Text transmission method and device, storage medium and computer equipment

Abstract

According to the text transmission method and device, the storage medium and the computer equipment provided by the invention, after the to-be-transmitted text, the current network state and the current system load are obtained, the text feature corresponding to the to-be-transmitted text can be determined firstly, and then the current network state, the current system load and the text feature are divided into the condition variables of multiple dimensions; and determining a target parameter combination corresponding to the condition variable of each dimension, then performing packaging, priority queuing and compression processing on the to-be-transmitted text according to the target parameter combination, and finally sending the compressed to-be-transmitted text to a receiving end. In the process, by monitoring the network state in real time, dynamically selecting the transmission protocol and adjusting the parameters and automatically switching to the low-bandwidth-demand protocol when the network is congested, the delay and packet loss rate are effectively reduced, and the transmission stability is guaranteed. In addition, transmission priorities can be allocated according to text features, and a proper compression mode is selected, so that the overall transmission efficiency is improved.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of data processing, and particularly relates to a text transmission method and device, a storage medium and a computer device. BACKGROUND

[0002] In the current digital information interaction scene, text transmission is widely used in instant messaging, file transmission, cloud storage and other fields. With the explosive growth of data volume, the complexity of network environment and the improvement of user requirements for transmission efficiency and stability, how to efficiently and reliably transmit text has become a technical problem to be solved.

[0003] In the existing text transmission process, text transmission is generally based on fixedly configured parameters. Before transmitting the text, fixed transmission protocols (such as TCP, UDP), transmission priorities, data compression rates and other parameters are preset. Taking instant messaging applications as an example, in the text transmission process, regardless of network bandwidth fluctuations, server load conditions and differences in the importance of text content, the same parameters are used for transmission. The existing text transmission method based on fixed parameter configuration has the following shortcomings:

[0004] (1) Lack of flexibility: fixed parameters cannot adapt to dynamic changes in network environment, such as when the network is congested, a high-bandwidth-demand transmission protocol is still used, which can easily lead to increased transmission delay or even packet loss.

[0005] (2) Low resource utilization: the same transmission priority is used for different importance levels of text, which causes critical information to be unable to be delivered in time, and non-critical information to occupy a large amount of network resources.

[0006] (3) Poor transmission efficiency: fixed data compression methods can waste computing resources when over-compressing highly repetitive text, and can affect transmission efficiency when under-compressing complex text. SUMMARY

[0007] The present application aims to at least solve one of the above technical defects, particularly the technical defects of lack of flexibility, low resource utilization and low transmission efficiency in the prior art text transmission method based on fixed parameter configuration.

[0008] The present application provides a text transmission method, which comprises:

[0009] obtaining a to-be-transmitted text, a current network state and a current system load, and determining a text feature corresponding to the to-be-transmitted text;

[0010] dividing the current network state, the current system load and the text feature into condition variables of multiple dimensions;

[0011] Determine the target parameter combination corresponding to the conditional variables of each dimension, encapsulate, prioritize and compress the text to be transmitted according to the target parameter combination, and send the compressed text to be transmitted to the receiving end.

[0012] Optionally, determining a text feature corresponding to the text to be transmitted includes:

[0013] Extracting the text length and keywords of the text to be transmitted to obtain an extraction result;

[0014] Performing a semantic importance evaluation on the text to be transmitted to obtain an evaluation result;

[0015] Determine text features of the text to be transmitted based on the extraction result and the evaluation result.

[0016] Optionally, dividing the current network state, the current system load, and the text features into conditional variables of multiple dimensions includes:

[0017] The current network status, the current system load and the text features are input into a preset dynamic trade-off model, and the current network status, the current system load and the text features are subjected to conditional variable standardization processing through the dynamic trade-off model to obtain conditional variables of multiple dimensions.

[0018] Optionally, determining target parameter combinations corresponding to conditional variables of various dimensions includes:

[0019] Matching the condition variables of each dimension with a preset rule base to obtain a matching result, wherein the preset rule base contains parameter combinations under different condition combinations;

[0020] When the matching result is that a corresponding parameter combination is matched, the matched parameter combination is used as the target parameter combination corresponding to the conditional variables of each dimension;

[0021] When the matching result is that no corresponding parameter combination is matched, a target parameter combination corresponding to the conditional variables of each dimension is generated through a multi-objective optimization algorithm.

[0022] Optionally, generating target parameter combinations corresponding to conditional variables of various dimensions through a multi-objective optimization algorithm includes:

[0023] The constraints are shortest transmission time, highest stability, and resource utilization not exceeding the preset system threshold;

[0024] Constructing a multi-objective function based on the conditional variables of each dimension and the constraint conditions, and optimizing and solving the multi-objective function to obtain a solution result;

[0025] A target parameter combination corresponding to the conditional variables of each dimension is generated according to the solution result.

[0026] Optionally, the target parameter combination includes a target transmission protocol, a target transmission priority, a target compression algorithm, and a compression ratio;

[0027] The encapsulating, prioritizing, and compressing the text to be transmitted according to the target parameter combination to obtain a compressed text to be transmitted includes:

[0028] Encapsulating the text to be transmitted according to the target transmission protocol;

[0029] Prioritize the encapsulated text to be transmitted according to the target transmission priority;

[0030] When the encapsulated text to be transmitted is placed at the head of the queue, the encapsulated text to be transmitted is compressed according to the target compression algorithm and compression ratio to obtain the compressed text to be transmitted.

[0031] Optionally, the method further includes:

[0032] Obtaining feedback information sent by the receiving end;

[0033] After performing parameter association on the feedback information, an optimized text is generated;

[0034] Optimizing and adjusting parameters in the dynamic trade-off model according to the optimization text;

[0035] Alternatively, the rules in the rule base are optimized and adjusted according to the optimization text.

[0036] Optionally, generating an optimized text after associating parameters with the feedback information includes:

[0037] Dividing the feedback information into multiple data types and determining data indicators corresponding to each data type;

[0038] After cleaning and standardizing the data indicators corresponding to various data types, the standardized data indicators are associated with the conditional parameters when transmitting the text, and optimized text is generated.

[0039] Optionally, optimizing and adjusting parameters in the dynamic trade-off model according to the optimization text includes:

[0040] The parameters in the dynamic trade-off model are optimized and adjusted according to the preset parameter optimization strategy and the optimization text.

[0041] Optionally, optimizing and adjusting the rules in the rule base according to the optimization text includes:

[0042] According to the optimized text, the priority of the rules in the rule base is adjusted, new rules are generated, and / or conflicting rules are resolved.

