A big data-based message reminding method and system

Abstract

The application provides a message reminding method and system based on big data, and relates to the technical field of message reminding.The method comprises the following steps: obtaining historical chat records of a user and a plurality of video files to be received in the chat records; using a variational autoencoder to generate a plurality of user interest pictures based on the historical chat records of the user; displaying the plurality of user interest pictures on a user software interface and obtaining a plurality of target interest pictures selected by the user; determining a plurality of video files that need to be reminded based on the plurality of video files to be received in the chat records and the plurality of target interest pictures selected by the user; receiving and storing the video files that need to be reminded and reminding the user of messages.The method can intelligently screen out video files that the user may be interested in and remind the user of messages.

Description

Technical Field

[0001] This invention relates to the field of message notification technology, and specifically to a message notification method and system based on big data. Background Technology

[0002] With the rapid development of internet and mobile communication technologies, social media and instant messaging applications have become an indispensable part of people's daily lives. Users receive a massive amount of messages daily on these platforms, including numerous video files. Without effective filtering and management of these videos, users may experience information overload and struggle to find content that truly interests them.

[0003] Therefore, how to effectively manage and filter these video files so that they can promptly attract users' attention is an urgent problem to be solved. Summary of the Invention

[0004] The main technical problem this invention addresses is how to intelligently filter out video files that users may be interested in and send them notifications.

[0005] According to a first aspect, the present invention provides a message notification method based on big data, comprising: acquiring a user's historical chat history and multiple video files to be received in the chat history; generating multiple user interest images based on the user's historical chat history using a variational autoencoder; displaying the multiple user interest images on the user's software interface, and acquiring multiple target interest images selected by the user; determining multiple video files that need to be notified via message notification based on the multiple video files to be received in the chat history and the multiple target interest images selected by the user; receiving and storing the video files that need to be notified via message notification and notifying the user via message notification.

[0006] In one possible implementation, determining the multiple video files requiring message notification based on the multiple video files to be received in the chat history and the multiple target interest images selected by the user includes:

[0007] Based on the interest determination model, the multiple video files to be received in the chat history and the multiple target interest images selected by the user are processed to determine multiple key frames of each video file to be received and the user interest of each video file to be received.

[0008] A message notification graph structure is constructed, which includes multiple nodes and multiple edges between the nodes. The multiple nodes include a user node and multiple video file nodes. Each video file node establishes an edge with the user node. The node features of the user node include multiple target interest images selected by the user. The node features of each video file node include multiple keyframes of each video file to be received. The features of the edge established between each video file node and the user node include the user interest level of each video file to be received.

[0009] The message notification graph structure is processed using a graph convolutional network to determine multiple video files that require message notifications.

[0010] In one possible implementation, the input of the variational autoencoder is the user's historical chat history, and the output of the variational autoencoder is multiple images of user interest.

[0011] In one possible implementation, the interest determination model is a recurrent neural network model. The input of the interest determination model is multiple video files to be received in the chat history and multiple target interest images selected by the user. The output of the interest determination model is multiple keyframes of each video file to be received and the user's interest score for each video file to be received.

[0012] According to a second aspect, the present invention provides a message notification system based on big data, comprising:

[0013] The acquisition module is used to acquire the user's historical chat history and multiple video files to be received from the chat history;

[0014] The generation module is used to generate multiple user interest images based on the user's historical chat records using a variational autoencoder;

[0015] The display module is used to display the multiple user interest images on the user software interface and to acquire the multiple target interest images selected by the user.

[0016] The determination module is used to determine multiple video files that need to be notified of messages based on multiple video files to be received in the chat history and multiple target interest images selected by the user;

[0017] The reminder module is used to receive and store the video files that require message reminders and to remind the user.

[0018] In one possible implementation, the probability value determination module is further configured to:

[0019] Based on the interest determination model, the multiple video files to be received in the chat history and the multiple target interest images selected by the user are processed to determine multiple key frames of each video file to be received and the user interest of each video file to be received.

[0020] A message notification graph structure is constructed, which includes multiple nodes and multiple edges between the nodes. The multiple nodes include a user node and multiple video file nodes. Each video file node establishes an edge with the user node. The node features of the user node include multiple target interest images selected by the user. The node features of each video file node include multiple keyframes of each video file to be received. The features of the edge established between each video file node and the user node include the user interest level of each video file to be received.

