A data transmission system based on artificial intelligence services

Abstract

The application discloses a data transmission system based on artificial intelligence service, and relates to the technical field of data transmission.The system comprises a data acquisition module, a word library construction module, a word segmentation comparison module, a transmission analysis module, a conflict analysis module and a transmission optimization module.The data acquisition module performs word segmentation on an artificial intelligence service data packet to construct a word segmentation data group.The model construction module constructs a key word library containing weights according to a scene.The word segmentation comparison module calculates word frequency and weight ratio to obtain a word group priority value to determine a data packet transmission sequence.The transmission analysis module compares word frequency of data packets with network maximum byte values to determine conflicts.The conflict analysis module calculates vector dot products and continuous invalid word frequency Wx to determine the priority sequence of equivalent data packets.The transmission optimization module optimizes data packets and sends the data packets to a network.The application can optimize data packet transmission, improve efficiency and provide a basis for data transmission optimization.

Description

Technical Field

[0001] This invention relates to the field of data transmission technology, and more specifically to a data transmission system based on artificial intelligence services. Background Technology

[0002] With the widespread application of artificial intelligence services, data packets need to be effectively processed and optimized during data transmission to ensure timely and accurate transmission of important information, avoid network congestion and conflicts, and improve overall system performance.

[0003] In existing technologies, data transmission systems do not intelligently prioritize data packets based on their importance. For example, in medical data transmission, current technologies may simply transmit packets according to their arrival order or a first-come, first-served principle, without considering critical factors such as the patient's potential condition. This could lead to delays in processing packets with potentially serious conditions, affecting the timeliness of medical treatment.

[0004] In the face of limited network bandwidth and multiple data packets competing for transmission resources simultaneously, existing technologies often lack effective conflict handling mechanisms. They may not analyze the byte values ​​of data packets, making it impossible to determine whether sending multiple data packets of equal importance simultaneously exceeds the network's carrying capacity, nor can they make reasonable selections or adjustments based on the importance of the data packets.

[0005] Therefore, the present invention provides a data transmission system based on artificial intelligence services. Summary of the Invention

[0006] The purpose of this invention is to provide a data transmission system based on artificial intelligence services to solve at least one of the aforementioned problems in the prior art.

[0007] In a first aspect, the present invention provides a data transmission system based on artificial intelligence services, comprising:

[0008] Data acquisition module: Acquires data packets from artificial intelligence services, performs word segmentation on the text in the data packets, constructs word segmentation data groups, and organizes the word segmentation data groups of the data packets;

[0009] Lexicon building module: Based on the scenario of artificial intelligence services, it builds a key lexicon to provide a basis for word group comparison in data packages;

[0010] The word segmentation and comparison module performs numerical analysis on the word segmentation data groups and key word groups in the lexicon to obtain the word group priority value Cy. Based on the word group priority value Cy, it constructs a data sending sequence and determines the transmission order of data packets.

[0011] Transmission Analysis Module: Data packets with the same priority value Cy in the data transmission sequence are marked as equivalent data packets. The byte values ​​of all equivalent data packets are obtained and compared with the maximum byte value sent simultaneously by the current transmission network to determine whether conflict handling of equivalent data packets is required.

[0012] Conflict Analysis Module: If conflict handling is required for equivalent data packets, construct the word frequency vector A and weight vector B of the equivalent data packets and perform numerical analysis to construct a transmission optimization sequence and determine the priority order of equivalent data packets;

[0013] Transmission optimization module: Based on the transmission optimization sequence, invalid word frequency groups are removed from the data packets of the transmission optimization sequence. Then, the word groups of the processed data packets of the transmission optimization sequence are analyzed to obtain optimized data packets, and the optimized data packets are sent.

[0014] Secondly, the present invention provides a data transmission method based on artificial intelligence services, including:

[0015] Step 1: Obtain data packets from the artificial intelligence service, perform word segmentation on the text in the data packets, construct word segmentation data groups, and organize the word segmentation data groups in the data packets;

[0016] Step 2: Based on the scenario of artificial intelligence services, construct a key word library to provide a basis for word comparison of data packets;

[0017] Step 3: Compare and analyze the word segmentation data sets and key word groups to obtain the word frequency (TF). i For each data packet, the term frequency TF i and weight Qz j Numerical analysis is performed to obtain the phrase priority value Cy. Based on the phrase priority value Cy, a data transmission sequence is constructed to determine the transmission order of data packets.