[0043] The present application also provides a text transmission device, comprising:

[0044] A data acquisition module, configured to acquire the text to be transmitted, the current network status, and the current system load, and determine text features corresponding to the text to be transmitted;

[0045] A condition variable division module, configured to divide the current network state, the current system load, and the text features into condition variables of multiple dimensions;

[0046] The text transmission module is used to determine the target parameter combination corresponding to the conditional variables of each dimension, encapsulate, prioritize and compress the text to be transmitted according to the target parameter combination, and send the compressed text to be transmitted to the receiving end.

[0047] The present application also provides a computer-readable storage medium, which stores computer-readable instructions. When the computer-readable instructions are executed by one or more processors, the one or more processors execute the steps of the text transmission method described in any one of the above embodiments.

[0048] The present application also provides a computer device, comprising: one or more processors, and a memory;

[0049] The memory stores computer-readable instructions, and when the computer-readable instructions are executed by the one or more processors, the steps of the text transmission method described in any one of the above embodiments are performed.

[0050] It can be seen from the above technical solutions that the embodiments of the present application have the following advantages:

[0051] The text transmission method, device, storage medium and computer equipment provided by the present application can first determine the text features corresponding to the text to be transmitted after obtaining the text to be transmitted, the current network status and the current system load; then, divide the current network status, the current system load and the text features into conditional variables of multiple dimensions; then, determine the target parameter combination corresponding to the conditional variables of each dimension, and encapsulate, prioritize and compress the text to be transmitted according to the target parameter combination, and finally send the compressed text to be transmitted to the receiving end. In this process, the present application monitors the network status in real time, dynamically selects the transmission protocol and adjusts the parameters, and automatically switches to a low-bandwidth requirement protocol when the network is congested, effectively reducing delays and packet loss rates, and ensuring transmission stability. In addition, the present application can also assign transmission priorities based on text features to ensure that key information is transmitted first; at the same time, select a suitable compression method based on text features to avoid resource waste, thereby improving overall transmission efficiency. BRIEF DESCRIPTION OF THE DRAWINGS

[0052] In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the following briefly introduces the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are only 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 labor.

[0053] Figure 1 A flowchart of a text transmission method provided in an embodiment of the present application;

[0054] Figure 2 A flow chart of the target parameter combinations corresponding to the shortcomings and conditional variables of each dimension provided in the embodiment of the present application;

[0055] Figure 3 A diagram showing the structure of the data collection, dynamic parameter balance, text transmission, feedback and optimization process provided in the embodiment of the present application;

[0056] Figure 4 A schematic diagram of the structure of a text transmission device provided in an embodiment of the present application;

[0057] Figure 5 A schematic diagram of the internal structure of a computer device provided in an embodiment of the present application. DETAILED DESCRIPTION

[0058] With reference to the drawings of the embodiments of the present application, the technical solutions in the embodiments of the present application will be clearly and completely described. Obviously, the described embodiments are only a part of the embodiments of the present application, but not all the embodiments of the present application. Based on the embodiments of the present application, all the other embodiments obtained by those of ordinary skill in the art without creative effort should fall into the scope of the present application.

[0059] In one embodiment, as shown in Figure 1 Figure 1 a flowchart of a text transmission method provided by the embodiments of the present application; the present application provides a text transmission method, which can include:

[0060] S110: obtaining a to-be-transmitted text, a current network state and a current system load, and determining a text feature corresponding to the to-be-transmitted text.

[0061] In this step, when the to-be-transmitted text is obtained, the current network state and the current system load can be obtained together. In this way, the transmission parameters can be dynamically adjusted according to the current network state, the current system load and the text feature of the to-be-transmitted text, so as to achieve the optimal balance of efficiency, stability and resource utilization rate in the text transmission process, thereby solving the problems of poor transmission flexibility, resource waste and low efficiency in the prior art.

[0062] Among them, the to-be-transmitted text obtained by the present application refers to any form of data information, including but not limited to text, picture, audio, video, etc. These information are encapsulated in the form of text for transmission in the network. The current network state includes but is not limited to network bandwidth, delay, packet loss rate and other parameters, which reflect the transmission capacity and stability of the current network. The current system load refers to the current running state of the text transmission device, including but not limited to CPU usage, memory occupancy and other parameters, which reflect the current resource occupation of the text transmission device.

[0063] Further, when determining the text feature corresponding to the to-be-transmitted text, the text feature includes but is not limited to text length, keyword, semantic importance and the like, which reflect the content characteristics and importance of the to-be-transmitted text. By comprehensively considering these factors, the present application can dynamically select the most suitable transmission parameters to achieve efficient and stable transmission of the text.

[0064] S120: dividing the current network state, the current system load and the text feature into multiple dimensional condition variables.

[0065] ​In this step, after obtaining the text to be transmitted, the current network status and the current system load through S110 and determining the text features corresponding to the text to be transmitted, the present application can divide the current network status, the current system load and the text features into conditional variables of multiple dimensions, so that the conditional variables of each dimension can be combined to determine the final target parameter combination.

[0066] Specifically, after determining the current network status, the current system load and the text features corresponding to the text to be transmitted, the present application can divide the current network status into conditional variables of multiple dimensions such as bandwidth, delay, packet loss rate, etc., and can also divide the current system load into conditional variables of multiple dimensions such as text length, keyword type, semantic importance, etc. The current system load can also be divided into conditional variables of multiple dimensions such as CPU usage and memory occupancy. In this way, conditional variables of each dimension corresponding to the current network status, current system load and text features can be obtained.

[0067] It is understandable that the condition variables divided by this application include not only the corresponding condition types, but also the corresponding values, such as bandwidth of 5Mbps, latency of 100ms, packet loss rate of 2%, CPU utilization of 60%, memory occupancy of 40%, text length of 1000 words, keyword type of business-related, semantic importance of high, etc. By refining the current network status, current system load, and text features into condition variables of multiple dimensions, this application can more accurately describe the current transmission environment and provide more detailed data support for the subsequent determination of the target parameter combination.

[0068] S130: Determine the target parameter combination corresponding to the conditional variables of each dimension, encapsulate, prioritize, and compress the text to be transmitted according to the target parameter combination, and send the compressed text to be transmitted to the receiving end.

[0069] In this step, after dividing the current network status, current system load and text features into conditional variables of multiple dimensions through S120, the present application can also determine the target parameter combination corresponding to the conditional variables of each dimension, so that the text to be transmitted can be encapsulated, prioritized and compressed according to the target parameter combination, and the compressed text to be transmitted can be sent to the receiving end.

