A 5G message, SMS and voice integrated information access method and system

Abstract

The present application provides a method and system for fusion information access of 5G messages, SMS and voice, which relates to the field of information processing technology. Among them, the method includes: obtaining the information content to be sent, the user's real-time geographic location information and the initial multimodal user portrait; optimizing the initial multimodal user portrait using the information content to be sent and the user's real-time geographic location information to obtain the target multimodal user portrait; parsing the information content to obtain the parsing result, and based on the target multimodal user portrait and the parsing result, using the reinforcement learning framework to determine the target communication strategy corresponding to the information content; based on the target communication strategy, using a dynamic resource allocation algorithm to determine the target communication resource allocation plan according to the user's real-time geographic location information and network conditions; based on the target communication resource allocation plan, executing the information transmission process, and introducing a differential privacy protection mechanism to encrypt the information transmission process. This application improves the accuracy of information content access.

Description

Technical Field

[0001] The embodiments of the present application relate to the field of information processing technology, and in particular to a method and system for delivering integrated information of 5G messages, SMS and voice. Background Art

[0002] With the rapid development of mobile Internet, 5G messaging, SMS, voice and other communication methods have been widely used in various application scenarios. For example, in intelligent transportation systems, road condition information needs to be transmitted in real time; in medical rescue scenarios, emergency notifications must be quickly conveyed to relevant personnel; in corporate communication, efficient information transmission can improve work efficiency. These application scenarios have put forward higher requirements for information access in terms of accuracy, timeliness and personalization.

[0003] Existing information access methods are usually based on fixed rules or a single communication method, and have problems with poor accuracy and low effectiveness of information access. They are also difficult to quickly adapt to changes in complex network environments, resulting in slow response speeds and waste or insufficient network resources. There is a lack of effective security measures in the information transmission process, which can easily lead to the leakage of sensitive information, especially in cases involving personal privacy or sensitive content. Summary of the invention

[0004] The embodiments of the present application provide a method and system for fusion information access of 5G messages, SMS and voice, so as to solve the problems existing in the prior art such as poor information access accuracy, slow response speed, insufficient resource utilization and poor information transmission security.

[0005] In a first aspect, an embodiment of the present application provides a method for fusion information access of 5G messages, SMS and voice, including:

[0006] Obtaining the content of the information to be sent, the real-time geographic location information of the user, and an initial multimodal user portrait constructed based on the user's behavior pattern and communication preference information; optimizing the initial multimodal user portrait using the content of the information to be sent and the real-time geographic location information of the user to obtain a target multimodal user portrait;

[0007] Parsing the information content to obtain a parsing result, and determining a target communication strategy corresponding to the information content using a reinforcement learning framework based on the target multimodal user portrait and the parsing result;

[0008] Based on the target communication strategy, a dynamic resource allocation algorithm is adopted to determine a target communication resource allocation scheme according to the user's real-time geographic location information and network conditions, wherein the target communication resource allocation scheme is a resource allocation scheme for 5G messages, SMS and voice;

[0009] Based on the target communication resource allocation scheme, an information transmission process is executed, and a differential privacy protection mechanism is introduced to encrypt the information transmission process.

[0010] Optionally, parsing the information content to obtain a parsing result, and determining a target communication strategy corresponding to the information content using a reinforcement learning framework based on the target multimodal user portrait and the parsing result, includes:

[0011] Semantically encoding each modal content in the information content to obtain a coding result corresponding to each modal content, and fusing the coding results corresponding to all modal contents to obtain a coding fusion result;

[0012] Extracting features from the coding fusion results to obtain information content features, and using an intention recognition algorithm to recognize the intention reflected by the information content to be sent based on the information content features and the coding fusion results; the information content features include: keywords, named entities, sentiment tendencies, themes, and objects;

[0013] According to the coding fusion result and the intention reflected by the content of the information to be sent, an emotion vector for reflecting the user's emotion is determined by using an emotion evaluation algorithm, wherein the emotion vector includes multiple emotion dimensions, and the emotion vector and the information content feature together constitute the analysis result; the intention includes at least one of the following: notification, request, command and suggestion;

[0014] Based on the emotion vector and other factors, a reward function is defined by a reward mechanism; the other factors include: service quality, cost and risk, which are used to quantify the effect of the communication action, and the communication action includes: communication mode selection result, communication priority setting information, communication time window configuration information and personalized configuration information;

[0015] Constructing a state transition probability matrix, wherein the state transition probability matrix is ​​used to represent the probability of transferring to another state after performing a communication action in one state, wherein the state includes a target multimodal user profile, information content characteristics, network conditions, and historical feedback information;

[0016] Based on the state transition probability matrix and the current state of the user, the best action is selected from the communication action library using the reward function, and the best action is the target communication strategy;

[0017] Based on the state transition probability matrix, the reinforcement learning framework is used to perform state estimation to obtain the state estimation result. If the state estimation result indicates that the target communication strategy does not achieve the expected communication effect or there is a potential risk, the target communication strategy is adjusted to optimize the communication effect of the target communication strategy.

[0018] Optionally, semantic encoding is performed on each modal content in the information content to obtain an encoding result corresponding to each modal content, and the encoding results corresponding to all modal contents are merged to obtain an encoding fusion result, including:

[0019] Using a bidirectional encoder representation model, a convolutional neural network model, a self-supervised learning model and a time series analysis model, the text content, the image content, the audio content and the video content in the information content are semantically encoded respectively to obtain encoding results corresponding to the text content, the image content, the audio content and the video content;

[0020] Weighted fusion of the encoding results corresponding to the text content, the image content, the audio content and the video content to obtain a weighted fusion result;

[0021] Based on the user's real-time geographic location information, perception parameters, and the interactive response results between the geographic location information and the perception parameters, context-aware data is constructed, and the weighted fusion result and the context-aware data are fused using a Bayesian network to obtain a coded fusion result; the perception parameters include environmental intensity, noise level, user biometrics, and user behavior patterns.

[0022] Optionally, the context-aware data is constructed based on the user's real-time geographic location information, perception parameters, and the interactive response results between the geographic location information and the perception parameters, and the weighted fusion result and the context-aware data are fused using a Bayesian network to obtain a coded fusion result; the perception parameters include environmental intensity, noise level, user biometrics, and user behavior patterns, including:

[0023] Use decision trees to build an interactive response model, and use the interactive response model to identify the interactive response results between geographic location information and perception parameters;

[0024] Extracting geographic location features, perception parameter features, and interactive response features from the user's real-time geographic location information, perception parameters, and interactive response results, respectively, and fusing the geographic location features, perception parameter features, and interactive response features to obtain context-aware data;

[0025] Defining nodes of a Bayesian network, wherein the nodes include geographic location nodes, perception parameter nodes, and interactive response nodes;

[0026] The causal relationship between nodes is identified according to the domain knowledge of information access, and the statistical dependency between nodes is analyzed. Based on the causal relationship and the statistical dependency, the weighted fusion result and the context-aware data are fused to obtain the encoded fusion result.