[0018] Step 4: Data packets with the same priority value Cy in the data transmission sequence are marked as equivalent data packets. The byte values ​​of all equivalent data packets are obtained and compared with the maximum byte value sent simultaneously by the current transmission network to determine whether conflict handling of equivalent data packets is required.

[0019] Step 5: If conflict resolution is required for equivalent data packets, construct the word frequency vector A and weight vector B of the equivalent data packets. Perform numerical calculations on word frequency vector A and weight vector B to obtain the word group vector dot product Dj. Construct a word frequency run analysis model for word frequency vector A to obtain the continuous invalid word frequency Wx. Combine the word group vector dot product Dj and the continuous invalid word frequency Wx for numerical analysis to construct a transmission optimization sequence and determine the priority order of equivalent data packets.

[0020] Step 6: Based on the transmission optimization sequence, invalid word frequency groups are removed from the data packets of the transmission optimization sequence. Sentence analysis is performed on the word groups of the processed data packets of the transmission optimization sequence to obtain optimized data packets. The optimized data packets are then sent.

[0021] The beneficial effects of this invention are:

[0022] 1. This invention, by constructing a key vocabulary library and calculating word frequencies and weights, can identify word groups in data packets that are closely related to specific scenarios. Because the key vocabulary library has higher weights and the word groups have higher priority values ​​after word frequency calculation, they will be prioritized for transmission. This content-based priority determination avoids the blindness of traditional transmission methods that rely solely on the order of data packet arrival or simple rules for sorting.

[0023] 2. The transmission analysis module compares the data transmission sequence, specifically the byte values ​​of data packets with the same term priority value, with the maximum byte value simultaneously transmitted on the current network. This allows for timely detection of potential transmission conflicts. For example, when network bandwidth is limited, and multiple data packets compete for transmission resources with a total byte value potentially exceeding network capacity, the system can accurately identify and activate the conflict resolution mechanism. Based on the conflict analysis results, a transmission optimization sequence is constructed to further divide equivalent data packets and determine their priority. During processing, the system eliminates invalid term frequencies, reduces data packet size, and improves transmission efficiency. Simultaneously, by rationally arranging the transmission order of data packets, important data is prioritized for network transmission, further optimizing the overall performance of data transmission. Attached Figure Description

[0024] To more clearly illustrate the technical solutions in the embodiments of the present invention, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0025] Figure 1 This is a flowchart of the phrase extraction and analysis method provided in Embodiment 1 of the present invention;

[0026] Figure 2 This is a flowchart of the data analysis and optimization method provided in Embodiment 2 of the present invention;

[0027] Figure 3 This is a block diagram of a data transmission system based on artificial intelligence services provided by the present invention;

[0028] Figure 4 This is a schematic diagram of the structure of the computer device provided in Embodiment 4 of the present invention. Detailed Implementation

[0029] To enable those skilled in the art to better understand the present invention, the technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings of the embodiments of the present invention. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort should fall within the scope of protection of the present invention.

[0030] Example 1

[0031] Figure 1 This is a flowchart of a phrase extraction and analysis method provided in Embodiment 1 of the present invention. This embodiment of the invention is applicable to the analysis of data packets generated by artificial intelligence services. The phrase extraction and analysis method can be executed by a data transmission system based on artificial intelligence services. This data transmission system based on artificial intelligence services can be implemented by software and / or hardware, and can be configured in a computer device. Optionally, the computer device can be an electronic device, such as a laptop, desktop computer, or smart tablet, etc. This embodiment of the invention does not impose any limitations on this.