[0070] To elaborate, when determining the target parameter combination corresponding to the conditional variables of each dimension, the present application can first try to match the conditional variables of each dimension with the preset rule base. The rule base pre-stores the optimal parameter combinations under a variety of conditional combinations. These parameter combinations are based on a large amount of historical data and experiments, and can ensure the optimal balance of transmission efficiency, stability and resource utilization in different transmission environments. When the match is successful, that is, when a parameter combination that fully matches or is closest to the current conditional variable is found, the present application will directly use this parameter combination as the target parameter combination.

[0071] However, due to the complexity and variability of actual transmission environments, the preset rule base may not cover all possible condition combinations. In this case, when the matching result is that no corresponding parameter combination is found, this application will use other methods to generate target parameter combinations. These target parameter combinations include but are not limited to target transmission protocol, target transmission priority, target compression algorithm and compression ratio, etc.

[0072] Subsequently, the present application can encapsulate, prioritize, and compress the text to be transmitted according to these target parameter combinations. The encapsulation process can package the text to be transmitted into a format that meets the protocol requirements according to the target transmission protocol; priority queuing places the text to be transmitted into the corresponding transmission queue based on the target transmission priority to ensure that critical information can be transmitted first; compression processing can use the target compression algorithm and compression rate to compress the text to be transmitted to reduce the amount of transmitted data and improve transmission efficiency. Finally, the present application can send the compressed text to be transmitted to the receiving end, completing the entire transmission process.

[0073] In addition, when the receiving end of the present application receives the compressed text to be transmitted, it can decompress, decode and reassemble it to restore the original text to be transmitted. During the decompression process, the receiving end can perform reverse operations based on the compression algorithm and compression rate used during compression to restore the data before compression. The decoding process is to parse the data according to the target transmission protocol used during encapsulation to ensure the integrity and accuracy of the data. The reassembly process is to sort the received data according to the priority order during transmission to restore the original data order. Through this series of processing, the receiving end can accurately obtain the text to be transmitted sent by the sending end, and realize efficient and stable transmission of text.

[0074] In the above embodiment, after obtaining the text to be transmitted, the current network status and the current system load, the text features corresponding to the text to be transmitted can be determined first; then, the current network status, the current system load and the text features are divided into conditional variables of multiple dimensions; then, the target parameter combination corresponding to the conditional variables of each dimension is determined, and the text to be transmitted is encapsulated, prioritized and compressed according to the target parameter combination, and finally the compressed text to be transmitted is sent to the receiving end. In this process, the present application monitors the network status in real time, dynamically selects the transmission protocol and adjusts the parameters, and automatically switches to a low-bandwidth demand protocol when the network is congested, effectively reducing delays and packet loss rates, and ensuring transmission stability. In addition, the present application can also assign transmission priorities based on text features to ensure that key information is transmitted first; at the same time, a suitable compression method is selected according to the text features to avoid waste of resources, thereby improving overall transmission efficiency.

[0075] In one embodiment, determining the text features corresponding to the text to be transmitted in S110 may include:

[0076] S111: extracting the text length and keywords of the text to be transmitted to obtain an extraction result.

[0077] S112: Evaluate the semantic importance of the text to be transmitted to obtain an evaluation result.

[0078] S113: Determine text features of the text to be transmitted based on the extraction result and the evaluation result.

[0079] In this embodiment, when determining the text features corresponding to the text to be transmitted, the text length and keyword extraction can be performed on the text to be transmitted. Text length analysis can provide a preliminary understanding of the text data volume, providing a reference for subsequent compression processing. Keyword extraction selects representative and important words from the text. These words often reflect the theme and content of the text, which helps in the subsequent determination of text features and the allocation of transmission priorities.

[0080] Among them, when extracting the text length of the text to be transmitted, this application can be completed by counting the number of characters, bytes or vocabulary of the text. When counting the number of characters, the number of all visible characters in the text (such as Chinese characters, letters, punctuation marks, etc.) can be directly calculated. When counting the number of bytes, the actual occupied storage space can be calculated based on the encoding format of the text (such as UTF-8, GBK), which is suitable for evaluating the amount of transmitted data. When counting the vocabulary, the text can be divided into words using word segmentation technology (such as Jieba word segmentation) to count the number of unique words or the total number of words.

[0081] Further, when extracting keywords from the text to be transmitted, the application can use statistical methods such as TF-IDF algorithm: by calculating the term frequency (TF) and inverse document frequency (IDF), high-frequency and low-frequency words in other texts are filtered out. For example, in technical documents, words such as "dynamic balance" and "transmission parameters" may be extracted as keywords due to high TF-IDF values. Or word frequency statistics: directly count the high-frequency words in the text, suitable for short texts or clearly defined content. It can also be based on graph model methods such as TextRank algorithm: the text is regarded as a graph structure, the words are regarded as nodes, the weight is calculated through the co-occurrence relationship between nodes, and the important words are iteratively filtered out, similar to the principle of web page ranking. Or domain dictionary matching: combined with a pre-set domain dictionary (such as a communication technology dictionary), professional terms are identified as keywords through regular expression or string matching. The specific method used can be selected according to the actual situation, and is not limited here.

[0082] Secondly, the application can evaluate the semantic importance of the text to be transmitted. This step is to evaluate the semantic importance of each part of the text by deeply understanding and analyzing the content of the text. This usually involves natural language processing techniques and algorithms, which can identify key information, important points or conclusions in the text, etc., thereby providing a strong basis for subsequent transmission priority allocation.

[0083] Among them, when evaluating the semantic importance of the text to be transmitted, the application can use rule-based semantic analysis, for example, the application can first assign weights to keywords: according to the importance of keywords in the field (such as "transmission protocol" has higher weight than "ordinary text"), the semantic importance is calculated by combining the frequency and weight of keywords; then determine the sentence position feature: the title, the first sentence of the paragraph or the sentence containing the transition word (such as "key" "must") is usually given higher weight. Of course, the application can also use machine learning methods, for example, the application can use a text classification model: by training a classifier (such as SVM, BERT) to determine whether the text belongs to the "key information", "ordinary information" or "secondary information" category, for example, the semantic importance of emergency notification text is higher than that of daily chat content; or use sentiment analysis and intent recognition: analyze the sentiment (such as urgency, warning) or intent (such as instruction, request) of the text, and evaluate its importance to the receiving end; it can also use structure feature analysis: perform syntactic analysis (such as extracting subject-predicate-object structure) on the text, and identify sentences containing core actions or objects; for long texts, topic modeling (such as LDA) can be used to extract main topics and evaluate the importance of the topics. The specific method used can be selected according to the actual situation, and is not limited here.

