringtones and a high probability of calling the target user are identified as displayable friends. This list is then displayed to the target user during the video ringtone setting process, facilitating recommendations of video ringtones that are commonly liked by both the target user and friends with a high willingness to set ringtones and a high probability of calling the target user, thereby further stimulating the ringtone setting conversion rate of the target user and / or their friends.

[0106] Understandably, after the ringtone server sends the target user's list of visible friends to the target user's terminal, the terminal can then display this list. It is also understandable that the list of visible friends obtained by the terminal device represents the target user's list of visible friends at the current time.

[0107] For example, such as Figure 5 As shown, Figure 5 The diagram illustrates a user terminal display interface, which shows a list of friends and a prompt message: "Don't know which video ringback tones to set? Select a few of your friends and see what they like."

[0108] Step S302: The candidate video ringback tone with the highest recommendation value is determined as the target video ringback tone.

[0109] In step S202, the ringtone server can recommend the target video ringtone to the target user and / or their friends.

[0110] In this embodiment of the disclosure, when a target user sets a video ringback tone, a target video ringback tone can be determined and recommended to the target user. Since the target video ringback tone is determined based on the user characteristics of the target user and the user characteristics of the target user's friends, the target video ringback tone can also be recommended to the friends when they set a video ringback tone. It can be understood that if the target video ringback tone is determined based on the target friends selected by the target user, then the target video ringback tone can be recommended to the target friends.

[0111] In one alternative implementation, the target video ringback tone is determined by the ringtone server upon receiving a video ringback tone setting operation sent by the target user's user terminal. The ringtone server can then recommend the target video ringback tone to the target user's user terminal, whereby the user terminal can display the target video ringback tone in response to receiving it from the ringtone server.

[0112] In one optional implementation, the process of displaying the target video ringback tone on the user terminal may include: displaying the target video ringback tone and its recommended value. Displaying the target video ringback tone and its recommended value on the user terminal allows the target user to easily see how much their friends like the target video ringback tone.

[0113] like Figure 6 As shown, Figure 6 A schematic diagram of a user terminal display interface is shown, which displays the target video ringback tone and the recommended value of the target video ringback tone: friend preference index 5.

[0114] In one optional implementation, after the user terminal displays the target video ringback tone, in response to receiving a video ringback tone update operation, a target video ringback tone update operation is sent to the ringtone server. This causes the ringtone server to determine the candidate video ringback tone with the smallest difference from the recommended value of the target video ringback tone as the updated target video ringback tone, and send the updated target video ringback tone to the user terminal. Furthermore, the user terminal can display the updated target video ringback tone in response to receiving it from the ringtone server. If the user is not satisfied with the currently recommended target video ringback tone, candidate video ringback tones that are most favored by the target user and their friends, excluding the target video ringback tone, can be recommended to the target user.

[0115] In one alternative implementation, after determining the target video ringback tone, the ringtone server can recommend the target video ringback tone to the user terminal of the friend user after obtaining the video ringback tone settings sent by the friend user's user terminal. The friend user can be the target friend user.

[0116] It should be noted that, in this embodiment of the disclosure, after the ringtone server determines the target user's list of friends, it can determine the total value of at least one user indicator that can display friends in the list of friends, and obtain the target user's value indicator value. The value indicator value is used to characterize the value of video ringtone distribution. The higher the user's value indicator value, the greater the benefit to the user from distributing video ringtones.

[0117] For example, if the target user's friend list includes K friend users, then the value index V(U0) of the target user U0 can include:

[0118]

[0119] Where Value(U) k ) represents the user metrics of the kth friend among K friend users.

[0120] The foregoing primarily describes the solutions provided by the embodiments of this disclosure from the perspective of the ringtone server. It is understood that, in order to achieve the above functions, the ringtone server includes corresponding hardware structures and / or software modules for executing each function. Those skilled in the art should readily recognize that, based on the units and algorithm steps of the various examples described in conjunction with the embodiments disclosed herein, this disclosure can be implemented in hardware or a combination of hardware and computer software. Whether a function is executed in hardware or by computer software driving hardware depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of this disclosure.

[0121] This disclosure embodiment can divide the ringtone server into functional units according to the above method example. For example, it can divide each function into a separate functional module, or it can integrate two or more functions into one processing module. The integrated module can be implemented in hardware or as a software functional module. It should be noted that the module division in this disclosure embodiment is illustrative and only represents one logical functional division. In actual implementation, there may be other division methods.