[0027] Optionally, the reward function adopts the following formula:

[0028] ;

[0029] in, For the current state Perform communication actions The reward value obtained after is the service quality, which is used to reflect the current status Perform communication actions Quality of service, is the cost, used to reflect the current status Perform communication actions The cost, is the risk, used to reflect the current status Perform communication actions risk, is the sentiment vector, which reflects the current state Perform communication actions emotional state, and are the weights of the corresponding factors, which are dynamically adjusted through the online learning mechanism. and They are the parameters used to adjust the service quality assessment model, cost assessment model, risk assessment model and sentiment assessment model respectively.

[0030] Optionally, the dynamic resource allocation algorithm includes a multi-objective optimization algorithm and a load balancing algorithm; the dynamic resource allocation algorithm is used based on the target communication strategy to determine the target communication resource allocation scheme according to the user's real-time geographic location information and network conditions, including:

[0031] Determine the network conditions of the area based on the real-time geographic location information of the user, the network conditions including: signal strength limit, bandwidth limit and delay limit;

[0032] Generate an initial resource allocation plan using a multi-objective optimization algorithm based on the communication mode selection result, communication priority setting information, communication time window configuration information and personalized configuration information in the target communication strategy;

[0033] Use a load balancing algorithm to adjust the initial resource allocation plan to obtain an adjusted resource allocation plan;

[0034] Identify the available resources for 5G messages, SMS and voice in the communication resource pool, and based on the available resources for 5G messages, SMS and voice, as well as signal strength limitations, bandwidth limitations and delay limitations, verify the feasibility of the adjusted resource allocation scheme. If the verification passes, determine the adjusted resource allocation scheme as the target communication resource allocation scheme.

[0035] Optionally, during the information transmission process, the method further includes:

[0036] Constructing a monitoring indicator system; the monitoring indicator system includes the following monitoring indicators: information transmission success rate, user response time, network delay and resource utilization;

[0037] Based on the monitoring indicator system, when the value of any monitoring indicator exceeds the corresponding safety value range, the monitoring indicator whose value exceeds the corresponding safety value range is determined as an abnormal monitoring indicator, and the early warning mechanism corresponding to the abnormal monitoring indicator is triggered.

[0038] In the second aspect, the embodiment of the present application provides a 5G message, SMS and voice integrated information access system, including:

[0039] An acquisition optimization module is used to acquire the information content to be sent, the user's real-time geographic location information, and an initial multimodal user portrait constructed based on the user's behavior pattern and communication preference information; using the information content to be sent and the user's real-time geographic location information, the initial multimodal user portrait is optimized to obtain a target multimodal user portrait;

[0040] A parsing and determining module, used to parse the information content, obtain a parsing result, and determine a target communication strategy corresponding to the information content using a reinforcement learning framework based on the target multimodal user portrait and the parsing result;

[0041] An adjustment module is used to determine a target communication resource allocation scheme based on the target communication strategy and using a dynamic resource allocation algorithm according to the user's real-time geographic location information and network conditions, wherein the target communication resource allocation scheme is a resource allocation scheme for 5G messages, SMS, and voice;

[0042] The transmission module is used to execute the information transmission process based on the target communication resource allocation scheme, and introduce a differential privacy protection mechanism to encrypt the information transmission process.

[0043] In a third aspect, an embodiment of the present application provides a computing device, comprising a processing component and a storage component; the storage component stores one or more computer instructions; the one or more computer instructions are used to be called and executed by the processing component to implement a method for fusion information access of 5G messages, SMS and voice as described in any one of the first aspects.

[0044] In a fourth aspect, an embodiment of the present application provides a computer storage medium storing a computer program. When the computer program is executed by a computer, it implements a method for fusion information access of 5G messages, SMS and voice as described in any one of the first aspects.

[0045] In an embodiment of the present application, a method for integrated information access of 5G messages, text messages and voice is provided, including: obtaining information content to be sent, real-time geographic location information of the user, and an initial multimodal user portrait constructed based on user behavior patterns and communication preference information; optimizing the initial multimodal user portrait using the information content to be sent and the real-time geographic location information of the user to obtain a target multimodal user portrait; parsing the information content to obtain a parsing result, and determining a target communication strategy corresponding to the information content using a reinforcement learning framework based on the target multimodal user portrait and the parsing result; based on the target communication strategy, a dynamic resource allocation algorithm is used to determine a target communication resource allocation scheme according to the user's real-time geographic location information and network conditions, and the target communication resource allocation scheme is a resource allocation scheme for 5G messages, text messages and voice; based on the target communication resource allocation scheme, an information transmission process is executed, and a differential privacy protection mechanism is introduced to encrypt the information transmission process.

[0046] The embodiment of the present application obtains the information content to be sent and the user's real-time geographic location information, optimizes the initial multimodal user portrait, obtains the target multimodal user portrait, ensures that the target communication strategy can adapt to the dynamic changes of the user, and improves the accuracy of information access. The optimal communication strategy is determined by using the reinforcement learning framework, and the emotion vector, intent recognition results and other factors (such as service quality, cost, and risk) are comprehensively considered to achieve highly personalized communication decisions. Based on the dynamic resource allocation algorithm, the maximum and flexibility of resource utilization are achieved, and the differential privacy protection mechanism is introduced in the information transmission process to encrypt the information transmission process to ensure the security and privacy of user data and prevent the leakage of sensitive information. The embodiment of the present application introduces a real-time monitoring mechanism to dynamically optimize resource allocation according to the actual communication effect to ensure the efficiency and reliability of information transmission. This embodiment introduces the goals of four dimensions of service quality, cost, risk and emotional state, and ensures the comprehensiveness and accuracy of the evaluation through a weighted combination. The weights are dynamically adjusted through an online learning mechanism, so that the system can be continuously optimized according to real-time feedback, adapt to different application scenarios and user demand changes, and provide more personalized services.

[0047] These and other aspects of the present application will become more clearly understood in the description of the following embodiments. BRIEF DESCRIPTION OF THE DRAWINGS

[0048] In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, a brief introduction will be given below to the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0049] Figure 1 A flowchart of a method for delivering 5G messages, SMS and voice integrated information provided in an embodiment of the present application;

[0050] Figure 2 A structural diagram of a 5G message, SMS and voice integrated information access system provided in an embodiment of the present application;

[0051] Figure 3 A schematic diagram of the structure of a computing device provided in an embodiment of the present application. DETAILED DESCRIPTION

[0052] In order to enable those skilled in the art to better understand the solution of the present application, the technical solution in the embodiments of the present application will be clearly and completely described below in conjunction with the drawings in the embodiments of the present application.