[0032] Data acquisition module: Acquires data packets from artificial intelligence services, performs word segmentation on the text in the data packets, constructs word segmentation data groups, and organizes the word segmentation data groups of the data packets;

[0033] Obtain data packets from artificial intelligence services, perform word segmentation on the text in the data packets, and construct word segmentation data groups;

[0034] In some embodiments, a JSON-formatted data packet sent by an artificial intelligence service is obtained, and the JSON-formatted string is converted into a Python list using the built-in Python json.load() function. The string is then stored in the list, and the jieba library for natural language processing is used to perform word segmentation on the text in the list.

[0035] For example, if the data packet obtained in the artificial intelligence service is analyzed and processed to obtain a list of strings, and the text in the list is "patients need help", the jieba library is used to perform word segmentation on the sentence "patients need help", select the word groups, and construct the word segmentation data group ["patient", "need", "help"].

[0036] Lexicon building module: Based on the scenario of artificial intelligence services, it builds a key lexicon to provide a basis for word group comparison in data packages;

[0037] Based on AI service scenarios, a key vocabulary database is constructed, which contains multiple word phrases.

[0038] Based on the varying severity of terms within the key terminology database, different weights are assigned, and the weight of each term in the key terminology database is labeled as Qz. j， Where j is the number of the phrase in the key vocabulary;

[0039] It should be noted that the key term library and its weight are set by professionals in the field based on their experience. For example, in an AI online consultation platform, building a medical-related key term library is beneficial for making a preliminary judgment on the patient's critical condition and provides a basis for the word comparison of subsequent data packets.

[0040] For example, the key vocabulary for AI services in a medical setting could be set as [low-grade fever, persistent, palpitations, ..., shortness of breath], where "low-grade fever" is a phrase in the key vocabulary.

[0041] Word segmentation comparison module: compares and analyzes word segments and key word groups in the vocabulary database to obtain word frequency (TF). i For each data packet, the term frequency TF i and weight Qz j Numerical analysis is performed to obtain the phrase priority value Cy. Based on the phrase priority value Cy, a data transmission sequence is constructed to determine the transmission order of data packets.

[0042] Based on the segmented data set and the key vocabulary, the frequency of word phrases in the segmented data set appearing in the key vocabulary is calculated, i.e., the term frequency (TF). i , where i is the word group number in the word segmentation data group;

[0043] The word segmentation data group is labeled Cz, and each word in the word segmentation data group is labeled Cz. i ;

[0044] For example, for the word segmentation data group ["patient", "need", "treatment"], Cz1 ​​represents "treatment";

[0045] Through formula Obtain the term frequency (TF) of a single word phrase in the key vocabulary within the segmented data set. i , where n i This represents the word group Cz in the segmented data set. i The number of times it appears in the key vocabulary, n j This indicates the number of all phrases within the key vocabulary.

[0046] For example, if the word "sore" in the segmented data group ["patient", "need", "treatment"] appears 3 times in the key word library, and the total number of word groups in the key word library is 300, then the word frequency is 3 / 500;

[0047] Each word frequency TF in the segmented data group iPerform summation to obtain the word frequency sum;

[0048] Obtain the number of word groups in the segmented data set, calculate the ratio of the sum of word frequencies to the number of word groups in the segmented data set, and label the word frequency ratio as TFb;

[0049] If a word phrase in a segmented data set exists in the key vocabulary, the weight of that word phrase in the key vocabulary will be Qz. j Perform summation to obtain the sum of word segmentation weights;

[0050] The word segmentation weight is calculated as a ratio to the number of word segments in the word segmentation data group to obtain the word segmentation weight ratio, which is then labeled as Qzb.

[0051] Based on the word frequency ratio TFb and the phrase weight ratio Qzb, the phrase priority value Cy is calculated;

[0052] Through the formula: Obtain the phrase priority value Cy, where a=1.065 and b=1.253;

[0053] For example, in a medical scenario, the word segmentation data group Cz = ["fever", "cough", "headache", "fatigue"] has a key word library of ["high fever", "persistent cough", "severe headache", "extreme fatigue", "low fever", "occasional cough", "mild headache", "slight fatigue"], and the weights of each word group in the key word library are [0.8, 0.7, 0.6, 0.5, 0.3, 0.2, 0.1, 0.1], respectively. The total number of word groups in the key word library is 8. The calculated word frequency ratio TFb is 0.25, the word group weight ratio Qzb is 0.825, and the word group priority value Cy is 1.150 calculated by the formula.

