Internet advertisement putting method and system based on artificial intelligence

Abstract

The invention relates to an Internet advertisement putting method and system based on artificial intelligence, and relates to the field of advertisement putting, and the method comprises the steps: classifying a first video through a classifier when the first video is uploaded, and obtaining a first video type; configuring a first advertisement category for the first video category; and when the first video is clicked and played by the first user, randomly matching the to-be-delivered advertisement from the pre-stored advertisement according to the first advertisement category, and delivering the to-be-delivered advertisement to the first user. According to the method and the device, the technical problem of inaccurate advertisement putting caused by inaccurate data identification before putting in the prior art is solved, and the technical effects of quickly and accurately matching the video content with the advertisement category and realizing accurate advertisement putting are achieved.

Description

Technical Field

[0001] The present invention relates to the field of advertising delivery, and in particular to an artificial intelligence-based Internet advertising delivery method and system. Background Art

[0002] With the rapid development of Internet technology, video has become one of the important information carriers on the Internet. As a common form of advertising, video advertising is favored by advertisers because of its vivid images and strong communication power. However, existing video advertising delivery technologies generally have problems such as low relevance between advertisements and video content and insufficient accuracy of advertising delivery. This is because existing technologies are difficult to accurately identify the core information of video content and cannot achieve accurate matching of video content and advertisements, resulting in inaccurate advertising delivery. Summary of the invention

[0003] The present invention aims to solve the technical problem in the prior art of inaccurate data recognition before advertising, which leads to inaccurate advertising delivery, and provides an Internet advertising delivery based on artificial intelligence to solve the problem.

[0004] The technical solution of the present invention to solve the above technical problems is as follows: In a first aspect, the present invention provides an artificial intelligence-based Internet advertising delivery method, comprising: when a first video is uploaded, classifying the first video through a classifier to obtain a first video category; configuring a first advertisement category for the first video category; when the first video is clicked and played by a first user, randomly matching an advertisement to be delivered from a pre-stored advertisement according to the first advertisement category and delivering it to the first user.

[0005] In a second aspect, the present invention provides an artificial intelligence-based Internet advertising delivery system, comprising: a video classification module, used to classify the first video through a classifier to obtain a first video category when a first video is uploaded; an advertising configuration module, used to configure a first advertising category for the first video category; an advertising delivery module, used to randomly match the advertisement to be delivered from the pre-stored advertisements according to the first advertising category and deliver it to the first user when the first video is clicked to play by the first user.

[0006] In a third aspect, the present application provides an electronic device comprising: a processor; a memory for storing processor executable instructions; wherein the processor is used to execute an artificial intelligence-based Internet advertising delivery method provided in the present application.

[0007] In a fourth aspect, the present application provides a computer-readable storage medium storing a computer program, which is used to execute an artificial intelligence-based Internet advertising delivery method provided by the present application.

[0008] The beneficial effects of the present invention are: When the first video is uploaded, the first video is classified by a classifier to obtain a first video category, and the content category of the video is automatically identified to provide a basis for subsequent advertisement matching; the first advertisement category is configured for the first video category, so that videos of the same category can correspond to advertisements of the same category, thereby improving the relevance of advertisement delivery; when the first video is clicked and played by the first user, advertisements to be delivered are randomly matched from pre-stored advertisements according to the first advertisement category and delivered to the first user, and advertisements related to the video category are quickly matched and delivered immediately when the user clicks on the video, thereby improving the timeliness and accuracy of advertisement delivery. BRIEF DESCRIPTION OF THE DRAWINGS

[0009] Figure 1 A schematic diagram of a flow chart of an artificial intelligence-based Internet advertising delivery method provided by the present invention; Figure 2 A schematic diagram of the structure of an artificial intelligence-based Internet advertising delivery system provided by the present invention; Figure 3 A schematic diagram of the structure of an electronic device provided by the present invention; Figure 4 A schematic diagram of the structure of a computer-readable storage medium provided by the present invention.

[0010] In the accompanying drawings, the components represented by each reference numeral are listed as follows: video classification module 11, advertisement configuration module 12, advertisement delivery module 13, electronic device 500, memory 510, processor 520, computer program 511, computer readable storage medium 600, computer program 611. DETAILED DESCRIPTION

[0011] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative work are within the scope of protection of the present invention.