[0122] An exemplary embodiment of this disclosure provides a video ringback tone recommendation device, which can be a user terminal or a chip applied to a user terminal. Figure 7 A schematic block diagram of the functional modules of a video ringback tone recommendation device according to an exemplary embodiment of the present disclosure is shown. Figure 7 As shown, the video ringback tone recommendation device 700 includes:

[0123] The first determining module 701 is configured to determine a target video ringback tone from the plurality of optional video ringback tones based on the ringback tone characteristics of a plurality of optional video ringback tones, the user characteristics of each friend of the target user, and the user characteristics of the target user.

[0124] The recommendation module 702 is configured to recommend the target video ringback tone to the target user and / or the friend user.

[0125] Optionally, the first determining module 701 is configured as follows:

[0126] Based on the ringback tone characteristics of multiple optional video ringback tones, the user characteristics of each friend user, and the user characteristics of the target user, multiple candidate video ringback tones and a recommendation value for each candidate video ringback tone are determined. The multiple candidate video ringback tones are optional video ringback tones that the friend user is interested in, or optional video ringback tones that both the target user and the friend user are interested in. The recommendation value is used to characterize the degree of interest of multiple friend users in the candidate video ringback tones.

[0127] The candidate video ringback tone with the highest recommendation value is selected as the target video ringback tone.

[0128] Optionally, the first determining module 701 is configured to:

[0129] Based on the user characteristics of each friend user, the user characteristics of the target user, the ringback tone characteristics of multiple optional video ringback tones, and a pre-trained recall model, multiple candidate video ringback tones are determined.

[0130] The user characteristics of each friend user, as well as the ringback tone characteristics of the multiple candidate video ringback tones, are input into a pre-trained interest value prediction model to determine the interest index value of each friend user for each of the candidate video ringback tones.

[0131] Based on each friend user's interest index value for each of the candidate video ringback tones, and each friend user's user index, a recommendation value for each candidate video ringback tone is determined. The friend user's user index is determined based on the friend user's ringback tone setting willingness coefficient and the call probability value of the friend user calling the target user.

[0132] Optionally, the device further includes a second determining module 703, configured to:

[0133] Based on the historical ringback tone settings information of each friend user associated with the target user, and a pre-trained ringback tone setting intention prediction model, determine the ringback tone setting intention coefficient of each friend user;

[0134] Based on the sequence of friends who have called the target user and a pre-trained call probability prediction model, determine the call probability value of each friend calling the target user;

[0135] Based on the ringback tone setting willingness coefficient of each friend user and the call probability value of each friend user calling the target user, the user indicators of each friend user are determined.

[0136] Optionally, the second determining module 703 is configured as follows:

[0137] Input the ringback tone setting information of each friend user in multiple historical time units before the current time unit into the ringback tone setting intention prediction model associated with the current time unit to obtain the first ringback tone setting intention coefficient of each friend user in the current time unit.

[0138] Based on the first ringback tone setting willingness coefficient of each friend user in the current unit of time, determine the ringback tone setting willingness coefficient of each friend user in the current unit of time.

[0139] Optionally, the device further includes a first prediction module 704, configured to:

[0140] The ringback tone setting information of each friend user in multiple historical time units before the current time unit is input into the ringback tone setting intention prediction model associated with multiple target time units to obtain the second ringback tone setting intention coefficient of each friend user in each target time unit, wherein the target time unit is later than the current time unit;

[0141] The step of determining the ringback tone setting willingness coefficient for each friend user in the current unit of time based on the first ringback tone setting willingness coefficient for each friend user in the current unit of time includes:

[0142] The maximum value of the first ringback tone setting willingness coefficient of each friend user in the current unit time and the second ringback tone setting willingness coefficient of each friend user in each target unit time is determined as the ringback tone setting willingness coefficient of each friend user in the current unit time.

[0143] Optionally, the second determining module 703 is configured to:

[0144] The sequence of friends who called the target user before the current unit time is input into the call probability prediction model associated with the current unit time to obtain the first call probability value of each friend calling the target user at the current unit time;

[0145] Based on the first call probability value of each friend user calling the target user in the current unit time, the call probability value of each friend user calling the target user in the current unit time is determined.