[0021] The message notification graph structure is processed using a graph convolutional network to determine multiple video files that require message notifications.

[0022] In one possible implementation, the input of the variational autoencoder is the user's historical chat history, and the output of the variational autoencoder is multiple images of user interest.

[0023] In one possible implementation, the interest determination model is a recurrent neural network model. The input of the interest determination model is multiple video files to be received in the chat history and multiple target interest images selected by the user. The output of the interest determination model is multiple keyframes of each video file to be received and the user's interest score for each video file to be received.

[0024] According to a third aspect, embodiments of the present invention provide an electronic device, including: a processor; a memory; and a computer program; wherein the computer program is stored in the memory and configured to be executed by the processor to implement the method as described above, the method including: acquiring a user's historical chat history and a plurality of video files to be received in the chat history; generating a plurality of user interest images based on the user's historical chat history using a variational autoencoder; displaying the plurality of user interest images on a user software interface and acquiring a plurality of target interest images selected by the user; determining a plurality of video files that need to be notified via message based on the plurality of video files to be received in the chat history and the plurality of target interest images selected by the user; receiving and storing the video files that need to be notified via message and notifying the user via message.

[0025] According to the fourth aspect, this embodiment provides a computer-readable storage medium storing a computer program thereon. When executed by a processor, the program implements the aforementioned message notification method based on big data. The method includes: acquiring a user's historical chat history and multiple video files to be received in the chat history; generating multiple user interest images based on the user's historical chat history using a variational autoencoder; displaying the multiple user interest images on a user's software interface and acquiring multiple target interest images selected by the user; determining multiple video files that need to be notified via message notification based on the multiple video files to be received in the chat history and the multiple target interest images selected by the user; receiving and storing the video files that need to be notified via message notification and notifying the user via message notification.

[0026] This invention provides a message notification method and system based on big data. The method includes acquiring a user's historical chat history and multiple video files to be received from the chat history; generating multiple user-interested images using a variational autoencoder based on the user's historical chat history; displaying the multiple user-interested images on the user's software interface and acquiring multiple target interest images selected by the user; determining multiple video files that need to be notified via message notification based on the multiple video files to be received from the chat history and the multiple target interest images selected by the user; receiving and storing the video files that need to be notified via message notification and notifying the user via message notification. This method can intelligently filter out video files that the user may be interested in and provide message notifications. Attached Figure Description

[0027] Figure 1 This is a schematic diagram illustrating an application scenario of a message notification method based on big data, provided by an embodiment of the present invention.

[0028] Figure 2 A flowchart illustrating a message notification method based on big data provided in an embodiment of the present invention;

[0029] Figure 3 This invention provides a schematic diagram of a process for determining multiple video files that require message notifications, as provided in an embodiment of the invention.

[0030] Figure 4 This is a schematic diagram of a message notification system based on big data provided in an embodiment of the present invention;

[0031] Figure 5 A schematic diagram of an electronic device provided in an embodiment of the present invention;

[0032] Figure 6 This is a schematic diagram of the structure of an electronic device provided in an embodiment of the present invention. Detailed Implementation

[0033] The present invention will now be described in further detail with reference to specific embodiments and accompanying drawings. Similar elements in different embodiments are referred to by associated similar element reference numerals. In the following embodiments, many details are described to facilitate a better understanding of the invention. However, those skilled in the art will readily recognize that some features may be omitted in different situations, or may be replaced by other elements, materials, or methods. In some cases, certain operations related to the present invention are not shown or described in the specification. This is to avoid obscuring the core parts of the invention with excessive description. For those skilled in the art, detailed description of these related operations is not necessary; they can fully understand the related operations based on the description in the specification and general technical knowledge in the art.

[0034] Figure 1 This is a schematic diagram illustrating an application scenario of a message notification method based on big data, provided in an embodiment of the present invention. Figure 1 The application scenarios of big data-based message notification methods can include servers 11, networks 12, terminals 13, and storage devices 14.

[0035] In some embodiments, server 11 may be a single server or a group of servers. Server 11 can access information and / or data stored in terminal 13 or storage device 14 via network 12. In some embodiments, server 11 may be used to perform... Figure 2 The message notification method based on big data is shown in the figure.

[0036] Network 12 can facilitate the exchange of information and / or data. In some embodiments, network 12 can be any form of wired or wireless network, or any combination thereof.

