Data processing method, device, electronic device and storage medium

Abstract

The present application discloses a data processing method, device, electronic device and storage medium. The method includes: obtaining live broadcast data of a target live broadcast room; identifying a live broadcast event occurring in the target live broadcast room according to the live broadcast data; determining a target product when the identified live broadcast event is a product switching event; the target product is a product being displayed in the target live broadcast room; determining a target advertisement creative corresponding to the target product; the target advertisement creative is used to deliver a target advertisement to a target device; when the target device detects a click operation on the target advertisement, the target live broadcast room can be entered through the target device.

Description

Technical Field

[0001] The present application relates to the field of computer technology, and in particular to a data processing method, device, electronic device and storage medium. Background Art

[0002] With the popularization of the fourth generation of mobile communication technology (4G) and the fifth generation of mobile communication technology (5G), the era of short videos and self-media has arrived. Among them, live broadcasting is an important carrier. It can meet the marketing needs of merchants by having anchors bring goods in the live broadcast room, and put promotional advertisements on user terminals as the traffic entrance of the live broadcast room. In order to improve the user's viewing experience, how to improve the correlation between promotional advertisements and live broadcast content has become an urgent problem to be solved. Summary of the invention

[0003] To solve related technical problems, the embodiments of the present application provide a data processing method, device, electronic device and storage medium.

[0004] The technical solution of the embodiment of the present application is implemented as follows:

[0005] The present application provides a data processing method, including:

[0006] Get the live broadcast data of the target live broadcast room;

[0007] According to the live broadcast data, identifying a live broadcast event occurring in the target live broadcast room;

[0008] When the identified live broadcast event is a product switching event, a target product is determined; the target product is the product being displayed in the target live broadcast room;

[0009] Determine the target advertising creative corresponding to the target product; the target advertising creative is used to deliver the target advertisement to the target device; when the target device detects a click operation on the target advertisement, the target live broadcast room can be entered through the target device.

[0010] In the above solution, the step of identifying the live broadcast event occurring in the target live broadcast room according to the live broadcast data includes:

[0011] Using the live broadcast data, determine whether the target live broadcast room meets the triggering conditions of various types of live broadcast events, and obtain a determination result;

[0012] Based on the judgment result, a live broadcast event occurring in the target live broadcast room is identified.

[0013] In the above solution, the live broadcast data includes the picture to be processed; and the method of using the live broadcast data to determine whether the target live broadcast room meets the triggering conditions of various types of live broadcast events includes:

[0014] Comparing the image to be processed with images in a first database to obtain a first comparison result; the first database includes a plurality of images and a type of live event corresponding to each image;

[0015] Based on the first comparison result, determine whether the target live broadcast room meets the trigger conditions of each type of live broadcast event; wherein, when the first comparison result indicates that the similarity between the image to be processed and the target image in the first database is greater than a first threshold, it is determined that the target live broadcast room meets the trigger conditions of the live broadcast event of the type corresponding to the target image.

[0016] In the above scheme, the live broadcast data includes a video clip to be processed; and using the live broadcast data to determine whether the target live broadcast room meets the triggering conditions of various types of live broadcast events includes:

[0017] Decomposing the video clip to be processed into a plurality of pictures to be processed to obtain a sequence of pictures to be processed;

[0018] Comparing the to-be-processed picture sequence with a picture sequence in a second database to obtain a second comparison result; the second database includes a plurality of picture sequences and a type of live event corresponding to each picture sequence;

[0019] Based on the second comparison result, determine whether the target live broadcast room meets the trigger conditions of each type of live broadcast event; wherein, when the second comparison result indicates that the similarity between the image sequence to be processed and the target image sequence in the second database is greater than a second threshold, it is determined that the target live broadcast room meets the trigger conditions of the live broadcast event of the corresponding type of the target image sequence.

[0020] In the above solution, the step of determining the target advertisement creative corresponding to the target product includes:

[0021] Receiving an advertisement request from the target device;

[0022] Based on the resource position identifier carried in the advertisement request, a target advertisement creative corresponding to the target commodity is determined; the resource position identifier represents the form in which the target device presents the target advertisement.

[0023] In the above solution, the step of determining the target advertisement creative corresponding to the target product based on the resource location identifier carried by the advertisement request includes:

[0024] Based on the resource position identifier, the target advertising creative corresponding to the target product is determined from a third database; the third database contains multiple advertising creatives corresponding to each product among all the products that need to be displayed in the target live broadcast room; the advertising presentation forms corresponding to each of the multiple advertising creatives corresponding to each product are different; each advertising creative contained in the third database has passed a preset review process.

[0025] In the above solution, the step of determining the target advertisement creative corresponding to the target product based on the resource location identifier carried by the advertisement request includes:

[0026] Based on the resource location identifier, multiple target creative elements corresponding to the target product are determined from the fourth database; the target advertising creativity is synthesized using the multiple target creative elements; the fourth database contains all the creative elements that can be used in the target live broadcast room; each creative element contained in the fourth database has passed a preset review process.

[0027] The present application also provides a data processing device, including:

[0028] An acquisition unit, used to acquire live broadcast data of a target live broadcast room;

[0029] A first processing unit, configured to identify a live broadcast event occurring in the target live broadcast room according to the live broadcast data;

[0030] The second processing unit is configured to determine a target product when the identified live broadcast event is a product switching event; the target product is the product being displayed in the target live broadcast room;

[0031] The third processing unit is used to determine the target advertising creative corresponding to the target product; the target advertising creative is used to deliver the target advertisement to the target device; when the target device detects a click operation on the target advertisement, the target live broadcast room can be entered through the target device.

[0032] An embodiment of the present application further provides an electronic device, comprising: a processor and a memory for storing a computer program that can be run on the processor, wherein the processor is used to execute the steps of any of the above methods when running the computer program.

[0033] An embodiment of the present application further provides a storage medium on which a computer program is stored. When the computer program is executed by a processor, the steps of any of the above methods are implemented.

[0034] The data processing method, device, electronic device and storage medium provided by the embodiment of the present application obtain the live broadcast data of the target live broadcast room; identify the live broadcast event occurring in the target live broadcast room according to the live broadcast data; when the identified live broadcast event is a product switching event, determine the target product; the target product is the product being displayed in the target live broadcast room; determine the target advertising creative corresponding to the target product; the target advertising creative is used to deliver the target advertisement to the target device; when the target device detects a click operation on the target advertisement, it can enter the target live broadcast room through the target device. The scheme of the embodiment of the present application, when the live broadcast event occurring in the target live broadcast room is identified as a product switching event according to the live broadcast data of the target live broadcast room, determine the target advertising creative corresponding to the product being displayed in the target live broadcast room for delivering the target advertisement to the target device; in this way, the promotional advertisement (i.e., the target advertisement) of the live broadcast room presented by the user terminal (i.e., the target device) can be dynamically updated as the product being displayed in the live broadcast room is switched, i.e., the promotional advertisement of the live broadcast room can be synchronized with the live broadcast content. In other words, the promotional advertisement of the live broadcast room can be strongly related to the live broadcast content, thereby improving the user's viewing experience of the live broadcast. BRIEF DESCRIPTION OF THE DRAWINGS

[0035] Figure 1 A flowchart of a data processing method according to an embodiment of the present application;

[0036] Figure 2 This is a schematic diagram of the interaction between the live broadcast center and the advertising system in the application embodiment of the present application;

[0037] Figure 3 This is a schematic diagram of the data flow of the interaction between the live broadcast center and the advertising system in the application embodiment of the present application;

[0038] Figure 4 This is a flow chart of offline data training in an application example of the present application;

[0039] Figure 5 This is a schematic diagram of the data processing flow of the advertising system of the application embodiment of the present application;

[0040] Figure 6 This is a schematic diagram of the structure of a data processing device according to an embodiment of the present application;

[0041] Figure 7 It is a schematic diagram of the structure of an electronic device according to an embodiment of the present application. DETAILED DESCRIPTION

[0042] The present application is further described in detail below in conjunction with the accompanying drawings and embodiments.