[0053] In some of the processes described in the specification and claims of this application and the above-mentioned figures, multiple operations that appear in a specific order are included, but it should be clearly understood that these operations may not be executed in the order in which they appear in this article or executed in parallel. The sequence numbers of the operations, such as 11, 12, etc., are only used to distinguish between different operations, and the sequence numbers themselves do not represent any execution order. In addition, these processes may include more or fewer operations, and these operations may be executed in sequence or in parallel. It should be noted that the descriptions of "first", "second", etc. in this article are used to distinguish different messages, devices, modules, etc., do not represent the order of precedence, and do not limit the "first" and "second" to different types.

[0054] The following will be combined with the drawings in the embodiments of the present application to clearly and completely describe the technical solutions in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative work are within the scope of protection of this application.

[0055] Figure 1 A flowchart of a method for delivering 5G messages, SMS and voice integrated information provided in an embodiment of the present application, such as Figure 1 As shown, the method includes:

[0056] S11. Obtain the information content to be sent, the user's real-time geographic location information, and an initial multimodal user portrait constructed based on the user's behavior pattern and communication preference information; use the information content to be sent and the user's real-time geographic location information to optimize the initial multimodal user portrait to obtain a target multimodal user portrait.

[0057] Among them, information content refers to the specific information that needs to be delivered to the user, including text, images, audio, video and other forms. Real-time geographic location information refers to the specific location information of the user, usually including geographic coordinates such as latitude, longitude, altitude, etc. Real-time geographic location information can be used to evaluate network conditions (such as signal strength, bandwidth, latency), and optimize communication strategies in combination with user behavior patterns. User behavior patterns can be used to predict users' future behaviors and needs, making communication strategies more personalized and accurate. Communication preference information is the user's preference for different communication methods (such as 5G messages, SMS, and voice) extracted from multi-channel feedback information, ensuring that the selected communication method is in line with the user's habits and preferences and improving user satisfaction. Multimodal user portraits contain user feature descriptions in multiple dimensions, such as behavior patterns, communication preferences, emotional states, etc., providing comprehensive user background information for subsequent communication strategy formulation, ensuring that the strategy can adapt to the dynamic changes of users. The target multimodal user portrait is an optimized user portrait that more accurately reflects the user's current status and needs, ensuring that the communication strategy can better adapt to the user's real-time situation and improve the accuracy and timeliness of information delivery.

[0058] Optionally, this embodiment may adopt a hidden Markov model to analyze the user's historical interaction records to obtain the user's behavior pattern, and extract the user's communication preference information from multi-channel feedback information based on a Bayesian network.

[0059] S12. Analyze the information content to obtain the analysis results, and determine the target communication strategy corresponding to the information content using the reinforcement learning framework based on the target multimodal user portrait and the analysis results.

[0060] The parsing result may be a result generated after semantic encoding and feature extraction of the information content, including keywords, named entities, sentiment tendencies, themes, objects, etc., which are used to identify the specific intent of the information content and guide the subsequent communication strategy formulation. The reinforcement learning framework is a machine learning method that learns the optimal decision-making strategy through interaction with the environment. According to the target multimodal user portrait and the parsing results, the best action is selected from multiple candidate communication actions to ensure the optimization of the communication strategy. The target communication strategy may include the best communication method, priority setting, time window configuration, and personalized configuration determined by the reinforcement learning framework.

[0061] S13. Based on the target communication strategy, a dynamic resource allocation algorithm is adopted to determine the target communication resource allocation plan according to the user's real-time geographic location information and network conditions. The target communication resource allocation plan is a resource allocation plan for 5G messages, SMS and voice.

[0062] Among them, the dynamic resource allocation algorithm is an algorithm that can dynamically adjust resource allocation according to real-time data, which can include multi-objective optimization algorithms and load balancing algorithms to ensure that under different network conditions, the resource allocation scheme can maximize the use of existing resources and improve communication efficiency. The target communication resource allocation scheme is the optimal resource allocation scheme determined according to the target communication strategy and real-time network conditions, including the resource allocation of 5G messages, SMS and voice, which is used to ensure the maximization and flexibility of resource utilization during information transmission and avoid waste or shortage of resources.

[0063] S14. Based on the target communication resource allocation scheme, the information transmission process is executed, and a differential privacy protection mechanism is introduced to encrypt the information transmission process.

[0064] Among them, the information transmission process is the process of actually executing information transmission according to the target communication resource allocation plan to ensure that the information can be delivered to the user efficiently and securely. The differential privacy protection mechanism is a privacy protection technology that protects the privacy of user data by adding noise and other methods, ensures data security during information transmission, prevents sensitive information leakage, and enhances user trust. Encryption processing is to encrypt the information transmission process to ensure the security of the information content, prevent the information from being stolen or tampered with during the transmission process, and ensure information security.

[0065] By executing steps S11 to S14, the embodiment of the present application can more accurately understand the needs and preferences of users by optimizing multimodal user portraits, ensuring that information can be conveyed to users in the most effective way at the most appropriate time. The application of the reinforcement learning framework makes communication strategies more intelligent and personalized, and improves user satisfaction with the information delivery process. The dynamic resource allocation algorithm ensures the maximization and flexibility of resource utilization, avoids waste or shortage of resources, and improves the overall efficiency of the reach system. The application of differential privacy protection mechanism and encryption processing enhances the security of the information transmission process, protects user privacy, and improves user trust.

[0066] In some possible embodiments, during the information transmission process, the method further includes:

[0067] Step 141, construct a monitoring indicator system; the monitoring indicator system includes the following monitoring indicators: information transmission success rate, user response time, network delay and resource utilization.

[0068] Among them, the monitoring indicator system is a set of indicators used to evaluate the performance and effect of the information transmission process in real time, providing comprehensive performance data to help the system promptly discover potential problems and take corresponding measures. The information transmission success rate refers to the ratio of the number of successfully transmitted information to the total number of sent information. It is used to measure the effectiveness of information transmission and ensure that most information can be accurately delivered to users. User response time can refer to the time interval from the issuance of information to the user's response. It is used to reflect the timeliness of information transmission and the user's attention, and help optimize the time window for information transmission. Network latency refers to the time required for information to travel from the sender to the receiver, usually in milliseconds. It is used to evaluate the impact of network conditions on information transmission and ensure low-latency and efficient communication. Resource utilization refers to the actual use of various communication resources in the system (such as 5G messages, SMS, and voice). It is used to measure the rationality of resource allocation, avoid resource waste or shortage, and improve overall efficiency.

[0069] Step 142: Based on the monitoring indicator system, when the value of any monitoring indicator exceeds the corresponding safety value range, the monitoring indicator whose value exceeds the corresponding safety value range is determined as an abnormal monitoring indicator, and the early warning mechanism corresponding to the abnormal monitoring indicator is triggered.

