A 5G message-based information interaction method and system

Through the built-in sensor data analysis and deep learning algorithm of the device, situational awareness and personalized interaction in 5G message services are realized, solving the problems of single-information and low sorting efficiency of information display, and improving the reliability of user experience and information transmission.

CN118714523BActive Publication Date: 2024-11-15FUJIAN BOSHITONG INFORMATION CO LTD
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202411180675.9
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Filing Date
2024-08-27
Publication Date
2024-11-15
Estimated Expiration
2044-08-27

AI Technical Summary

Technical Problem

5G message services are single in information display and interaction, which is difficult to adapt to user diverse situations and network fluctuations, resulting in reduced experience quality, low personalized sorting efficiency, and lack of precise matching driven by deep learning.

Method used

Through the device's built-in sensor data analysis, the user situation identification model is used to generate situation tags, automatically adjust the message display form and interaction method, perform multi-modal orchestration according to the user's equipment capabilities and network conditions, and use deep learning algorithms to review and personalize the ordering. , send optimized message queues through 5G network.

Benefits of technology

It improves the relevance and timeliness of information push, achieves efficient communication, adapts to different equipment capabilities, enhances user experience, ensures smooth information transmission, improves the reliability and efficiency of interaction, reduces information overload, and optimizes user experience.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN118714523B_ABST
    Figure CN118714523B_ABST
Patent Text Reader

Abstract

The present invention discloses a method and system for information interaction based on 5G messages, which relates to the field of mobile communication technology, including: using the sensor data built into the device to analyze the usage mode of the device, and outputting the context label of the user through the user context recognition model; based on the context label, automatically adjusting the message display form and interaction mode; multi-modal arrangement of the message content according to the user device capability and network status; deep learning algorithm review of the arranged multi-modal message content, and personalized sorting of the reviewed and approved messages according to user preferences, and outputting the message queue that has passed the review and completed the sorting; sending the message queue to the receiver through the 5G network, and the receiving end performing context analysis again, and outputting the message display optimized according to the receiver's context, completing the interaction process. The present invention not only significantly improves the intelligence level and user experience of 5G message services, but also opens up a new direction for the development of future information interaction technology.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention relates to the field of mobile communication technologies, and in particular to a 5G message-based information interaction method and system. Background Art

[0002] Against the backdrop of the rapid development of information technology, the commercial deployment of the fifth generation mobile communication technology (5G), with its ultra-high speed, low latency, and high connection density, has brought revolutionary changes to the field of information interaction. However, how to efficiently and intelligently utilize the advantages of 5G technology to improve the intelligence level of information interaction has become a key issue that needs to be addressed urgently.

[0003] Although 5G messaging services have made great progress in transmission speed and capacity, they still face challenges: the singleness of information display and interaction fails to fully adapt to the diverse situations of users, reducing the quality of experience; content arrangement and adaptation methods are relatively rigid, making it difficult to dynamically respond to differences in user device performance and network fluctuations, affecting the best display effects; information review and personalized sorting technologies are lagging behind, automated review is inefficient, and it is difficult to process large-scale data; personalized sorting simply relies on user preference models and lacks precise matching driven by deep learning. Summary of the invention

[0004] In view of the above existing problems, the present invention is proposed.

[0005] Therefore, the present invention provides an information interaction method, device, computer equipment and storage medium based on 5G messages to solve the problems of insufficient adaptability of information interaction scenarios and slow response to personalized needs in the 5G environment.

[0006] In order to solve the above technical problems, the present invention provides the following technical solutions:

[0007] In a first aspect, an embodiment of the present invention provides an information interaction method based on 5G messages, which includes analyzing the usage mode of the device using sensor data built into the device, and outputting a context label of the user through a user context recognition model;

[0008] Automatically adjust message display and interaction methods based on context tags;

[0009] Perform multi-modal arrangement of message content based on user device capabilities and network conditions;

[0010] The deep learning algorithm is used to review the content of the arranged multimodal messages, and the approved messages are sorted according to user preferences, and a queue of approved and sorted messages is output;

[0011] The message queue is sent to the receiver via the 5G network. The receiver performs context analysis again and outputs the message display optimized according to the receiver's context, completing the interaction process.

[0012] As a preferred solution of the information interaction method based on 5G messages described in the present invention, the usage mode of the user device includes the screen on and off time, application usage frequency and charging status.