[0043] Usually, a live broadcast will promote a variety of products, but the promotional ads of the live broadcast are fixed, that is, the promotional ads only contain a limited number of product images. Therefore, when a user enters the live broadcast room because of a product that he or she is interested in in the promotional ads, the product being displayed in the live broadcast room may not be the product that the user is interested in. It can be seen that the promotional ads are not synchronized with the live broadcast content, that is, the degree of relevance between the promotional ads and the live broadcast content is low, which seriously affects the user's viewing experience of the live broadcast.

[0044] Based on this, in various embodiments of the present application, when the live broadcast event occurring in the target live broadcast room is identified as a product switching event based on the live broadcast data of the target live broadcast room, the target advertising creative corresponding to the product being displayed in the target live broadcast room is determined for delivering the target advertisement to the target device; in this way, the promotional advertisement of the live broadcast room (i.e., the target advertisement) presented by the user terminal (i.e., the target device) can be dynamically updated as the product being displayed in the live broadcast room is switched, that is, the promotional advertisement of the live broadcast room can be synchronized with the live broadcast content. In other words, the promotional advertisement of the live broadcast room can be strongly related to the live broadcast content, thereby enhancing the user's viewing experience of the live broadcast.

[0045] The present application embodiment provides a data processing method, which is applied to an electronic device (such as a server), such as Figure 1 As shown, the method includes:

[0046] Step 101: Acquire live broadcast data of a target live broadcast room;

[0047] Step 102: Identify a live broadcast event occurring in the target live broadcast room according to the live broadcast data;

[0048] Step 103: When the identified live event is a product switching event, determine the target product;

[0049] Here, the target product is the product being displayed in the target live broadcast room;

[0050] Step 104: Determine the target advertising creative corresponding to the target product;

[0051] Here, the target advertising creative is used to deliver the target advertisement to the target device; when the target device detects a click operation on the target advertisement, the target live broadcast room can be entered through the target device.

[0052] In actual application, the target device can be understood as a terminal, and can also be called a user equipment (UE); a user can enter the target live broadcast room by clicking on the target advertisement on the target device.

[0053] In actual application, after the electronic device determines the target advertising creative corresponding to the target product, it can directly use the determined target advertising creative to deliver the target advertisement to the target device; or, the electronic device can send the determined target advertising creative to other devices (such as an advertising demand-side platform (DSP)), and the other devices will use the target advertising creative to deliver the target advertisement to the target device.

[0054] In actual application, after the target advertisement is delivered to the target device, the user can enter the target live broadcast room by clicking the target advertisement on the target device. It can be understood that the target advertisement is used to guide the user to enter the target live broadcast room.

[0055] In step 101, in actual application, the live broadcast data may include at least one of the following:

[0056] Live text;

[0057] Live pictures;

[0058] Live video clips.

[0059] Here, the live video clip can be understood as a continuous plurality of frames of live pictures.

[0060] Specifically, the electronic device can obtain live broadcast data through voice recognition when the host of the target live broadcast room is broadcasting, that is, obtain live broadcast text; or, the electronic device can also obtain live broadcast data through picture capture when the host of the target live broadcast room is broadcasting, that is, obtain live broadcast pictures; of course, the electronic device can also obtain live broadcast data through video capture when the host of the target live broadcast room is broadcasting, that is, obtain live video clips.

[0061] In step 102, in actual application, trigger conditions of different types of live events may be preset, and the live events occurring in the target live room may be identified by using the live data and the preset trigger conditions of different types of live events.

[0062] Based on this, in one embodiment, identifying the live broadcast event occurring in the target live broadcast room according to the live broadcast data may include:

[0063] Using the live broadcast data, determine whether the target live broadcast room meets the triggering conditions of various types of live broadcast events, and obtain a determination result;

[0064] Based on the judgment result, a live broadcast event occurring in the target live broadcast room is identified.

[0065] In actual application, the types of live events may include: lottery events, red envelope events, product display events, try-on events, food tasting events and product switching events, etc. Different trigger conditions can be set for different types of live events. In addition, for different types of live data (i.e., live text, live pictures and live video clips), the trigger conditions of each type of live event may also contain data in different forms. Specifically, when the live data contains live text, the trigger conditions of each type of live event may contain keywords and / or key words; when the live data contains live pictures or live video clips, the trigger conditions of each type of live event may contain the host's action description information, and the action description information may contain at least one of the following:

[0066] Action description text;

[0067] Action description pictures;

[0068] Action description video clips.

[0069] Here, the live broadcast data includes a live broadcast picture or a live broadcast video clip, which can be understood as the live broadcast data includes one live broadcast picture or multiple (ie, at least two) live broadcast pictures. The action description video clip can be understood as a continuous multiple-frame action description picture.

[0070] The triggering condition is described below by way of example with respect to the commodity switching event:

[0071] When the live broadcast data contains live broadcast text, the trigger condition of the product switching event may include keywords or key words such as "change", "next", "another", etc.; when the number of corresponding keywords or key words contained in the live broadcast text (that is, the keywords or key words contained in the trigger condition of the product switching event) is greater than the third threshold, it can be determined that the target live broadcast room meets the trigger condition of the product switching event, that is, it can be identified that the product switching event occurs in the target live broadcast room.

[0072] In the case where the live broadcast data includes a live broadcast picture or a live broadcast video clip, the triggering condition of the product switching event may include at least one of the following:

[0073] Action description text: The anchor moves a product out of the camera and brings another product from outside the camera to the camera;

[0074] Action description image corresponding to the action of "the anchor moves a product out of the camera and brings another product from outside the camera to the camera";

[0075] An action description video clip corresponding to the action of "the host moves a product out of the camera and brings another product from outside the camera to the camera."

[0076] When the triggering condition of the product switching event includes the action description text, the live broadcast picture or live video clip can be analyzed by a pre-trained analysis model to obtain the description text of the live broadcast picture or live video clip, and the description text output by the analysis model can be compared with the action description text for similarity by a pre-trained text comparison model. When the similarity output by the text comparison model is greater than a fourth threshold, it can be determined that the target live broadcast room meets the triggering condition of the product switching event, that is, it can be identified that the product switching event has occurred in the target live broadcast room.