[0012] In the description of the present invention, the terms "first" and "second" are used for descriptive purposes only and should not be understood as indicating or implying relative importance or implicitly indicating the number of the indicated technical features. Thus, the features defined as "first" and "second" may explicitly or implicitly include one or more of the features. In the description of the present invention, the meaning of "plurality" is two or more, unless otherwise clearly and specifically defined.

[0013] In the description of the present invention, the term "for example" is used to mean "used as an example, illustration or explanation". Any embodiment described as "for example" in the present invention is not necessarily to be construed as being more preferred or advantageous than other embodiments. The following description is given to enable any technician in the field to implement and use the present invention. In the following description, details are listed for the purpose of explanation. It should be understood that a person of ordinary skill in the art can recognize that the present invention can be implemented without using these specific details. In other examples, well-known structures and processes will not be elaborated in detail to avoid obscuring the description of the present invention with unnecessary details. Therefore, the present invention is not intended to be limited to the embodiments shown, but is consistent with the widest scope consistent with the principles and features disclosed in the present invention.

[0014] Embodiment 1: like Figure 1 As shown, an embodiment of the present invention provides an Internet advertising delivery method based on artificial intelligence, including: S1: When a first video is uploaded, the first video is classified by a classifier to obtain a first video category.

[0015] Specifically, when a user or a video platform uploads a first video, a pre-trained classifier is used to analyze and process the first video. By obtaining the video frame information of the first video and extracting features from the video frame, a feature vector of the first video is obtained. The extracted feature vector is then input into the classifier for classification and judgment, thereby obtaining the category to which the first video belongs, that is, obtaining the first video category. Among them, the classifier adopts a video classification algorithm, such as a convolutional neural network model based on deep learning. The classifier is trained with a large amount of video data so that it can quickly and accurately determine the category to which the video belongs. In practical applications, common video categories are defined and labeled in advance, such as entertainment, news, sports, education, etc., and the classifier is trained accordingly. After obtaining the first video category, it is stored for subsequent advertising matching and delivery processes. By identifying and recording video categories, a data foundation is laid for achieving accurate advertising delivery.

[0016] S2: configuring a first advertisement category for the first video category.

[0017] Specifically, a database of correspondences between video categories and advertising categories is pre-established. The database records one or more advertising categories corresponding to each video category. For example, sports videos are configured with advertisements related to sporting goods, sports drinks, etc.; education videos are configured with advertisements related to online courses, supplementary teaching books, etc. When the video category of the first video is obtained, the correspondence database is automatically searched to find one or more advertising categories that match the first video category as the first advertising category. The first advertising category is semantically associated with the first video category, which improves the pertinence and accuracy of advertising delivery.

[0018] In another embodiment, the advertiser or administrator may manually specify the first advertisement category corresponding to the first video category. Through manual intervention, the configuration relationship between the video category and the advertisement category may be adjusted more flexibly and dynamically.

[0019] Regardless of whether automatic matching or manual configuration is used, establishing a correspondence between the first video category and the first advertisement category and configuring the first advertisement category for the first video category can provide a basis for subsequent advertisement delivery, so as to quickly respond and deliver when an advertisement request is received.

[0020] S3: When the first video is clicked by the first user to play, the advertisement to be delivered is randomly matched from the pre-stored advertisements according to the first advertisement category and delivered to the first user.

[0021] Specifically, a large amount of candidate advertising data is pre-stored, and each advertising data includes information such as advertising content and advertising category. These candidate advertisements can be provided and uploaded by advertisers, or obtained from advertising platforms or advertising networks. When the first user clicks to play the first video on the video platform, the play request is received, and the first video category of the first video and the corresponding configured first advertising category are obtained. Then, one or more advertisements matching the first advertising category are screened out from the pre-stored candidate advertisements to form a set of candidate advertisements. Next, an advertisement is randomly selected from the set of candidate advertisements as the advertisement to be delivered. The random selection method can avoid the frequent recurrence of the same advertisement, which may cause aesthetic fatigue to users, and on the other hand, it can also give different advertisements equal opportunities to be displayed, thereby improving the participation and satisfaction of advertisers. Afterwards, the randomly selected advertisement to be delivered is pushed to the first user, so that the first user receives relevant advertising information while watching the first video.