[0013] As a preferred solution of the information interaction method based on 5G messages of the present invention, the specific operation steps of outputting the user's context label through the user context recognition model are as follows:

[0014] Clean the collected raw data, remove outliers, perform time series smoothing to reduce noise interference, and convert various types of data into a unified format;

[0015] Integrate the preprocessed data into feature vectors;

[0016] Load the user context recognition model that has been trained with a large number of labeled data sets, input the feature vector into the model, and output the predicted probability of each context category;

[0017] According to the probability distribution of the model output, the context label with the highest probability is selected as the user's current context.

[0018] As a preferred solution of the information interaction method based on 5G messages of the present invention, the message display form and interaction mode are automatically adjusted based on the context label. The specific operation steps are as follows:

[0019] Build a situational response strategy library, including user situations such as driving, meeting, and home;

[0020] The policy library contains specific rule sets for the user scenario of driving, including:

[0021] Priority adjustment: Urgent messages have higher priority than non-urgent messages;

[0022] Automatic screen brightness adjustment: reduce screen brightness when driving at night to prevent glare from affecting driving safety;

[0023] Simplified interface: The interface layout is simplified, complex operations are removed, and key function buttons for answering and rejecting calls and voice message control are retained;

[0024] According to the highest probability situation tag system, the corresponding strategy in the situation response strategy library is automatically triggered;

[0025] Immediately switch the user interface to minimalist mode, collapse non-essential text input areas, and enlarge key buttons;

[0026] The system automatically turns on the text-to-speech function, sets the default message receiving mode to voice playback, and configures the voice recognition module to stand by, ready to receive user voice commands;

[0027] Based on a preset set of specific rules, the system automatically screens and distinguishes received messages into urgent and non-urgent ones. Non-urgent text messages are automatically converted into voice broadcasts, while urgent messages are alerted to the user through special prompt sounds and a brief text summary preview is given;

[0028] Based on the user's real-time geographic location, the system uses geo-fencing technology to circle a certain area around the user and use this as the boundary to filter relevant information;

[0029] Extract the merchant discounts, event information, and traffic condition updates related to the geographic location from the database, and use the collaborative filtering algorithm to combine the user's historical preferences and behavior patterns to score the relevance of each content;

[0030] Sort the content according to the ratings, and select the most relevant information to form a personalized recommendation list;

[0031] Text-to-speech conversion of titles and summaries of recommended content;

[0032] The system monitors user voice commands and performs corresponding operations based on the commands.

[0033] As a preferred solution of the information interaction method based on 5G messages of the present invention, the message content is multi-modally arranged according to the user equipment capability and network status. The specific operation steps are as follows:

[0034] Query user device information through API to monitor current network connection speed and stability;

[0035] When the device supports AR and the network condition is good, add 3D models to provide high-definition image and video experience;

[0036] When device resources are limited, reduce the use of images and videos;

[0037] Use different resolutions and compression standards according to the network bandwidth level;

[0038] According to the importance of information and the immediate needs of users, text, voice, image, video and AR content are arranged in layers. Key information is displayed in the form of text and voice first, and enhanced content is used as auxiliary information to provide an immersive experience.

[0039] Design intuitive interactive entrances for each modal content to enable users to operate conveniently in different situations;

[0040] During the playback of the message content, the background continuously monitors the CPU, GPU load and network traffic;

[0041] When performance bottlenecks and network fluctuations are detected, content quality is automatically reduced until environmental conditions improve and the original settings are restored.

[0042] As a preferred solution of the information interaction method based on 5G messages described in the present invention, the deep learning algorithm is used to review the arranged multimodal message content, and the approved messages are sorted according to the user's preferences, and the message queue that has passed the review and completed the sorting is output. The specific operation steps are as follows:

[0043] Build a deep learning audit model to split the multimodal message content into text segments, image frames, and video segments, and feed them into the audit model for analysis;

[0044] Identify sensitive information in text, images, videos, and AR content, and set sensitivity thresholds for each type of content. Content below the threshold is directly approved, while content above the threshold is subject to manual review and direct rejection.