[0146] Optionally, the device further includes a second prediction module 705, configured to:

[0147] The sequence of friends who called the target user before the current unit time is input into the call probability prediction model associated with multiple target unit times to obtain the second call probability value of each friend calling the target user in each target unit time.

[0148] The step of determining the call probability value of each friend user calling the target user in the current unit time based on the first call probability value of each friend user calling the target user in the current unit time includes:

[0149] The maximum value between the first call probability value of each friend user calling the target user in the current unit time and the second call probability value of each friend user calling the target user in each target unit time is determined as the call probability value of each friend user calling the target user in the current unit time.

[0150] Optionally, the first determining module 701 is configured to:

[0151] In response to obtaining the target user's video ringback tone settings, determine the target user's friend list;

[0152] If a selection operation is obtained for at least one target friend user in the friend user list, then the target video ringback tone is determined from the multiple optional video ringback tones based on the ringback tone characteristics of multiple optional video ringback tones, the user characteristics of each target friend user, and the user characteristics of the target user.

[0153] Figure 8 A schematic block diagram of a chip according to an exemplary embodiment of the present disclosure is shown. Figure 8 As shown, the chip 800 includes one or more processors 801 and a communication interface 802. The communication interface 802 can support the server in performing the data transmission and reception steps in the video ringback tone recommendation method described above, and the processor 801 can support the server in performing the data processing steps in the video ringback tone recommendation method described above.

[0154] Optional, such as Figure 8As shown, the chip 800 also includes a memory 803, which may include read-only memory and random access memory, and provides operation instructions and data to the processor. A portion of the memory may also include non-volatile random access memory (NVRAM).

[0155] In some implementations, such as Figure 8 As shown, processor 801 executes corresponding operations by calling operation instructions stored in memory (which may be stored in the operating system). Processor 801 controls the processing operations of any terminal device; processor can also be called a central processing unit (CPU). Memory 803 may include read-only memory and random access memory, and provides instructions and data to processor 2201. A portion of memory 803 may also include NVRAM. For example, in applications, memory, communication interfaces, and other components are coupled together via a bus system, which may include, in addition to a data bus, a power bus, a control bus, and a status signal bus, etc. However, for clarity, in... Figure 8 The general labeled all buses as Bus System 804.

[0156] The methods disclosed in the embodiments of this disclosure can be applied to a processor or implemented by a processor. The processor may be an integrated circuit chip with signal processing capabilities. During implementation, each step of the above methods can be completed by integrated logic circuits in the processor's hardware or by instructions in software form. The processor can be a general-purpose processor, a digital signal processor (DSP), an ASIC, a field-programmable gate array (FPGA), or other programmable logic devices, discrete gate or transistor logic devices, or discrete hardware components. It can implement or execute the methods, steps, and logic block diagrams disclosed in the embodiments of this disclosure. The general-purpose processor can be a microprocessor or any conventional processor. The steps of the methods disclosed in the embodiments of this disclosure can be directly embodied in the execution of a hardware decoding processor, or executed by a combination of hardware and software modules in the decoding processor. The software modules can reside in random access memory, flash memory, read-only memory, programmable read-only memory, electrically erasable programmable memory, registers, or other mature storage media in the art. This storage medium is located in memory; the processor reads information from the memory and, in conjunction with its hardware, completes the steps of the above methods.

[0157] Exemplary embodiments of this disclosure also provide an electronic device, including: at least one processor; and a memory communicatively connected to the at least one processor. The memory stores a computer program executable by the at least one processor, the computer program being executed by the at least one processor to cause the electronic device to perform a method according to an embodiment of this disclosure.

[0158] Exemplary embodiments of this disclosure also provide a non-transitory computer-readable storage medium storing a computer program, wherein the computer program, when executed by a computer's processor, is used to cause the computer to perform a method according to embodiments of this disclosure.

[0159] Exemplary embodiments of this disclosure also provide a computer program product, including a computer program, wherein, when executed by a processor of a computer, the computer program is used to cause the computer to perform a method according to an embodiment of this disclosure.