[0054] Obtain the phrase priority value Cy for each data packet, sort the phrase priority values ​​Cy of each data packet in descending order to obtain the data transmission sequence;

[0055] The technical solution of this embodiment is as follows: Data packets are obtained from an artificial intelligence service; the text in the data packets is segmented into words to construct a segmented data set; based on the scenario of the artificial intelligence service, a key word library is constructed; and the segmented data set and the key word library are compared and analyzed to obtain the word frequency (TF). i For each data packet, the term frequency TF i and weight Qz j Numerical analysis is performed to obtain the phrase priority value Cy. Based on the phrase priority value Cy, a data transmission sequence is constructed. Based on the data transmission sequence, the priority of data packets is determined, providing a data basis for subsequent data packet transmission optimization analysis.

[0056] Example 2

[0057] Figure 2 This is a flowchart of the data analysis and optimization method according to an embodiment of the present invention, applicable to the data analysis and optimization of a data transmission system based on artificial intelligence services:

[0058] Transmission Analysis Module: Data packets with the same priority value Cy in the data transmission sequence are marked as equivalent data packets. The byte values ​​of all equivalent data packets are obtained and compared with the maximum byte value sent simultaneously by the current transmission network to determine whether conflict handling of equivalent data packets is required.

[0059] Based on the data transmission sequence, data packets with the same phrase priority value Cy in the data transmission sequence are marked as equivalent data packets;

[0060] The file.read() function in Python is used to obtain the byte values ​​of all equivalent data packets. The byte values ​​of all equivalent data packets are then summed to obtain the total data byte value.

[0061] Get the maximum number of bytes that can be sent simultaneously on the current network;

[0062] It should be noted that the maximum number of bytes that the current transmission network can send simultaneously is obtained from the operating system's configuration file;

[0063] Compare the total number of data bytes with the maximum number of bytes to determine whether conflict resolution is needed for data packets of equal value.

[0064] If the total data byte value is greater than or equal to the maximum byte value, it means that the byte value of data packets with the same phrase priority value Cy exceeds the maximum byte value that the system allows to send at the same time, and conflict handling is required for the data packets of the same value;

[0065] If the total data byte value is less than the maximum byte value, it means that the byte value of data packets with the same phrase priority value Cy does not exceed the maximum byte value that the system allows to send at the same time, and there is no need to handle the conflict of equal data packets;

[0066] Conflict Analysis Module: If conflict handling is required for equivalent data packets, construct the word frequency vector A and weight vector B of the equivalent data packets. Perform numerical calculations on word frequency vector A and weight vector B to obtain the word vector dot product Dj. Construct a word frequency run analysis model for word frequency vector A to obtain the continuous invalid word frequency Wx. Combine the word vector dot product Dj and the continuous invalid word frequency Wx for numerical analysis to construct a transmission optimization sequence and determine the priority order of equivalent data packets.

[0067] If it is necessary to handle conflicts between equivalent data packets, construct the word frequency vector A of the equivalent data packets based on the equivalent data packets;

[0068] Mark the equivalent data packet as Dzi , where i represents the word number in the data packet;

[0069] Obtain the term frequency (TF) of each word group in the segmented data group corresponding to the equivalent data package. i Construct the word frequency vector A=[TF1,TF2,...,TF] i ];

[0070] Get the weight Qz of each word group in the key vocabulary in the word segmentation data group corresponding to the equivalent data package. j Construct a weight vector B=[Qz1,Qz2,...,Qz j ];

[0071] Based on the word frequency vector A=[TF1,TF2,...,TF] i The weight vector B = [Qz1, Qz2, ..., Qz] j Calculate the dot product of the vectors to obtain the dot product of the word vectors, Dj;

[0072] Through the formula: Obtain the dot product of the word vectors, Dj;