[0045] For content that contains sensitive information but does not constitute a violation, the system automatically adds a sensitive information warning label and prompts the user to confirm whether to continue sending;

[0046] Based on the user's historical browsing, liking, and sharing behavior data, collaborative filtering is used to establish a user preference model;

[0047] A multi-objective optimization algorithm is used to generate a personalized ranking list;

[0048] According to the personalized sorting results, the message contents with audit status and sensitive information labels are organized into message queues. Through the RESTful API interface, the message queues are connected to the message sending service and prepared for distribution to the target user group.

[0049] As a preferred solution of the information interaction method based on 5G messages of the present invention, wherein: a message queue is sent to a receiver via a 5G network, the receiver performs context analysis again, and the output is a message display optimized according to the receiver's context, completing the interaction process. The specific operation steps are as follows:

[0050] Dynamically select the optimal path based on the receiver's location and network conditions;

[0051] Push the message queue to the recipient’s 5G messaging application through the MQTT protocol;

[0052] The receiving application monitors the user's device status in real time and analyzes the user's preferred message receiving method in a specific situation based on historical usage habits;

[0053] Using logistic regression model, we can quickly determine the most appropriate display mode based on the collected information and convert multimedia content;

[0054] Adaptively adjust the message display interface layout for different devices and scenarios.

[0055] In a second aspect, the present invention provides an information interaction system based on 5G messages, comprising:

[0056] Context awareness and data collection module: Analyzes the usage pattern of the device using the sensor data built into the device, and outputs the user's context label through the user context recognition model;

[0057] Context-adaptive interaction design module: automatically adjusts message display and interaction methods based on context tags;

[0058] Multimodal content optimization and arrangement module: multimodal arrangement of message content based on user device capabilities and network conditions;

[0059] Intelligent content review and personalized sorting module: The deep learning algorithm is used to review the content of the arranged multi-modal messages, and the approved messages are personalized and sorted according to user preferences, and the queue of approved and sorted messages is output;

[0060] 5G network transmission and receiving end adaptation module: The message queue is sent to the receiver through the 5G network. The receiving end performs situational analysis again and outputs the message display optimized according to the receiver's situation, completing the interaction process.

[0061] In a third aspect, an embodiment of the present invention provides a computer device, comprising a memory and a processor, wherein the memory stores a computer program, wherein: when the computer program is executed by the processor, any step of the information interaction method based on 5G messages as described in the first aspect of the present invention is implemented.

[0062] In a fourth aspect, an embodiment of the present invention provides a computer-readable storage medium on which a computer program is stored, wherein: when the computer program is executed by a processor, any step of the information interaction method based on 5G messages as described in the first aspect of the present invention is implemented.

[0063] The beneficial effects of the present invention are: using the built-in sensor data of the device to analyze the usage pattern of the device, and through the context recognition model, to achieve accurate context label generation, improve the relevance and timeliness of information push, lay the foundation for personalized interaction, solve the problems of weak context perception and low matching degree of traditional push, and enhance user experience; according to the context label, automatically optimize the interface and interaction method to achieve efficient communication; according to the user device and network conditions, dynamically arrange multimodal message content to ensure smooth information transmission, adapt to different device capabilities, and improve the reliability and efficiency of interaction; use deep learning to review content and sort it according to user preferences, which not only strengthens information security, but also realizes content personalization, reduces information overload, optimizes user experience, enhances user stickiness, and promotes effective information communication. BRIEF DESCRIPTION OF THE DRAWINGS

[0064] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the accompanying drawings required for use in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other accompanying drawings can be obtained based on these accompanying drawings without paying creative work.

[0065] Figure 1 This is a flowchart of the information interaction method based on 5G messages in Example 1.

[0066] Figure 2 This is the sensitivity threshold determination diagram in Example 1. DETAILED DESCRIPTION

[0067] In order to make the above-mentioned objects, features and advantages of the present invention more obvious and easy to understand, the specific implementation methods of the present invention are described in detail below in conjunction with the accompanying drawings.

[0068] In the following description, many specific details are set forth to facilitate a full understanding of the present invention, but the present invention may also be implemented in other ways different from those described herein, and those skilled in the art may make similar generalizations without violating the connotation of the present invention. Therefore, the present invention is not limited to the specific embodiments disclosed below.

[0069] Secondly, the term "one embodiment" or "embodiment" as used herein refers to a specific feature, structure, or characteristic that may be included in at least one implementation of the present invention. The term "in one embodiment" that appears in different places in this specification does not necessarily refer to the same embodiment, nor does it refer to a separate or selective embodiment that is mutually exclusive with other embodiments.

