A method, device and equipment for displaying recommended information in a live broadcast room

Abstract

The present application provides a method, device and equipment for displaying recommended information of a live broadcast room, which is used to provide a personalized live broadcast room recommendation mechanism for users and improve the recommendation accuracy of the live broadcast room. The present application provides a method for displaying recommended information of a live broadcast room, including: when a user logs in to an application of a live broadcast platform currently running on a client, outputting a preference survey page to the user; obtaining the recommendation position preference features input by the user based on the preference survey page, the recommendation position preference features including image position features and image style features; determining a target live broadcast room that matches the image style features from the current live broadcast room set of the live broadcast platform; adjusting the default display page of the live broadcast platform to a target display page corresponding to the image position features; and displaying the recommended information of the target live broadcast room on the target display page.

Description

Technical Field

[0001] The present application relates to the field of information, and specifically to a method, device and equipment for displaying recommended information in a live broadcast room. Background Art

[0002] As an emerging form of information dissemination, online live streaming has the characteristics of online communication and live broadcast. In recent years, it has become a part of the daily life of the majority of users, covering many fields such as training, shopping, games, sports and so on.

[0003] The audience, or user, can load the corresponding application program through the live broadcast client on the user equipment (UE), access the live broadcast platform, and select a live broadcast room in the platform to watch the live broadcast content. In this process, the user can enter the specific room number of the live broadcast room to open the corresponding live broadcast room for viewing, or click and open the live broadcast room recommended by the live broadcast platform on the application interface for viewing.

[0004] During the research process of existing related technologies, the inventors found that for the live broadcast rooms recommended by the live broadcast platform on the application interface, in addition to some live broadcast rooms that are at the top of the rankings, there are also some live broadcast rooms recommended to users under the recommendation mechanism. However, in actual applications, these live broadcast rooms recommended to users often have low user click-through rates. Obviously, their recommendation accuracy is not good and there is a problem of low matching between them and users. Summary of the invention

[0005] The present application provides a method, device and equipment for displaying live broadcast room recommendation information, which is used to provide a personalized live broadcast room recommendation mechanism for users and improve the recommendation accuracy of the live broadcast room.

[0006] In a first aspect, the present application provides a method for displaying recommended information in a live broadcast room, the method comprising:

[0007] When the user logs in to the application of the live broadcast platform currently running on the client, a preference survey page is output to the user;

[0008] Obtaining the recommended position preference features input by the user based on the preference survey page, the recommended position preference features including image position features and image style features;

[0009] Determine a target live broadcast room that matches the image style feature from the current live broadcast room set of the live broadcast platform;

[0010] Adjust the default display page of the live broadcast platform to the target display page corresponding to the image position feature;

[0011] The recommended information of the target live broadcast room is displayed on the target display page.

[0012] In combination with the first aspect of the present application, in a first possible implementation of the first aspect of the present application, obtaining the recommendation position preference feature input by the user based on the preference survey page includes:

[0013] Receive the answer information entered by the user on the preference survey page;

[0014] Based on the answer information, the preference characteristics of the recommended positions are analyzed.

[0015] In combination with the first possible implementation of the first aspect of the present application, in a second possible implementation of the first aspect of the present application, when a user logs in to an application of a live broadcast platform currently running on a client, a preference survey page is output to the user, including:

[0016] When the user logs in to the application of the live broadcast platform currently running on the client, sample recommendation information of the live broadcast room is output to the user in the preference survey page according to different image position features and different image style features;

[0017] Receive the user's answer information for the preference survey page, including:

[0018] The user's attention to different image position features and different image style features is received as answer information.

[0019] In combination with the first possible implementation of the first aspect of the present application, in a third possible implementation of the first aspect of the present application, when a user logs in to an application of a live broadcast platform currently running on a client, a preference survey page is output to the user, including:

[0020] When the user logs in to the application of the live broadcast platform currently running on the client, sample recommendation information of the live broadcast room is output to the user in the preference survey page according to different image position features and different image style features;

[0021] Receive the user's answer information for the preference survey page, including:

[0022] Under the user's confirmation, a facial image of the user while viewing the sample recommendation information is captured by a camera;

[0023] Under image recognition processing, the face image is tracked by the eyes, and the user's attention to different image position features and different image style features is determined based on the face tracking results as answer information.

[0024] In combination with the first aspect of the present application, in a fourth possible implementation manner of the first aspect of the present application, determining a target live broadcast room that matches the image style feature from a current live broadcast room set of the live broadcast platform includes:

[0025] Determine the first live broadcast room that reaches a preset click rate from the current live broadcast room set of the live broadcast platform;

[0026] In the viewing records opened by the user, determine the second live broadcast room that the user has watched in the historical time period and meets the preset conditions;

[0027] In the first live broadcast room, according to the similarity between the working status characteristics of each live broadcast room and the working status characteristics of the second live broadcast room, the third live broadcast room is selected as the target live broadcast room, and the corresponding similarity of the third live broadcast room is within the preset similarity range.

[0028] In combination with the fourth possible implementation of the first aspect of the present application, in a fifth possible implementation of the first aspect of the present application, the determination process of the similarity specifically includes:

[0029] Calculate the Euclidean distance and cosine similarity between two working state features whose similarity is to be calculated respectively. The Euclidean distance can be calculated by a Euclidean distance calculation formula or a standardized Euclidean distance calculation formula, and the pre-similarity can be calculated by a cosine similarity calculation formula or a normalized pre-similarity calculation formula;

[0030] The calculated Euclidean distance and cosine similarity are added together to obtain the final similarity.

[0031] In combination with the fifth possible implementation of the first aspect of the present application, in a sixth possible implementation of the first aspect of the present application, the determination process of the second live broadcast room includes:

[0032] Calculate the proportion of the user's viewing time in the current live broadcast room in the first live broadcast room;

[0033] If the viewing time ratio is greater than the preset value, the user is determined to belong to the current live broadcast room, and the user group watching the current live broadcast room is circled;

[0034] Calculate the total viewing time of each potential live broadcast room based on the viewing records opened by each user in the user group;

[0035] The ranking is calculated based on the total viewing time of each potential live broadcast room, and the live broadcast room with the best ranking or within the preset ranking range is determined as the second live broadcast room;

[0036] Add the calculated Euclidean distance and cosine similarity to get the final similarity, including:

[0037] The final similarity is obtained by calculating the Euclidean distance, cosine similarity and the total viewing time of the second live broadcast room.

[0038] In a second aspect, the present application provides a device for displaying recommended information in a live broadcast room, the device comprising:

[0039] An output unit, used to output a preference survey page to the user when the user logs into the application of the live broadcast platform currently running on the client;

[0040] An acquisition unit, used to acquire the recommendation position preference features input by the user based on the preference survey page, the recommendation position preference features including image position features and image style features;

[0041] A determination unit, used to determine a target live broadcast room that matches the image style feature from a current live broadcast room set of the live broadcast platform;

[0042] An adjustment unit, used to adjust a default display page of the live broadcast platform to a target display page corresponding to the image position feature;

[0043] The display unit is used to display the recommendation information of the target live broadcast room on the target display page.