[0160] refer to Figure 9 The present invention describes a structural block diagram of an electronic device 900 that can serve as a server or client of the present disclosure, which is an example of a hardware device that can be applied to various aspects of the present disclosure. The electronic device is intended to represent various forms of digital electronic computer devices, such as laptop computers, desktop computers, workstations, personal digital assistants, servers, blade servers, mainframe computers, and other suitable computers. The electronic device can also represent various forms of mobile devices, such as personal digital processors, cellular phones, smartphones, wearable devices, and other similar computing devices. The components shown herein, their connections and relationships, and their functions are merely illustrative and are not intended to limit the implementation of the present disclosure described and / or claimed herein.

[0161] like Figure 9 As shown, the electronic device 900 includes a computing unit 901, which can perform various appropriate actions and processes according to a computer program stored in a read-only memory (ROM) 902 or a computer program loaded from a storage unit 908 into a random access memory (RAM) 903. The RAM 903 may also store various programs and data required for the operation of the device 900. The computing unit 901, ROM 902, and RAM 903 are interconnected via a bus 904. An input / output (I / O) interface 905 is also connected to the bus 904.

[0162] Multiple components in electronic device 900 are connected to I / O interface 905, including: input unit 906, output unit 907, storage unit 908, and communication unit 909. Input unit 906 can be any type of device capable of inputting information to electronic device 900. Input unit 906 can receive input digital or character information and generate key signal inputs related to user settings and / or function control of electronic device. Output unit 907 can be any type of device capable of presenting information and may include, but is not limited to, a display, speaker, video / audio output terminal, vibrator, and / or printer. Storage unit 904 may include, but is not limited to, disk and optical disk. Communication unit 909 allows electronic device 900 to exchange information / data with other devices through computer networks such as the Internet and / or various telecommunications networks, and may include, but is not limited to, modems, network cards, infrared communication devices, wireless communication transceivers, and / or chipsets, such as Bluetooth™ devices, WiFi devices, WiMax devices, cellular communication devices, and / or the like.

[0163] The computing unit 901 can be a variety of general-purpose and / or special-purpose processing components with processing and computing capabilities. Some examples of the computing unit 901 include, but are not limited to, a central processing unit (CPU), a graphics processing unit (GPU), various special-purpose artificial intelligence (AI) computing chips, various computing units running machine learning model algorithms, a digital signal processor (DSP), and any suitable processor, controller, microcontroller, etc. The computing unit 901 performs the various methods and processes described above. For example, in some embodiments, the methods of the exemplary embodiments of this disclosure can be implemented as a computer software program tangibly contained in a machine-readable medium, such as storage unit 908. In some embodiments, part or all of the computer program can be loaded and / or installed on the electronic device 900 via ROM 902 and / or communication unit 909. In some embodiments, the computing unit 901 can be configured to perform the methods of the exemplary embodiments of this disclosure by any other suitable means (e.g., by means of firmware).

[0164] The program code used to implement the methods of this disclosure may be written in any combination of one or more programming languages. This program code may be provided to a processor or controller of a general-purpose computer, special-purpose computer, or other programmable data processing apparatus, such that when executed by the processor or controller, the program code causes the functions / operations specified in the flowcharts and / or block diagrams to be implemented. The program code may be executed entirely on a machine, partially on a machine, as a standalone software package partially on a machine and partially on a remote machine, or entirely on a remote machine or server.

[0165] In the context of this disclosure, a machine-readable medium can be a tangible medium that may contain or store a program for use by or in conjunction with an instruction execution system, apparatus, or device. A machine-readable medium can be a machine-readable signal medium or a machine-readable storage medium. A machine-readable medium can be, but is not limited to, electronic, magnetic, optical, electromagnetic, infrared, or semiconductor systems, apparatus, or devices, or any suitable combination of the foregoing. More specific examples of machine-readable storage media include electrical connections based on one or more wires, portable computer disks, hard disks, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), optical fiber, portable compact disk read-only memory (CD-ROM), optical storage devices, magnetic storage devices, or any suitable combination of the foregoing.

[0166] As used in this disclosure, the terms "machine-readable medium" and "computer-readable medium" refer to any computer program product, device, and / or apparatus (e.g., disk, optical disk, memory, programmable logic device (PLD)) for providing machine instructions and / or data to a programmable processor, including machine-readable media that receive machine instructions as machine-readable signals. The term "machine-readable signal" refers to any signal for providing machine instructions and / or data to a programmable processor.