[0077] In the case where the triggering condition of the product switching event includes the action description picture, since the live video clip can be understood as a continuous multi-frame live picture, the live picture or live video clip can be compared with the action description picture for similarity through a pre-trained picture comparison model. When the similarity output by the picture comparison model is greater than the fifth threshold, it can be determined that the target live broadcast room meets the triggering condition of the product switching event, that is, it can be identified that the product switching event has occurred in the target live broadcast room.

[0078] In the case where the triggering condition of the product switching event includes the action description video clip, since the live video clip can be understood as a continuous multiple-frame live picture, and the action description video clip can be understood as a continuous multiple-frame action description picture, the live picture or live video clip can also be compared with the action description video clip for similarity through the picture comparison model. When the similarity output by the picture comparison model is greater than the sixth threshold, it can be determined that the target live broadcast room meets the triggering condition of the product switching event, that is, it can be identified that the product switching event has occurred in the target live broadcast room.

[0079] In actual application, the third threshold, the fourth threshold, the fifth threshold and the sixth threshold can be set according to requirements; in addition, the fourth threshold, the fifth threshold and the sixth threshold can be the same or different.

[0080] In actual application, in order to improve the efficiency of judging whether the target live broadcast room meets the triggering conditions of various types of live broadcast events, when the triggering conditions of various types of live broadcast events include action description pictures, the action description pictures corresponding to various types of live broadcast events can be stored in a database (referred to as the first database in the subsequent description); when the live broadcast data includes live broadcast pictures, the live broadcast pictures can be compared with the pictures in the first database, and based on the comparison results, it can be judged whether the target live broadcast room meets the triggering conditions of various types of live broadcast events.

[0081] Based on this, in one embodiment, the live broadcast data may include a to-be-processed picture (i.e., the live broadcast picture); accordingly, using the live broadcast data to determine whether the target live broadcast room meets the triggering conditions of various types of live broadcast events may include:

[0082] Comparing the image to be processed with the image in the first database (i.e., the action description image) to obtain a first comparison result; the first database includes multiple images and the type of live event corresponding to each image;

[0083] Based on the first comparison result, determine whether the target live broadcast room meets the trigger conditions of each type of live broadcast event; wherein, when the first comparison result indicates that the similarity between the image to be processed and the target image in the first database is greater than a first threshold, it is determined that the target live broadcast room meets the trigger conditions of the live broadcast event of the type corresponding to the target image.

[0084] In actual application, the image comparison model can be used to compare the image to be processed with the images in the first database. In other words, the image to be processed and each image in the first database are input into the image comparison model to obtain the similarity between the image to be processed and each image in the first database.

[0085] In actual application, the first threshold can be set according to requirements; in addition, the first threshold, the fourth threshold, the fifth threshold and the sixth threshold can be the same or different.

[0086] In actual application, in order to improve the efficiency of judging whether the target live broadcast room meets the triggering conditions of various types of live broadcast events, when the triggering conditions of various types of live broadcast events include action description video clips, for each action description video clip, the corresponding action description video clip can be decomposed into multiple action description pictures, and an action description picture sequence corresponding to the corresponding action description video clip is obtained, and the action description picture sequence can be understood as a collection of multiple action description pictures; and the action description picture sequence corresponding to each type of live broadcast event can be stored in a database (referred to as the second database in the subsequent description). When the live broadcast data includes a live video clip, the live video clip can be decomposed into a live picture sequence, and the live picture sequence can be compared with the action description picture sequence in the second database, and based on the comparison result, it is judged whether the target live broadcast room meets the triggering conditions of various types of live broadcast events.

[0087] Based on this, in one embodiment, the live broadcast data may include a video segment to be processed (i.e., the live broadcast video segment); accordingly, using the live broadcast data to determine whether the target live broadcast room meets the triggering conditions of various types of live broadcast events may include:

[0088] Decomposing the to-be-processed video clip into a plurality of to-be-processed pictures (i.e., the live pictures) to obtain a to-be-processed picture sequence (i.e., the live picture sequence);

[0089] Comparing the to-be-processed picture sequence with a picture sequence in the second database (i.e., the action description picture sequence) to obtain a second comparison result; the second database includes a plurality of picture sequences and a type of live event corresponding to each picture sequence;

[0090] Based on the second comparison result, determine whether the target live broadcast room meets the trigger conditions of each type of live broadcast event; wherein, when the second comparison result indicates that the similarity between the image sequence to be processed and the target image sequence in the second database is greater than a second threshold, it is determined that the target live broadcast room meets the trigger conditions of the live broadcast event of the corresponding type of the target image sequence.

[0091] In actual application, the picture comparison model can be used to compare the picture sequence to be processed with the picture sequence in the second database. In other words, the picture sequence to be processed and each picture sequence in the second database are input into the picture comparison model to obtain the similarity between the picture sequence to be processed and each picture sequence in the second database.

[0092] In actual application, the second threshold can be set according to demand; in addition, the first threshold, the second threshold, the fourth threshold, the fifth threshold and the sixth threshold can be the same or different.

[0093] In actual application, in order to improve the rationality and accuracy of the triggering conditions of various types of live broadcast events, the historical live broadcast data of the anchor of the target live broadcast room (which may include at least one of the historical live broadcast text, historical live broadcast pictures and historical live broadcast video clips) can be obtained as sample data. The sample data is used in combination with the deep learning network to train a feature extraction model for live broadcast events, and the features extracted by the feature extraction model are descriptively represented (that is, data vectorization is performed) to obtain the triggering conditions of various types of live broadcast events, that is, the first database and the second database are obtained.

[0094] In actual application, in order to improve the accuracy of the judgment result when using the live broadcast data to judge whether the target live broadcast room meets the trigger conditions of various types of live broadcast events, the acquisition of the live broadcast data of the target live broadcast room may include: using the feature extraction model to acquire the live broadcast data of the target live broadcast room. Specifically, the candidate live broadcast data of the live broadcast room may be acquired, and the candidate live broadcast data may include live broadcast pictures and / or live broadcast video clips. The candidate live broadcast data is input into the feature extraction model, and the features extracted by the feature extraction model are descriptively represented, so that the live broadcast data can be obtained.

[0095] In step 103, in actual application, after identifying the occurrence of the product switching event in the target live broadcast room, the target product can be determined using the live broadcast data. For example, the name of the target product can be extracted by performing text analysis on the live broadcast text; for another example, the target product can be identified in the live broadcast picture or the live broadcast video clip.

[0096] In step 104, in actual application, when a user accesses media (such as an application (APP)) through a terminal, for an application scenario in which the electronic device directly uses a determined target advertising creative to deliver a target advertisement to the terminal, the terminal can send an advertisement request to the electronic device when the user triggers an advertisement delivery condition (such as APP opening the screen, APP homepage browsing, etc.); for an application scenario in which the electronic device sends the determined target advertising creative to other devices (such as an advertising DSP) and other devices deliver a target advertisement to the terminal using the target advertising creative, the terminal can send an advertisement request to other devices when the user triggers an advertisement delivery condition, and the other devices then forward the advertisement request to the electronic device. Accordingly, the electronic device can determine the target advertising creative corresponding to the target product based on the advertisement request.