[0044] In conjunction with the second aspect of the present application, in a first possible implementation manner of the second aspect of the present application, the acquiring unit is specifically configured to:

[0045] Receive the answer information entered by the user on the preference survey page;

[0046] Based on the answer information, the preference characteristics of the recommended positions are analyzed.

[0047] In combination with the first possible implementation of the second aspect of the present application, in a second possible implementation of the second aspect of the present application, the output unit is specifically used to:

[0048] When the user logs in to the application of the live broadcast platform currently running on the client, sample recommendation information of the live broadcast room is output to the user in the preference survey page according to different image position features and different image style features;

[0049] Acquisition unit, specifically used for:

[0050] The user's attention to different image position features and different image style features is received as answer information.

[0051] In combination with the first possible implementation of the second aspect of the present application, in a third possible implementation of the second aspect of the present application, the output unit is specifically used to:

[0052] When the user logs in to the application of the live broadcast platform currently running on the client, sample recommendation information of the live broadcast room is output to the user in the preference survey page according to different image position features and different image style features;

[0053] Acquisition unit, specifically used for:

[0054] Under the user's confirmation, a facial image of the user while viewing the sample recommendation information is captured by a camera;

[0055] Under image recognition processing, the face image is tracked by the eyes, and the user's attention to different image position features and different image style features is determined based on the face tracking results as answer information.

[0056] In combination with the second aspect of the present application, in a fourth possible implementation manner of the second aspect of the present application, the determining unit is specifically configured to:

[0057] Determine the first live broadcast room that reaches a preset click rate from the current live broadcast room set of the live broadcast platform;

[0058] In the viewing records opened by the user, determine the second live broadcast room that the user has watched in the historical time period and meets the preset conditions;

[0059] In the first live broadcast room, according to the similarity between the working status characteristics of each live broadcast room and the working status characteristics of the second live broadcast room, the third live broadcast room is selected as the target live broadcast room, and the corresponding similarity of the third live broadcast room is within the preset similarity range.

[0060] In combination with the fourth possible implementation of the second aspect of the present application, in a fifth possible implementation of the second aspect of the present application, the determination process of the similarity specifically includes:

[0061] Calculate the Euclidean distance and cosine similarity between two working state features whose similarity is to be calculated respectively. The Euclidean distance can be calculated by a Euclidean distance calculation formula or a standardized Euclidean distance calculation formula, and the pre-similarity can be calculated by a cosine similarity calculation formula or a normalized pre-similarity calculation formula;

[0062] The calculated Euclidean distance and cosine similarity are added together to obtain the final similarity.

[0063] In combination with the fifth possible implementation of the second aspect of the present application, in a sixth possible implementation of the second aspect of the present application, the determination process of the second live broadcast room includes:

[0064] Calculate the proportion of the user's viewing time in the current live broadcast room in the first live broadcast room;

[0065] If the viewing time ratio is greater than the preset value, the user is determined to belong to the current live broadcast room, and the user group watching the current live broadcast room is circled;

[0066] Calculate the total viewing time of each potential live broadcast room based on the viewing records opened by each user in the user group;

[0067] The ranking is calculated based on the total viewing time of each potential live broadcast room, and the live broadcast room with the best ranking or within the preset ranking range is determined as the second live broadcast room;

[0068] The similarity determination process specifically includes:

[0069] The final similarity is obtained by calculating the Euclidean distance, cosine similarity and the total viewing time of the second live broadcast room.

[0070] In a third aspect, the present application provides a device for displaying recommended information in a live broadcast room, including a processor and a memory, wherein a computer program is stored in the memory, and when the processor calls the computer program in the memory, the method provided in the first aspect of the present application or any possible implementation method of the first aspect of the present application is executed.

[0071] In a fourth aspect, the present application provides a computer-readable storage medium, which stores multiple instructions, and the instructions are suitable for a processor to load to execute the method provided by the first aspect of the present application or any possible implementation of the first aspect of the present application.

[0072] From the above content, it can be concluded that the present application has the following beneficial effects:

[0073] For the live broadcast room recommendation scenario of the live broadcast platform, the present application obtains the recommendation position preference features input by the user based on the preference survey page by outputting a preference survey page to the user. The recommendation position preference features specifically include image position features and image style features. On the one hand, a matching target live broadcast room is determined from the current live broadcast room set of the live broadcast platform based on the image style features. On the other hand, the default display page of the live broadcast platform is adjusted to the corresponding target display page based on the image position features. At this time, the recommendation information of the target live broadcast room is displayed on the target display page. During the recommendation process, since the recommendation information of the target live broadcast room is output based on the user's recommendation position preference features, it is more in line with the current user's operating habits, and the user's acceptance is significantly improved, thereby realizing a personalized live broadcast room recommendation mechanism for users, providing more matching judgment information for user decision-making, thereby improving the recommendation accuracy of the live broadcast room, and at the same time improving resource utilization under limited recommendation resources, meeting the recommendation needs of both the live broadcast platform and the user. BRIEF DESCRIPTION OF THE DRAWINGS

[0074] In order to more clearly illustrate the technical solutions in the embodiments of the present application, the drawings required for use in the description of the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present application. For those skilled in the art, other drawings can be obtained based on these drawings without creative work.

[0075] Figure 1 A flowchart of a method for displaying recommended information in the live broadcast room of this application;

[0076] Figure 2 A schematic diagram of a scenario for the recommended position of this application;

[0077] Figure 3 A schematic diagram of another scenario for the recommended position for this application;

[0078] Figure 4 A schematic diagram of another scenario for the recommended position for this application;

[0079] Figure 5 A schematic diagram of a process for preference survey processing for this application;

[0080] Figure 6 Another flow diagram for preference survey processing for this application;

[0081] Figure 7 A schematic diagram of a process for determining the target live broadcast room for this application;

[0082] Figure 8 A structural diagram of a display device for the recommended information of the live broadcast room of this application;

[0083] Fig. 9 A structural diagram of a device for displaying recommended information in the live broadcast room of this application. DETAILED DESCRIPTION

[0084] The following will be combined with the drawings in the embodiments of the present application to clearly and completely describe the technical solutions in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative work are within the scope of protection of this application.

[0085] The terms "first", "second", etc. in the specification and claims of this application and the above-mentioned drawings are used to distinguish similar objects, and are not necessarily used to describe a specific order or sequence. It should be understood that the data used in this way can be interchangeable where appropriate, so that the embodiments described here can be implemented in an order other than that illustrated or described here. In addition, the terms "including" and "having" and any of their variations are intended to cover non-exclusive inclusions, for example, a process, method, system, product or device that includes a series of steps or modules is not necessarily limited to those steps or modules clearly listed, but may include other steps or modules that are not clearly listed or inherent to these processes, methods, products or devices. The naming or numbering of steps in this application does not mean that the steps in the method flow must be executed in the time / logical sequence indicated by the naming or numbering. The process steps that have been named or numbered can change the execution order according to the technical purpose to be achieved, as long as the same or similar technical effects can be achieved.