[0070] Example 1, reference Figure 1 and Figure 2 , which is the first embodiment of the present invention, and provides an information interaction method based on 5G messages, comprising the following steps:

[0071] S1. Use the built-in sensor data of the device to analyze the usage pattern of the user's device, including the screen on and off time, application usage frequency and charging status.

[0072] Furthermore, the GPS module is used to obtain precise longitude and latitude coordinates, and base station positioning technology (such as A-GPS) is combined to enhance the position accuracy indoors or in poor signal environments. The frequency is set to update once a minute to balance accuracy and energy consumption;

[0073] Use sensors such as accelerometers and gyroscopes to determine the motion state of the device (such as stillness, walking, running, driving), and use motion pattern recognition algorithms to improve recognition accuracy;

[0074] Analyze the list of recently used apps, identify high-frequency app categories (such as social, news, and health), and the distribution of active time within the app to reflect the user's immediate interests.

[0075] S2. Output the user's context label through the user context recognition model. The specific operation steps are as follows:

[0076] Clean the collected raw data, remove outliers (such as obviously wrong geographic location information), and perform time series smoothing to reduce noise interference;

[0077] Convert various data into a unified format, such as converting geographic locations into standardized geocodes, unifying timestamps into UTC standard time, and encoding movement status into predefined categories;

[0078] Integrate the preprocessed data into feature vectors, where each feature corresponds to a data type, such as geographic location features, time features, etc.

[0079] Load the user context recognition model that has been trained with a large number of labeled data sets (including geographic location, time, motion status, application usage and corresponding context labels);

[0080] The feature vector is input into the model. The model adopts a multi-layer neural network structure, including an embedding layer, a hidden layer (using the ReLU activation function), and an output layer (the Softmax function is used for probability assignment), and outputs the predicted probability of each situation category;

[0081] According to the probability distribution of the model output, the context label with the highest probability is selected as the user's current context.

[0082] S3. Automatically adjust the message display format and interaction method based on the context label. The specific steps are as follows:

[0083] Build a situational response strategy library, including user situations such as driving, meeting, and home;

[0084] The policy library contains specific rule sets for the user scenario of driving, including:

[0085] Priority adjustment: Urgent messages have higher priority than non-urgent messages;

[0086] Automatic screen brightness adjustment: reduce screen brightness when driving at night to prevent glare from affecting driving safety;

[0087] Simplified interface: The interface layout is simplified, complex operations are removed, and key function buttons for answering and rejecting calls and voice message control are retained;

[0088] According to the highest probability situation tag system, the corresponding strategy in the situation response strategy library is automatically triggered;

[0089] Immediately switch the user interface to minimalist mode, collapse non-essential text input areas, and enlarge key buttons (such as answer / reject call, voice message control) to reduce eye movement time;

[0090] The system automatically turns on the text-to-speech function, sets the default message receiving mode to voice playback, and configures the voice recognition module to stand by, ready to receive user voice commands;

[0091] Based on a preset set of specific rules, the system automatically screens and distinguishes received messages into urgent and non-urgent ones. Non-urgent text messages are automatically converted into voice broadcasts, while urgent messages are alerted to the user through special prompt sounds and a brief text summary preview is given;

[0092] Based on the user's real-time geographic location, the system uses geo-fencing technology to circle a certain area around the user and use this as the boundary to filter relevant information;

[0093] Extract the merchant discounts, event information, and traffic condition updates related to the geographic location from the database, and use the collaborative filtering algorithm to combine the user's historical preferences and behavior patterns to score the relevance of each content;

[0094] Sort the content by ratings and select the most relevant information to form a personalized recommendation list, which includes short titles, audio summaries and quick navigation options;

[0095] Text-to-speech conversion of recommended content titles and summaries ensures that users can obtain important information through hearing while driving, while keeping their hands and eyes focused on the driving task;

[0096] The system listens to user voice commands, such as "Ignore", "Remind me later" or "More details", and performs corresponding operations based on the commands, such as skipping the current message, setting a reminder or voice broadcasting of detailed information.