[0037] Terminal 13 may refer to one or more terminal devices used by a user. In some embodiments, terminal 13 may include one or more combinations of mobile devices, tablet computers, laptop computers, etc.

[0038] Storage device 14 can store data and / or instructions, for example, storage device 14 can store data instructions based on big data message notification methods.

[0039] In this embodiment of the invention, the following are provided: Figure 2 The above describes a message notification method based on big data, which includes steps S1 to S5:

[0040] Step S1: Obtain the user's historical chat history and multiple video files to be received from the chat history.

[0041] A user's chat history is a record of all chat messages a user has sent and received over a period of time, including various types of messages such as text, images, audio, and video. For example, user A may have sent multiple messages in their past chat history, such as "The weather is so nice today" and "This photo is beautiful," and received some video files.

[0042] The video files to be received are those that the user has not yet viewed or processed in the current chat history. For example, user A received three video files in the current chat history: "video1.mp4", "video2.mp4", and "video3.mp4".

[0043] Step S2: Based on the user's historical chat records, use a variational autoencoder to generate multiple user interest images.

[0044] The input to the variational autoencoder is the user's historical chat history, and the output of the variational autoencoder is multiple images of the user's interests.

[0045] User interest images are generated by a variational autoencoder based on the user's historical chat history, and they reflect the user's interests and preferences. For example, the generated user interest images may include scenery, people, activities, etc., that the user likes.

[0046] A Variational Autoencoder (VAE) is a generative model that learns a latent representation of data and generates new samples from it. A VAE consists of an encoder and a decoder. The encoder learns the distribution of the data and finds its latent structure. The decoder generates new samples with similar features from the latent structure learned by the encoder. The encoder transforms user history chat logs into latent variables. These latent variables capture key features and patterns in the chat logs, such as common words, topics, and sentiment. The decoder generates new samples from the latent variables sampled from the latent space. These new samples appear as new images in the data space that are similar to the original data.

[0047] User chat history contains conversations between users at different times and in different contexts, encompassing information about their interests, hobbies, and emotions. Variational autoencoders (VAEs), as generative models, can generate new samples with similar characteristics from a latent space. This means that even if the original data (user chat history) doesn't directly contain images, a VAE can learn the latent structure of the data and generate images related to the user's interests. For example, if a user frequently discusses scenery and travel, a VAE can generate images depicting beautiful landscapes; if a user frequently discusses sports and fitness, a VAE can generate images depicting sports activities.

[0048] Step S3: Display the multiple user interest images on the user software interface and obtain the multiple target interest images selected by the user.

[0049] The user interface is the interface through which users interact with applications, such as the main screen of a mobile application.

[0050] The target interest image is the image selected by the user from a pool of generated images representing user interests; it represents the image the user is currently most interested in. For example, the user selected two target interest images from the generated images: "beautiful scenery" and "sports competition."

[0051] Step S4: Based on the multiple video files to be received in the chat history and the multiple target interest images selected by the user, determine the multiple video files that need to be notified.

[0052] In some embodiments, Figure 3 This is a flowchart illustrating a process for determining multiple video files that require message notifications, as provided in an embodiment of the present invention. The determination of these multiple video files includes steps S21-S23:

[0053] Step S21: Based on the interest determination model, process the multiple video files to be received in the chat history and the multiple target interest images selected by the user to determine multiple key frames of each video file to be received and the user interest of each video file to be received.

[0054] The interest determination model is a recurrent neural network (RNN) model. The input to the RNN model consists of multiple video files to be received from the chat history and multiple target interest images selected by the user. The output of the RNN model consists of multiple keyframes for each video file to be received and the user's interest score for each video file. The RNN model can process sequential data, capture sequence information, and output results based on the relationships between preceding and following data within the sequence. By processing multiple video files to be received from chat history over a continuous time period using an RNN model, the output can comprehensively consider the relationships between sequences at various time points, making the output features more accurate and comprehensive.

[0055] Keyframes are representative frames extracted from video files and can be used to reflect the video content.

[0056] User interest score is a numerical value used to measure the degree of interest a user has in a particular video file.

[0057] In some embodiments, the interest determination model includes a keyframe determination layer, a keyframe correlation determination layer, and a user interest determination layer. Each of these layers comprises a recurrent neural network. The input to the keyframe determination layer is multiple video files to be received from the chat history, and the output is multiple keyframes from each video file to be received. The input to the keyframe correlation determination layer is multiple keyframes from each video file to be received and multiple target interest images selected by the user. The output is the correlation degree between each keyframe and each target interest image selected by the user. The input to the user interest determination layer is the correlation degree between each keyframe and each target interest image selected by the user, and the output is the user interest score for each video file to be received.