[0086] The division of modules in this application is a logical division. There may be other division methods when it is implemented in actual applications. For example, multiple modules can be combined or integrated into another system, or some features can be ignored or not executed. In addition, the mutual coupling or direct coupling or communication connection shown or discussed can be through some interfaces, and the indirect coupling or communication connection between modules can be electrical or other similar forms, which are not limited in this application. In addition, the modules or submodules described as separate components may or may not be physically separated, may or may not be physical modules, or may be distributed in multiple circuit modules, and some or all of the modules may be selected according to actual needs to achieve the purpose of the present application.

[0087] Before introducing the method for displaying live broadcast room recommendation information provided by this application, the background content involved in this application is first introduced.

[0088] The method, device and computer-readable storage medium for displaying live broadcast room recommendation information provided in the present application can be applied to a display device for live broadcast room recommendation information to provide users with a personalized live broadcast room recommendation mechanism and improve the recommendation accuracy of the live broadcast room.

[0089] The method for displaying recommended information in a live broadcast room mentioned in this application can be executed by a recommendation device in the live broadcast room, or a server, physical host, or UE and other devices integrated with the display device for recommended information in the live broadcast room. The device can be implemented in hardware or software, and the UE can be a terminal device such as a smart phone, tablet computer, laptop computer, desktop computer, or personal digital assistant (PDA). The display device for recommended information in the live broadcast room can be set in the form of a device cluster.

[0090] Specifically, taking the application scenario of a local UE as an example, when a user triggers the running of a live broadcast platform application installed locally on the UE, the live broadcast platform can be accessed through the application, and the live broadcast room to be watched can be selected in the application interface, and clicked to watch. The application running on the UE can recommend live broadcast rooms on the application interface according to the recommended position preference characteristics configured by the user by applying the display method of live broadcast room recommendation information provided by the present application.

[0091] Taking the application scenario of the server as an example, when the user triggers the application of the live broadcast platform running locally on the UE, the live broadcast platform can be accessed through the application, and the live broadcast room to be watched can be selected in the application interface, and clicked to watch. The server behind the live broadcast platform can trigger the local UE to recommend the live broadcast room according to the recommendation position preference characteristics configured by the user on the application interface by applying the display method of the live broadcast room recommendation information provided by this application.

[0092] Taking the application scenario of local UE combined with server as an example, when the user triggers the application of the live broadcast platform running locally on the UE, the live broadcast platform can be accessed through the application, and the live broadcast room to be watched can be selected in the application interface, and clicked to watch. The server behind the live broadcast platform can determine the recommendation position preference characteristics configured by the user, and then the local UE recommends the live broadcast room based on the recommendation position preference characteristics.

[0093] It can be seen that the application scenarios of the method for displaying recommended information in the live broadcast room of the present application, or the display device for recommending information in the live broadcast room of the present application, have a variety of possible application forms, which can be adjusted according to actual needs and are not limited here.

[0094] At the same time, it should be noted that the application of the live broadcast platform mentioned in this application can not only be an application running locally on the UE, such as an independently running application (Application, APP), a mini-program, a built-in program in other APPs, etc., but can also be a web program triggered by the UE and running on the server. For example, the UE can access the web program provided by the live broadcast platform to the user through a browser application, so that the user can watch the live broadcast online through the browser application.

[0095] Next, we will introduce the method for displaying live broadcast room recommendation information provided by this application.

[0096] First, see Figure 1 , Figure 1 A flow chart of a method for displaying recommended information in a live broadcast room of the present application is shown. The method for displaying recommended information in a live broadcast room provided by the present application may specifically include the following steps S101 to S105:

[0097] Step S101, when a user logs into an application of a live broadcast platform currently running on a client, a preference survey page is output to the user;

[0098] In the present application, it can be understood that the recommendation of the live broadcast room is based on the individual user, so that different personalized recommendation mechanisms can be implemented for different users.

[0099] In contrast, the process of personalized recommendations in the live broadcast room involves the identification of individual users, which is based on the user's personal account.

[0100] It is easy to understand that after a user registers a personal account on the live broadcast platform, a unique account identification (ID) can be generated for the personal account, such as "102d5056g098f897y". The account ID can be a digital string, or an alphabetic string, or a combination of numbers and letters, etc.

[0101] In addition, the account ID involved in this application can be a unique ID generated when a personal account is generated, or it can be a unique ID generated or recorded by a user in the process of using a personal account, such as a nickname, mobile phone number, etc.

[0102] It can be seen here that the account ID referred to in this application can be not only identification information directly bound to the user's personal account, but also identification information indirectly bound to the user's personal account.

[0103] After the user logs in to the application of the live broadcast platform currently running on the client and confirms the user identity through the account ID, a preference survey page can be output to the user to trigger subsequent personalized recommendation processing.

[0104] Step S102, obtaining the recommended position preference features input by the user based on the preference survey page, the recommended position preference features including image position features and image style features;

[0105] The recommendation position preference feature is the preference feature of the user for the corresponding recommendation position when the application recommends the live broadcast room on the application interface. It can specifically include the image position feature and the image style feature of the recommendation position. The recommendation position can be understood as the location of the live broadcast room when the application recommends the live broadcast room on the application interface, or it can be understood as a recommended resource of the live broadcast room on the application interface. It corresponds to different scenarios of local UE watching live broadcasts. For example, it can be divided into the slide show recommendation position on the homepage of the Web end, the banner recommendation position on the Web end, and the banner recommendation position on the mobile end. The example scenarios can be referred to Figures 2 to 4 Schematic diagrams of different scenarios of the recommended positions of this application are shown.

[0106] It can be understood that when the live broadcast platform recommends a live broadcast room to users in the recommendation position on the application interface, it exists in the form of an image. This application believes that the image may have differences in the performance effects in the position outside the image and the style inside the image. Therefore, this application extracts the image position features and image style features to characterize the recommendation form that users can intuitively view in the process of recommending the live broadcast room.

[0107] Image position features can be directly expressed in the form of pixel positions, such as "from the left to the right, the x1th pixel to the x2th pixel, from the top to the bottom, the y1th pixel to the y2th pixel"; or they can be expressed in the form of a recommendation module (which can be understood as a recommendation control on the application interface, with a specific deployment position or occupied position), such as "the third recommendation module".

[0108] Image style features can be specifically characterized by elements that may be involved in the image, such as the text content style, text style, font style, image body style, image background style, etc. in the recommended information.

[0109] It can be understood that the above-mentioned feature expressions are only examples. In practical applications, the specific feature expressions can be adjusted according to actual needs, and this application does not make any specific restrictions here.

[0110] Further explanation of the source of obtaining the recommendation position preference characteristics:

[0111] The acquisition of the recommendation position preference feature, in this application, can be understood as the feature obtained by the user's self-selection in the preference survey interface displayed on the live broadcast platform. This scenario, in layman's terms, is similar to the questionnaire survey scenario, that is, the acquisition of the recommendation position preference feature can be:

[0112] Outputting a preference survey page to the user;

[0113] Receive the answer information entered by the user on the preference survey page;

[0114] Based on the answer information, the preference characteristics of the recommended positions are analyzed.