[0097] S4. Perform multi-modal arrangement of message content according to user device capabilities and network conditions. The specific operation steps are as follows:

[0098] Query user device information through API, including but not limited to screen size, resolution, processor model, GPU capability, and whether it supports AR and other functions, monitor the current network connection speed and stability, and use HTTP Speed+Mobility or similar algorithms to evaluate real-time bandwidth and classify it into three levels: low, medium, and high;

[0099] When the device supports AR and the network condition is good, 3D models are added. For high-resolution large-screen devices, high-definition image and video experience is provided while ensuring content quality.

[0100] When device resources are limited, focus on optimizing text and voice content, reduce the use of images and videos, and ensure smooth basic functions;

[0101] Different resolutions and compression standards are used according to the network bandwidth level. For example, when the network condition is poor, it automatically switches to H.264 encoding, reduces the resolution to 480p, and increases the compression ratio; when the network condition is good, it uses HEVC encoding to maintain 1080p or higher resolution, ensuring image quality while reducing latency;

[0102] According to the importance of information and the immediate needs of users, text, voice, image, video and AR content are arranged in layers. Key information is presented in the form of text and voice first, and enhanced content (such as video and AR) is used as auxiliary information to provide an immersive experience.

[0103] Design intuitive interactive entrances for each modal content, such as touch, voice commands, or gesture control, so that users can operate conveniently in different situations;

[0104] During the playback of the message content, the background continuously monitors the CPU, GPU load and network traffic to ensure a smooth user experience without lags;

[0105] When performance bottlenecks and network fluctuations are detected, the content quality is automatically reduced, such as lowering video resolution and pausing AR rendering, until the original settings are restored after environmental conditions improve.

[0106] S5. Perform a deep learning algorithm review on the arranged multimodal message content, sort the approved messages according to user preferences, and output the approved and sorted message queue. The specific operation steps are as follows:

[0107] Build a deep learning audit model to split the multimodal message content into text segments, image frames, and video segments, and feed them into the audit model for analysis;

[0108] Identify sensitive information in text, images, videos, and AR content, and set sensitivity thresholds for each type of content. Content below the threshold is directly approved, while content above the threshold is subject to manual review and direct rejection.

[0109] For content that contains sensitive information but does not constitute a violation, such as screenshots of personal medical records uploaded by users, the system automatically adds a sensitive information warning label and prompts the user to confirm whether to continue sending;

[0110] Based on the user's historical browsing, liking, and sharing behavior data, collaborative filtering is used to establish a user preference model;

[0111] Use multi-objective optimization algorithms (such as weighted linear combination or Pareto frontier exploration) to comprehensively consider factors such as content novelty, popularity, and relevance to user interests to generate a personalized ranking list;

[0112] According to the personalized sorting results, the message contents with audit status and sensitive information labels are organized into message queues. Through the RESTful API interface, the message queues are connected to the message sending service and prepared for distribution to the target user group.

[0113] Use object detection technology to identify sensitive image elements;

[0114] Identify inappropriate content such as improper behavior through spatiotemporal feature analysis;

[0115] Convert AR elements into 3D models or keyframes and apply image recognition technology for review.

[0116] Furthermore, the sensitivity threshold α is specifically divided as follows:

[0117] When 0≤α<0.6, it is a safe area, no sensitive information is detected in the content, no special processing is required, and it passes the review directly;

[0118] When 0.6≤α<0.8, the warning area indicates that the content contains slightly sensitive information or potential risks. For text, images, and videos, more detailed inspection is required. For AR content, the comprehensive impact should be analyzed.

[0119] When 0.8≤α<0.9, in the review area, if the content contains obvious sensitive information, such as text, images, videos, and AR content, manual review will be performed to determine whether it can be published or needs to be modified;

[0120] When 0.9≤α<1.0, the region is rejected, which clearly contains serious violations or highly sensitive information and is directly refused to be released.

[0121] S6. Send the message queue to the receiver through the 5G network. The receiver performs context analysis again and outputs the message display optimized according to the receiver's context to complete the interaction process. The specific operation steps are as follows:

[0122] Dynamically select the optimal path based on the receiver's location and network conditions to reduce packet loss and delay and improve transmission efficiency;

[0123] Push the message queue to the recipient’s 5G messaging application through the MQTT protocol;

[0124] The receiving application monitors the user's device status in real time, including but not limited to geographic location, device type, screen size, network connection status and activity mode (such as walking, driving, and stationary), and analyzes the user's preferred message receiving method (text, voice, image or video) in specific situations based on historical usage habits;

[0125] The logistic regression model is used to quickly determine the most appropriate display mode based on the collected information and perform multimedia content conversion. For example, when it is detected that the user is driving, the model automatically selects voice broadcast as the preferred display form;

[0126] Adaptively adjust the message display interface layout for different devices and scenarios to ensure user-friendliness. For example, prioritize displaying core information summaries on small-screen devices.