[0084] Finally, based on the extraction and evaluation results, this application can comprehensively determine the textual features of the text to be transmitted. These textual features include, but are not limited to, text length, keyword type, keyword frequency, and semantic importance, which together constitute a comprehensive description and characterization of the text to be transmitted. For example, in an instant messaging system, a technical notice is transmitted: "Please update core transmission protocol parameters immediately to avoid data packet loss due to network congestion." This application can extract the text length, such as counting characters to determine if it is approximately 30 words, and the byte count is determined based on the encoding. This application can also use TF-IDF to extract "update," "transmission protocol," "network congestion," and "data packet loss." Because it contains keywords such as "immediately" and "core" and has an instruction-like sentence structure, this application can be identified as high-importance text using rules or models. These textual features can more accurately grasp the content and characteristics of the text, providing strong support for subsequent dynamic trade-off model parameter selection and rule base rule development.

[0085] In one embodiment, dividing the current network status, the current system load, and the text features into conditional variables of multiple dimensions in S120 may include:

[0086] S121: Input the current network status, the current system load and the text features into a preset dynamic trade-off model, and perform conditional variable normalization processing on the current network status, the current system load and the text features through the dynamic trade-off model to obtain conditional variables of multiple dimensions.

[0087] In this embodiment, after determining the current network status, current system load, and text features corresponding to the text to be transmitted, the present application can design a preset dynamic trade-off model to divide them into conditional variables of multiple dimensions. This model can receive the current network status, current system load, and text features as input and perform conditional variable standardization on them. The purpose of this step is to convert various types of data (such as network bandwidth, CPU usage, text length, etc.) into numerical values ​​or classifications with a unified scale and comparability to facilitate subsequent parameter matching and selection.

[0088] Specifically, the dynamic trade-off model can first pre-process the input data, such as data cleaning, missing value filling, outlier detection and processing, to ensure the accuracy and completeness of the data. Subsequently, the model can standardize the data, including data normalization, discretization or classification coding. Normalization is to scale the data to a specific range (such as between 0 and 1) to facilitate comparison of data of different dimensions. Discretization is to divide the data into a limited number of categories or intervals to facilitate subsequent classification and rule matching. Classification coding is to convert text features (such as keyword type, semantic importance) into numerical variables to facilitate model processing and calculation.

[0089] After normalization by the dynamic trade-off model, the current network status, current system load, and text features are converted into conditional variables of multiple dimensions. These conditional variables not only have a unified scale and comparability, but also more accurately describe the current transmission environment and text characteristics. For example, network bandwidth can be divided into three categories: "low bandwidth," "medium bandwidth," and "high bandwidth"; CPU utilization can be normalized to a value between 0 and 1; and text length can be divided into categories such as "short text," "medium-long text," and "long text."

[0090] In this way, through the standardization of the dynamic trade-off model, this application can obtain conditional variables of various dimensions corresponding to the current network status, current system load, and text characteristics. These conditional variables provide more accurate and detailed data support for the subsequent determination of target parameter combinations, helping to achieve efficient and stable text transmission.

[0091] In one embodiment, Figure 2 As shown, Figure 2 A flowchart of a target parameter combination corresponding to a disadvantage and conditional variables in each dimension provided in an embodiment of the present application; determining the target parameter combination corresponding to the conditional variables in each dimension in S130 may include:

[0092] S131: Match the condition variables of each dimension with a preset rule base to obtain a matching result, wherein the preset rule base contains parameter combinations under different condition combinations.

[0093] S132: When the matching result is that a corresponding parameter combination is matched, the matched parameter combination is used as a target parameter combination corresponding to the conditional variables of each dimension.

[0094] S133: When the matching result is that no corresponding parameter combination is matched, a target parameter combination corresponding to the conditional variables of each dimension is generated through a multi-objective optimization algorithm.

[0095] In this embodiment, when determining the target parameter combination corresponding to the condition variables of each dimension, the application can attempt to match the condition variables of each dimension with a preset rule base. This step is based on a rule base containing optimal parameter combinations under various condition combinations, which are derived based on a large amount of historical data and experiments, aiming to ensure the optimal balance of transmission efficiency, stability, and resource utilization under different transmission environments.

[0096] Among them, the rule base of the application is stored in the form of "condition-action" pair (IF-THEN), and the network state, text features, and system load are divided into condition variables of multiple dimensions, each variable sets a threshold or interval, forms a combined condition, and configures parameter combinations under different condition combinations.

[0097] For example, for the network stability + high importance text scenario, the trigger condition configured in the rule base of the application can be: bandwidth > 10Mbps, delay < 50ms, packet loss rate < 1%, semantic importance = high, text length = 1-10KB, CPU usage < 30%, memory occupancy < 40%. The parameter configuration can be: transmission protocol: TCP; priority: highest level; compression algorithm: Zstd (balance speed and compression rate); compression rate: level 6 (balance compression ratio and computing resources). Among them, TCP can ensure reliability and avoid loss of high importance text, high priority can ensure priority transmission, and Zstd algorithm can provide efficient compression when resources are sufficient.

[0098] For the network congestion + short text notification scenario, the application can configure the trigger condition as: bandwidth < 1Mbps, delay > 200ms, packet loss rate > 5%, semantic importance = medium, text length < 1KB, CPU usage > 70%, memory occupancy > 80%. The parameter configuration can be: transmission protocol: UDP; priority: medium level; compression algorithm: Snappy (lightweight compression); compression rate: level 3 (reduce computing overhead). Since UDP has no connection feature to reduce delay, it is suitable for short text, medium priority can avoid preempting critical resources, and Snappy algorithm has fast compression speed, suitable for high load system.

[0099] For the system high load + long text transmission scenario, the trigger condition configured by the application can be: bandwidth = medium (1-10Mbps), delay = 50-200ms, packet loss rate = 1-5%, semantic importance = low, text length > 10KB, CPU usage > 70%, memory occupancy > 80%. The parameter configuration can be: transmission protocol: TCP (enable congestion control); priority: lowest level; compression algorithm: LZ4 (extremely fast compression)

[0100] Compression rate: Level 1 (minimizes CPU usage). This allows TCP congestion control to adapt to moderate networks, while maintaining a low priority to avoid impacting core tasks. The LZ4 algorithm achieves compression speeds of GB / s, making it suitable for quickly processing long text files under high load.