[0115] It is understandable that the preference survey page is set based on the subsequent recommendation position preference features, for example, the user's preferred image position, image style, etc. can be asked.

[0116] The recommendation position preference features obtained based on the answer information entered by the user can be directly extracted from the user's answer information. At this time, the optional options of the recommendation position preference features should directly appear or be prompted in the preference survey page, or the information should be directly filled in by the user.

[0117] If the recommendation position preference feature is indirectly extracted from the user's answer information, it can be further analyzed and processed based on the user's answer information.

[0118] For example, based on the recommendation position preference features a1, a2, and a3 of multiple small categories, the recommendation position preference feature A above these recommendation position preference features, that is, the recommendation position preference feature of the large category, can be determined; for another example, multiple similar recommendation position preference features a1 and b2 can be merged into a unified recommendation position preference feature C; for another example, based on the recommendation position preference feature d2, it is predicted that the user also has a potential recommendation position preference feature g5.

[0119] In addition, in an exemplary implementation, the questionnaire survey scenario can also be set dynamically, see Figure 5 A schematic diagram of a process flow of the preference survey processing of the present application is shown. The preference survey processing involved in the present application may include the following steps S501 and S502:

[0120] Step S501: when a user logs into an application of a live broadcast platform currently running on a client, sample recommendation information of a live broadcast room is output to the user in a preference survey page according to different image position features and different image style features;

[0121] It can be understood that in addition to the basic blank items for users to fill in the recommendation position preference characteristics according to their own viewing habits, the present application can also provide dynamic sample display scenarios, so that users can determine the recommendation position preference characteristics based on their own viewing habits based on a more intuitive experience.

[0122] In this dynamic sample display scenario, the sample recommendation information of the live broadcast room can be output on the application interface in different display modes according to different preset image position features and / or image style features.

[0123] Step S502, receiving the user's attention to different image position features and different image style features as answer information.

[0124] In the dynamic sample display scenario, this application also introduces an attention mechanism, whereby users fill in the corresponding attention based on their short-term memory of the different sample recommendation information displayed above. In this way, the user's interest, acceptance or preference for different recommendation position preference features, that is, different image position features and different image style features, can be determined concisely and clearly.

[0125] Exemplarily, the dynamic sample display scene may specifically be a nine-grid short-term display scene.

[0126] Among them, in the preference survey scenario of image position features, the user can click on the pop-up window to confirm the answer, and then the nine-square grid short-term memory answering tool interface will be presented.

[0127] Specifically, the recommended information of the live broadcast room will be displayed randomly in several pre-set locations;

[0128] Then, the default setting is to disappear after 2 seconds of exposure, and ask the user:

[0129] 1) Among the images just presented, which modules in which position impressed you the most?

[0130] 2) Please tick the reason that impressed you most;

[0131] Then you can proceed to the next question. After answering all questions, click Submit to end the questionnaire.

[0132] The code design involved may include the following:

[0133] 1. Read user information

[0134]

[0135]

[0136] 2. Timed switching of pictures in different locations

[0137]

[0138] 3. Implementation of the answering interface

[0139] Use the multiple-selection box to select the area that impressed you the most

[0140] lab=tk.Label(root,text='Please select the area that impressed you the most:')

[0141] lab.grid(row=0,column=6,sticky=tk.W)

[0142] var1 = tk.IntVar()

[0143] Answer the reason part

[0144] lab2 = tk.Label (root, text = 'The reason that impressed you most is:')

[0145] lab2.grid(row=0,column=8,sticky=tk.W,padx=10)

[0146] frm2 = tk.Frame(root)

[0147] var10 = tk.IntVar() # Set different variables. By default, the variable option is set to 1 to indicate the selected state, otherwise it is set to 0.

[0148] 4. Statistics of answer results

[0149]

[0150]

[0151] Among them, in the questionnaire survey scenario with image style features, users can click on the pop-up window to confirm the answer, and then the nine-square grid short-term memory answering tool interface will be presented.

[0152] Specifically, material pictures may be presented, and the user may be asked to select “interested” or “not interested” in the material based on their first reaction.

[0153] Present another set of comparison images and ask users to choose the image they are more interested in based on their first reaction.

[0154] The code design involved may include the following:

[0155] 1. For picture switching, please refer to the code principle of the timed switching of pictures at different positions involved in the questionnaire survey scenario of the above image position characteristics.

[0156] 2. Get interesting and uninteresting key information

[0157]

[0158]

[0159] 3. Statistical answer results

[0160]

[0161] It can be understood that manually filling in the attention paid to different image position features and different image style features by users is easy to implement technically and apply, and has a better advantage in application cost.

[0162] Furthermore, in yet another exemplary implementation, in a dynamic sample display scenario, in addition to manually completing the questionnaire answers based on the dynamically displayed images, the questionnaire answers can also be automatically completed based on the dynamically displayed images.

[0163] See also Figure 6 Another flowchart of the preference survey process of the present application is shown. The preference survey process involved in the present application may include the following steps S601 to S603:

[0164] Step S601: when a user logs into an application of a live broadcast platform currently running on a client, sample recommendation information of a live broadcast room is output to the user in a preference survey page according to different image position features and different image style features;

[0165] It can be understood that in addition to the basic blank items for users to fill in the recommendation position preference characteristics according to their own viewing habits, the present application can also provide dynamic sample display scenarios, so that users can determine the recommendation position preference characteristics based on their own viewing habits based on a more intuitive experience.

[0166] In this dynamic sample display scenario, the sample recommendation information of the live broadcast room can be output on the application interface in different display modes according to different preset image position features and / or image style features.

[0167] Step S602, under the user's confirmation, photographing a facial image of the user in the process of viewing the sample recommendation information through a camera;

[0168] It can be understood that the present application may also consider using eye tracking technology to automatically determine the user's attention level to the sample recommendation information.

[0169] First of all, face tracking technology starts from the image processing level. Therefore, considering user privacy, after user confirmation, the camera is used to capture the user's facial image while he or she is viewing the sample recommendation information.

[0170] Among them, the camera can be a camera configured for the display device of the recommended information in the live broadcast room, or a camera other than the display device of the recommended information in the live broadcast room, or even a camera configured on other devices other than the display device of the recommended information in the live broadcast room. The specific details can be adjusted according to the actual application scenario.

[0171] Step S603, under the image recognition processing, the human face image is tracked by the human eye, and the user's attention to different image position features and different image style features is determined according to the face tracking result as the answer information.

[0172] Eye tracking, also known as eyeball tracking, can be understood as tracking the image features of the human eye to capture subtle changes in the user's eyes, thereby determining the focus position of the human eye.

[0173] Naturally, the shorter the time from the appearance of sample feature information to the user's attention to the sample feature information, the greater the user's attention can be determined to be.