[0127] This embodiment also provides an information interaction system based on 5G messages, including a context perception and data collection module: using the sensor data built into the device to analyze the usage mode of the device, and outputting the context label of the user through the user context recognition model;

[0128] Context-adaptive interaction design module: automatically adjusts message display and interaction methods based on context tags;

[0129] Multimodal content optimization and arrangement module: multimodal arrangement of message content based on user device capabilities and network conditions;

[0130] Intelligent content review and personalized sorting module: The deep learning algorithm is used to review the content of the arranged multi-modal messages, and the approved messages are personalized and sorted according to user preferences, and the queue of approved and sorted messages is output;

[0131] 5G network transmission and receiving end adaptation module: The message queue is sent to the receiver through the 5G network. The receiving end performs situational analysis again and outputs the message display optimized according to the receiver's situation, completing the interaction process.

[0132] This embodiment also provides a computer device, which is suitable for the information interaction method based on 5G messages, including: a memory and a processor; the memory is used to store computer executable instructions, and the processor is used to execute computer executable instructions to implement the information interaction method based on 5G messages proposed in the above embodiment.

[0133] The computer device may be a terminal, and the computer device includes a processor, a memory, a communication interface, a display screen and an input device connected through a system bus. Among them, the processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system and a computer program. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The communication interface of the computer device is used to communicate with an external terminal in a wired or wireless manner, and the wireless manner can be achieved through WIFI, an operator network, NFC (near field communication) or other technologies. The display screen of the computer device may be a liquid crystal display screen or an electronic ink display screen, and the input device of the computer device may be a touch layer covered on the display screen, or a key, trackball or touchpad provided on the housing of the computer device, or an external keyboard, touchpad or mouse, etc.

[0134] This embodiment also provides a storage medium on which a computer program is stored. When the program is executed by a processor, the information interaction method based on 5G messages proposed in the above embodiment is implemented; the storage medium can be implemented by any type of volatile or non-volatile storage device or a combination thereof, such as static random access memory (Static Random Access Memory, referred to as SRAM), electrically erasable programmable read-only memory (Electrically Erasable Programmable Read-Only Memory, referred to as EEPROM), erasable programmable read-only memory (Erasable Programmable Read Only Memory, referred to as EPROM), programmable read-only memory (Programmable Red-Only Memory, referred to as PROM), read-only memory (Read-Only Memory, referred to as ROM), magnetic storage, flash memory, disk or optical disk.

[0135] In summary, the present invention utilizes the built-in sensor data of the device to analyze the usage pattern of the device, and utilizes the context recognition model to realize the generation of accurate context labels, thereby improving the relevance and timeliness of information push, laying the foundation for personalized interaction, solving the problems of weak context perception and low matching degree of traditional push, and enhancing user experience; according to the context labels, the interface and interaction mode are automatically optimized to achieve efficient communication; according to the user device and network conditions, the multimodal message content is dynamically arranged to ensure smooth information transmission, adapt to the capabilities of different devices, and improve the reliability and efficiency of interaction; deep learning is used to review the content and sort it according to user preferences, which not only strengthens information security, but also realizes content personalization, reduces information overload, optimizes user experience, enhances user stickiness, and promotes effective information communication.

[0136] Example 2

[0137] Referring to Table 1, which is the second embodiment of the present invention, in order to further verify the advancement of the present invention, experimental simulation data of the information interaction method based on 5G messages are provided.

[0138] First, the experiment selected representative user devices, including high-performance smartphones, mid-range tablets, and basic feature phones, to evaluate the interaction performance under different device capabilities.

[0139] In addition, the experiment also considered different geographical locations such as city centers, suburbs, and mobile transportation, as well as network conditions during peak and non-peak hours to fully examine the adaptability of the system.