[0073] For each segmented data group, obtain the word frequency vector A=[TF1,TF2,...,TF1]. i [Chinese word frequency TF] i A value of 0 indicates a word or phrase, which is then marked as an invalid word frequency phrase. The word frequency TF is... i Phrases with a non-zero frequency are marked as valid frequency phrases;

[0074] A word frequency run analysis model is constructed using the run test method to obtain the number of runs r of invalid word frequency groups in word frequency vector A;

[0075] Specifically, the construction process of the word frequency runs analysis model is as follows:

[0076] S1. Initialize the number of runs r=0 and locate the run to the initial word frequency TF1 of the word frequency vector A;

[0077] S2. Starting from the second term frequency TF2 in term frequency vector A, traverse term frequency vector A;

[0078] If the term frequency TF2 is an invalid term frequency group, the run count r is incremented by 1, and the next term frequency TF3 of the term frequency vector A is traversed.

[0079] If the term frequency TF2 is a valid term frequency group, store the run number r in the run data group, set the run number r to 0, and continue to traverse the next term frequency TF3 of the term frequency vector A;

[0080] S3. Repeat steps S1 and S2 until the last term frequency TF of the term frequency vector A is obtained. iThis yields the run-length data set.

[0081] Summing the run counts r in the run data set yields the run sum, which is the sum of the consecutive occurrences of invalid word frequency groups.

[0082] The sum of the occurrences of invalid word frequency groups is denoted as the continuous invalid word frequency Wx;

[0083] Calculate the high-frequency value Gp of a word group based on the continuous invalid word frequency Wx and the dot product of word group vectors Dj;

[0084] Through the formula: Obtain the high-frequency value Gp of the phrase, where c=0.46 and d=0.64;

[0085] It should be noted that the consecutive invalid word frequency Wx represents the number of times the invalid word frequency group appears consecutively in the word frequency vector A. The higher the consecutive invalid word frequency Wx is, the more invalid word frequency groups there are in the word frequency vector A, indicating that the information that is not related to or is not important to the key word library is more concentrated in the word groups included in the data package.

[0086] The dot product of word vectors Dj reflects the comprehensive correlation between the word frequency of words in the segmented data group and the weight of the corresponding word group in the key vocabulary. By calculating the dot product of word vectors Dj and the continuous invalid word frequency Wx, the high frequency value of word group Gp is ​​obtained. The high frequency value of word group Gp reflects the correlation between the data packet and the key vocabulary and the distribution of invalid word frequency groups. The priority order between data packets of equal value can be determined by the high frequency value of word group Gp, and the data packets with transmission conflicts can be further resolved.

[0087] Obtain the high-frequency value Gp of all equivalent data packets, construct a transmission optimization sequence according to the high-frequency value Gp in descending order, and determine the priority order of equivalent data packets;

[0088] It should be noted that by constructing a transmission optimization sequence, equivalent data packets can be further divided to determine their priority order.

[0089] Transmission optimization module: Based on the transmission optimization sequence, invalid word frequency groups are removed from the data packets of the transmission optimization sequence, and sentence analysis is performed on the word groups of the processed data packets of the transmission optimization sequence to obtain optimized data packets, and the optimized data packets are sent.

[0090] According to the order of the transmission optimization sequence, the data packets of the transmission optimization sequence are obtained sequentially, and invalid word frequency groups and duplicate word groups are removed from the data packets of the transmission optimization sequence to obtain the data packets to be optimized;

[0091] The word segmentation data groups in the data package to be optimized are used to perform word entity recognition using a natural language processing (NLP) model to identify the part-of-speech and grammatical relationships of the words.

[0092] Among them, the parts of speech of phrases include: subject, predicate, object, and adverbial words; the grammatical relations include: subject-predicate relation and verb-object relation.

[0093] It should be noted that in Natural Language Processing (NLP) models, NER technology is used to identify named entities in text, such as people's names, disease names, etc.

[0094] The word groups in the segmented data group are reorganized according to the order of adverbial, subject, predicate, object and word usage to obtain the segmented data group with the sentence sorting completed. The data group corresponding to the segmented data group with the sentence sorting completed is marked as the optimized data group, i.e. the optimized data group.