[0174] Similarly, the longer the user focuses on the same sample feature information, the greater the user's attention the sample feature information can attract.

[0175] Among them, the image recognition processing involved can generally be achieved through artificial intelligence (AI) technology, that is, human eye tracking is performed on the input face image through a trained neural network model.

[0176] The neural network model may be a YOLOv3 model, a ResNet model, an R-CNN model, a Fast R-CNN model, a Faster R-CNN model, a Mask R-CNN model, an SSD model, or other different types of models.

[0177] The training process of the model may include: inputting different face images marked with corresponding eye tracking results into the initial neural network model, performing forward propagation, optimizing model parameters according to the output eye tracking results, and performing back propagation. When the preset training requirements such as the number of training times, training duration, and recognition accuracy are met, the training is completed and the obtained model can be put into practical use.

[0178] It can be understood that the user's attention to different image position features and different image style features determined by eye tracking is automatically determined by a machine, and therefore has better objectivity and accuracy than being determined and manually filled in by the user himself, and can further reduce the manual operations required by the user.

[0179] After obtaining the user's recommendation position preference characteristics, the recommendation information of the target live broadcast room can be output to the user in the recommendation method corresponding to the recommendation position preference characteristics.

[0180] It can be understood that the recommendation information is displayed in the form of an image in combination with the content of the image. During the display process, the image position feature and the image style feature in the recommendation position preference feature are followed.

[0181] It can be understood that the main consideration of this application is to improve the recommendation accuracy of the live broadcast room in the recommendation form (image form) by combining the recommendation position preference characteristics.

[0182] The recommended target live broadcast room itself can be recommended by the live broadcast platform to multiple users simultaneously, or in other words, the same live broadcast room can be recommended among users.

[0183] In further research, we can continue to optimize and improve the selection strategy of the recommended live broadcast room itself, and recommend personalized live broadcast rooms to individual users, continue to improve user acceptance, and thus improve the accuracy of recommendations.

[0184] The target live broadcast room is recommended according to the recommendation position preference feature. On the one hand, it is considered from the perspective of the live broadcast room itself. The recommended target live broadcast room follows the image style feature in the recommendation position preference feature. On the other hand, the position of the recommendation information of the target live broadcast room on the application interface follows the image position feature summarized by the recommendation position preference feature. That is, after step S102, the following steps S103 and S104 are also included:

[0185] Step S103, determining a target live broadcast room that matches the image style feature from the current live broadcast room set of the live broadcast platform;

[0186] It can be understood that the target live broadcast room that matches the image style features in the recommendation position preference features can be specifically understood as the image style of the recommended information of the target live broadcast room, which conforms to the image style features in the recommendation position preference features.

[0187] In this process, the image style of the recommended information of the target live broadcast room can be pre-configured, or can be configured in real time according to the image style features in the current recommendation position preference features.

[0188] The live broadcast rooms in the current live broadcast room set of the live broadcast platform generally refer to the live broadcast rooms that are in the broadcast state. It can be all the live broadcast rooms that are in the broadcast state on the current live broadcast platform, or it can be some of the live broadcast rooms that are in the broadcast state on the current live broadcast platform, for example, all the live broadcast rooms that are in the broadcast state and have high popularity on the current live broadcast platform. The specific matching range can be determined according to the actual situation and is not limited here.

[0189] Step S104, adjusting the default display page of the live broadcast platform to a target display page corresponding to the image position feature;

[0190] Compared with the page layout of each application interface of the live broadcast platform in the prior art, for example, the div+css code of the web page is fixed, which leads to a fixed page layout. The fixed page can be understood as the default page mentioned in this application, or a page that is difficult to configure as flexibly as in this application.

[0191] The present application considers a more flexible and effective processing method for the image position of the recommended information, that is, according to the image position features in the user's recommendation position preference features, it can not only be understood as selecting an adaptive page layout position, but also can be understood as adjusting the page layout of the default page. This flexibly and effectively complies with the image position features in the recommendation position preference features and complies with the current user's preference habits.

[0192] For example, user A likes the recommendation position to be located in the bottom corner of the application interface. Then the recommendation information of the target live broadcast room that the server returns to user A will be located in the bottom corner of the application interface when it is displayed on the application interface, while the recommendation position of the original default page may be located next to the current video playback interface.

[0193] The adjustment of the page layout may include not only the adjustment of the position of the recommendation position of the fixed window size, but also the adjustment of the position of the recommendation position and its window size, and further, the adjustment of other elements in the page.

[0194] Step S105, display the recommended information of the target live broadcast room on the target display page.

[0195] Subsequently, after determining the image position of the target live broadcast room and its recommended information through the user's current recommendation position preference characteristics, the recommended information of the target live broadcast room can be displayed at the specific image position of the recommended information in the adaptively adjusted target display page, thereby completing the personalized recommendation mechanism for the live broadcast room.

[0196] It can be seen from the above embodiments that in the recommendation process of the present application, since the recommendation information of the target live broadcast room is output according to the user's recommendation position preference characteristics, it is more in line with the current user's operating habits and the user's acceptance is significantly improved, thereby realizing a personalized live broadcast room recommendation mechanism for users, providing more matching judgment information for user decision-making, thereby improving the recommendation accuracy of the live broadcast room, and at the same time improving resource utilization under limited recommendation resources, thereby meeting the recommendation needs of both the live broadcast platform and the user.

[0197] In addition, there may be further optimization settings for the recommended target live broadcast room itself, and this content will continue to be explained.

[0198] In yet another exemplary implementation, see Figure 7 The present application shows a flow chart of determining a target live broadcast room, which determines a target live broadcast room that matches the image style feature from the current live broadcast room set of the live broadcast platform, and specifically may include the following steps S701 to S703:

[0199] Step S701, determining the first live broadcast room on the live broadcast platform that reaches a preset click rate;

[0200] Step S702, determining the second live broadcast room that the user has watched in the historical time period and meets the preset conditions in the viewing records opened by the user account;

[0201] Step S703, in the first live broadcast room, according to the similarity between the working status characteristics of each live broadcast room and the working status characteristics of the second live broadcast room, select the third live broadcast room as the target live broadcast room, and the similarity corresponding to the third live broadcast room is within the preset similarity range.

[0202] It can be understood that the target live broadcast room based on user's personal customization can be determined in this application based on the user's personal viewing habits and the viewing habits of other users. In this way, it can take into account the user's personal acceptance level and combine the viewing habits of major users to determine the live broadcast room that matches the user and that the user has never been exposed to, has less exposure to, or is of potential interest.

[0203] In the process of determining the target live broadcast room, for individual users, a representative first live broadcast room can be selected from the live broadcast rooms watched in the historical period from the user's personal viewing records. The second live broadcast room can be understood as the user's affiliated live broadcast room, the user's affiliated anchor, etc. Generally, the live broadcast room in the first place or the front row can be determined as the first live broadcast room based on the viewing time or the number of views.