[0101] Furthermore, when the match is successful, that is, when a parameter combination that fully matches or is closest to the current condition variable is found, the present application can directly use this parameter combination as the target parameter combination. This method can quickly and accurately determine the target parameter combination, avoid complex calculations and optimization processes, and improve transmission efficiency. For example, when the present application transmits a technical document containing urgent instructions (text length 5KB, semantic importance = high), the network bandwidth suddenly drops from 10Mbps to 2Mbps (packet loss rate rises to 3%), and the server CPU utilization rate rises from 20% to 85%. The rule matching process of the present application is as follows:

[0102] Initially, configure TCP+Zstd level 6 for "network stability+high importance";

[0103] After the network status changes, the rule base matches "Bandwidth < 10 Mbps, Packet Loss Rate > 1%, and CPU > 70%."

[0104] Trigger parameter adjustments: protocol switched to UDP (low latency), compression algorithm changed to LZ4 level 3 (reduces CPU load), priority kept as highest (due to high semantic importance).

[0105] However, faced with the complexity and variability of actual transmission environments, the preset rule base may not cover all possible condition combinations. In this case, when the matching result is that no corresponding parameter combination is found, this application will instead use a multi-objective optimization algorithm to generate a target parameter combination. This step takes into account multiple transmission goals, such as transmission efficiency, stability, and resource utilization, and generates the optimal parameter combination by weighing these goals. Although this method is relatively complex, it can flexibly cope with various transmission environments and ensure the efficiency and stability of the transmission process.

[0106] In summary, when determining the target parameter combination, this application takes into account both the fast matching of the preset rule base and the flexible response of the multi-objective optimization algorithm, thereby achieving efficient and stable transmission in different transmission environments.

[0107] In one embodiment, generating target parameter combinations corresponding to conditional variables of various dimensions through a multi-objective optimization algorithm in S133 may include:

[0108] S1331: The constraints are shortest transmission time, highest stability, and resource usage not exceeding a preset system threshold.

[0109] S1332: Construct a multi-objective function based on the conditional variables of each dimension and the constraint conditions, and optimize and solve the multi-objective function to obtain a solution result.

[0110] S1333: Generate target parameter combinations corresponding to the conditional variables of each dimension according to the solution results.

[0111] In this embodiment, when the preset rule base cannot match a parameter combination that fully corresponds to the current conditional variables, this application can use a multi-objective optimization algorithm to generate a target parameter combination. The core of this step is to construct a multi-objective function that can reflect multiple transmission objectives (transmission time, stability, resource utilization), and then find the optimal parameter combination through optimization.

[0112] Specifically, this application can first define the various objectives of the multi-objective function based on constraints such as minimizing transmission time, maximizing stability, and ensuring that resource utilization does not exceed a preset system threshold. These objectives can include the inverse of transmission time (to maximize transmission speed), a quantitative indicator of stability (such as a negative exponential function of packet loss rate to minimize the impact of packet loss), and a linear function of resource utilization (to minimize CPU and memory usage).

[0113] This application then integrates these variables and objectives into a unified multi-objective function based on the conditional variables of various dimensions (such as network bandwidth, latency, packet loss rate, text length, semantic importance, etc.) and the aforementioned constraints. This function will be a complex nonlinear expression designed to balance the conflicts and trade-offs between multiple transmission objectives.

[0114] After constructing multiple objective functions, this application can employ an appropriate optimization algorithm to solve these functions. This algorithm can be an evolutionary computation technique such as a genetic algorithm, particle swarm optimization, or simulated annealing, or a gradient-based optimization method such as the conjugate gradient method or Newton's method. The specific choice depends on the complexity of the problem and the efficiency of the solution, and is not limited here. The goal of the optimization algorithm is to find the parameter combination that optimizes the multiple objective functions (i.e., minimizes transmission time, maximizes stability, and minimizes resource utilization) while satisfying all constraints.

[0115] After optimization, this application can obtain one or more optimal solutions (i.e., target parameter combinations). These solutions may correspond to different transmission strategies and configuration parameters, such as transmission protocols, priorities, compression algorithms, and compression ratios. Based on actual needs and environmental changes, this application can select the most appropriate solution as the target parameter combination and apply it to subsequent text transmission processes.

[0116] In this way, the application can generate the optimal parameter combination flexibly when facing complex and changeable transmission environment, and ensure efficient and stable transmission of the text.

[0117] In one embodiment, the target parameter combination can include a target transmission protocol, a target transmission priority, a target compression algorithm, and a compression rate.

[0118] In S130, the target parameter combination is used to package, prioritize, and compress the text to be transmitted, and a compressed text to be transmitted is obtained, which can include:

[0119] S131: The text to be transmitted is packaged according to the target transmission protocol.

[0120] S132: The packaged text to be transmitted is prioritized according to the target transmission priority.

[0121] S133: When the packaged text to be transmitted is arranged at the head of the queue, the packaged text to be transmitted is compressed according to the target compression algorithm and the compression rate, and a compressed text to be transmitted is obtained.

[0122] In this embodiment, after determining the target parameter combination, the application can package, prioritize, and compress the text to be transmitted according to these parameters. This step aims to configure the optimal transmission parameters for the text to be transmitted according to network status, system load, and text characteristics, so as to ensure efficient and stable transmission of the text to the target receiving end.

[0123] Specifically, the packaging process refers to packaging the text to be transmitted into a data packet that conforms to the protocol specification according to the target transmission protocol. Different transmission protocols have different data packet formats and requirements, so the application can perform corresponding packaging processing according to the target transmission protocol. This step ensures that the data packet can be correctly transmitted on the network and correctly parsed by the target receiving end.

[0124] Prioritizing refers to sorting the packaged text to be transmitted according to the target transmission priority. In the case of network congestion or high system load, texts with high priority can be transmitted first, thereby ensuring timely delivery of critical information. The application can flexibly cope with different transmission environments and requirements by setting appropriate priorities for target texts.

[0125] The compression process refers to compressing the packaged text to be transmitted according to the target compression algorithm and the compression rate. Compression algorithms can reduce the data volume of the text, thereby reducing transmission time and resource occupancy. The compression rate determines the trade-off between compression effect and computational overhead. The application can minimize the impact on system resources while ensuring transmission efficiency by selecting appropriate compression algorithms and compression rates for target texts.

[0126] When the encapsulated text to be transmitted reaches the head of the queue, the application can compress it according to the target compression algorithm and compression ratio to obtain the compressed text to be transmitted. This step ensures that the text has been optimized before transmission, which can improve transmission efficiency and stability while ensuring transmission quality.

[0127] After encapsulation, priority queuing, and compression, the compressed text to be transmitted will be sent to the target receiving end according to the target transmission protocol. At the receiving end, the receiving device will parse, decompress, and reassemble the received data packets to restore the original text information.