[0097] Based on this, in one embodiment, determining the target advertisement creative corresponding to the target product may include:

[0098] Receiving an advertisement request from the target device;

[0099] Based on the resource position identifier carried in the advertisement request, a target advertisement creative corresponding to the target commodity is determined; the resource position identifier represents the form in which the target device presents the target advertisement.

[0100] Here, receiving the advertisement request of the target device may include:

[0101] directly receiving an advertisement request from the target device; or,

[0102] An advertisement request of the target device forwarded by other devices is received.

[0103] In actual application, the target device may present the target advertisement in a form including text, pictures, video clips, etc.; the target device may present the target advertisement in a different form according to different advertisement delivery conditions triggered by the user.

[0104] In actual application, in order to improve the timeliness of updating live broadcast promotional advertisements, before the live broadcast of the target live broadcast room begins, multiple advertising creatives corresponding to each of all the products that need to be displayed in the target live broadcast room can be generated, and each advertising creative in the multiple advertising creatives corresponding to each product corresponds to a different advertising presentation form, and all the generated advertising creatives are reviewed in advance; in this way, the target advertising creative corresponding to the target product can be quickly determined, thereby improving the timeliness of updating the live broadcast promotional advertisements.

[0105] Based on this, in one embodiment, determining the target advertisement creative corresponding to the target product based on the resource location identifier carried by the advertisement request may include:

[0106] Based on the resource position identifier, the target advertising creative corresponding to the target product is determined from a third database; the third database contains multiple advertising creatives corresponding to each product among all the products that need to be displayed in the target live broadcast room; the advertising presentation forms corresponding to each of the multiple advertising creatives corresponding to each product are different; each advertising creative contained in the third database has passed a preset review process.

[0107] In actual application, the audit content of the preset audit process can be set according to needs, such as auditing whether the advertising creativity involves pornography and / or violence.

[0108] In actual application, in order to save storage resources, it is also possible to pre-store and review only all the creative elements that can be used in the target live broadcast room.

[0109] Based on this, in one embodiment, determining the target advertisement creative corresponding to the target product based on the resource location identifier carried by the advertisement request may include:

[0110] Based on the resource location identifier, multiple target creative elements corresponding to the target product are determined from the fourth database; the target advertising creativity is synthesized using the multiple target creative elements; the fourth database contains all the creative elements that can be used in the target live broadcast room; each creative element contained in the fourth database has passed a preset review process.

[0111] In actual application, the creative elements may include text (i.e. copywriting), pictures and video clips, etc.; the text may contain a web link (i.e. a landing page for jumping to the live broadcast room), etc.

[0112] In actual application, the method of specifically determining the target advertisement creative corresponding to the target commodity can be set according to demand.

[0113] The data processing method provided by the embodiment of the present application obtains the live broadcast data of the target live broadcast room; identifies the live broadcast event occurring in the target live broadcast room according to the live broadcast data; when the identified live broadcast event is a product switching event, determines the target product; the target product is the product being displayed in the target live broadcast room; determines the target advertising creative corresponding to the target product; the target advertising creative is used to deliver the target advertisement to the target device; when the target device detects a click operation on the target advertisement, it can enter the target live broadcast room through the target device. The scheme of the embodiment of the present application, when the live broadcast event occurring in the target live broadcast room is identified as a product switching event according to the live broadcast data of the target live broadcast room, determines the target advertising creative corresponding to the product being displayed in the target live broadcast room for delivering the target advertisement to the target device; in this way, the promotional advertisement (i.e., the target advertisement) of the live broadcast room presented by the user terminal (i.e., the target device) can be dynamically updated with the switching of the products being displayed in the live broadcast room, i.e., the promotional advertisement of the live broadcast room can be synchronized with the live broadcast content. In other words, the promotional advertisement of the live broadcast room can be strongly related to the live broadcast content, thereby improving the user's viewing experience of the live broadcast.

[0114] The present application is further described in detail below in conjunction with application examples.

[0115] In this application embodiment, through the interaction between the live broadcast center and the advertising system, dynamic switching of live broadcast promotional advertisements is achieved in the live broadcast scenario.

[0116] Combine the following Figure 2 and Figure 3 Describe the functions of the live broadcast center and advertising system.

[0117] In this application example, Figure 2 and Figure 3As shown, the live broadcast center is used to train offline data. Specifically, the live broadcast center can obtain the offline recorded broadcast data of the target anchor (i.e., the above-mentioned historical live broadcast data) as sample data, and use the sample data in combination with the deep learning network to train the feature extraction model of the live broadcast event, and perform descriptive representation of the features extracted by the feature extraction model (i.e., perform data vectorization processing), and obtain a picture set and a video set. After obtaining the picture set, the live broadcast center can classify the pictures according to the type of live broadcast event, that is, mark the type of live broadcast event corresponding to each picture, and obtain a picture data set (i.e., the above-mentioned first database); correspondingly, after obtaining the video set, the live broadcast center can classify the video clips according to the type of live broadcast event, that is, mark the type of live broadcast event corresponding to each video clip, and obtain a video data set (i.e., the above-mentioned second database).

[0118] Here, the features extracted by the feature extraction model can be understood as the host's actions, such as picking up goods, putting down goods, etc. Different actions correspond to different live events, such as lottery events, red envelope events, product display events, try-on events, tasting events, and product switching events. In addition, the features extracted by the feature extraction model are descriptively represented, which can be understood as representing the pictures or picture sequences (i.e., video clips) corresponding to the actions in the form of coordinates or matrices; marking the type of live events corresponding to each picture and each video clip can be understood as defining the trigger conditions of each type of live events. In this way, the picture data set and the video data set can be used as the data source for the live broadcast center to judge the online behavior of the host (i.e., the basis for online behavior judgment). When the host's behavior meets the trigger conditions, the corresponding live broadcast event is triggered (i.e., the live broadcast center recognizes the corresponding live broadcast event).

[0119] It should be noted that, in this application embodiment, training offline data can be understood as determining the image data set and the video data set.

[0120] In this application example, Figure 2 and Figure 3 As shown, the live broadcast center is also used for online behavior judgment. Specifically, the live broadcast center can obtain online live broadcast data (i.e., obtain live broadcast data in real time), use the feature extraction model to extract key images and image sequences (i.e., video clips) in the live broadcast data, and identify the live broadcast events corresponding to the key images and image sequences by performing similarity comparison between the extracted key images and images in the image data set, and performing similarity comparison between the extracted image sequences and image sequences in the video data set, that is, judging the live broadcast events triggered by the host's actions (i.e., behaviors) in real time.

[0121] In this application example, Figure 2 and Figure 3As shown, when the live broadcast center recognizes a product switching event, the identifier of the product currently being displayed in the live broadcast room can be sent to the advertising system. The advertising system includes two subsystems: a creative generation system and an advertising playback system. The creative generation system is used to synthesize advertising ideas based on the product identifier, using creative elements such as titles, pictures, videos, and landing pages (i.e., network links used to jump to the live broadcast room), and send the generated advertising ideas to the advertising playback system, which plays live broadcast promotional ads (i.e., places ads) on the media side based on the advertising ideas. In other words, the advertising playback system can display advertising ideas on the media side through user exposure. In this way, when users see products of interest in the live broadcast promotional ads on the media side, they can click on the ads to be directed (i.e., jump) to the live broadcast room, and the live broadcast room is currently displaying products of interest to the user.