[0204] For major users, the viewing habits of major users can be directly reflected from the original first live broadcast room with a higher click-through rate on the live broadcast platform. These live broadcast rooms have higher traffic, retention and other indicators. Therefore, major users generally have a high degree of acceptance and interest, and they can be regarded as representative high-quality live broadcast rooms.

[0205] In this way, the first live broadcast room can be used as the selection range, and the second live broadcast room can be used as the selection basis. A matching third live broadcast room can be selected from the first live broadcast room and recommended as the target live broadcast room.

[0206] The matching process is specifically based on the working status characteristics of the live broadcast room (characterizing different elements of the live broadcast room, configured in the form of vectors), combined with the similarity between the working status feature vectors for matching. The higher the similarity, the higher the matching degree.

[0207] In actual applications, a corresponding number of target live broadcast rooms can be selected based on the preset number of target live broadcast rooms (i.e., the number of recommended resource bits configured on the application interface).

[0208] Of course, in addition to the preset number, the preset similarity range can also be set to other ranges. For example, as long as there is a live broadcast room with a higher similarity than the preset number, it can be recommended as the third live broadcast room. If the number is large, it can be recommended in turn according to the similarity ranking under the condition of a fixed preset number. Alternatively, the live broadcast rooms that meet the fixed preset number can be randomly selected for recommendation. It can be seen that the preset similarity range can be flexibly adjusted in actual applications, and no specific limitation is made here.

[0209] For example, the similarity is calculated as follows:

[0210] Euclidean distance

[0211] Assuming that the working status characteristics of two live broadcast rooms (tentatively n dimensions) are p (p1, p2, p3, p4...) and q (q1, q2, q3, q4,...), the similarity between the two can be calculated using the Euclidean distance similarity method:

[0212]

[0213] Improved standardized Euclidean distance

[0214] Standardized Euclidean distance is an improvement to the shortcomings of simple Euclidean distance (the distribution of components in each dimension is different). Each component is standardized. Assuming that the mean of the sample set X is m and the standard deviation is s, the "standardized variable" of X is expressed as:

[0215]

[0216] That is, the standardized value = (the value before standardization - the mean of the component) / the standard deviation of the component.

[0217] Assuming that the working status characteristics of two live broadcast rooms (temporarily set to n dimensions) are a(a1, a2, …, an) and b(b1, b2, …, bn), the similarity between the two can be calculated using the standardized Euclidean distance:

[0218]

[0219] Cosine similarity

[0220] Assuming that the working status characteristics of the two live broadcast rooms (tentatively set to 2 dimensions) are a(x1,y1) and b(x2,y2), respectively, under the similarity calculation method of cosine similarity, the similarity between the two can be:

[0221]

[0222] If it is n-dimensional, the similarity between the two can be:

[0223]

[0224] Normalized cosine similarity

[0225] Cosine similarity is more about distinguishing differences in direction, and is not sensitive to absolute values, so this application can also introduce modified cosine similarity, or normalized cosine similarity, which is an algorithm for normalizing cosine similarity. The formula is as follows:

[0226]

[0227] Furthermore, in another exemplary implementation, considering that Euclidean distance (including standardized Euclidean distance) and cosine similarity (including normalized cosine similarity) have certain data differences in specific applications, or that each has certain limitations, for example:

[0228] In the Euclidean distance algorithm, each coordinate contributes equally to the Euclidean distance, without considering the impact of overall variation on the distance, which results in the size of the "distance" being related to the unit of the indicator;

[0229] In the normalized cosine similarity algorithm, although the influence of the absolute value difference of each vector is solved by normalization, when normalizing and calculating the R mean, if there is a missing value in the vector, the vector is ignored instead of being set to 0, which is unreasonable in certain scenarios.

[0230] Therefore, the present application also considers the characteristics of the final similarity interval Euclidean distance and cosine similarity, and takes the sum of the two as the final similarity to provide data support for matching judgment.

[0231] That is, the similarity determination process specifically includes:

[0232] Calculate the Euclidean distance and cosine similarity between two working state features whose similarity is to be calculated respectively. The Euclidean distance can be calculated by a Euclidean distance calculation formula or a standardized Euclidean distance calculation formula, and the pre-similarity can be calculated by a cosine similarity calculation formula or a normalized pre-similarity calculation formula;

[0233] The calculated Euclidean distance and cosine similarity are added together to obtain the final similarity.

[0234] Taking the standardized Euclidean distance and the normalized cosine similarity as examples, the final similarity ranking can be specifically expressed as:

[0235]

[0236] Among them, the rank() function ranks the values ​​in ascending order (the larger the value, the lower the ranking and the lower the matching degree). k is the kth index of the first live broadcast room, i k is the kth index of the second live broadcast room i, s k is the standard deviation; R is the mean of each vector, R u，A , R u，i Same as above.

[0237] In this setting, the two similarity algorithms, standardized Euclidean distance and normalized cosine similarity, are combined, which not only considers the absolute value of the indicator, but also the normalization of the indicator and the impact of missing values, optimizing the limitations of the two algorithms to a certain extent.

[0238] Furthermore, in another exemplary implementation, the user's personal second live broadcast room may also be one or more live broadcast rooms determined according to the ranking strategy.

[0239] The Euclidean-cosine combination algorithm considers similarity more from the dimension of the live broadcast room (which can also be understood as the anchor dimension. Generally, the live broadcast room and the anchor are in a one-to-one binding relationship). This application can further consider the similarity between users. In this case, the determination scheme of the second live broadcast room is as follows:

[0240] 1. Calculate the proportion of user i's viewing time in the current live broadcast room in the first live broadcast room R i :

[0241] R i=t i / T i

[0242] Among them, t i is the viewing time of user i in the current room, T i The total viewing time of user i on the live broadcast platform.

[0243] 2. If R i ≥0.5, then user i is determined to belong to the current live broadcast room, and the user group C watching the current live broadcast room is selected A .

[0244] 3. According to user group C A The viewing records opened by each user are used to obtain the total viewing time T of each potential live broadcast room B :

[0245]

[0246] in, User Group C A The viewing time of medium users in potential live broadcast rooms.

[0247] 4. Calculate the ranking of potential live broadcast rooms based on the total viewing time of each potential live broadcast room, and determine the live broadcast room with the best ranking or within the preset ranking range as the second live broadcast room:

[0248] Rank=rank(T B )

[0249] The rank() function ranks the values ​​in descending order (the larger the value, the higher the ranking and the higher the matching degree)

[0250] In this case, the second live broadcast room determined according to the ranking (first place or the top few places) also has the characteristics of collaborative filtering, which can better determine the second live broadcast room that is representative for the user i.

[0251] At the same time, the collaborative filtering element can also be introduced when matching the user's personal second live broadcast room with the first live broadcast room of the live broadcast platform. For example, the total viewing time of the second live broadcast room involved in the ranking of potential live broadcast rooms can be used as a ranking element in the matching process.

[0252] Still taking the standardized Euclidean distance and the normalized cosine similarity as an example, the final similarity ranking can be specifically expressed as:

[0253]

[0254] In another exemplary implementation, the target live broadcast room can also be directly obtained from the live broadcast platform based on the matching of the recommendation position preference characteristics.