[0167] To provide interaction with a user, the systems and techniques described herein can be implemented on a computer having: a display device for displaying information to the user (e.g., a CRT (cathode ray tube) or LCD (liquid crystal display) monitor); and a keyboard and pointing device (e.g., a mouse or trackball) through which the user provides input to the computer. Other types of devices can also be used to provide interaction with the user; for example, feedback provided to the user can be any form of sensory feedback (e.g., visual feedback, auditory feedback, or tactile feedback); and input from the user can be received in any form (including sound input, voice input, or tactile input).

[0168] The systems and technologies described herein can be implemented in computing systems that include backend components (e.g., as a data server), or computing systems that include middleware components (e.g., an application server), or computing systems that include frontend components (e.g., a user computer with a graphical user interface or web browser through which a user can interact with embodiments of the systems and technologies described herein), or any combination of such backend, middleware, or frontend components. The components of the system can be interconnected via digital data communication of any form or medium (e.g., a communication network). Examples of communication networks include local area networks (LANs), wide area networks (WANs), and the Internet.

[0169] Computer systems can include clients and servers. Clients and servers are generally located far apart and typically interact through communication networks. Client-server relationships are created by computer programs running on the respective computers and having a client-server relationship with each other.

[0170] In the above embodiments, implementation can be achieved, in whole or in part, through software, hardware, firmware, or any combination thereof. When implemented using software, it can be implemented, in whole or in part, as a computer program product. The computer program product includes one or more computer programs or instructions. When the computer program or instructions are loaded and executed on a computer, the processes or functions described in the embodiments of this disclosure are performed, in whole or in part. The computer can be a general-purpose computer, a special-purpose computer, a computer network, a terminal, a user equipment, or other programmable device. The computer program or instructions can be stored in a computer-readable storage medium or transferred from one computer-readable storage medium to another. For example, the computer program or instructions can be transferred from one website, computer, server, or data center to another website, computer, server, or data center via wired or wireless means. The computer-readable storage medium can be any available medium accessible to a computer or a data storage device such as a server or data center integrating one or more available media. The available medium can be a magnetic medium, such as a floppy disk, hard disk, or magnetic tape; it can also be an optical medium, such as a digital video disc (DVD); or it can be a semiconductor medium, such as a solid-state drive (SSD).

[0171] Although this disclosure has been described in conjunction with specific features and embodiments, it will be apparent that various modifications and combinations can be made therein without departing from the spirit and scope of this disclosure. Accordingly, this specification and drawings are merely exemplary illustrations of the disclosure as defined by the appended claims and are to be considered as covering any and all modifications, variations, combinations, or equivalents within the scope of this disclosure. It is obvious that those skilled in the art can make various alterations and modifications to this disclosure without departing from its spirit and scope. Thus, this disclosure is also intended to include any such modifications and modifications that fall within the scope of the claims of this disclosure and their equivalents.

Claims

1. A video ringback tone recommendation method, characterized in that, include: Based on the ringback tone characteristics of multiple optional video ringback tones, the user characteristics of each friend of the target user, and the user characteristics of the target user, the target video ringback tone is determined from the multiple optional video ringback tones; Recommend the target video ringback tone to the target user and / or the friend user; The method further includes, before determining the target video ringback tone from the multiple optional video ringback tones based on the ringback tone characteristics of multiple optional video ringback tones, the user characteristics of each friend of the target user, and the user characteristics of the target user: Based on the historical ringback tone settings information of each friend user associated with the target user, and a pre-trained ringback tone setting intention prediction model, the ringback tone setting intention coefficient of each friend user is determined; Based on the sequence of friends who have called the target user and a pre-trained call probability prediction model, determine the call probability value of each friend calling the target user; Based on the ringback tone setting willingness coefficient of each friend user and the call probability value of each friend user calling the target user, the user indicators of each friend user are determined; The step of determining the ringback tone setting intention coefficient for each friend user based on their historical ringback tone setting information and a pre-trained ringback tone setting intention prediction model, includes: Input the ringback tone setting information of each friend user in multiple historical time units before the current time unit into the ringback tone setting intention prediction model associated with the current time unit to obtain the first ringback tone setting intention coefficient of each friend user in the current time unit. Based on the first ringback tone setting willingness coefficient of each friend user in the current unit of time, determine the ringback tone setting willingness coefficient of each friend user in the current unit of time.