[0140] A training set containing 10,000 labeled data was constructed, covering user behavior patterns in various situations, for training the user situation recognition model. The model uses an LSTM neural network structure and achieves an accuracy rate of 92% through cross-validation. The details are shown in the following table:

[0141] Table 1. 5G information interaction method performance evaluation experiment record

[0142]

[0143] Through the comparative analysis of the above experimental data, it can be seen that the embodiment fully demonstrates the innovation and advantages of the information interaction method based on 5G messages. First, the accuracy of the context label is high. For example, in the test object 001, through the comprehensive analysis of the real-time geographic location (city center) and application activity records (shopping applications), the system accurately identified the "mobile shopping" context, and then automatically adjusted it to the display form of voice + short text summary, which greatly improved the convenience of users when walking. The user satisfaction score reached 4.7, proving the effectiveness of personalized adjustment.

[0144] Secondly, the adaptability to different devices and network environments has been verified. For example, for test subject 003, under the conditions of basic feature phones and mobile networks, the system intelligently selected pure voice broadcasting, reducing the demand for device resources. Although the personalized sorting score was relatively low (6.9), the user satisfaction score showed that this method could still meet the basic needs in specific situations (3.9 points).

[0145] Furthermore, the personalized sorting algorithm significantly improved the user experience. For subject 004, under a high-performance mobile phone and a stable home network environment, combined with the user's historical preferences, the system provided an immersive experience of high-definition video + AR elements for the work-related "home office" scenario. The personalized sorting score was as high as 9.2, and the user satisfaction score also reached the highest 4.9 points, proving the successful application of deep learning auditing and personalized sorting algorithms.

[0146] To sum up, this embodiment not only verifies the feasibility of the invention in practical applications, but also shows its significant advantages in adapting to user device capabilities, optimizing content display, and improving user interaction efficiency. Compared with the prior art, it significantly enhances the intelligence and personalization level of 5G message services and improves user experience, thereby strongly proving the creativity and novelty of the invention.

[0147] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention rather than to limit it. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical solutions of the present invention may be modified or replaced by equivalents without departing from the spirit and scope of the technical solutions of the present invention, which should all be included in the scope of the claims of the present invention.

Claims

1. A 5G message-based information interaction method, characterized in that: include, Analyze the usage pattern of the device using the sensor data built into the device, and output the user's context label through the user context recognition model; Automatically adjust message display and interaction methods based on context tags; Perform multi-modal arrangement of message content based on user device capabilities and network conditions; The deep learning algorithm is used to review the content of the arranged multimodal messages, and the approved messages are sorted according to user preferences, and a queue of approved and sorted messages is output; The message queue is sent to the receiver via the 5G network. The receiver performs context analysis again and outputs the message display optimized according to the receiver's context, completing the interaction process. According to the user's device capabilities and network conditions, the message content is arranged in multiple modes. The specific steps are as follows: Query user device information through API to monitor current network connection speed and stability; When the device supports AR and the network condition is good, add 3D models to provide high-definition image and video experience; When device resources are limited, reduce the use of images and videos; Use different resolutions and compression standards according to the network bandwidth level; According to the importance of information and the immediate needs of users, text, voice, image, video and AR content are arranged in layers. Key information is displayed in the form of text and voice first, and enhanced content is used as auxiliary information to provide an immersive experience. Design intuitive interactive entrances for each modal content to enable users to operate conveniently in different situations; During the playback of the message content, the background continuously monitors the CPU, GPU load and network traffic; When performance bottlenecks and network fluctuations are detected, the content quality is automatically reduced until the environmental conditions improve and the original settings are restored; The deep learning algorithm is used to review the content of the arranged multimodal messages, and the approved messages are sorted according to user preferences. The queue of messages that have passed the review and have been sorted is output. The specific steps are as follows: Build a deep learning audit model to split the multimodal message content into text segments, image frames, and video segments, and feed them into the audit model for analysis; Identify sensitive information in text, images, videos, and AR content, and set sensitivity thresholds for each type of content. Content below the threshold is directly approved, while content above the threshold is subject to manual review and direct rejection. For content that contains sensitive information but does not constitute a violation, the system automatically adds a sensitive information warning label and prompts the user to confirm whether to continue sending; Based on the user's historical browsing, liking, and sharing behavior data, collaborative filtering is used to establish a user preference model; A multi-objective optimization algorithm is used to generate a personalized ranking list; According to the personalized sorting results, the message contents with audit status and sensitive information labels are organized into message queues. Through the RESTful API interface, the message queues are connected to the message sending service and prepared for distribution to the target user group.