[0122] Here, the advertising system can generate advertising creatives in two ways:

[0123] Method 1: Generate advertising creatives in advance. Specifically, before the target live broadcast begins, the advertising system can generate multiple advertising creatives corresponding to each product among all the products that need to be displayed in this live broadcast. Each advertising creative in the multiple advertising creatives corresponding to each product corresponds to a different advertising presentation form (i.e., advertising exposure method); in addition, the advertising system can also pre-review all generated advertising creatives (for example, review whether the advertising creatives involve pornography and / or violence). After receiving the product identifier, the advertising system can directly search for the target advertising creative in the pre-generated and reviewed advertising creatives for playback; in this way, it can quickly determine the advertising creative corresponding to the product currently being displayed in the live broadcast room, thereby improving the timeliness of updating the live broadcast promotional advertisements.

[0124] Method 2: Generate advertising creatives dynamically (i.e., in real time). Specifically, after receiving the product identification, the advertising system can obtain the corresponding advertising elements in real time to synthesize advertising creatives; in this way, the storage resources of the advertising system can be saved. Here, the advertising elements can also be reviewed in advance to improve the efficiency of synthesizing advertising creatives.

[0125] In this application example, Figure 4 As shown, the live broadcast center trains offline data, which may specifically include the following steps:

[0126] Step 401: Acquire data samples; then execute step 402.

[0127] Specifically, the live broadcast center can obtain the offline recorded broadcast data of the target anchor as sample data, and mark key pictures (i.e. pictures that can reflect the anchor's actions) in the sample data. The offline recorded broadcast data can include text, pictures, voice, video and other data.

[0128] Step 402: Use a deep learning network to extract features from data samples; then execute step 403.

[0129] Specifically, the live broadcast center can build a deep learning network through the image classification deep learning toolbox, use the sample data obtained in step 401, combine it with the deep learning network to train a feature extraction model for the live broadcast event, and use the feature extraction model to extract features of the sample data (such as picture color, lighting, anchor's actions, etc.).

[0130] Here, when the live broadcast center constructs a deep learning network, it can set the target features of the deep learning network; the target features can be understood as the standard for the feature extraction model to identify and classify the sample data; the target features can include the overall movements of the anchor (such as the movements of the head, hands, and body swing amplitude), the movements of a specific part (such as hand movements), and the environment of the live broadcast room (such as picture color, lighting, etc.). The feature extraction model can analyze the target features in the sample data, identify and classify the same or similar features in the sample data, and output the results (i.e., feature data).

[0131] In actual application, the anchor will have some fixed actions for specific behaviors during the live broadcast, such as holding the product with one hand and the other hand with the palm facing up when displaying the product; at this time, the same or similar actions can be regarded as the same behavior (such as the behavior of displaying the product). The feature extraction model can identify and classify the same or similar actions in the sample data based on the target features. In other words, after the sample data containing text, pictures, voice and video are input into the feature extraction model, the feature extraction model can aggregate and classify the sample data, and output feature data that also contains text, pictures, voice and video.

[0132] Step 403: perform data vectorization processing on the features extracted by the deep learning network; then execute step 404.

[0133] Specifically, the live broadcast center can perform data vectorization processing on the feature data output by the feature extraction model, that is, generate a descriptive representation of the feature data, which can also be understood as realizing a descriptive expression of the feature data. For example, the image can be represented in the form of coordinates or matrices as the basis for subsequent processing.

[0134] Step 404: Classify and define the feature data to determine the image data set and the video data set.

[0135] Specifically, the live broadcast center can classify and define the feature data in step 403, that is, first divide the feature data into a picture set and a video set according to the data type, and then define (i.e., label) the type of live broadcast event corresponding to each picture (the type of live broadcast event can be determined based on the target feature) to obtain a picture data set; and define (i.e., label) the type of live broadcast event corresponding to each video clip to obtain a video data set.

[0136] Here, since the data obtained through deep learning (i.e., the feature data) is homogeneous but has no business meaning, it is necessary to redefine the meaning of the feature data, that is, to mark the type of live event corresponding to each picture and video clip, so as to provide a basis for subsequent online behavior judgment.

[0137] In this application embodiment, when the live broadcast center makes online behavior judgments, it can obtain live broadcast data as source data in real time by intercepting key frames or video clips, and can judge whether the source data meets the triggering conditions of various types of live broadcast events by comparing the source data with the data in the image data set or the video data set for similarity. In the case where the source data has a high degree of similarity with the data corresponding to the product switching event in the image data set or the video data set (for example, greater than the above-mentioned first threshold or second threshold), the live broadcast center can determine that the source data meets the triggering conditions of the product switching event, and can send a command containing the product identification to the advertising system to instruct the advertising system to complete the switching of the live broadcast promotional advertisement.

[0138] Here, when the source data is compared with the data in the image dataset or video dataset for similarity, it can include image-based comparison and video-based comparison. When the similarity comparison is based on the image, since only the similarity of two images needs to be compared, the implementation is simple and efficient, but the probability of misjudgment is high. When the similarity comparison is based on the video, since it is necessary to consider the continuous state before and after the action (i.e., the change state), in other words, it is necessary to compare the similarity of multiple frames of images, so it is difficult to implement and time-consuming, but the accuracy is high.

[0139] Specifically, when performing similarity comparison based on images, the live broadcast center can capture the key frame (i.e., the above-mentioned image to be processed) as the retrieval target in real time, query the category to which the key image belongs in the image data set, and determine whether the key image meets the triggering conditions of the corresponding live broadcast event; in other words, the live broadcast center can determine that the key image meets the triggering conditions of the live broadcast event corresponding to the target when the similarity between the key image and the target image in the image data set is greater than a first threshold.

[0140] When performing similarity comparison based on the video, the live broadcast center can capture key video clips (i.e. the above-mentioned video clips to be processed) in real time, decompose the key video clips into key picture sequences (i.e. the above-mentioned picture sequences to be processed), and use the key picture sequences as retrieval targets, query the category to which the key picture sequences belong in the video data set, and determine whether the key picture sequences meet the triggering conditions of the corresponding live broadcast events; in other words, the live broadcast center can determine that the key picture sequence meets the triggering conditions of the live broadcast event corresponding to the target picture sequence when the similarity between the key picture sequence and the target picture sequence in the video data set is greater than a second threshold.

[0141] In actual application, the live broadcast center can use the similarity model (i.e., the above-mentioned picture comparison model) to realize online behavior judgment. The similarity model can include N pictures (N is an integer greater than 0), and the N pictures can represent M live events (M is an integer greater than 0). When the source data includes picture I, that is, when it is necessary to determine the live event corresponding to picture I, the live event corresponding to picture I can be expressed by the following formula:

[0142] E=max(average(sim(I,N))) (1)

[0143] Among them, I represents the image to be identified contained in the source data (i.e., the image to be processed mentioned above); E represents the live event corresponding to the image I; N represents the image sample, i.e., the image in the image data set; sim represents the similarity function; average represents the average function; and max represents the maximum value.