[0128] In one embodiment, Figure 3 As shown, Figure 3 This is a diagram showing the structure of the data collection, dynamic parameter trade-off, text transmission, feedback and optimization process provided in the embodiment of the present application; the method may also include:

[0129] S140: Acquire feedback information sent by the receiving end.

[0130] S141: After performing parameter association on the feedback information, an optimized text is generated.

[0131] S142: Optimize and adjust parameters in the dynamic trade-off model according to the optimization text.

[0132] S143: Alternatively, the rules in the rule base are optimized and adjusted according to the optimization text.

[0133] In this embodiment, Figure 3 As shown, after the text transmission is completed, in order to ensure the continuous optimization and improvement of the transmission effect, this application also designs a feedback and optimization mechanism. This mechanism enables the transmission solution to continuously learn and adapt to the changing environment, and continuously optimizes the transmission performance in long-term use, which is more advantageous than the fixed parameter method.

[0134] Specifically, this application can obtain feedback information sent by the receiving end, including but not limited to key indicators such as text transmission success rate, packet loss rate, transmission time, receiving end system load, etc. This feedback data provides valuable reference for subsequent optimization.

[0135] Next, the application can associate parameters with these feedback information. This step aims to match the various indicators in the feedback information with the parameter settings during the transmission process, thereby identifying the key factors affecting the transmission effect. For example, if a transmission has a high packet loss rate, the application will associate and check the network bandwidth, transmission protocol, compression algorithm and other parameter settings at that time to analyze whether these factors have caused the decline in transmission quality.

[0136] After completing the parameter association, this application can generate an optimization document based on the analysis results, such as a condition-result pair. In addition, the optimization document of this application can also include adjustment suggestions for transmission parameters, or optimization suggestions for the dynamic trade-off model and rule base. For example, if the analysis finds that the TCP protocol performs poorly in a congested network, this application can recommend switching to the UDP protocol or adjusting TCP's congestion control parameters.

[0137] Subsequently, this application can optimize and adjust the parameters in the dynamic trade-off model or the rules in the rule base based on this optimization text. This step aims to continuously adjust and optimize the transmission strategy based on actual transmission results and feedback information to adapt to the ever-changing network environment and system load.

[0138] In this way, this application can not only achieve efficient and stable transmission of text, but also continuously learn and optimize transmission strategies to ensure the best transmission effect under any circumstances.

[0139] In one embodiment, after the feedback information is associated with parameters in S141, generating an optimized text may include:

[0140] S1411: Divide the feedback information into multiple data types, and determine data indicators corresponding to each data type.

[0141] S1412: After cleaning and standardizing the data indicators corresponding to various data types, the standardized data indicators are associated with the conditional parameters when transmitting the text, and an optimized text is generated.

[0142] In this embodiment, during the process of parameter association of feedback information and generating optimized text, the application can classify the feedback information into multiple data types, such as transmission results, transmission efficiency, resource consumption, semantic matching, etc., and determine key data indicators corresponding to each data type, such as reception integrity (whether there is packet loss), data error rate, end-to-end latency, throughput, sender CPU usage, memory usage, and text restoration accuracy (such as the semantic integrity of instruction text). These indicators can comprehensively reflect various situations and factors in the transmission process, providing strong support for subsequent analysis and optimization.

[0143] Next, the application can clean and standardize the data indicators corresponding to various data types. This step aims to eliminate noise and outliers in the data to ensure data accuracy and reliability. At the same time, through standardization, data of different dimensions and ranges are unified to the same scale, facilitating subsequent data correlation and analysis.

[0144] After completing data cleaning and standardization, this application can correlate the standardized data metrics with the conditional parameters used during text transmission. This step is the core of parameter correlation, aiming to identify the key factors and parameter combinations that influence transmission effectiveness. By comparing and analyzing transmission results and data metrics under different conditions, the inherent connections and patterns between parameters and transmission effectiveness can be revealed.

[0145] Finally, this application generates an optimization document based on the results of the correlation analysis. This document not only includes condition-result pairs but also includes recommendations for adjusting transmission parameters and optimizing the dynamic trade-off model and rule base. These recommendations are based on actual transmission results and feedback, are targeted and actionable, and can provide strong guidance for subsequent transmission optimization.

[0146] In this way, the present application can achieve in-depth analysis and effective utilization of feedback information, continuously optimize transmission strategies, and improve transmission effects.

[0147] In one embodiment, optimizing and adjusting the parameters in the dynamic trade-off model according to the optimization text in S142 may include:

[0148] S1421: Optimize and adjust the parameters in the dynamic trade-off model according to a preset parameter optimization strategy and the optimization text.

[0149] In this embodiment, when optimizing and adjusting the parameters in the dynamic trade-off model according to the optimization text, the present application can obtain a pre-set parameter optimization strategy. The parameter optimization strategy refers to the method used when optimizing and adjusting the model parameters, such as a parameter adjustment method based on error backpropagation or a strategy optimization method based on reinforcement learning. Then, the present application can optimize and adjust the parameters in the dynamic trade-off model according to the parameter optimization strategy and the optimization text.

[0150] For example, when the parameter optimization strategy of this application is a parameter adjustment method based on error back propagation, this type of method is suitable for optimizing the internal algorithm parameters of the model (such as weight coefficients and thresholds). The specific implementation steps include the following:

[0151] Define the optimization goal:

[0152] Minimize packet loss rate (objective function: Loss = packet loss rate × importance weight);

[0153] Maximize the cost-effectiveness of throughput and resource consumption (objective function: Score = throughput / (CPU usage + memory usage)).

[0154] Parameter iterative optimization:

[0155] Adopting gradient descent algorithm, the parameters are calculated based on feedback data to adjust the direction;

[0156] Example: If it is found that the packet loss rate of high-importance text is high, increase the weight coefficient of "semantic importance" in protocol selection (for example, from 0.3 to 0.5).

[0157] When the parameter optimization strategy of this application is a reinforcement learning-based strategy optimization method, this method is applicable to systems that regard the dynamic trade-off process as a "state-action-reward" system. The specific implementation steps include the following:

[0158] State: network status + text features + system load;

[0159] Action: parameter combination (protocol / priority / compression rate);

[0160] Reward: Feedback score of data conversion (e.g., +10 for successful transmission, -5 for packet loss)

[0161] Training method:

[0162] Use Q-Learning or Deep Q Network (DQN) to train the model by optimizing samples from history.

[0163] Example: When the network is congested and the action of selecting UDP receives "Low Latency Bonus +8", the probability of selecting UDP in this state is increased.