[0070] Among them, the safe numerical range refers to a reasonable numerical range set for each monitoring indicator. If it exceeds this range, it is considered that there is a potential risk. It is used to define the boundaries of normal operation, help identify abnormal situations, and ensure the stable operation of the touch system. Abnormal monitoring indicators are used to quickly locate the problem, which is convenient for subsequent processing and optimization. The early warning mechanism refers to a set of response measures automatically triggered by the touch system when abnormal monitoring indicators are detected, including but not limited to alarm notifications, log records, resource adjustments, etc., which are used to promptly notify relevant personnel or systems to intervene to prevent the problem from further expanding and ensure the continuity and security of information transmission.

[0071] By executing steps 141 to 142, the embodiment of the present application monitors various key indicators in real time, can promptly discover and handle problems at the early stage of occurrence, reduce the scope of fault impact, and ensure the continuous and stable operation of the system. Monitor the success rate of information transmission and user response time to ensure that most information can be conveyed to users in a timely and accurate manner, and improve user experience. By monitoring resource utilization, the resource allocation strategy can be dynamically adjusted to avoid waste or insufficiency of resources and maximize resource utilization efficiency. When a monitoring indicator exceeds the safe numerical range, the early warning mechanism is immediately triggered to notify relevant personnel or systems to intervene to prevent the problem from expanding and ensure the security and continuity of information transmission. The complete monitoring indicator system and early warning mechanism provide detailed operating data, which is convenient for subsequent analysis and optimization, and improves the transparency and maintainability of the system. Therefore, this method not only enhances the security and reliability of the information transmission process, but also provides users with a more high-quality and stable communication service experience.

[0072] In a possible embodiment, S12, parsing the information content to obtain a parsing result, and determining a target communication strategy corresponding to the information content using a reinforcement learning framework based on the target multimodal user portrait and the parsing result, including:

[0073] Step 121: semantically encode each modal content in the information content to obtain the encoding result corresponding to each modal content, and fuse the encoding results corresponding to all modal contents to obtain the encoding fusion result. The modalities include text, image, audio and video.

[0074] As a possible implementation method, step 121, semantic encoding is performed on each modal content in the information content to obtain the encoding result corresponding to each modal content, and the encoding results corresponding to all modal contents are merged to obtain the encoding fusion result, including:

[0075] Step a1, using a bidirectional encoder representation model, a convolutional neural network model, a self-supervised learning model and a time series analysis model, respectively, the text content, image content, audio content and video content in the information content are semantically encoded, and the encoding results corresponding to the text content, image content, audio content and video content are obtained. Among them, the bidirectional encoder representation model is a natural language processing model that can simultaneously consider contextual information and generate more accurate word vector representations for semantic encoding of text content, capturing the deep semantic structure of the text, and improving the accuracy of text understanding. The convolutional neural network model is a deep learning model that is particularly good at processing image data. It extracts image features through the convolution layer and is used to semantically encode image content, identify key visual elements in the image, and enhance image understanding capabilities. The self-supervised learning model is a machine learning method that automatically learns useful feature representations from unlabeled data, and is used to semantically encode audio content. It can capture the key features of audio without a large amount of labeled data, thereby improving encoding efficiency and accuracy. The time series analysis model is a statistical model for processing and predicting time series data, which is used to semantically encode video content, capture the temporal relationship between video frames, and enhance the understanding of dynamic scenes. Semantic coding refers to converting information of different modalities into digital representations that can be understood and processed by computers, ensuring that multimodal information can be effectively fused and processed in subsequent steps. The coding result is a digital representation obtained after semantic coding, which reflects the main characteristics of the original information content and serves as the basis for subsequent fusion and analysis to ensure the accuracy and completeness of the information content.

[0076] Step a2: weighted fusion of the encoding results corresponding to the text content, image content, audio content and video content to obtain a weighted fusion result. Weighted fusion refers to assigning different weights according to the importance or relevance of each modal information, and merging multiple encoding results into a comprehensive result, to ensure that the fusion of multimodal information is more reasonable and effective, and to avoid excessive influence or neglect of a certain modal information. The weighted fusion result is the final encoding result obtained through weighted fusion, which integrates the content features of text, image, audio and video, and serves as the basis for subsequent context-aware data fusion to ensure the comprehensiveness and accuracy of the information content.

[0077] Step a3: Based on the user's real-time geographic location information, perception parameters, and the interactive response results between the geographic location information and the perception parameters, context-aware data is constructed, and the weighted fusion results and context-aware data are fused using a Bayesian network to obtain a coded fusion result; the perception parameters include environmental intensity, noise level, user biometrics, and user behavior patterns. Among them, context-aware data is data that reflects the user's current situation, which is formed by combining the user's real-time geographic location information, perception parameters, and their interactive response results. It provides rich background information for subsequent coding fusion and ensures that the communication strategy can adapt to the dynamic changes of the user. The coding fusion result is the final result after the weighted fusion result and context-aware data are fused through the Bayesian network. It serves as an important basis for determining the target communication strategy to ensure the accuracy and personalization of information transmission.

[0078] Specifically, step a3, based on the user's real-time geographic location information, perception parameters, and the interactive response results between the geographic location information and the perception parameters, construct context-aware data, and use a Bayesian network to fuse the weighted fusion result and the context-aware data to obtain a coded fusion result; the perception parameters include environmental intensity, noise level, user biometrics, and user behavior patterns, including:

[0079] Step a31, use the decision tree to build an interactive response model, and identify the interactive response results between the geographic location information and the perception parameters through the interactive response model. Among them, the decision tree is a machine learning model based on a tree structure, which is used to build an interactive response model. By analyzing the relationship between geographic location information and perception parameters, it predicts the response behavior of users under different conditions. The interactive response model is a model used to identify the interaction relationship between geographic location information and perception parameters, helping the access system understand the user's behavior pattern under specific geographic locations and environmental conditions, and providing a basis for subsequent feature extraction. The interactive response result is the result obtained by the interactive response model, which describes the specific interaction between geographic location information and perception parameters, and is used to further refine the context-aware data and enhance the understanding of the user's current state.

[0080] Step a32, extract geographic location features, perception parameter features and interactive response features from the user's real-time geographic location information, perception parameters, and interactive response results, respectively, and fuse the geographic location features, perception parameter features and interactive response features to obtain context-aware data. Among them, geographic location features can refer to features extracted from the user's real-time geographic location information, such as latitude and longitude, altitude, etc., which are used to describe the user's physical location and affect the choice of communication mode and resource allocation strategy. Perception parameter features are features extracted from perception parameters (such as environmental intensity, noise level, user biometrics and user behavior patterns) to reflect the specific situation of the user's environment and help the access system better understand the user's real-time status and needs. Interaction response features are features extracted from the interactive response results, which describe the specific interaction between geographic location information and perception parameters, enhance the understanding of user behavior patterns, and make context perception more accurate. Context-aware data is data that integrates geographic location features, perception parameter features and interactive response features, comprehensively reflects the user's current situation, and provides rich background information for subsequent coding fusion, ensuring that the communication strategy can adapt to the dynamic changes of the user.