[0022] By dynamically matching and delivering relevant advertisements according to the video category when the first user watches the first video, the timeliness and accuracy of advertisement delivery can be improved through personalized advertisement recommendation.

[0023] Further, classifying the first video by a classifier to obtain a first video category includes: Step S110: performing feature preprocessing on the first video to obtain a first video feature vector; Step S120: Processing the first video feature vector through a classifier to obtain the first video category.

[0024] In a feasible implementation, first, a feature preprocessing operation is performed on the first video, the purpose of which is to convert the video data into a feature vector representation suitable for input into the classifier. Various video feature extraction algorithms are used, such as frame difference method, optical flow method, HOG (Histogram of Oriented Gradients), SIFT (Scale Invariant Feature Transform), etc., to capture the temporal information and spatial information between video frames to form high-dimensional feature data. Next, the extracted feature data is preprocessed, such as normalization, dimensionality reduction, filtering and other operations, to remove redundancy and noise and improve the robustness and discriminability of the features. After preprocessing, the feature vector representation of the first video is obtained, which is recorded as the first video feature vector.

[0025] Afterwards, the first video feature vector is input into a pre-trained classifier, and the classifier calculates and judges to obtain the category label to which the video belongs, that is, the first video category. The classifier can use machine learning or deep learning algorithms, such as support vector machines, random forests, convolutional neural networks, long short-term memory networks, etc. It is trained through a large amount of labeled video sample data to learn the characteristic patterns and discrimination rules of different categories of videos. When the first video feature vector is input, the classifier calculates and classifies the features according to the learned model parameters, and outputs the category with the highest probability as the judgment result. For example, if the result output by the classifier is "Sports: 0.8, Entertainment: 0.1, News: 0.1", then the "Sports" with the largest probability value is taken as the category label of the first video, as the first video category.

[0026] By extracting features, preprocessing and classifier judgment of the first video, category information of the first video is obtained, laying a foundation for subsequent advertisement matching.

[0027] Further, performing feature preprocessing on the first video to obtain a first video feature vector includes: Step S111: obtaining a video frame at the tth moment and a video frame at the t+1th moment of the first video; Step S112: performing feature fusion on the video frame at time t and the video frame at time t+1 to obtain a first feature vector at time t+1; Step S113: performing time feature fusion on the first feature vector at time t+1 to obtain a second feature vector at time t+1; Step S114: until the second feature vector at time t+N is obtained, where N represents the total number of video frames minus one; Step S115: Through a fully connected layer, sequentially concatenate the first feature vector at time t+1 to the second feature vector at time t+N to obtain the first video feature vector.

[0028] In a preferred embodiment, first, two adjacent video frames in the first video are obtained in chronological order, and are respectively recorded as the video frame at time t and the video frame at time t+1, where t is the timestamp of the frame. Then, feature extraction and fusion processing are performed on the video frame at time t and the video frame at time t+1. Specifically, algorithms such as convolutional neural networks are used to extract feature vectors representing appearance information from the two video frames, and then the two feature vectors are fused according to a certain fusion strategy (such as splicing, summing, averaging, etc.) to obtain a fused feature vector, which is recorded as the first feature vector at time t+1. By making full use of the temporal continuity and visual correlation between adjacent video frames, a more comprehensive and robust feature representation is obtained by feature fusion.

[0029] Next, the first t+1 time feature vector is further fused with temporal features. Specifically, algorithms such as recurrent neural networks (such as LSTM, GRU, etc.) or temporal convolutional networks are used to perform sequence modeling and feature fusion on the first t+1 time feature vector and the feature vectors of several previous times, so as to capture the long-term dependency between video frames and obtain the fused feature vector, which is recorded as the second t+1 time feature vector. Repeat the process to process each pair of adjacent video frames in the first video in turn, and obtain a series of second time feature vectors, including the second t+1 time feature vector until the second t+N time feature vector is obtained, for a total of N feature vectors, where N is the total number of video frames of the first video minus one. For the feature vector at t+N, the processing flow is the same as that of the first t+1 time feature vector. Afterwards, the feature vectors at the first t+1 time to the second t+N time feature vector are sequentially spliced ​​or fused through the fully connected layer to form the first video feature vector. This feature vector contains the spatiotemporal feature information of the entire video and is used for subsequent video classification tasks.