[0255] That is, from the set of live broadcast rooms, a live broadcast room that has obtained a preset degree of attention according to the recommendation position preference characteristics in a historical time period or a predicted time period is determined as the target live broadcast room.

[0256] The preset attention can be reflected in indicators such as click-through rate, total viewing time, total number of viewers, etc., and the screening range can be set. For example, it can be set to a flexible range such as maximum attention, top three attention, attention threshold, etc.

[0257] In addition, when determining whether the degree of attention has reached the preset degree of attention, it can not only be determined based on the performance range of the live broadcast room in the historical time period, but also the performance information of the live broadcast room in the historical time period can be considered as a kind of time series information for data processing. In this way, the performance in the predicted time period can be predicted based on the extracted time series features, and the corresponding degree of attention can be determined. In this way, live broadcast rooms with better potential can also be predicted for recommendation.

[0258] The above is an introduction to the method for displaying recommended information in the live broadcast room provided by this application. In order to facilitate better implementation of the method for displaying recommended information in the live broadcast room provided by this application, this application also provides a device for displaying recommended information in the live broadcast room from the functional module level.

[0259] See also Figure 8 , Figure 8 This is a structural diagram of a display device for recommended information in a live broadcast room of the present application. In the present application, the display device 800 for recommended information in a live broadcast room may specifically include the following structure:

[0260] The output unit 801 is used to output a preference survey page to the user when the user logs in to the application of the live broadcast platform currently running on the client;

[0261] An acquisition unit 802 is used to acquire a recommendation position preference feature input by a user based on a preference survey page, wherein the recommendation position preference feature includes an image position feature and an image style feature;

[0262] A determination unit 803 is used to determine a target live broadcast room that matches the image style feature from the current live broadcast room set of the live broadcast platform;

[0263] An adjusting unit 804, configured to adjust a default display page of the live broadcast platform to a target display page corresponding to the image position feature;

[0264] The display unit 805 is used to display the recommendation information of the target live broadcast room on the target display page.

[0265] In an exemplary implementation, the acquiring unit 802 is specifically configured to:

[0266] Receive the answer information entered by the user on the preference survey page;

[0267] Based on the answer information, the preference characteristics of the recommended positions are analyzed.

[0268] In yet another exemplary implementation, the output unit 801 is specifically configured to:

[0269] When the user logs in to the application of the live broadcast platform currently running on the client, sample recommendation information of the live broadcast room is output to the user in the preference survey page according to different image position features and different image style features;

[0270] The acquisition unit 802 is specifically used for:

[0271] The user's attention to different image position features and different image style features is received as answer information.

[0272] In yet another exemplary implementation, the output unit 801 is specifically configured to:

[0273] When the user logs in to the application of the live broadcast platform currently running on the client, sample recommendation information of the live broadcast room is output to the user in the preference survey page according to different image position features and different image style features;

[0274] The acquisition unit 802 is specifically used for:

[0275] Under the user's confirmation, a facial image of the user while viewing the sample recommendation information is captured by a camera;

[0276] Under image recognition processing, the face image is tracked by the eyes, and the user's attention to different image position features and different image style features is determined based on the face tracking results as answer information.

[0277] In yet another exemplary implementation, the determining unit 803 is specifically configured to:

[0278] Determine the first live broadcast room that reaches a preset click rate from the current live broadcast room set of the live broadcast platform;

[0279] In the viewing records opened by the user, determine the second live broadcast room that the user has watched in the historical time period and meets the preset conditions;

[0280] In the first live broadcast room, according to the similarity between the working status characteristics of each live broadcast room and the working status characteristics of the second live broadcast room, the third live broadcast room is selected as the target live broadcast room, and the corresponding similarity of the third live broadcast room is within the preset similarity range.

[0281] In another exemplary implementation, the similarity determination process specifically includes:

[0282] Calculate the Euclidean distance and cosine similarity between two working state features whose similarity is to be calculated respectively. The Euclidean distance can be calculated by a Euclidean distance calculation formula or a standardized Euclidean distance calculation formula, and the pre-similarity can be calculated by a cosine similarity calculation formula or a normalized pre-similarity calculation formula;

[0283] The calculated Euclidean distance and cosine similarity are added together to obtain the final similarity.

[0284] In yet another exemplary implementation, the determination process of the second live broadcast room includes:

[0285] Calculate the proportion of the user's viewing time in the current live broadcast room in the first live broadcast room;

[0286] If the viewing time ratio is greater than the preset value, the user is determined to belong to the current live broadcast room, and the user group watching the current live broadcast room is circled;

[0287] Calculate the total viewing time of each potential live broadcast room based on the viewing records opened by each user in the user group;

[0288] The ranking is calculated based on the total viewing time of each potential live broadcast room, and the live broadcast room with the best ranking or within the preset ranking range is determined as the second live broadcast room;

[0289] The similarity determination process specifically includes:

[0290] The final similarity is obtained by calculating the Euclidean distance, cosine similarity and the total viewing time of the second live broadcast room.

[0291] This application also provides a display device for live broadcast room recommendation information from the hardware level, see Fig. 9 , Fig. 9 A schematic diagram of a structure of a display device for recommended information in a live broadcast room of the present application is shown. Specifically, the display device for recommended information in a live broadcast room of the present application may include a processor 901, a memory 902, and an input / output device 903. The processor 901 is used to execute a computer program stored in the memory 902 to implement the following Figure 1 The steps of the method for displaying recommended information in the corresponding embodiment of the live broadcast room; or, the processor 901 is used to execute the computer program stored in the memory 902 to implement the following Figure 8 Corresponding to the functions of each unit in the embodiment, the memory 902 is used to store the processor 901 executing the above Figure 1 The computer program required for the method of displaying recommended information in the live broadcast room in the corresponding embodiment.

[0292] Exemplarily, the computer program may be divided into one or more modules / units, one or more modules / units are stored in the memory 902, and executed by the processor 901 to complete the present application. One or more modules / units may be a series of computer program instruction segments capable of completing specific functions, and the instruction segments are used to describe the execution process of the computer program in the computer device.

[0293] The display device of the recommended information in the live broadcast room may include, but is not limited to, the processor 901, the memory 902, and the input-output device 903. Those skilled in the art will understand that the illustration is only an example of the display device of the recommended information in the live broadcast room, and does not constitute a limitation on the display device of the recommended information in the live broadcast room. It may include more or less components than shown in the illustration, or combine certain components, or different components. For example, the display device of the recommended information in the live broadcast room may also include a network access device, a bus, etc., and the processor 901, the memory 902, the input-output device 903, etc. are connected through the bus.

[0294] Processor 901 may be a central processing unit (CPU), or other general-purpose processors, digital signal processors (DSP), application-specific integrated circuits (ASIC), field-programmable gate arrays (FPGA) or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. A general-purpose processor may be a microprocessor or any conventional processor, etc. The processor is the control center of the display device for the recommended information in the live broadcast room, and uses various interfaces and lines to connect various parts of the entire device.