[0058] The keyframe extraction layer is specifically responsible for extracting keyframes from the video file. The output of this layer can be directly used to verify the effectiveness of the keyframe extraction.

[0059] The keyframe correlation determination layer is specifically responsible for calculating the correlation between keyframes and the user's selected target interest image. The output of this layer can be directly used to evaluate the degree of matching between keyframes and user interests.

[0060] The user interest determination layer is specifically responsible for calculating the user interest score for each video file based on the correlation between keyframes and target interest images. The output of this layer can be directly used to assess the user's level of interest in the video file.

[0061] Step S22: Construct a message notification graph structure. The message notification graph structure includes multiple nodes and multiple edges between the nodes. The multiple nodes include a user node and multiple video file nodes. Each video file node establishes an edge with the user node. The node features of the user node include multiple target interest images selected by the user. The node features of each video file node include multiple keyframes of each video file to be received. The features of the edge established between each video file node and the user node include the user interest level of each video file to be received.

[0062] The message notification graph structure is a graph used to represent the relationship between users and video files. The nodes in the graph include user nodes and video file nodes, and the edges represent the user's level of interest in the video file.

[0063] A node is a vertex in the graph, representing a user or a video file.

[0064] An edge is a line connecting two nodes, representing the user's level of interest in the video file.

[0065] Node features are the attributes of each node. The features of a user node include multiple target interest images selected by the user, while the features of a video file node include multiple keyframes.

[0066] Edge features are the attributes of the edges connecting two nodes, representing the user's level of interest in the video file.

[0067] For example, the user node is user A, and the video file nodes are the video files "video1.mp4", "video2.mp4", and "video3.mp4". The user node feature is the target interest image selected by user A, such as "beautiful scenery" and "sports competition". The video file node feature is the keyframe of the video file "video1.mp4". The edge feature is that user A's user interest score for the video file "video1.mp4" is 0.85.

[0068] Step S23: Process the message reminder graph structure based on the graph convolutional network to determine multiple video files that need to be reminded.

[0069] Graph Convolutional Networks (GCNs) are deep learning models used to process graph-structured data. The input to a GCN is the message notification graph structure, and the output is multiple video files for which message notifications are needed. GCNs process nodes and edges in the graph through convolution operations, learning feature representations of nodes. Convolution operations can capture local dependencies between nodes, thus providing a better understanding of information within the graph structure.

[0070] The message notification graph structure visually represents the relationship between users and video files through nodes and edges. Edges between user nodes and video file nodes represent the user's level of interest in the video file, making the relationship clear and straightforward. User node features include multiple selected images of interest, while video file node features include multiple keyframes. These features integrate multifaceted information about both the user and the video file. The edge features, representing the user's level of interest in the video file, help to more accurately capture the relationship between the user and the video file. The message notification graph structure is a type of graph data, well-suited for processing with graph convolutional networks. Graph convolutional networks can capture the complex relationships within the graph structure, thereby more accurately identifying the video files that require message notifications.

[0071] Step S5: Receive and store the video file that needs to be notified, and send a notification to the user.

[0072] For example, video files that require notifications can be downloaded and stored on the user's device.

[0073] The application's notification function can also inform users that there are new video files to view.

[0074] Based on the same inventive concept Figure 4 This is a schematic diagram of a big data-based message notification system provided in an embodiment of the present invention. The big data-based message notification system includes:

[0075] Module 41 is used to acquire the user's historical chat history and multiple video files to be received from the chat history;

[0076] The generation module 42 is used to generate multiple user interest images based on the user's historical chat records using a variational autoencoder.

[0077] Display module 43 is used to display the multiple user interest images on the user software interface and to acquire the multiple target interest images selected by the user.

[0078] The determination module 44 is used to determine multiple video files that need to be notified of messages based on multiple video files to be received in the chat history and multiple target interest images selected by the user;

[0079] The reminder module 45 is used to receive and store the video file that needs to be reminded and to remind the user.