[0030] Further, feature fusion is performed on the video frame at time t and the video frame at time t+1 to obtain a first feature vector at time t+1, including: Step S1121: performing optical flow analysis on the video frame at time t and the video frame at time t+1 to obtain an optical flow image; Step S1122: performing pixel difference analysis on the video frame at time t and the video frame at time t+1 to obtain a pixel difference image; Step S1123: stitching the optical flow image and the pixel difference image to obtain a stitched image; Step S1124: extracting features from the stitched image to obtain the first feature vector at time t+1.

[0031] In a preferred embodiment, when the feature fusion is performed on the video frame at the tth moment and the video frame at the t+1th moment to obtain the first feature vector at the t+1th moment, first, the pixel motion information between the video frame at the tth moment and the video frame at the t+1th moment is calculated using an optical flow analysis algorithm, such as the Lucas-Kanade algorithm, the Horn-Schunck algorithm, etc. Optical flow is a vector field that describes the pixel motion pattern in an image sequence, reflecting the correspondence and motion trend between video frames. Through optical flow analysis, an optical flow image between two video frames is obtained, in which the value of each pixel point represents the displacement vector of the point between the two frames. The optical flow image can effectively capture the motion information in the video and provide useful prior knowledge for subsequent feature extraction. Next, a pixel difference image is obtained by calculating the difference between the corresponding pixel points of the video frame at the tth moment and the video frame at the t+1th moment. Specifically, the pixel value of the video frame at the tth moment is subtracted from the pixel value of the video frame at the tth moment by the pixel value of the video frame at the t+1th moment to obtain a difference image, and then the absolute value is taken to obtain a pixel difference image. The pixel difference image reflects the pixel changes between two video frames, highlighting the moving areas and foreground objects to capture the changes in video content.

[0032] After obtaining the optical flow image and the pixel difference image, the two images are spliced ​​to form a spliced ​​image. Among them, the splicing method can be horizontal splicing, vertical splicing or channel splicing, etc., which is selected according to the specific features. By splicing the optical flow image and the pixel difference image, a more comprehensive feature representation is obtained by using motion information and pixel change information at the same time. After that, feature extraction algorithms such as convolutional neural networks are used to learn and extract features of the spliced ​​image. Through multi-layer convolution and pooling operations, the convolutional neural network automatically learns the hierarchical features in the image, from low-level edges and textures to high-level semantic concepts. The spliced ​​image is input into the pre-trained convolutional neural network, and after forward propagation calculation, a fixed-length feature vector is obtained, that is, the first t+1 time feature vector. This feature vector integrates the information of the optical flow image and the pixel difference image, and can better characterize the content changes and semantic information between adjacent video frames.

[0033] Furthermore, feature extraction is performed on the spliced ​​image to obtain the first feature vector at time t+1, including: Step S11241: inputting the stitched image into three parallel convolutional feature extraction networks to obtain three stitched image feature vectors; Step S11242: flattening the three stitched image feature vectors to obtain three flattened feature vectors; Step S11243: splicing the three flattened feature vectors according to RGB channels to obtain a spliced ​​feature vector; Step S11244: fusing the concatenated feature vectors through a fully connected neural network to obtain the first feature vector at time t+1, wherein the fully connected neural network includes a plurality of fully connected layers, and the sizes of the plurality of fully connected layers decrease successively.

[0034] In a preferred embodiment, when extracting features from a spliced ​​image, first, the obtained spliced ​​image is simultaneously input into three parallel convolutional feature extraction networks. The three networks have the same structure but independent parameters, and the purpose is to extract feature information of the spliced ​​image from different scales and angles. Specifically, each convolutional feature extraction network is composed of multiple convolutional layers, pooling layers, and activation functions, and the spliced ​​image is mapped to a low-dimensional feature space through layer-by-layer feature transformation and abstraction. The three parallel convolutional feature extraction networks are calculated in parallel to obtain three spliced ​​image feature vectors of the same dimension, which are respectively recorded as spliced ​​image feature vector 1, spliced ​​image feature vector 2, and spliced ​​image feature vector 3. Next, the three spliced ​​image feature vectors obtained are flattened. Specifically, the multi-dimensional tensor form of each feature vector is converted into a one-dimensional vector form, keeping the order of the eigenvalues ​​unchanged. After flattening, three flattened feature vectors are obtained, which are respectively recorded as flattened feature vector 1, flattened feature vector 2, and flattened feature vector 3. The flattening operation can unify feature vectors from different sources into the same representation form, preparing for subsequent feature fusion.