[0081] Step a33, define the nodes of the Bayesian network, which include geographic location nodes, perception parameter nodes and interactive response nodes. Among them, the Bayesian network is a probabilistic graph model used to represent the causal relationship and statistical dependency between variables. By modeling the relationship between each node, the user's behavior and needs can be more accurately predicted and the communication strategy can be optimized. The node is the basic unit in the Bayesian network, representing a specific variable or feature, and is used to build a model basis for causal relationships and statistical dependencies. Among them, the geographic location node represents the geographic location characteristics of the user. The perception parameter node represents the perception parameter characteristics of the user, and the interactive response node represents the interactive response characteristics of the user in a specific context.

[0082] Step a34, identify the causal relationship between each node according to the information access domain knowledge, analyze the statistical dependency between each node, and fuse the weighted fusion result and context-aware data based on the causal relationship and statistical dependency to obtain the coding fusion result. Among them, the causal relationship describes the direct influence relationship between variables, that is, how the change of one variable leads to the change of another variable, helping the access system to understand the intrinsic connection between various factors and improve the accuracy of prediction. Statistical dependency describes the correlation between variables. Even if there is no direct causal relationship, there may be some statistical association, which is used to supplement the causal relationship and make the model more comprehensive and accurate. The weighted fusion result is the result obtained by weighted fusion of the coding results corresponding to the text content, image content, audio content and video content, ensuring that the fusion of multimodal information is more reasonable and effective. The coding fusion result is the result of the final fusion of the weighted fusion result and the context-aware data, which fully reflects the information content and user context, and serves as an important basis for the formulation of subsequent communication strategies to ensure the accuracy and personalization of information transmission.

[0083] By executing steps a31 to a34, the embodiment of the present application constructs an interactive response model and extracts multiple features, so that the system can more comprehensively understand the real-time status and needs of the user, ensuring that the communication strategy can adapt to the dynamic changes of the user. The causal relationship and statistical dependency between nodes are modeled using a Bayesian network, which improves the accuracy of predicting user behavior and needs and enhances the adaptive ability of the system. The coding fusion result comprehensively considers the information content and user context to ensure that the information can be conveyed to the user in the most effective way at the most appropriate time, thereby improving the user experience. By real-time monitoring and analysis of the user's geographic location information and perception parameters, the system can quickly respond to environmental changes and ensure the continuity and stability of the communication process. Combined with geographic location, perception parameters and interactive response results, the system can provide users with more personalized services, meet the needs and preferences of different users, and improve user satisfaction. Based on context-aware data and coding fusion results, the system can more reasonably allocate communication resources, avoid waste or shortage of resources, and improve overall resource utilization efficiency.

[0084] By executing steps a1 to a3, the embodiment of the present application adopts a variety of coding models to ensure that the semantic coding of text, image, audio and video content is more accurate and comprehensive. Through weighted fusion, the consistency and coordination of multimodal information are further enhanced. Combined with the user's real-time geographic location information, perception parameters and interactive response results, the system can more comprehensively understand the user's real-time status and needs, ensure that the communication strategy can adapt to the user's dynamic changes, and improve the accuracy and timeliness of information transmission. The causal relationship and statistical dependency between nodes are modeled using a Bayesian network, which improves the accuracy of user behavior and demand prediction, enhances the system's adaptive ability, and ensures the optimization of communication strategies. Based on the coding fusion results, the system can more reasonably allocate communication resources, avoid resource waste or shortage, maximize resource utilization efficiency, and improve overall communication efficiency. Combined with geographic location, perception parameters and interactive response results, the system can provide users with more personalized services, meet the needs and preferences of different users, and improve user satisfaction. By real-time monitoring and analysis of the user's geographic location information and perception parameters, the system can quickly respond to environmental changes, ensure the continuity and stability of the communication process, and improve the reliability of the system and user experience.

[0085] Step 122: extract features from the encoding fusion results to obtain information content features, and use an intent recognition algorithm based on the information content features and the encoding fusion results to identify the intent reflected in the information content to be sent; information content features include: keywords, named entities, sentiment tendencies, themes, and objects.

[0086] Step 123: Based on the coding fusion result and the intention reflected by the information content to be sent, an emotion evaluation algorithm is used to determine an emotion vector used to reflect the user's emotion. The emotion vector includes multiple emotion dimensions, and the emotion vector and the information content characteristics together constitute the analysis result; the intention includes at least one of the following: notification, request, command and suggestion.

[0087] Step 124: Based on the emotion vector and other factors, a reward function is defined using a reward mechanism; other factors include: service quality, cost and risk, which are used to quantify the effect of communication actions, and communication actions include: communication mode selection results, communication priority setting information, communication time window configuration information and personalized configuration information.

[0088] It should be understood that a single indicator is difficult to fully evaluate the communication effect. Therefore, the design introduces four dimensions: service quality, cost, risk, and emotional state to ensure the comprehensiveness and accuracy of the evaluation. In practical applications, it is necessary to find a balance between multiple goals such as service quality, cost control, and risk management. Through weighted combination, the system can weigh different goals and achieve the most optimized communication strategy. The communication environment and user needs are dynamically changing, and fixed weights cannot adapt to all situations. Therefore, a dynamic adjustment mechanism is adopted so that the system can flexibly adjust the weights of various factors according to actual conditions to ensure the optimal configuration. The introduction of emotional vectors enables the system to better understand the user's emotional state, provide more personalized services, meet the needs and preferences of different users, and improve user satisfaction. The introduction of risk factors ensures that the system can effectively identify and avoid potential security risks during information transmission and ensure the security and privacy of user data.

[0089] Exemplarily, the reward function adopts the following formula:

[0090] ;

[0091] in, For the current state Perform communication actions The reward value obtained after is the service quality, which is used to reflect the current status Perform communication actions Quality of service, is the cost, used to reflect the current status Perform communication actions The cost, is the risk, used to reflect the current status Perform communication actions risk, is the sentiment vector, which reflects the current state Perform communication actions emotional state, and are the weights of the corresponding factors, which are dynamically adjusted through the online learning mechanism. and are parameters used to adjust the service quality assessment model, cost assessment model, risk assessment model and sentiment assessment model. , , and They are calculated through a service quality evaluation model, a cost evaluation model, a risk evaluation model and a sentiment evaluation model respectively, and this embodiment does not specifically limit the specific forms of the above models.