[0295] The memory 902 can be used to store computer programs and / or modules. The processor 901 implements various functions of the computer device by running or executing the computer programs and / or modules stored in the memory 902 and calling the data stored in the memory 902. The memory 902 can mainly include a program storage area and a data storage area, wherein the program storage area can store an operating system, at least one application required for a function, etc.; the data storage area can store data created according to the use of a display device for recommending information in the live broadcast room, etc. In addition, the memory can include a high-speed random access memory, and can also include a non-volatile memory, such as a hard disk, a memory, a plug-in hard disk, a smart memory card (Smart Media Card, SMC), a secure digital (Secure Digital, SD) card, a flash card (Flash Card), at least one disk storage device, a flash memory device, or other volatile solid-state storage devices.

[0296] When the processor 901 is used to execute the computer program stored in the memory 902, the following functions can be implemented:

[0297] When the user logs in to the application of the live broadcast platform currently running on the client, a preference survey page is output to the user;

[0298] Obtaining the recommended position preference features input by the user based on the preference survey page, the recommended position preference features including image position features and image style features;

[0299] Determine a target live broadcast room that matches the image style feature from the current live broadcast room set of the live broadcast platform;

[0300] Adjust the default display page of the live broadcast platform to the target display page corresponding to the image position feature;

[0301] The recommended information of the target live broadcast room is displayed on the target display page.

[0302] Those skilled in the art can clearly understand that for the convenience and simplicity of description, the specific working process of the display device, equipment and corresponding units of the live broadcast room recommendation information described above can refer to the following. Figure 1 The description of the method for displaying the recommended information in the live broadcast room in the corresponding embodiment will not be repeated here.

[0303] A person of ordinary skill in the art will appreciate that all or part of the steps in the various methods of the above embodiments may be completed by instructions, or by controlling related hardware through instructions. The instructions may be stored in a computer-readable storage medium and loaded and executed by a processor.

[0304] To this end, the present application provides a computer-readable storage medium, in which a plurality of instructions are stored, and the instructions can be loaded by a processor to execute the present application as follows: Figure 1 The steps in the method for displaying recommended information in the corresponding embodiment of the live broadcast room can be referred to as follows for specific operations. Figure 1 The description of the method for displaying recommended information in the live broadcast room in the corresponding embodiment will not be repeated here.

[0305] The computer-readable storage medium may include: a read-only memory (ROM), a random access memory (RAM), a magnetic disk or an optical disk, etc.

[0306] Due to the instructions stored in the computer-readable storage medium, the present application can be executed. Figure 1 The steps in the method for displaying recommended information in the corresponding embodiment of the live broadcast room can thus be implemented as follows: Figure 1 The beneficial effects that can be achieved by the method for displaying recommended information in the live broadcast room in the corresponding embodiment are detailed in the previous description and will not be repeated here.

[0307] The above is a detailed introduction to the display method, device, equipment and computer-readable storage medium for the live broadcast room recommendation information provided by the present application. Specific examples are used in this article to illustrate the principles and implementation methods of the present application. The description of the above embodiments is only used to help understand the method of the present application and its core idea; at the same time, for technical personnel in this field, according to the idea of ​​the present application, there will be changes in the specific implementation method and application scope. In summary, the content of this specification should not be understood as a limitation on the present application.

Claims

1. A method for displaying recommended information in a live broadcast room, characterized in that: The method comprises: When a user logs in to an application of a live broadcast platform currently running on a client, a preference survey page is output to the user; Acquire the recommended position preference features input by the user based on the preference survey page, wherein the recommended position preference features include image position features and image style features; Determine a target live broadcast room that matches the image style feature from a current live broadcast room set of the live broadcast platform; Adjusting the default display page of the live broadcast platform to a target display page corresponding to the image position feature; Displaying the recommended information of the target live broadcast room on the target display page; The obtaining of the recommended position preference characteristics input by the user based on the preference survey page includes: Receiving answer information entered by the user on the preference survey page; Based on the answer information, analyzing and obtaining the recommendation position preference characteristics; When the user logs in to the application of the live broadcast platform currently running on the client, outputting a preference survey page to the user includes: When a user logs in to an application of a live broadcast platform currently running on a client, sample recommendation information of a live broadcast room is output to the user in the preference survey page according to different image position features and different image style features; The receiving the answer information entered by the user on the preference survey page includes: receiving the user's attention to the different image position features and the different image style features as the question answering information; When the user logs in to the application of the live broadcast platform currently running on the client, outputting a preference survey page to the user includes: When a user logs in to an application of a live broadcast platform currently running on a client, sample recommendation information of a live broadcast room is output to the user in the preference survey page according to different image position features and different image style features; The receiving the answer information entered by the user on the preference survey page includes: Under the determination of the user, capturing a facial image of the user in the process of viewing the sample recommendation information through a camera; Under the image recognition processing, the human face image is subjected to eye tracking, and the user's attention to the different image position features and the different image style features is determined according to the face tracking result as the answer information; The determining a target live broadcast room matching the image style feature from the current live broadcast room set of the live broadcast platform includes: Determine, from the current set of live broadcast rooms on the live broadcast platform, a first live broadcast room that reaches a preset click rate; Determine, from the viewing records opened by the user, a second live broadcast room that the user has viewed in a historical time period and meets a preset condition; In the first live broadcast room, according to the similarity between the working status characteristics of each live broadcast room and the working status characteristics of the second live broadcast room, a third live broadcast room is selected as the target live broadcast room, and the similarity corresponding to the third live broadcast room is within a preset similarity range; The similarity determination process specifically includes: Respectively calculating the Euclidean distance and cosine similarity between two working state features whose similarities are to be calculated, wherein the Euclidean distance is calculated by a Euclidean distance calculation formula or a standardized Euclidean distance calculation formula, and the cosine similarity is calculated by a cosine similarity calculation formula or a normalized cosine similarity calculation formula; Adding the calculated Euclidean distance and the cosine similarity to obtain a final similarity; The determination process of the second live broadcast room includes: Calculate the proportion of the viewing time of the current live broadcast room in the first live broadcast room watched by the user; If the viewing time ratio is greater than a preset value, it is determined that the user belongs to the current live broadcast room, and the user group watching the current live broadcast room is circled; Calculate the total viewing time of each potential live broadcast room based on the viewing records opened by each user in the user group; Calculate the ranking of each potential live broadcast room according to the total viewing time, and determine the live broadcast room with the best ranking or within a preset ranking range as the second live broadcast room; The calculated Euclidean distance and cosine similarity are added together to obtain a final similarity, including: The final similarity is obtained by calculating the Euclidean distance, the cosine similarity, and the total viewing time of the second live broadcast room.

2. A display device for recommended information in a live broadcast room, characterized in that: The method comprises a processor and a memory, wherein a computer program is stored in the memory, and the method according to claim 1 is executed when the processor calls the computer program in the memory.

3. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores a plurality of instructions, and the instructions are suitable for being loaded by a processor to execute the method of claim 1.

Images