[0035] Then, in the order of RGB channels, flattened feature vector 1, flattened feature vector 2 and flattened feature vector 3 are concatenated to obtain a concatenated feature vector with a length three times that of the original one, so as to summarize and integrate the feature information extracted by the three parallel convolutional feature extraction networks and obtain a more comprehensive and high-dimensional feature representation. The concatenated feature vector contains the semantic information of the concatenated image at different scales and angles, providing richer prior knowledge for subsequent feature fusion. After that, the concatenated feature vector is input into a pre-designed fully connected neural network, and the first feature vector at time t+1 is obtained through forward propagation calculation. The fully connected neural network consists of multiple fully connected layers, each of which contains several neurons, and performs nonlinear transformation and high-level abstraction on the input features. Unlike convolutional neural networks, each neuron in the fully connected layer is connected to all neurons in the previous layer, which can achieve global integration and fusion of features. In order to improve the effect of feature fusion and reduce the risk of overfitting, the fully connected neural network adopts a size-decreasing design, that is, the number of neurons in each fully connected layer decreases successively. Through this structural design, the dimension of the feature is gradually compressed, and the most discriminative and representative high-level features are extracted. In the last layer of the fully connected neural network, the output feature vector is the feature vector at the first time t+1, which integrates the semantic information of the spliced ​​image on different convolutional networks, different flattened feature vectors and different fully connected layers, and can better characterize the content association and action characteristics between adjacent video frames.

[0036] Further, performing time feature fusion on the first feature vector at time t+1 to obtain the feature vector at time t+1 includes: Step S1131: obtaining the first feature vector at time t+1 to the first feature vector at time t+L, where L represents a preset time step; Step S1132: Input the first feature vector at time t+1 to the first feature vector at time t+L into the LSTM neural network for processing to obtain the second feature vector at time t+1.

[0037] In a preferred embodiment, when performing time feature fusion on the first t+1 moment feature vector to obtain the second t+1 moment feature vector, first, from the obtained series of first moment feature vectors, a time window starting from the first t+1 moment feature vector and having a length of L is selected, and L continuous feature vectors are extracted therein to obtain the first t+1 moment feature vector until the first t+L moment feature vector. Among them, the time step L is a pre-set parameter that determines the length of the context information considered each time the time feature fusion is performed. By means of a sliding time window, local time series features are extracted at different time positions and used as the input sequence of the LSTM neural network.

[0038] Next, the obtained L first-time feature vectors are sequentially input into the pre-trained LSTM neural network in chronological order, and the second t+1 time feature vector is obtained through forward propagation calculation. The LSTM neural network is a recurrent neural network. By introducing a gating mechanism and a memory unit, it can effectively learn and store long-term dependent information. Compared with traditional recurrent neural networks, LSTM can alleviate the problems of gradient vanishing and gradient explosion. In the process of temporal feature fusion, the LSTM neural network gradually receives the L first-time feature vectors and accumulates and updates the timing information in the internal state. Through the control of the gating unit, the LSTM adaptively decides when to forget the past state, when to update the current state, and when to output the final feature representation. After being processed by the LSTM neural network, the feature vector output at the t+1th time is the second t+1 time feature vector. Compared with the first t+1 time feature vector, the second t+1 time feature vector not only contains the video frame information of the current time, but also integrates the context information of the previous L-1 times, which can better capture the temporal continuity and action evolution law in the video.

[0039] Through the fusion of time window extraction and LSTM neural network, the information integration of the time dimension of the first feature vector at time t+1 is realized, and the feature vector at time t+1 containing richer temporal semantics is obtained, which makes full use of the temporal continuity and motion evolution laws of video data, and provides important prior knowledge for subsequent video classification tasks.