[0164] In this way, the present application can flexibly adjust the parameters in the dynamic trade-off model based on actual transmission performance and feedback information to adapt to the ever-changing transmission environment and system load. This not only improves the flexibility and adaptability of transmission, but also ensures the continuous optimization and improvement of transmission performance.

[0165] In one embodiment, optimizing and adjusting the rules in the rule base according to the optimization text in S143 may include:

[0166] S1431: Adjust the priority of rules in the rule base according to the optimized text, generate new rules, and / or resolve conflicting rules.

[0167] In this embodiment, when optimizing and adjusting the rules in the rule base based on the optimization document, the present application may first analyze the adjustment suggestions or newly discovered transmission patterns proposed in the optimization document. These suggestions or patterns may involve various conditions and parameter settings during the transmission process, as well as their specific impact on the transmission effect.

[0168] Next, the application can adjust the rule priorities in the rule base based on these analysis results. In the rule base, different rules can correspond to different transmission strategies and condition judgments. By adjusting the priorities of the rules, the application can ensure that when facing a specific transmission environment, those strategies that have been verified and are better can be prioritized. For example, when the execution result of a rule is not as expected for N consecutive times (such as packet loss rate > 5%), the application can reduce the rule priority or add additional conditions, for example, the original rule "use UDP when packet loss rate > 3%" is adjusted to "use UDP when packet loss rate > 3% and text importance ≠ high" if it results in loss of high importance text.

[0169] In addition, the application can also generate new rules based on the optimized text. These new rules can cover transmission scenarios or condition combinations that have not been considered before, thereby further enriching and perfecting the content of the rule base. The generation of new rules not only improves the diversity and flexibility of transmission strategies, but also provides more possibilities for subsequent optimization. For example, the application can perform cluster analysis on the optimization samples to find new condition combination patterns. Example: when "bandwidth < 500kbps and text length > 5KB" is found, the transmission time is generally > 2s, automatically generate the rule: IF bandwidth < 500kbps AND text length > 5KB THEN, protocol = TCP (enable segmented transmission), compression algorithm = Brotli (high compression rate), priority = medium level (avoid long text from occupying resources). In addition, the application also supports administrators manually adding rules according to business needs (such as adding "financial transaction text forced to use TCP + highest priority").

[0170] At the same time, in the optimization process of the rule base, the application also needs to pay attention to resolving possible rule conflicts. Rule conflicts can be caused by contradictions between condition judgments or strategy choices of different rules. In order to eliminate these conflicts, the application can perform in-depth analysis on the conflicting rules, find their commonalities and differences, and accordingly perform operations such as merging, modifying or deleting. For example, the application can sort the priorities, such as semantic importance rule > network state rule > system load rule, or select rules with higher success rates by simulating execution based on historical data, to eliminate rule conflicts. This can ensure the consistency and accuracy of the rule base, thereby improving the reliability and stability of the transmission strategy.

[0171] For example, to optimize the transmission feedback of financial transaction instructions, the initial configuration is as follows: rule: "Financial text → TCP + highest priority + Zlib compression level 9"; model parameters: semantic importance weight = 0.4. Receiver feedback: 10% of transaction instructions timed out due to excessive compression (latency > 300ms). Sender resource data: CPU usage reached 90% at Zlib compression level 9, resulting in stuttering. This application uses the above process to reduce the semantic importance weight from 0.4 to 0.3, increase the "latency sensitivity" weight by 0.2, and add a new rule to the rule base: IF text type = financial transaction AND network latency > 200ms THEN, compression algorithm = LZ4 (compression speed increased by 5x), compression level = 5 (balancing compression ratio and speed).

[0172] In summary, this application continuously improves the adaptability and effectiveness of transmission strategies by optimizing and adjusting the rules in the rule base based on optimized text. This not only helps cope with complex and changing transmission environments, but also continuously optimizes transmission performance over long-term use, providing users with more efficient and stable text transmission services.

[0173] The text transmission device provided in an embodiment of the present application is described below. The text transmission device described below and the text transmission method described above can be referenced to each other.

[0174] In one embodiment, Figure 4 As shown, Figure 4 This is a structural diagram of a text transmission device provided in an embodiment of the present application. The present application also provides a text transmission device, which may include a data acquisition module 210, a conditional variable partitioning module 220, and a text transmission module 230, specifically including the following:

[0175] The data acquisition module 210 is used to acquire the text to be transmitted, the current network status and the current system load, and determine the text features corresponding to the text to be transmitted.

[0176] The condition variable division module 220 is configured to divide the current network status, the current system load, and the text features into condition variables of multiple dimensions.

[0177] The text transmission module 230 is used to determine the target parameter combination corresponding to the conditional variables of each dimension, encapsulate, prioritize and compress the text to be transmitted according to the target parameter combination, and send the compressed text to be transmitted to the receiving end.

[0178] In the above embodiment, after obtaining the text to be transmitted, the current network status and the current system load, the text features corresponding to the text to be transmitted can be determined first; then, the current network status, the current system load and the text features are divided into conditional variables of multiple dimensions; then, the target parameter combination corresponding to the conditional variables of each dimension is determined, and the text to be transmitted is encapsulated, prioritized and compressed according to the target parameter combination, and finally the compressed text to be transmitted is sent to the receiving end. In this process, the present application monitors the network status in real time, dynamically selects the transmission protocol and adjusts the parameters, and automatically switches to a low-bandwidth demand protocol when the network is congested, effectively reducing delays and packet loss rates, and ensuring transmission stability. In addition, the present application can also assign transmission priorities based on text features to ensure that key information is transmitted first; at the same time, a suitable compression method is selected according to the text features to avoid waste of resources, thereby improving overall transmission efficiency.

[0179] In one embodiment, the present application also provides a computer-readable storage medium, which stores computer-readable instructions. When the computer-readable instructions are executed by one or more processors, the one or more processors execute the steps of the text transmission method described in any of the above embodiments.

[0180] In one embodiment, the present application further provides a computer device, including: one or more processors, and a memory.

[0181] The memory stores computer-readable instructions, and when the computer-readable instructions are executed by the one or more processors, the steps of the text transmission method described in any one of the above embodiments are performed.

[0182] Schematically, as Figure 5 As shown, Figure 5 This is a schematic diagram of the internal structure of a computer device provided in an embodiment of the present application. The computer device 300 can be provided as a server. Figure 5 Computer device 300 includes a processing component 302, which further includes one or more processors, and a memory resource represented by memory 301 for storing instructions executable by processing component 302, such as an application. The application stored in memory 301 may include one or more modules, each corresponding to a set of instructions. In addition, processing component 302 is configured to execute the instructions to perform the text transmission method of any of the above embodiments.