[0080] Based on the same inventive concept, embodiments of the present invention provide an electronic device, such as... Figure 5 As shown, it includes:

[0081] The system includes: a processor 51; a memory 52; and a computer program; wherein the computer program is stored in the memory 52 and configured to be executed by the processor 51 to implement the big data-based message notification method provided above, the method including: acquiring a user's historical chat history and multiple video files to be received in the chat history; generating multiple user interest images based on the user's historical chat history using a variational autoencoder; displaying the multiple user interest images on the user's software interface and acquiring multiple target interest images selected by the user; determining multiple video files that need to be notified via message based on the multiple video files to be received in the chat history and the multiple target interest images selected by the user; receiving and storing the video files that need to be notified via message and notifying the user via message.

[0082] Based on the same inventive concept, this embodiment provides a computer-readable storage medium storing a computer program. When executed by processor 51, the program implements the aforementioned message notification method based on big data. The method includes: acquiring a user's historical chat history and multiple video files to be received in the chat history; generating multiple user interest images using a variational autoencoder based on the user's historical chat history; displaying the multiple user interest images on the user's software interface and acquiring multiple target interest images selected by the user; determining multiple video files that need to be notified via message based on the multiple video files to be received in the chat history and the multiple target interest images selected by the user; receiving and storing the video files that need to be notified via message and providing a message notification to the user.

[0083] The big data-based message notification method provided in this application can be applied to terminal devices (such as mobile phones), tablets, laptops, ultra-mobile personal computers (UMPCs), handheld computers, netbooks, personal digital assistants (PDAs), wearable devices (such as smartwatches, smart glasses, or smart helmets), augmented reality (AR) / virtual reality (VR) devices, smart home devices, in-vehicle computers, and other electronic devices. This application does not impose any limitations on this.

[0084] Taking mobile phone 100 as an example of the aforementioned electronic devices, Figure 6 A structural schematic diagram of mobile phone 100 is shown.

[0085] like Figure 6 As shown, the mobile phone 100 may include a processing module 110, an external memory interface 120, an internal memory 121, a universal serial bus (USB) interface 130, a charging management module 140, a power management module 141, a battery 142, an antenna 1, an antenna 2, a mobile communication module 150, a wireless communication module 160, an audio module 170, a speaker 170A, a receiver 170B, a microphone 170C, a headphone jack 170D, a sensor module 180, buttons 190, a motor 191, an indicator 192, a camera 193, a display screen 194, and a subscriber identification module (SIM) card interface 195, etc.

[0086] The processing module 110 may include one or more processing units, such as an application processor (AP), a modem processor, a graphics processing unit (GPU), an image signal processor (ISP), a controller, a memory, a video codec, a digital signal processor (DSP), a baseband processor, and / or a neural network processing unit (NPU). These different processing units may be independent devices or integrated into one or more processors.

[0087] The processing module 110 can be used to: acquire the user's historical chat history and multiple video files to be received in the chat history; generate multiple user interest images using a variational autoencoder based on the user's historical chat history; display the multiple user interest images on the user's software interface and acquire multiple target interest images selected by the user; determine multiple video files that need to be notified by messages based on the multiple video files to be received in the chat history and the multiple target interest images selected by the user; receive and store the video files that need to be notified by messages and notify the user by messages.

[0088] The basic concepts have been described above. Obviously, for those skilled in the art, the detailed disclosure above is merely illustrative and does not constitute a limitation of this specification. Although not explicitly stated herein, those skilled in the art may make various modifications, improvements, and corrections to this specification. Such modifications, improvements, and corrections are suggested in this specification and therefore remain within the spirit and scope of the exemplary embodiments described herein.

[0089] Furthermore, this specification uses specific terms to describe embodiments thereof. For example, "an embodiment," "one embodiment," and / or "some embodiments" refer to a particular feature, structure, or characteristic associated with at least one embodiment of this specification. Therefore, it should be emphasized and noted that references to "an embodiment," "one embodiment," or "an alternative embodiment" in different locations throughout this specification do not necessarily refer to the same embodiment. Moreover, certain features, structures, or characteristics in one or more embodiments of this specification can be appropriately combined.

[0090] Furthermore, unless expressly stated in the claims, the order of processing elements and sequences, the use of numbers and letters, or other names described in this specification are not intended to limit the order of the processes and methods described herein. Although various examples have been discussed in the foregoing disclosure of some embodiments of the invention that are currently considered useful, it should be understood that such details are for illustrative purposes only, and the appended claims are not limited to the disclosed embodiments; rather, the claims are intended to cover all modifications and equivalent combinations that conform to the spirit and scope of the embodiments described herein. For example, while the system components described above can be implemented using hardware devices, they can also be implemented solely using software solutions, such as installing the described system on existing servers or mobile devices.