[0040] Further, the first video feature vector is processed by a classifier to obtain the first video category, including: Step S121: obtaining a video feature vector dataset and a video category identification dataset; Step S122: supervising a classifier with the video category identification dataset, taking the video feature vector dataset as classifier input, and training the classifier.

[0041] In a feasible implementation, when the first video feature vector is processed by a classifier to obtain the first video category, first, a data set for training the classifier is prepared. The data set includes two parts: a video feature vector data set and a video category identification data set. Among them, the video feature vector data set is obtained by performing a feature preprocessing process on a large number of video samples. After feature extraction and fusion, each video sample forms a feature vector of fixed length, and the feature vectors of all video samples are aggregated together to form a video feature vector data set; the video category identification data set is label information corresponding to the video feature vector data set one by one. For each video sample, there is a predefined category identification, which indicates the semantic category to which the video belongs, for example, sports, news, entertainment, education, etc. The category identification of all video samples is aggregated together to form a video category identification data set. The acquisition of these two data sets can be completed by various methods such as manual annotation, web crawling, and public data sets. In order to ensure the generalization performance of the classifier, the video samples in the data set should cover different shooting scenes, action types, time lengths, etc. as much as possible, and the category distribution should be relatively balanced.

[0042] Next, the classifier is trained using the obtained video feature vector dataset and video category identification dataset. The classifier can use machine learning algorithms such as support vector machine, decision tree, naive Bayes, etc., or deep learning algorithms such as multi-layer perceptron, convolutional neural network, etc. During the training process, the classifier takes the video feature vector dataset as input and the video category identification dataset as the supervision signal. Through iterative optimization, the internal parameters and weights are continuously adjusted to minimize the error between the predicted category and the true category. In order to improve the generalization performance of the classifier and avoid overfitting, cross-validation is used to evaluate the model and select hyperparameters. The dataset is divided into training set, validation set and test set. Model training is performed on the training set, model selection is performed on the validation set, and the final performance of the model is evaluated on the test set. After sufficient training and tuning, the obtained classifier can effectively map the video feature vector to the corresponding video category identification, and realize the automatic classification of new unknown videos.

[0043] By constructing a video feature vector dataset and a video category identification dataset and using them to perform supervised learning on the classifier, a video classification model with superior performance is obtained to provide support for subsequent advertising delivery.

[0044] An artificial intelligence-based Internet advertising delivery method provided by an embodiment of the present invention has at least the following technical effects: When the first video is uploaded, the first video is classified by a classifier to obtain the first video category. The video content is identified by the classifier and the video category is obtained, laying the foundation for subsequent advertising matching. The first advertising category is configured for the first video category. By configuring the corresponding advertising category for each video category, it is ensured that videos of the same category are matched with similar advertisements with higher relevance, thereby improving the targeted advertising. When the first video is clicked and played by the first user, the advertisements to be delivered are randomly matched from the pre-stored advertisements according to the first advertising category and delivered to the first user. Advertisements of the corresponding category are quickly matched according to the advertising category, and the advertisements are pushed immediately when the user clicks the video, thereby improving the timeliness of advertising delivery and achieving precise advertising delivery.

[0045] Embodiment 2: like Figure 2 As shown, based on the same inventive concept as the artificial intelligence-based Internet advertising delivery method provided in Embodiment 1, the embodiment of the present invention further provides an artificial intelligence-based Internet advertising delivery system, including: A video classification module 11, wherein when a first video is uploaded, the video classification module is used to classify the first video through a classifier to obtain a first video category; An advertisement configuration module 12, the advertisement configuration module being used to configure a first advertisement category for the first video category; The advertisement delivery module 13 is used for delivering advertisements to the first user by randomly matching the advertisements to be delivered from the pre-stored advertisements according to the first advertisement category when the first video is clicked by the first user for playing.

[0046] Furthermore, the video classification module 11 includes the following execution steps: Performing feature preprocessing on the first video to obtain a first video feature vector; The first video feature vector is processed by a classifier to obtain the first video category.