2. The video ringback tone recommendation method as described in claim 1, characterized in that, The step of determining the target video ringback tone from the multiple optional video ringback tones based on the ringback tone characteristics of multiple optional video ringback tones, the user characteristics of each friend of the target user, and the user characteristics of the target user includes: Based on the ringback tone characteristics of multiple optional video ringback tones, the user characteristics of each friend user, and the user characteristics of the target user, multiple candidate video ringback tones and a recommendation value for each candidate video ringback tone are determined. The multiple candidate video ringback tones are optional video ringback tones that the friend user is interested in, or optional video ringback tones that both the target user and the friend user are interested in. The recommendation value is used to characterize the degree of interest of multiple friend users in the candidate video ringback tones. The candidate video ringback tone with the highest recommendation value is selected as the target video ringback tone.

3. The video ringback tone recommendation method as described in claim 2, characterized in that, The process of determining multiple candidate video ringback tones and a recommended value for each candidate video ringback tone based on the ringback tone characteristics of multiple selectable video ringback tones, the user characteristics of each friend user, and the user characteristics of the target user includes: Based on the user characteristics of each friend user, the user characteristics of the target user, the ringback tone characteristics of multiple optional video ringback tones, and a pre-trained recall model, multiple candidate video ringback tones are determined. The user characteristics of each friend user, as well as the ringback tone characteristics of the multiple candidate video ringback tones, are input into a pre-trained interest value prediction model to determine the interest index value of each friend user for each of the candidate video ringback tones. The recommended value for each candidate video ringback tone is determined based on the interest index value of each friend user for each candidate video ringback tone, and the user index of each friend user. The user index of each friend user is determined based on the ringback tone setting willingness coefficient of the friend user and the calling probability value of the friend user calling the target user.

4. The video ringback tone recommendation method as described in claim 1, characterized in that, The method further includes: The ringback tone setting information of each friend user in multiple historical time units before the current time unit is input into the ringback tone setting intention prediction model associated with multiple target time units to obtain the second ringback tone setting intention coefficient of each friend user in each target time unit, wherein the target time unit is later than the current time unit; The step of determining the ringback tone setting willingness coefficient for each friend user in the current unit of time based on the first ringback tone setting willingness coefficient for each friend user in the current unit of time includes: The maximum value of the first ringback tone setting willingness coefficient of each friend user in the current unit time and the second ringback tone setting willingness coefficient of each friend user in each target unit time is determined as the ringback tone setting willingness coefficient of each friend user in the current unit time.

5. The video ringback tone recommendation method as described in claim 1, characterized in that, The step of determining the call probability value of each friend user calling the target user based on the sequence of friends who have called the target user and a pre-trained call probability prediction model includes: The sequence of friends who called the target user before the current unit time is input into the call probability prediction model associated with the current unit time to obtain the first call probability value of each friend calling the target user at the current unit time. Based on the first call probability value of each friend user calling the target user in the current unit time, the call probability value of each friend user calling the target user in the current unit time is determined.

6. The video ringback tone recommendation method as described in claim 5, characterized in that, The method further includes: The sequence of friends who called the target user before the current unit time is input into the call probability prediction model associated with multiple target unit times to obtain the second call probability value of each friend calling the target user in each target unit time. The step of determining the call probability value of each friend user calling the target user in the current unit time based on the first call probability value of each friend user calling the target user in the current unit time includes: The maximum value between the first call probability value of each friend user calling the target user in the current unit time and the second call probability value of each friend user calling the target user in each target unit time is determined as the call probability value of each friend user calling the target user in the current unit time.

7. The video ringback tone recommendation method as described in claim 1, characterized in that, The step of determining the target video ringback tone from the multiple optional video ringback tones based on the ringback tone characteristics of multiple optional video ringback tones, the user characteristics of each friend of the target user, and the user characteristics of the target user includes: In response to obtaining the target user's video ringback tone settings, determine the target user's friend list; If a selection operation is obtained for at least one target friend user in the friend user list, then the target video ringback tone is determined from the multiple optional video ringback tones based on the ringback tone characteristics of multiple optional video ringback tones, the user characteristics of each target friend user, and the user characteristics of the target user.

8. An electronic device, characterized in that, include: processor; as well as Stored program memory, The program includes instructions that, when executed by the processor, cause the processor to perform the method as described in any one of claims 1-7.

Images