[0144] In practical applications, the similarity function may be a cosine theorem function, a Euclidean distance function, etc. When the similarity function is a cosine theorem function, it may be expressed by the following formula:

[0145]

[0146] Among them, V1 represents the vectorization result of image I; V2 represents the vectorization result of image N; V1×V2 represents the multiplication of two vectors; ‖V1‖ represents the length of vector V1; ‖V2‖ represents the length of vector V2.

[0147] When the similarity function is a Euclidean distance function, assuming that x (X1, X2 ... Xn) and y (Y1, Y2 ... Yn) are two points in an n-dimensional space, the distance between point x and point y can be expressed by the following formula:

[0148]

[0149] Where Xi represents the coordinates of point x; Yi represents the coordinates of point y. Based on formula (3), the similarity formula can be determined as:

[0150]

[0151] Among them, the value range of sim(I,N) is [0,1]. The larger the value, the smaller d is, which means the closer the distance is, and the higher the similarity is.

[0152] In this application example, Figure 5 As shown, the data processing process of the advertising system may specifically include the following steps:

[0153] Step 501: When the live broadcast center recognizes a product switching event, the identifier of the product currently being displayed in the live broadcast room is sent to the advertising system; after receiving the product identifier, the creative generation system in the advertising system searches for advertising ideas with an index of the product identifier in a pre-generated advertising creative database (i.e., the third database mentioned above); or, using the creative elements in the creative element database (i.e., the fourth database mentioned above), dynamically generate advertising ideas corresponding to the product identifier; then, execute step 502.

[0154] Specifically, the creative generation system can generate advertising creatives in two ways: pre-synthesizing advertising creatives and real-time synthesizing advertising creatives. Among them, pre-synthesizing advertising creatives means generating multiple advertising creatives corresponding to each of all the products that need to be displayed in this live broadcast before the target live broadcast starts, and each of the multiple advertising creatives corresponding to each product corresponds to a different advertising presentation form (i.e., advertising exposure method); in this way, the failure rate of generating advertising creatives is low, but it is easy to cause the problem of advertising creative expansion, that is, the generated advertising creatives will occupy too many storage resources. Real-time synthesizing advertising creatives means that when the creative generation system receives the creative generation instruction containing the product identification issued by the live broadcast center during the live broadcast, it will use the creative elements in the creative element database to generate advertising creatives in real time; in this way, the failure rate of generating advertising creatives is high.

[0155] In actual application, the specific method for generating advertising creatives can be set according to needs.

[0156] In actual application, the creative element database may include creative elements such as logos, landing pages, product identification numbers (IDs), pictures, videos, etc.

[0157] In this application embodiment, in order to meet the timeliness requirements of advertising creativity in live broadcast scenarios, the post-review of advertising creativity is decomposed into the pre-review of creative elements. That is to say, the review process of "generating advertising creativity first, and then reviewing the advertising creativity" in the relevant technology is modified to "first review the creative elements (for example, review whether the creative elements are pornographic and / or violent), and then use the creative elements that have passed the review to generate an advertising creative database and a creative element database" Review process; there is no need to review advertising creativity during the live broadcast process, which improves the timeliness of advertising creativity.

[0158] Step 502: The creative generation system sends the determined advertising creative to the advertising playback system, and the advertising playback system plays the live broadcast promotional advertisement on the media side according to the advertising creative.

[0159] Specifically, the advertising playback system can display the latest advertising ideas on the media side through user exposure. In this way, when a user sees a product of interest in a live broadcast promotion advertisement on the media side, he can click on the advertisement to be directed (i.e., jump) to the live broadcast room, and the live broadcast room is currently displaying the product of interest to the user.

[0160] In actual application, basic advertising creatives can be set for scenarios where creative generation fails. If advertising creatives are not successfully generated, or if advertising creatives that match the product identifiers are not found in the advertising creative database, the advertising playback system can play live promotional ads on the media side based on the basic advertising creatives.

[0161] The solution provided by this application embodiment has the following advantages:

[0162] First, it realizes the real-time perception of product switching events, and can dynamically update the advertising creative when identifying the product switching event in the live broadcast room; the promotional advertisements of the live broadcast room presented on the user terminal can be dynamically updated as the products being displayed in the live broadcast room are switched, that is, the promotional advertisements of the live broadcast room can be synchronized with the live broadcast content. In other words, the promotional advertisements of the live broadcast room can be strongly related to the live broadcast content, thereby improving the user's viewing experience of the live broadcast;

[0163] Second, the advertising creativity was broken down into creative elements, and the creative elements were reviewed in advance. The advertising creativity review process during the live broadcast was deleted, the advertising creativity update process during the live broadcast was simplified, and the timeliness of the advertising creativity was enhanced.

[0164] In order to implement the method of the embodiment of the present application, the embodiment of the present application also provides a data processing device, such as Figure 6 As shown, the device comprises:

[0165] An acquisition unit 601 is used to acquire live broadcast data of a target live broadcast room;

[0166] A first processing unit 602 is configured to identify a live broadcast event occurring in the target live broadcast room according to the live broadcast data;

[0167] The second processing unit 603 is used to determine a target product when the identified live broadcast event is a product switching event; the target product is the product being displayed in the target live broadcast room;

[0168] The third processing unit 604 is used to determine the target advertising creative corresponding to the target product; the target advertising creative is used to deliver the target advertisement to the target device; when the target device detects a click operation on the target advertisement, the target live broadcast room can be entered through the target device.

[0169] In one embodiment, the first processing unit 602 is specifically configured to:

[0170] Using the live broadcast data, determine whether the target live broadcast room meets the triggering conditions of various types of live broadcast events, and obtain a determination result;

[0171] Based on the judgment result, a live broadcast event occurring in the target live broadcast room is identified.

[0172] In one embodiment, the live broadcast data includes a picture to be processed; the first processing unit 602 is further configured to:

[0173] Comparing the image to be processed with images in a first database to obtain a first comparison result; the first database includes a plurality of images and a type of live event corresponding to each image;

[0174] Based on the first comparison result, determine whether the target live broadcast room meets the trigger conditions of each type of live broadcast event; wherein, when the first comparison result indicates that the similarity between the image to be processed and the target image in the first database is greater than a first threshold, it is determined that the target live broadcast room meets the trigger conditions of the live broadcast event of the type corresponding to the target image.

[0175] In one embodiment, the live broadcast data includes a video segment to be processed; the first processing unit 602 is further configured to:

[0176] Decomposing the video clip to be processed into a plurality of pictures to be processed to obtain a sequence of pictures to be processed;

[0177] Comparing the to-be-processed picture sequence with a picture sequence in a second database to obtain a second comparison result; the second database includes a plurality of picture sequences and a type of live event corresponding to each picture sequence;

[0178] Based on the second comparison result, determine whether the target live broadcast room meets the trigger conditions of each type of live broadcast event; wherein, when the second comparison result indicates that the similarity between the image sequence to be processed and the target image sequence in the second database is greater than a second threshold, it is determined that the target live broadcast room meets the trigger conditions of the live broadcast event of the corresponding type of the target image sequence.