[0047] Furthermore, the video classification module 11 further includes the following execution steps: Obtaining a video frame at time t and a video frame at time t+1 of the first video; Performing feature fusion on the video frame at time t and the video frame at time t+1 to obtain a first feature vector at time t+1; Performing time feature fusion on the first feature vector at time t+1 to obtain a second feature vector at time t+1; Until the second feature vector at time t+N is obtained, N represents the total number of video frames minus one; The first video feature vector is obtained by sequentially concatenating the first feature vector at time t+1 to the second feature vector at time t+N through a fully connected layer.

[0048] Furthermore, the video classification module 11 further includes the following execution steps: Performing optical flow analysis on the video frame at time t and the video frame at time t+1 to obtain an optical flow image; Performing pixel difference analysis on the video frame at time t and the video frame at time t+1 to obtain a pixel difference image; Stitching the optical flow image and the pixel difference image to obtain a stitched image; Perform feature extraction on the spliced ​​image to obtain the first feature vector at time t+1.

[0049] Furthermore, the video classification module 11 further includes the following execution steps: Inputting the stitched image into three parallel convolutional feature extraction networks to obtain three stitched image feature vectors; Flattening the three stitched image feature vectors to obtain three flattened feature vectors; Concatenate the three flattened feature vectors according to RGB channels to obtain a concatenated feature vector; The concatenated feature vectors are fused through a fully connected neural network to obtain the first feature vector at time t+1, wherein the fully connected neural network includes a plurality of fully connected layers, and the sizes of the plurality of fully connected layers decrease successively.

[0050] Furthermore, the video classification module 11 further includes the following execution steps: Obtaining the first feature vector at time t+1 to the first feature vector at time t+L, where L represents a preset time step; The first feature vector at time t+1 to the first feature vector at time t+L are input into the LSTM neural network for processing to obtain the second feature vector at time t+1.

[0051] Furthermore, the video classification module 11 further includes the following execution steps: Obtain a video feature vector dataset and a video category identification dataset; The classifier is supervised by the video category identification data set, and the video feature vector data set is used as the classifier input to train the classifier.

[0052] Embodiment three: See also Figure 3 , Figure 3 Schematic diagram of an electronic device provided by an embodiment of the present invention. Figure 3As shown, an electronic device 500 provided by an embodiment of the present invention includes a memory 510, a processor 520, and a computer program 511 stored in the memory 510 and executable on the processor 520. When the processor 520 executes the computer program 511, an artificial intelligence-based Internet advertising delivery method is implemented.

[0053] Embodiment 4: See also Figure 4 , Figure 4 A schematic diagram of an embodiment of a computer-readable storage medium provided in an embodiment of the present invention. Figure 4 As shown, this embodiment provides a computer-readable storage medium 600 on which a computer program 611 is stored. When the computer program 611 is executed by a processor, an artificial intelligence-based Internet advertising delivery method is implemented.

[0054] It should be noted that in the above embodiments, the description of each embodiment has its own emphasis, and for parts that are not described in detail in a certain embodiment, reference can be made to the relevant descriptions of other embodiments.

[0055] It will be appreciated by those skilled in the art that embodiments of the present invention may be provided as methods, systems, or computer program products. Therefore, the present invention may take the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware. Furthermore, the present invention may take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0056] The present invention is described with reference to flowcharts and / or block diagrams of methods, devices (systems), and computer program products according to embodiments of the present invention. It should be understood that each process and / or block in the flowchart and / or block diagram, as well as the combination of processes and / or blocks in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded computer, or other programmable data processing device to produce a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 A process or multiple processes and / or boxes Figure 1 A system that specifies the functions of a box or multiple boxes.

[0057] These computer program instructions may also be stored in a computer-readable memory that can direct a computer or other programmable data processing device to work in a specific manner, so that the instructions stored in the computer-readable memory produce a product including an instruction system, which is implemented in the process Figure 1A process or multiple processes and / or boxes Figure 1 A function specified in one or more boxes.

[0058] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operating steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing instructions for implementing the process. Figure 1 A process or multiple processes and / or boxes Figure 1 The steps for the functions specified in one or more boxes.

[0059] Although preferred embodiments of the present invention have been described, additional changes and modifications may occur to these embodiments once those skilled in the art understand the basic inventive concepts.