2. The information interaction method based on 5G messages according to claim 1, characterized in that: The usage pattern of the user device includes screen on / off time, application usage frequency and charging status.

3. The information interaction method based on 5G messages according to claim 1, characterized in that: The specific operation steps of outputting the user's context label through the user context recognition model are as follows: Clean the collected raw data, remove outliers, perform time series smoothing to reduce noise interference, and convert various types of data into a unified format; Integrate the preprocessed data into feature vectors; Load the user context recognition model that has been trained with a large number of labeled data sets, input the feature vector into the model, and output the predicted probability of each context category; According to the probability distribution of the model output, the context label with the highest probability is selected as the user's current context.

4. The information interaction method based on 5G messages according to claim 3, characterized in that: Automatically adjust the message display format and interaction method based on the context label. The specific steps are as follows: Build a situational response strategy library, including user situations such as driving, meeting, and home; The policy library contains specific rule sets for the user scenario of driving, including: Priority adjustment: Urgent messages have higher priority than non-urgent messages; Automatic screen brightness adjustment: reduce screen brightness when driving at night to prevent glare from affecting driving safety; Simplified interface: The interface layout is simplified, complex operations are removed, and key function buttons for answering and rejecting calls and voice message control are retained; According to the highest probability situation tag system, the corresponding strategy in the situation response strategy library is automatically triggered; Immediately switch the user interface to minimalist mode, collapse non-essential text input areas, and enlarge key buttons; The system automatically turns on the text-to-speech function, sets the default message receiving mode to voice playback, and configures the voice recognition module to stand by, ready to receive user voice commands; Based on a preset set of specific rules, the system automatically screens and distinguishes received messages into urgent and non-urgent ones. Non-urgent text messages are automatically converted into voice broadcasts, while urgent messages are alerted to the user through special prompt sounds and a brief text summary preview is given; Based on the user's real-time geographic location, the system uses geo-fencing technology to circle a certain area around the user and use this as the boundary to filter relevant information; Extract the merchant discounts, event information, and traffic condition updates related to the geographic location from the database, and use the collaborative filtering algorithm to combine the user's historical preferences and behavior patterns to score the relevance of each content; Sort the content according to the ratings, and select the most relevant information to form a personalized recommendation list; Text-to-speech conversion of titles and summaries of recommended content; The system monitors user voice commands and performs corresponding operations based on the commands.

5. The information interaction method based on 5G messages according to claim 4, characterized in that: The message queue is sent to the receiver via the 5G network. The receiver performs context analysis again and outputs the message display optimized according to the receiver's context to complete the interaction process. The specific steps are as follows: Dynamically select the optimal path based on the receiver's location and network conditions; Push the message queue to the recipient’s 5G messaging application through the MQTT protocol; The receiving application monitors the user's device status in real time and analyzes the user's preferred message receiving method in a specific situation based on historical usage habits; Using logistic regression model, we can quickly determine the most appropriate display mode based on the collected information and convert multimedia content; Adaptively adjust the message display interface layout for different devices and scenarios.

6. A 5G message-based information interaction system, based on the 5G message-based information interaction method according to any one of claims 1 to 5, characterized in that: include, Context awareness and data collection module: Analyzes the usage pattern of the device using the sensor data built into the device, and outputs the user's context label through the user context recognition model; Context-adaptive interaction design module: automatically adjusts message display and interaction methods based on context tags; Multimodal content optimization and arrangement module: multimodal arrangement of message content based on user device capabilities and network conditions; Intelligent content review and personalized sorting module: The deep learning algorithm is used to review the content of the arranged multi-modal messages, and the approved messages are personalized and sorted according to user preferences, and the queue of approved and sorted messages is output; 5G network transmission and receiving end adaptation module: The message queue is sent to the receiver through the 5G network. The receiving end performs situational analysis again and outputs the message display optimized according to the receiver's situation, completing the interaction process.

7. A computer device comprising a memory and a processor, wherein the memory stores a computer program, wherein: When the processor executes the computer program, the steps of the information interaction method based on 5G messages described in any one of claims 1 to 5 are implemented.

8. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the information interaction method based on 5G messages described in any one of claims 1 to 5 are implemented.