[0091] Similarly, it should be noted that, in order to simplify the description disclosed herein and thus aid in the understanding of one or more embodiments of the invention, the foregoing description of embodiments in this specification may sometimes combine multiple features into a single embodiment, drawing, or description thereof. However, this method of disclosure does not imply that the subject matter of this specification requires more features than those mentioned in the claims. In fact, the embodiments contain fewer features than all the features of a single embodiment disclosed above.

[0092] Finally, it should be understood that the embodiments described in this specification are merely illustrative of the principles of the embodiments described herein. Other variations may also fall within the scope of this specification. Therefore, alternative configurations of the embodiments described herein are intended to be illustrative rather than limiting, and should be considered consistent with the teachings of this specification. Accordingly, the embodiments described herein are not limited to those explicitly introduced and described herein.

Claims

1. A big data-based message alerting method, characterized by, The method comprises the following steps: obtaining a user historical chat record and a plurality of to-be-received video files in the chat record; generating a plurality of user interest pictures based on the user historical chat record using a variational autoencoder; displaying the plurality of user interest pictures on a user software interface and obtaining a plurality of target interest pictures selected by the user; determining a plurality of video files that need to be prompted based on the plurality of to-be-received video files in the chat record and the plurality of target interest pictures selected by the user, which comprises: processing the plurality of to-be-received video files in the chat record and the plurality of target interest pictures selected by the user based on an interest degree determination model to determine a plurality of key frames of each to-be-received video file and a user interest degree of each to-be-received video file, wherein the interest degree determination model is a recurrent neural network model, the input of the interest degree determination model is the plurality of to-be-received video files in the chat record and the plurality of target interest pictures selected by the user, the output of the interest degree determination model is the plurality of key frames of each to-be-received video file and the user interest degree of each to-be-received video file, and the interest degree determination model comprises a key frame determination layer, a key frame correlation degree determination layer and a user interest degree determination layer, all of which comprise a recurrent neural network, the input of the key frame determination layer is the plurality of to-be-received video files in the chat record, the output of the key frame determination layer is the plurality of key frames of each to-be-received video file, the input of the key frame correlation degree determination layer is the plurality of key frames of each to-be-received video file and the plurality of target interest pictures selected by the user, the output of the key frame correlation degree determination layer is the correlation degree between each key frame and each target interest picture selected by the user, the input of the user interest degree determination layer is the correlation degree between each key frame and each target interest picture selected by the user, and the output of the user interest degree determination layer is the user interest degree of each to-be-received video file; constructing a message prompt graph structure, which comprises a plurality of nodes and a plurality of edges between the nodes, the plurality of nodes comprising a user node and a plurality of video file nodes, each video file node establishing an edge with the user node, the node features of the user node comprising the plurality of target interest pictures selected by the user, the node features of each video file node comprising the plurality of key frames of each to-be-received video file, and the features of each video file node establishing an edge with the user node comprising the user interest degree of each to-be-received video file; processing the message prompt graph structure based on a graph convolution network to determine the plurality of video files that need to be prompted; receiving and storing the video files that need to be prompted and prompting the user.

2. The big data based message alerting method as claimed in claim 1, wherein, The input of the variational autoencoder is the user historical chat record, and the output of the variational autoencoder is a plurality of user interest pictures.