[0060] Obviously, those skilled in the art can make various changes and modifications to the present invention without departing from the spirit and scope of the present invention. Thus, if these modifications and variations of the present invention belong to the scope of the present invention and its equivalent technologies, the present invention is also intended to include these changes and variations.

Claims

1. An Internet advertising delivery method based on artificial intelligence, characterized in that: include: When the first video is uploaded, the first video is classified by a classifier to obtain a first video category; configuring a first advertisement category for the first video category; When the first video is clicked by the first user to play, randomly matching the advertisement to be delivered from the pre-stored advertisements according to the first advertisement category and delivering the advertisement to the first user; The step of classifying the first video by a classifier to obtain a first video category includes: Performing feature preprocessing on the first video to obtain a first video feature vector; The first video feature vector is processed by a classifier to obtain the first video category.

2. The method according to claim 1, characterized in that Performing feature preprocessing on the first video to obtain a first video feature vector includes: Obtaining a video frame at time t and a video frame at time t+1 of the first video; Performing feature fusion on the video frame at time t and the video frame at time t+1 to obtain a first feature vector at time t+1; Performing time feature fusion on the first feature vector at time t+1 to obtain a second feature vector at time t+1; Until the second feature vector at time t+N is obtained, N represents the total number of video frames minus one; The first video feature vector is obtained by sequentially concatenating the first feature vector at time t+1 to the second feature vector at time t+N through a fully connected layer.

3. The method according to claim 2, characterized in that Performing feature fusion on the video frame at time t and the video frame at time t+1 to obtain a first feature vector at time t+1 includes: Performing optical flow analysis on the video frame at time t and the video frame at time t+1 to obtain an optical flow image; Performing pixel difference analysis on the video frame at time t and the video frame at time t+1 to obtain a pixel difference image; Stitching the optical flow image and the pixel difference image to obtain a stitched image; Perform feature extraction on the spliced ​​image to obtain the first feature vector at time t+1.

4. The method according to claim 3, characterized in that Performing feature extraction on the spliced ​​image to obtain the first feature vector at time t+1 includes: Inputting the stitched image into three parallel convolutional feature extraction networks to obtain three stitched image feature vectors; Flattening the three stitched image feature vectors to obtain three flattened feature vectors; Concatenate the three flattened feature vectors according to RGB channels to obtain a concatenated feature vector; The concatenated feature vectors are fused through a fully connected neural network to obtain the first feature vector at time t+1, wherein the fully connected neural network includes a plurality of fully connected layers, and the sizes of the plurality of fully connected layers decrease successively.

5. The method according to claim 2, characterized in that Performing time feature fusion on the first feature vector at time t+1 to obtain the feature vector at time t+1 includes: Obtaining the first feature vector at time t+1 to the first feature vector at time t+L, where L represents a preset time step; The first feature vector at time t+1 to the first feature vector at time t+L are input into the LSTM neural network for processing to obtain the second feature vector at time t+1.

6. The method according to claim 1, characterized in that Processing the first video feature vector by a classifier to obtain the first video category includes: Obtain a video feature vector dataset and a video category identification dataset; The classifier is supervised by the video category identification data set, and the video feature vector data set is used as the classifier input to train the classifier.

7. An Internet advertising delivery system based on artificial intelligence, characterized in that: include: A video classification module, wherein when a first video is uploaded, the video classification module is used to classify the first video through a classifier to obtain a first video category; an advertisement configuration module, the advertisement configuration module being configured to configure a first advertisement category for the first video category; an advertisement delivery module, configured to deliver an advertisement to be delivered to the first user by randomly matching the advertisement to be delivered from the pre-stored advertisements according to the first advertisement category when the first video is clicked by the first user for playback; The step of classifying the first video by a classifier to obtain a first video category includes: Performing feature preprocessing on the first video to obtain a first video feature vector; The first video feature vector is processed by a classifier to obtain the first video category.

8. An electronic device, characterized in that: include: Memory for storing computer software programs; A processor is used to read and execute the computer software program, thereby implementing an artificial intelligence-based Internet advertising delivery method as described in any one of claims 1-6.

9. A non-transitory computer-readable storage medium, characterized in that: The storage medium stores a computer software program, which, when executed by a processor, implements an artificial intelligence-based Internet advertising delivery method as described in any one of claims 1 to 6.

Images