[0183] The computer device 300 may further include a power supply component 303 configured to perform power management of the computer device 300, a wired or wireless network interface 304 configured to connect the computer device 300 to a network, and an input / output (I / O) interface 305. The computer device 300 may operate based on an operating system stored in the memory 301, such as Windows Server™, Mac OS X™, Unix™, Linux™, Free BSD™, or the like.

[0184] Those skilled in the art will understand that Figure 5 The structure shown in the figure is only a block diagram of a part of the structure related to the solution of the present application, and does not constitute a limitation on the computer device to which the solution of the present application is applied. The specific computer device may include more or fewer components than shown in the figure, or combine certain components, or have a different component arrangement.

[0185] Finally, it should be noted that, in this document, relational terms such as first and second, etc., are used only to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any actual relationship or order between these entities or operations. Moreover, the terms "comprises," "comprising," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or device comprising a series of elements includes not only those elements, but also other elements not explicitly listed, or elements inherent to such process, method, article, or device. In the absence of further limitations, an element defined by the phrase "comprising a ..." does not exclude the presence of additional identical elements in the process, method, article, or device comprising the element.

[0186] The various embodiments in this specification are described in a progressive manner, and each embodiment focuses on the differences from other embodiments. The various embodiments can be combined as needed, and the same or similar parts can be referenced to each other.

[0187] The above description of the disclosed embodiments is intended to enable one skilled in the art to implement or use the present application. Various modifications to these embodiments will be readily apparent to one skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the present application. Therefore, the present application is not limited to the embodiments shown herein, but is intended to conform to the widest scope consistent with the principles and novel features disclosed herein.

Claims

1. A text transmission method, characterized in that: The method comprises: Obtaining the text to be transmitted, the current network status, and the current system load, and determining text features corresponding to the text to be transmitted; Dividing the current network state, the current system load, and the text features into conditional variables of multiple dimensions; Determine the target parameter combination corresponding to the conditional variables of each dimension, encapsulate, prioritize and compress the text to be transmitted according to the target parameter combination, and send the compressed text to be transmitted to the receiving end.

2. The text transmission method according to claim 1, characterized in that The determining of text features corresponding to the text to be transmitted includes: Extracting the text length and keywords of the text to be transmitted to obtain an extraction result; Performing a semantic importance evaluation on the text to be transmitted to obtain an evaluation result; Determine text features of the text to be transmitted based on the extraction result and the evaluation result.

3. The text transmission method according to claim 1, characterized in that The dividing the current network state, the current system load, and the text features into conditional variables of multiple dimensions includes: The current network status, the current system load and the text features are input into a preset dynamic trade-off model, and the current network status, the current system load and the text features are subjected to conditional variable standardization processing through the dynamic trade-off model to obtain conditional variables of multiple dimensions.

4. The text transmission method according to claim 1, wherein: Determining the target parameter combination corresponding to the conditional variables of each dimension includes: Matching the condition variables of each dimension with a preset rule base to obtain a matching result, wherein the preset rule base contains parameter combinations under different condition combinations; When the matching result is that a corresponding parameter combination is matched, the matched parameter combination is used as the target parameter combination corresponding to the conditional variables of each dimension; When the matching result is that no corresponding parameter combination is matched, a target parameter combination corresponding to the conditional variables of each dimension is generated through a multi-objective optimization algorithm.

5. The text transmission method according to claim 4, characterized in that: The method of generating target parameter combinations corresponding to conditional variables of each dimension through a multi-objective optimization algorithm includes: The constraints are shortest transmission time, highest stability, and resource utilization not exceeding the preset system threshold; Constructing a multi-objective function based on the conditional variables of each dimension and the constraint conditions, and optimizing and solving the multi-objective function to obtain a solution result; A target parameter combination corresponding to the conditional variables of each dimension is generated according to the solution result.

6. The text transmission method according to any one of claims 1 to 5, characterized in that: The target parameter combination includes a target transmission protocol, a target transmission priority, a target compression algorithm and a compression rate; The encapsulating, prioritizing, and compressing the text to be transmitted according to the target parameter combination to obtain a compressed text to be transmitted includes: Encapsulating the text to be transmitted according to the target transmission protocol; Prioritize the encapsulated text to be transmitted according to the target transmission priority; When the encapsulated text to be transmitted is placed at the head of the queue, the encapsulated text to be transmitted is compressed according to the target compression algorithm and compression ratio to obtain the compressed text to be transmitted.

7. The text transmission method according to claim 3 or 4, characterized in that: The method further comprises: Obtaining feedback information sent by the receiving end; After performing parameter association on the feedback information, an optimized text is generated; Optimizing and adjusting parameters in the dynamic trade-off model according to the optimization text; Alternatively, the rules in the rule base are optimized and adjusted according to the optimization text.

8. The text transmission method according to claim 7, characterized in that: After the feedback information is parameter-associated, generating an optimized text includes: Dividing the feedback information into multiple data types and determining data indicators corresponding to each data type; After cleaning and standardizing the data indicators corresponding to various data types, the standardized data indicators are associated with the conditional parameters when transmitting the text, and optimized text is generated.

9. The text transmission method according to claim 7, characterized in that: Optimizing and adjusting the parameters in the dynamic trade-off model according to the optimization text includes: The parameters in the dynamic trade-off model are optimized and adjusted according to the preset parameter optimization strategy and the optimization text.

10. The text transmission method according to claim 7, characterized in that: Optimizing and adjusting the rules in the rule base according to the optimization text includes: According to the optimized text, the priority of the rules in the rule base is adjusted, new rules are generated, and / or conflicting rules are resolved.

11. A text transmission device, characterized in that: include: A data acquisition module, configured to acquire the text to be transmitted, the current network status, and the current system load, and determine text features corresponding to the text to be transmitted; A condition variable division module, configured to divide the current network state, the current system load, and the text features into condition variables of multiple dimensions; The text transmission module is used to determine the target parameter combination corresponding to the conditional variables of each dimension, encapsulate, prioritize and compress the text to be transmitted according to the target parameter combination, and send the compressed text to be transmitted to the receiving end.

12. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores computer-readable instructions, and when the computer-readable instructions are executed by one or more processors, the one or more processors execute the steps of the text transmission method according to any one of claims 1 to 10.

13. A computer device, characterized in that: include: one or more processors, and memory; The memory stores computer-readable instructions, which, when executed by the one or more processors, execute the steps of the text transmission method according to any one of claims 1 to 10.

Images