[0179] In one embodiment, the device also includes a receiving unit for receiving an advertising request from the target device; accordingly, the third processing unit 604 is specifically used to determine the target advertising creative corresponding to the target product based on the resource position identifier carried by the advertising request; the resource position identifier represents the form in which the target device presents the target advertisement.

[0180] In one embodiment, the third processing unit 604 is also used to determine the target advertising creative corresponding to the target product from the third database based on the resource position identifier; the third database contains multiple advertising creatives corresponding to each product of all the products that need to be displayed in the target live broadcast room; the advertising presentation forms corresponding to each of the multiple advertising creatives corresponding to each product are different; each advertising creative contained in the third database has passed a preset review process.

[0181] In one embodiment, the third processing unit 604 is also used to determine, based on the resource position identifier, multiple target creative elements corresponding to the target product from the fourth database; synthesize the target advertising creativity using the multiple target creative elements; the fourth database contains all the creative elements that can be used in the target live broadcast room; each creative element contained in the fourth database has passed a preset review process.

[0182] In actual application, the acquisition unit 601, the first processing unit 602, the second processing unit 603 and the third processing unit 604 can be implemented by a processor in the data processing device; the receiving unit can be implemented by a communication interface in the data processing device.

[0183] It should be noted that: when the data processing device provided in the above embodiment processes data, only the division of the above program modules is used as an example. In actual application, the above processing can be assigned to different program modules as needed, that is, the internal structure of the device is divided into different program modules to complete all or part of the processing described above. In addition, the data processing device provided in the above embodiment and the data processing method embodiment belong to the same concept, and the specific implementation process is detailed in the method embodiment, which will not be repeated here.

[0184] Based on the hardware implementation of the above program modules, and in order to implement the method of the embodiment of the present application, the embodiment of the present application also provides an electronic device, such as Figure 7 As shown, the electronic device 700 includes:

[0185] Communication interface 701, capable of exchanging information with other electronic devices;

[0186] A processor 702, connected to the communication interface 701, to implement information interaction with other electronic devices, and used to execute the method provided by one or more of the above technical solutions when running a computer program;

[0187] The memory 703 stores computer programs that can be executed on the processor 702 .

[0188] Specifically, the processor 702 is configured to:

[0189] Get the live broadcast data of the target live broadcast room;

[0190] According to the live broadcast data, identifying a live broadcast event occurring in the target live broadcast room;

[0191] When the identified live broadcast event is a product switching event, a target product is determined; the target product is the product being displayed in the target live broadcast room;

[0192] Determine the target advertising creative corresponding to the target product; the target advertising creative is used to deliver the target advertisement to the target device; when the target device detects a click operation on the target advertisement, the target live broadcast room can be entered through the target device.

[0193] In one embodiment, the processor 702 is specifically configured to:

[0194] Using the live broadcast data, determine whether the target live broadcast room meets the triggering conditions of various types of live broadcast events, and obtain a determination result;

[0195] Based on the judgment result, a live broadcast event occurring in the target live broadcast room is identified.

[0196] In one embodiment, the live broadcast data includes a picture to be processed; the processor 702 is further configured to:

[0197] Comparing the image to be processed with images in a first database to obtain a first comparison result; the first database includes a plurality of images and a type of live event corresponding to each image;

[0198] Based on the first comparison result, determine whether the target live broadcast room meets the trigger conditions of each type of live broadcast event; wherein, when the first comparison result indicates that the similarity between the image to be processed and the target image in the first database is greater than a first threshold, it is determined that the target live broadcast room meets the trigger conditions of the live broadcast event of the type corresponding to the target image.

[0199] In one embodiment, the live broadcast data includes a video segment to be processed; the processor 702 is further configured to:

[0200] Decomposing the video clip to be processed into a plurality of pictures to be processed to obtain a sequence of pictures to be processed;

[0201] Comparing the to-be-processed picture sequence with a picture sequence in a second database to obtain a second comparison result; the second database includes a plurality of picture sequences and a type of live event corresponding to each picture sequence;

[0202] Based on the second comparison result, determine whether the target live broadcast room meets the trigger conditions of each type of live broadcast event; wherein, when the second comparison result indicates that the similarity between the image sequence to be processed and the target image sequence in the second database is greater than a second threshold, it is determined that the target live broadcast room meets the trigger conditions of the live broadcast event of the corresponding type of the target image sequence.

[0203] In one embodiment, the processor 702 is further configured to:

[0204] Receiving an advertisement request from the target device through the communication interface 701;

[0205] Based on the resource position identifier carried in the advertisement request, a target advertisement creative corresponding to the target commodity is determined; the resource position identifier represents the form in which the target device presents the target advertisement.

[0206] In one embodiment, the processor 702 is further used to determine, based on the resource position identifier, a target advertising creative corresponding to the target product from a third database; the third database contains multiple advertising creatives corresponding to each product of all the products that need to be displayed in the target live broadcast room; the advertising presentation forms corresponding to each of the multiple advertising creatives corresponding to each product are different; and each advertising creative contained in the third database has passed a preset review process.

[0207] In one embodiment, the processor 702 is also used to determine, based on the resource position identifier, multiple target creative elements corresponding to the target product from a fourth database; synthesize the target advertising creativity using the multiple target creative elements; the fourth database contains all the creative elements that can be used in the target live broadcast room; each creative element contained in the fourth database has passed a preset review process.

[0208] It should be noted that: the specific process of the processor 702 performing the above operations is detailed in the method embodiment, which will not be repeated here.

[0209] Of course, in actual application, the various components in the electronic device 700 are coupled together through the bus system 704. It can be understood that the bus system 704 is used to realize the connection and communication between these components. In addition to the data bus, the bus system 704 also includes a power bus, a control bus, and a status signal bus. However, for the sake of clarity, Figure 7 Various buses are labeled as bus system 704 .

[0210] The memory 703 in the embodiment of the present application is used to store various types of data to support the operation of the electronic device 700. Examples of such data include: any computer program used to operate on the electronic device 700.

[0211] The method disclosed in the above embodiment of the present application can be applied to the processor 702, or implemented by the processor 702. The processor 702 may be an integrated circuit chip with signal processing capabilities. In the implementation process, each step of the above method can be completed by the hardware integrated logic circuit or software instructions in the processor 702. The above processor 702 may be a general-purpose processor, a digital signal processor (DSP, Digital Signal Processor), or other programmable logic devices, discrete gates or transistor logic devices, discrete hardware components, etc. The processor 702 can implement or execute the various methods, steps and logic block diagrams disclosed in the embodiments of the present application. The general-purpose processor may be a microprocessor or any conventional processor, etc. In combination with the steps of the method disclosed in the embodiment of the present application, it can be directly embodied as a hardware decoding processor to execute, or it can be executed by a combination of hardware and software modules in the decoding processor. The software module can be located in a storage medium, which is located in the memory 703, and the processor 702 reads the information in the memory 703 and completes the steps of the above method in combination with its hardware.