3. A big data based message alerting system characterized in that, The method comprises the following steps: The acquisition module is configured to acquire a user historical chat record and a plurality of to-be-received video files in the chat record; The generation module is configured to generate a plurality of user interest pictures based on the user historical chat record using a variational autoencoder; The display module is configured to display the plurality of user interest pictures on a user software interface and acquire a plurality of target interest pictures selected by the user; The determination module is configured to determine a plurality of video files that need to be prompted based on the plurality of to-be-received video files in the chat record and the plurality of target interest pictures selected by the user, and the determination module is further configured to: determine a plurality of key frames of each to-be-received video file and a user interest degree of each to-be-received video file based on the plurality of to-be-received video files in the chat record and the plurality of target interest pictures selected by the user using an interest degree determination model, the interest degree determination model being a recurrent neural network model, an input of the interest degree determination model being the plurality of to-be-received video files in the chat record and the plurality of target interest pictures selected by the user, and an output of the interest degree determination model being the plurality of key frames of each to-be-received video file and the user interest degree of each to-be-received video file, the interest degree determination model including a key frame determination layer, a key frame correlation degree determination layer, and a user interest degree determination layer, the key frame determination layer, the key frame correlation degree determination layer, and the user interest degree determination layer all including a recurrent neural network, an input of the key frame determination layer being the plurality of to-be-received video files in the chat record, an output of the key frame determination layer being the plurality of key frames of each to-be-received video file, an input of the key frame correlation degree determination layer being the plurality of key frames of each to-be-received video file and the plurality of target interest pictures selected by the user, an output of the key frame correlation degree determination layer being a correlation degree between each key frame and each target interest picture selected by the user, an input of the user interest degree determination layer being the correlation degree between each key frame and each target interest picture selected by the user, and an output of the user interest degree determination layer being the user interest degree of each to-be-received video file; construct a message prompt graph structure, the message prompt graph structure including a plurality of nodes and a plurality of edges between the plurality of nodes, the plurality of nodes including a user node and a plurality of video file nodes, each video file node establishing an edge with the user node, a node feature of the user node including the plurality of target interest pictures selected by the user, a node feature of each video file node including the plurality of key frames of each to-be-received video file, and a feature of each video file node establishing an edge with the user node including the user interest degree of each to-be-received video file; determine the plurality of video files that need to be prompted based on the message prompt graph structure using a graph convolution network; The prompting module is configured to receive and store the plurality of video files that need to be prompted and prompt the user.

4. The big data based message alerting system as claimed in claim 3 wherein, An input of the variational autoencoder is the user historical chat record, and an output of the variational autoencoder is the plurality of user interest pictures.

5. An electronic device, comprising: The processor is configured to: acquire a user historical chat record and a plurality of to-be-received video files in the chat record; generate a plurality of user interest pictures based on the user historical chat record using a variational autoencoder; display the plurality of user interest pictures on a user software interface and acquire a plurality of target interest pictures selected by the user; determine a plurality of video files that need to be prompted based on the plurality of to-be-received video files in the chat record and the plurality of target interest pictures selected by the user; determine a plurality of key frames of each to-be-received video file and a user interest degree of each to-be-received video file based on the plurality of to-be-received video files in the chat record and the plurality of target interest pictures selected by the user using an interest degree determination model, the interest degree determination model being a recurrent neural network model, an input of the interest degree determination model being the plurality of to-be-received video files in the chat record and the plurality of target interest pictures selected by the user, and an output of the interest degree determination model being the plurality of key frames of each to-be-received video file and the user interest degree of each to-be-received video file, the interest degree determination model including a key frame determination layer, a key frame correlation degree determination layer, and a user interest degree determination layer, the key frame determination layer, the key frame correlation degree determination layer, and the user interest degree determination layer all including a recurrent neural network, an input of the key frame determination layer being the plurality of to-be-received video files in the chat record, an output of the key frame determination layer being the plurality of key frames of each to-be-received video file, an input of the key frame correlation degree determination layer being the plurality of key frames of each to-be-received video file and the plurality of target interest pictures selected by the user, an output of the key frame correlation degree determination layer being a correlation degree between each key frame and each target interest picture selected by the user, an input of the user interest degree determination layer being the correlation degree between each key frame and each target interest picture selected by the user, and an output of the user interest degree determination layer being the user interest degree of each to-be-received video file; construct a message prompt graph structure, the message prompt graph structure including a plurality of nodes and a plurality of edges between the plurality of nodes, the plurality of nodes including a user node and a plurality of video file nodes, each video file node establishing an edge with the user node, a node feature of the user node including the plurality of target interest pictures selected by the user, a node feature of each video file node including the plurality of key frames of each to-be-received video file, and a feature of each video file node establishing an edge with the user node including the user interest degree of each to-be-received video file; determine the plurality of video files that need to be prompted based on the message prompt graph structure using a graph convolution network; receive and store the plurality of video files that need to be prompted and prompt the user. and a computer program; wherein the computer program is stored in the memory and configured to be executed by the processor to implement the big data-based message reminding method according to any one of claims 1 to 2.

6. A computer-readable storage medium having stored thereon a computer program, characterized in that, The program is executed by the processor to implement the big data-based message reminding method according to any one of claims 1 to 2.

Images