[0092] Based on the above reward formula, this embodiment can achieve multi-dimensional comprehensive evaluation, comprehensively consider service quality, cost, risk and emotional state, ensure that information transmission is not only efficient but also safe and personalized, thereby improving the overall experience and satisfaction of users. and Through dynamic adjustment through online learning mechanism, the system can continuously optimize according to real-time feedback and adapt to different application scenarios and changes in user needs. By incorporating multiple key factors into the reward function, the system can make better communication strategy choices in complex environments, ensuring that each communication can achieve the best results.

[0093] Step 125: construct a state transition probability matrix. The state transition probability matrix is ​​used to represent the probability of transferring to another state after performing a communication action in one state. The state includes the target multimodal user portrait, information content characteristics, network conditions and historical feedback information.

[0094] Step 126: Based on the state transition probability matrix and the current state of the user, the best action is selected from the communication action library using the reward function, and the best action is the target communication strategy.

[0095] Step 127: Based on the state transition probability matrix, a reinforcement learning framework is used to perform state estimation to obtain a state estimation result. If the state estimation result indicates that the target communication strategy does not achieve the expected communication effect or there is a potential risk, the target communication strategy is adjusted to optimize the communication effect of the target communication strategy.

[0096] By executing steps 121 to 127, the embodiment of the present application performs semantic encoding on text, image, audio and video content, and integrates the encoding results, so that the system can fully understand and parse the information content, ensuring the integrity and accuracy of information transmission. By extracting information content features such as keywords, named entities, emotional tendencies, themes and objects, and combining the intention recognition algorithm, the system can accurately identify the specific intentions of the information to be sent (such as notifications, requests, commands and suggestions), so as to formulate a more appropriate communication strategy. The emotional evaluation algorithm is used to determine the emotional vector that reflects the user's emotions, including multiple emotional dimensions. This enables the system to better understand the user's emotional state, provide more humane services, and improve the user's acceptance and willingness to interact. Combined with the emotional vector and other factors (such as service quality, cost and risk), the system can generate personalized communication strategies for each user, meet the needs and preferences of different users, and improve user satisfaction. By constructing a state transition probability matrix and using the reinforcement learning framework for state estimation, the system can continuously optimize the target communication strategy after executing the communication action, ensuring that each communication can achieve the best effect, and improving the system's adaptability and robustness. Through the state transition probability matrix and reward function, the system can monitor the communication effect in real time, discover potential problems in time and take corresponding measures to ensure the continuity and stability of the communication process. Based on the state transition probability matrix and the current state of the user, the best action is selected from the communication action library to ensure the maximum and flexibility of resource utilization, avoid waste or shortage of resources, and improve the overall communication efficiency. By introducing cost and risk factors into the reward function, the system can effectively control costs and reduce potential risks while ensuring service quality, ensuring the security and reliability of the communication process.

[0097] In some possible embodiments, the dynamic resource allocation algorithm includes a multi-objective optimization algorithm and a load balancing algorithm; therefore, S13, based on the target communication strategy, the dynamic resource allocation algorithm is adopted to determine the target communication resource allocation scheme according to the user's real-time geographic location information and network conditions, including:

[0098] Step 131. Determine the network conditions in the area based on the user's real-time geographic location information. The network conditions include: signal strength limit, bandwidth limit and delay limit. Real-time geographic location information refers to the specific location information of the user's current location, usually including geographic coordinates such as longitude and latitude, altitude, etc., which is used to help evaluate the network conditions in the user's area and ensure that the communication strategy can adapt to the dynamic changes of the user. Network conditions are a set of parameters that describe the network performance in the user's area. Signal strength limit refers to the strength of the network signal, which is used to affect the stability and reliability of communication. Bandwidth limit refers to the available network bandwidth, which is used to affect the data transmission speed and capacity. Delay limit refers to the time required for information to travel from the sender to the receiver, which is used to affect the real-time nature of communication.

[0099] Step 132: Generate an initial resource allocation plan using a multi-objective optimization algorithm based on the communication mode selection result, communication priority setting information, communication time window configuration information, and personalized configuration information in the target communication strategy. The target communication strategy is the best communication mode, priority setting, time window configuration, and personalized configuration determined by the reinforcement learning framework, which is used to guide the generation of resource allocation plans and ensure the accuracy and timeliness of information transmission. The multi-objective optimization algorithm is an algorithm that can weigh and optimize between multiple goals, such as genetic algorithms, particle swarm optimization, etc., which is used to comprehensively consider multiple factors such as service quality, cost, and risk to generate the optimal initial resource allocation plan. The initial resource allocation plan is a preliminary resource allocation plan generated according to the target communication strategy, which provides a basis for subsequent adjustments and ensures the rationality and flexibility of resource utilization.

[0100] Step 133: Use a load balancing algorithm to adjust the initial resource allocation plan to obtain an adjusted resource allocation plan. The load balancing algorithm is an algorithm for allocating tasks or traffic between multiple servers or resource nodes to ensure load balancing of each node, to avoid overuse or idleness of a resource, and to improve overall resource utilization efficiency. The adjusted resource allocation plan is the final resource allocation plan adjusted by the load balancing algorithm, to ensure more reasonable and efficient resource allocation, and to avoid waste or insufficiency of resources.

[0101] Step 134, identify the available resources of 5G messages, SMS and voice in the communication resource pool, and verify the feasibility of the adjusted resource allocation scheme based on the available resources of 5G messages, SMS and voice, as well as the signal strength limit, bandwidth limit and delay limit. If the verification is passed, the adjusted resource allocation scheme is determined as the target communication resource allocation scheme. Among them, the communication resource pool is a collection of all resources that can be used for communication, such as 5G messages, SMS and voice, etc., which provides sufficient resource guarantee for resource allocation and ensures the continuity and stability of information transmission. The available resource amount can refer to the number of currently available communication resources, such as remaining bandwidth, unused channels, etc., which is used to evaluate the feasibility of the resource allocation scheme and ensure sufficient resources. Feasibility verification is to conduct a comprehensive evaluation of the adjusted resource allocation scheme to ensure that it meets the actual network conditions and resource constraints, ensure that the final resource allocation scheme can be effectively implemented, and avoid problems caused by insufficient resources or unsatisfied network conditions. The target communication resource allocation scheme is the final resource allocation scheme determined after feasibility verification, which serves as the basis for executing the information transmission process to ensure the efficiency and reliability of information transmission.

[0102] By executing steps 131 to 134, the embodiment of the present application identifies the amount of available resources in the communication resource pool, and performs feasibility verification in combination with network conditions to ensure that the adjusted resource allocation plan can be effectively implemented, avoid problems caused by insufficient resources or unsatisfactory network conditions, and improve the reliability and security of the system. Among them, a multi-objective optimization algorithm is used to comprehensively consider factors such as service quality, cost, and risk to generate an initial resource allocation plan to ensure maximum and flexible resource utilization. The application of the load balancing algorithm enables the system to flexibly adjust resource allocation under different conditions, ensure maximum and flexible resource utilization, avoid waste or shortage of resources, and improve overall communication efficiency.