[0212] In an exemplary embodiment, the electronic device 700 can be implemented by one or more application specific integrated circuits (ASICs), DSPs, programmable logic devices (PLDs), complex programmable logic devices (CPLDs), field programmable gate arrays (FPGAs), general-purpose processors, controllers, microcontrollers (MCUs), microprocessors, or other electronic components to execute the aforementioned method.

[0213] It can be understood that the memory 703 of the embodiment of the present application can be a volatile memory or a non-volatile memory, and can also include both volatile and non-volatile memories. Among them, the non-volatile memory can be a read-only memory (ROM), a programmable read-only memory (PROM), an erasable programmable read-only memory (EPROM), an electrically erasable programmable read-only memory (EEPROM), a magnetic random access memory (FRAM), a flash memory, a magnetic surface memory, an optical disc, or a compact disc read-only memory (CD-ROM); the magnetic surface memory can be a disk memory or a tape memory. The volatile memory can be a random access memory (RAM), which is used as an external cache. By way of example and not limitation, many forms of RAM are available, such as static random access memory (SRAM), synchronous static random access memory (SSRAM), dynamic random access memory (DRAM), synchronous dynamic random access memory (SDRAM), double data rate synchronous dynamic random access memory (DDRSDRAM), enhanced synchronous dynamic random access memory (ESDRAM), synchronous link dynamic random access memory (SLDRAM, SyncLink Dynamic Random Access Memory), and direct RAMbus random access memory (DRRAM, Direct Rambus Random Access Memory).The memory described in the embodiments of the present application is intended to include, but is not limited to, these and any other suitable types of memory.

[0214] In an exemplary embodiment, the present application also provides a storage medium, namely a computer storage medium, specifically a computer-readable storage medium, for example, a memory 703 storing a computer program, and the computer program can be executed by a processor 702 of an electronic device 700 to complete the steps of the aforementioned method. The computer-readable storage medium can be a memory such as FRAM, ROM, PROM, EPROM, EEPROM, Flash Memory, magnetic surface storage, optical disk, or CD-ROM.

[0215] It should be noted that: "first", "second", etc. are used to distinguish similar objects, and are not necessarily used to describe a specific order or sequence.

[0216] In addition, the technical solutions described in the embodiments of the present application can be combined arbitrarily without conflict.

[0217] The above description is only a preferred embodiment of the present application and is not intended to limit the protection scope of the present application.

Claims

1. A data processing method, characterized in that: include: Acquire live broadcast data of a target live broadcast room, wherein the live broadcast data includes at least one of live broadcast text, live broadcast picture, and live broadcast video clip; Using the live broadcast data and pre-set trigger conditions of different types of live broadcast events, the live broadcast event occurring in the target live broadcast room is identified; wherein for different types of live broadcast data, the trigger conditions of each type of live broadcast event include data in different forms; When the identified live broadcast event is a product switching event, a target product is determined; the target product is the product being displayed in the target live broadcast room; Determine the target advertising creative corresponding to the target product; the target advertising creative is used to deliver the target advertisement to the target device; when the target device detects a click operation on the target advertisement, the target live broadcast room can be entered through the target device.

2. The method according to claim 1, characterized in that The using the live broadcast data and pre-set trigger conditions of different types of live broadcast events to identify the live broadcast event occurring in the target live broadcast room includes: Using the live broadcast data, determine whether the target live broadcast room meets the triggering conditions of various types of live broadcast events, and obtain a determination result; Based on the judgment result, a live broadcast event occurring in the target live broadcast room is identified.

3. The method according to claim 2, characterized in that The live broadcast data includes live broadcast pictures; and using the live broadcast data to determine whether the target live broadcast room meets the triggering conditions of various types of live broadcast events includes: Comparing the live broadcast picture with pictures in a first database to obtain a first comparison result; the first database includes a plurality of pictures and a type of live broadcast event corresponding to each picture; Based on the first comparison result, determine whether the target live broadcast room meets the trigger conditions of each type of live broadcast event; wherein, when the first comparison result indicates that the similarity between the live broadcast picture and the target picture in the first database is greater than a first threshold, it is determined that the target live broadcast room meets the trigger conditions of the live broadcast event of the type corresponding to the target picture.

4. The method according to claim 2, characterized in that: The live broadcast data includes live broadcast video clips; The using the live broadcast data to determine whether the target live broadcast room meets the triggering conditions of various types of live broadcast events includes: Decomposing the live video clip into a plurality of live pictures to obtain a live picture sequence; Comparing the live broadcast picture sequence with a picture sequence in a second database to obtain a second comparison result; the second database includes a plurality of picture sequences and a type of live broadcast event corresponding to each picture sequence; Based on the second comparison result, determine whether the target live broadcast room meets the trigger conditions of each type of live broadcast event; wherein, when the second comparison result indicates that the similarity between the live broadcast picture sequence and the target picture sequence in the second database is greater than a second threshold, it is determined that the target live broadcast room meets the trigger conditions of the live broadcast event of the corresponding type of the target picture sequence.

5. The method according to any one of claims 1 to 4, characterized in that: The step of determining a target advertisement creative corresponding to the target product includes: Receiving an advertisement request from the target device; Based on the resource position identifier carried in the advertisement request, a target advertisement creative corresponding to the target commodity is determined; the resource position identifier represents the form in which the target device presents the target advertisement.

6. The method according to claim 5, characterized in that The determining, based on the resource location identifier carried in the advertisement request, a target advertisement creative corresponding to the target product, includes: Based on the resource position identifier, the target advertising creative corresponding to the target product is determined from a third database; the third database contains multiple advertising creatives corresponding to each product among all the products that need to be displayed in the target live broadcast room; the advertising presentation forms corresponding to each of the multiple advertising creatives corresponding to each product are different; each advertising creative contained in the third database has passed a preset review process.

7. The method according to claim 5, characterized in that The determining, based on the resource location identifier carried in the advertisement request, a target advertisement creative corresponding to the target product, includes: Based on the resource location identifier, multiple target creative elements corresponding to the target product are determined from the fourth database; the target advertising creativity is synthesized using the multiple target creative elements; the fourth database contains all the creative elements that can be used in the target live broadcast room; each creative element contained in the fourth database has passed a preset review process.

8. A data processing device, characterized in that: include: An acquisition unit, configured to acquire live broadcast data of a target live broadcast room, wherein the live broadcast data includes at least one of live broadcast text, live broadcast pictures, and live broadcast video clips; A first processing unit is used to identify the live broadcast event occurring in the target live broadcast room by using the live broadcast data and pre-set trigger conditions of different types of live broadcast events; wherein for different types of live broadcast data, the trigger conditions of each type of live broadcast event include data in different forms; The second processing unit is configured to determine a target product when the identified live broadcast event is a product switching event; the target product is the product being displayed in the target live broadcast room; The third processing unit is used to determine the target advertising creative corresponding to the target product; the target advertising creative is used to deliver the target advertisement to the target device; when the target device detects a click operation on the target advertisement, the target live broadcast room can be entered through the target device.

9. An electronic device, characterized in that: include: A processor and a memory for storing a computer program that can be run on the processor, wherein the processor executes the steps of the method according to any one of claims 1 to 7 when running the computer program.

10. A storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 7 are implemented.

Images