[0103] Figure 2 A structural diagram of a 5G message, SMS and voice integrated information access system provided in an embodiment of the present application, such as Figure 2 As shown, the system includes:

[0104] The acquisition optimization module 21 is used to obtain the information content to be sent, the real-time geographic location information of the user, and the initial multimodal user portrait constructed based on the user behavior pattern and communication preference information; using the information content to be sent and the real-time geographic location information of the user, the initial multimodal user portrait is optimized to obtain the target multimodal user portrait;

[0105] The parsing and determining module 22 is used to parse the information content, obtain the parsing result, and determine the target communication strategy corresponding to the information content by using the reinforcement learning framework based on the target multimodal user portrait and the parsing result;

[0106] An adjustment module 23 is used to determine a target communication resource allocation scheme based on a target communication strategy and a dynamic resource allocation algorithm according to the user's real-time geographic location information and network conditions. The target communication resource allocation scheme is a resource allocation scheme for 5G messages, SMS, and voice.

[0107] The transmission module 24 is used to execute the information transmission process based on the target communication resource allocation scheme, and introduce a differential privacy protection mechanism to encrypt the information transmission process.

[0108] Figure 2 The 5G message, SMS and voice integrated information access system can perform Figure 1 The implementation principle and technical effect of the 5G message, SMS and voice integrated information access method described in the illustrated embodiment will not be repeated. The specific way in which each module and unit performs operations in the 5G message, SMS and voice integrated information access system in the above embodiment has been described in detail in the embodiment of the method, and will not be elaborated here.

[0109] In one possible design, Figure 2The 5G message, SMS and voice integrated information access system of the embodiment shown can be implemented as a computing device, such as Figure 3 As shown, the computing device may include a storage component 31 and a processing component 32 .

[0110] The storage component 31 stores one or more computer instructions, wherein the one or more computer instructions are called and executed by the processing component 32 .

[0111] The processing component 32 is used to: obtain the information content to be sent, the user's real-time geographic location information, and an initial multimodal user portrait constructed based on the user's behavior pattern and communication preference information; optimize the initial multimodal user portrait using the information content to be sent and the user's real-time geographic location information to obtain a target multimodal user portrait; parse the information content to obtain a parsing result, and based on the target multimodal user portrait and the parsing result, use a reinforcement learning framework to determine the target communication strategy corresponding to the information content; based on the target communication strategy, use a dynamic resource allocation algorithm to determine the target communication resource allocation plan according to the user's real-time geographic location information and network conditions, and the target communication resource allocation plan is a resource allocation plan for 5G messages, SMS and voice; based on the target communication resource allocation plan, execute the information transmission process, and introduce a differential privacy protection mechanism to encrypt the information transmission process.

[0112] The processing component 32 may include one or more processors to execute computer instructions to complete all or part of the steps in the above method. Of course, the processing component may also be implemented by one or more application-specific integrated circuits (ASIC), digital signal processors (DSP), digital signal processing devices (DSPD), programmable logic devices (PLD), field programmable gate arrays (FPGA), controllers, microcontrollers, microprocessors or other electronic components to perform the above method.

[0113] The storage component 31 is configured to store various types of data to support operations at the terminal. The storage component can be implemented by any type of volatile or non-volatile storage device or a combination thereof, such as random access memory (RAM), static random access memory (SRAM), electrically erasable programmable read-only memory (EEPROM), erasable programmable read-only memory (EPROM), programmable read-only memory (PROM), read-only memory (ROM), magnetic memory, flash memory, magnetic disk or optical disk.

[0114] Of course, the computing device may also include other components, such as input / output interfaces, display components, communication components, etc.

[0115] The input / output interface provides an interface between the processing component and the peripheral interface module, which may be an output device, an input device, etc.

[0116] The communication component is configured to facilitate, among other things, wired or wireless communications between the computing device and other devices.

[0117] Among them, the computing device can be a physical device or an elastic computing host provided by a cloud computing platform, etc. In this case, the computing device can refer to a cloud server, and the above-mentioned processing components, storage components, etc. can be basic server resources rented or purchased from the cloud computing platform.

[0118] The present application also provides a computer storage medium storing a computer program, wherein the computer program can achieve the above-mentioned Figure 1 The illustrated embodiment shows a method for delivering information by integrating 5G messages, SMS and voice.

[0119] Those skilled in the art can clearly understand that, for the convenience and brevity of description, the specific working processes of the systems and units described above can refer to the corresponding processes in the aforementioned method embodiments and will not be repeated here.

[0120] The system embodiment described above is merely illustrative, wherein the units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place, or they may be distributed on multiple network units. Some or all of the modules may be selected according to actual needs to achieve the purpose of the solution of this embodiment. Those of ordinary skill in the art may understand and implement it without creative work.

[0121] Through the description of the above implementation methods, those skilled in the art can clearly understand that each implementation method can be implemented by means of software plus a necessary general hardware platform, and of course, can also be implemented by hardware. Based on this understanding, the above technical solution is essentially or the part that contributes to the prior art can be embodied in the form of a software product, and the computer software product can be stored in a computer-readable storage medium, such as ROM / RAM, a disk, an optical disk, etc., including a number of instructions for a computer device (which can be a personal computer, a server, or a network device, etc.) to execute the methods described in each embodiment or some parts of the embodiments.

[0122] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present application, rather than to limit it. Although the present application has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some of the technical features therein. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present application.

Claims

1. A method for delivering 5G messages, SMS and voice information, characterized in that: include: Obtain the content of the information to be sent, the user's real-time geographic location information, and an initial multimodal user profile based on the user's behavior pattern and communication preference information; Using the information content to be sent and the user's real-time geographic location information, the initial multimodal user portrait is optimized to obtain the target multimodal user portrait; Parsing the information content to obtain a parsing result, and determining a target communication strategy corresponding to the information content using a reinforcement learning framework based on the target multimodal user portrait and the parsing result; Based on the target communication strategy, a dynamic resource allocation algorithm is adopted to determine a target communication resource allocation scheme according to the user's real-time geographic location information and network conditions, wherein the target communication resource allocation scheme is a resource allocation scheme for 5G messages, SMS and voice; Based on the target communication resource allocation scheme, an information transmission process is executed, and a differential privacy protection mechanism is introduced to encrypt the information transmission process.

2. The method according to claim 1, characterized in that The step of parsing the information content to obtain a parsing result, and determining a target communication strategy corresponding to the information content by using a reinforcement learning framework based on the target multimodal user portrait and the parsing result, includes: Semantically encoding each modal content in the information content to obtain a coding result corresponding to each modal content, and fusing the coding results corresponding to all modal contents to obtain a coding fusion result; Extracting features from the coding fusion results to obtain information content features, and using an intention recognition algorithm to recognize the intention reflected by the information content to be sent based on the information content features and the coding fusion results; the information content features include: keywords, named entities, sentiment tendencies, themes, and objects; According to the coding fusion result and the intention reflected by the content of the information to be sent, an emotion vector for reflecting the user's emotion is determined by using an emotion evaluation algorithm, wherein the emotion vector includes multiple emotion dimensions, and the emotion vector and the information content feature together constitute the analysis result; the intention includes at least one of the following: notification, request, command and suggestion; Based on the emotion vector and other factors, a reward function is defined by a reward mechanism; the other factors include: service quality, cost and risk, which are used to quantify the effect of the communication action, and the communication action includes: communication mode selection result, communication priority setting information, communication time window configuration information and personalized configuration information; Constructing a state transition probability matrix, wherein the state transition probability matrix is ​​used to represent the probability of transferring to another state after performing a communication action in one state, wherein the state includes a target multimodal user profile, information content characteristics, network conditions, and historical feedback information; Based on the state transition probability matrix and the current state of the user, the best action is selected from the communication action library using the reward function, and the best action is the target communication strategy; Based on the state transition probability matrix, the reinforcement learning framework is used to perform state estimation to obtain the state estimation result. If the state estimation result indicates that the target communication strategy does not achieve the expected communication effect or there is a potential risk, the target communication strategy is adjusted to optimize the communication effect of the target communication strategy.

3. The method according to claim 2, characterized in that Semantically encode each modal content in the information content to obtain an encoding result corresponding to each modal content, and fuse the encoding results corresponding to all modal contents to obtain an encoding fusion result, including: Using a bidirectional encoder representation model, a convolutional neural network model, a self-supervised learning model and a time series analysis model, the text content, the image content, the audio content and the video content in the information content are semantically encoded respectively to obtain encoding results corresponding to the text content, the image content, the audio content and the video content; Weighted fusion of the encoding results corresponding to the text content, the image content, the audio content and the video content to obtain a weighted fusion result; Based on the user's real-time geographic location information, perception parameters, and the interactive response results between the geographic location information and the perception parameters, context-aware data is constructed, and the weighted fusion result and the context-aware data are fused using a Bayesian network to obtain a coded fusion result; the perception parameters include environmental intensity, noise level, user biometrics, and user behavior patterns.

4. The method according to claim 3, characterized in that: The context-aware data is constructed based on the user's real-time geographic location information, perception parameters, and the interactive response results between the geographic location information and the perception parameters, and the weighted fusion result and the context-aware data are fused using a Bayesian network to obtain a coded fusion result; The perception parameters include environmental intensity, noise level, user biometrics and user behavior patterns, including: Use decision trees to build an interactive response model, and use the interactive response model to identify the interactive response results between geographic location information and perception parameters; Extracting geographic location features, perception parameter features, and interactive response features from the user's real-time geographic location information, perception parameters, and interactive response results, respectively, and fusing the geographic location features, perception parameter features, and interactive response features to obtain context-aware data; Defining nodes of a Bayesian network, wherein the nodes include geographic location nodes, perception parameter nodes, and interactive response nodes; The causal relationship between nodes is identified according to the domain knowledge of information access, and the statistical dependency between nodes is analyzed. Based on the causal relationship and the statistical dependency, the weighted fusion result and the context-aware data are fused to obtain the encoded fusion result.

5. The method according to claim 2, characterized in that: The reward function uses the following formula: ； in, For the current state Perform communication actions The reward value obtained after is the service quality, which is used to reflect the current status Perform communication actions Quality of service, is the cost, used to reflect the current status Perform communication actions The cost, is the risk, used to reflect the current status Perform communication actions risk, is the sentiment vector, which reflects the current state Perform communication actions emotional state, and are the weights of the corresponding factors, which are dynamically adjusted through the online learning mechanism. and They are the parameters used to adjust the service quality assessment model, cost assessment model, risk assessment model and sentiment assessment model respectively.

6. The method according to claim 1, characterized in that The dynamic resource allocation algorithm includes a multi-objective optimization algorithm and a load balancing algorithm; the dynamic resource allocation algorithm is used based on the target communication strategy to determine the target communication resource allocation scheme according to the user's real-time geographic location information and network conditions, including: Determine the network conditions of the area based on the real-time geographic location information of the user, the network conditions including: signal strength limit, bandwidth limit and delay limit; Generate an initial resource allocation plan using a multi-objective optimization algorithm based on the communication mode selection result, communication priority setting information, communication time window configuration information and personalized configuration information in the target communication strategy; Use a load balancing algorithm to adjust the initial resource allocation plan to obtain an adjusted resource allocation plan; Identify the available resources for 5G messages, SMS and voice in the communication resource pool, and based on the available resources for 5G messages, SMS and voice, as well as signal strength limitations, bandwidth limitations and delay limitations, verify the feasibility of the adjusted resource allocation scheme. If the verification passes, determine the adjusted resource allocation scheme as the target communication resource allocation scheme.

7. The method according to claim 6, characterized in that During the information transmission process, the method further includes: Constructing a monitoring indicator system; the monitoring indicator system includes the following monitoring indicators: information transmission success rate, user response time, network delay and resource utilization; Based on the monitoring indicator system, when the value of any monitoring indicator exceeds the corresponding safety value range, the monitoring indicator whose value exceeds the corresponding safety value range is determined as an abnormal monitoring indicator, and the early warning mechanism corresponding to the abnormal monitoring indicator is triggered.

8. A 5G message, SMS and voice integrated information access system, characterized in that: include: An acquisition optimization module is used to obtain the content of the information to be sent, the real-time geographic location information of the user, and the initial multimodal user portrait constructed based on the user behavior pattern and communication preference information; Using the information content to be sent and the user's real-time geographic location information, the initial multimodal user portrait is optimized to obtain the target multimodal user portrait; A parsing and determining module, used to parse the information content, obtain a parsing result, and determine a target communication strategy corresponding to the information content using a reinforcement learning framework based on the target multimodal user portrait and the parsing result; An adjustment module is used to determine a target communication resource allocation scheme based on the target communication strategy and using a dynamic resource allocation algorithm according to the user's real-time geographic location information and network conditions, wherein the target communication resource allocation scheme is a resource allocation scheme for 5G messages, SMS, and voice; The transmission module is used to execute the information transmission process based on the target communication resource allocation scheme, and introduce a differential privacy protection mechanism to encrypt the information transmission process.

9. A computing device, characterized in that It includes a processing component and a storage component; the storage component stores one or more computer instructions; the one or more computer instructions are used to be called and executed by the processing component to implement a 5G message, SMS and voice integrated information access method as described in any one of claims 1 to 7.

10. A computer storage medium, characterized in that: A computer program is stored, and when the computer program is executed by a computer, a 5G message, SMS and voice integrated information access method as described in any one of claims 1 to 7 is implemented.

Images