[0095] Use the internet to send optimized data packets;

[0096] The technical solution of this embodiment is as follows: The byte values ​​of data packets with the same word frequency (TF) in the data transmission sequence are compared and analyzed with the maximum byte value simultaneously transmitted by the current transmission network to determine whether conflict handling of equivalent data packets is required. If conflict handling is required, a word frequency vector A and a weight vector B of the equivalent data packets are constructed. Numerical calculations are performed on word frequency vector A and weight vector B to obtain the word vector dot product Dj. A word frequency run-length analysis model is constructed on word frequency vector A to obtain continuous invalid word frequencies Wx. Numerical analysis is performed on the word vector dot product Dj and continuous invalid word frequencies Wx to construct a transmission optimization sequence, determine the priority order of equivalent data packets, and based on the transmission optimization sequence, invalid word frequency groups are removed from the data packets of the transmission optimization sequence. Sentence analysis is then performed on the word groups of the processed data packets of the transmission optimization sequence to obtain optimized data packets. The optimized data packets are then sent. By rationally arranging the transmission order of data packets, important data can pass through the network with priority, further optimizing the overall performance of data transmission.

[0097] Example 3

[0098] A data transmission method based on artificial intelligence services, comprising:

[0099] Step 1: Obtain data packets from the artificial intelligence service, perform word segmentation on the text in the data packets, construct word segmentation data groups, and organize the word segmentation data groups in the data packets;

[0100] Step 2: Based on the scenario of artificial intelligence services, construct a key word library to provide a basis for word comparison of data packets;

[0101] Step 3: Compare and analyze the word segmentation data sets and key word groups to obtain the word frequency (TF). i For each data packet, the term frequency TF i and weight Qz j Numerical analysis is performed to obtain the phrase priority value Cy. Based on the phrase priority value Cy, a data transmission sequence is constructed to determine the transmission order of data packets.

[0102] Step 4: Data packets with the same priority value Cy in the data transmission sequence are marked as equivalent data packets. The byte values ​​of all equivalent data packets are obtained and compared with the maximum byte value sent simultaneously by the current transmission network to determine whether conflict handling of equivalent data packets is required.

[0103] Step 5: If conflict resolution is required for equivalent data packets, construct the word frequency vector A and weight vector B of the equivalent data packets. Perform numerical calculations on word frequency vector A and weight vector B to obtain the word group vector dot product Dj. Construct a word frequency run analysis model for word frequency vector A to obtain the continuous invalid word frequency Wx. Combine the word group vector dot product Dj and the continuous invalid word frequency Wx for numerical analysis to construct a transmission optimization sequence and determine the priority order of equivalent data packets.

[0104] Step 6: Based on the transmission optimization sequence, invalid word frequency groups are removed from the data packets of the transmission optimization sequence. Sentence analysis is performed on the word groups of the processed data packets of the transmission optimization sequence to obtain optimized data packets. The optimized data packets are then sent.

[0105] Example 4

[0106] Reference Figure 4 The present invention also provides a computer device 3, including: a memory 302 and a processor 301, and a computer program 303 stored on the memory 302. When the computer program 303 is executed on the processor 301, it implements a data transmission method based on artificial intelligence services as described in any of the above methods.

[0107] The computer device 3 may be a desktop computer, laptop, handheld computer, or cloud server, etc. The computer device 3 may include, but is not limited to, a processor 301 and a memory 302. Those skilled in the art will understand.

[0108] Figure 4 The computer device 3 is merely an example and does not constitute a limitation on the computer device 3. It may include more or fewer components than shown in the figure, or combine certain components, or different components, such as input / output devices, network access devices, etc.

[0109] The processor 301 may be a Central Processing Unit (CPU), or it may be other general-purpose processors, digital signal processors (DSPs), application-specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. A general-purpose processor may be a microprocessor or any conventional processor.

[0110] In some embodiments, the memory 302 may be an internal storage unit of the computer device 3, such as a hard disk or memory of the computer device 3. In other embodiments, the memory 302 may be an external storage device of the computer device 3, such as a plug-in hard disk, smart media card (SMC), secure digital (SD) card, flash card, etc., equipped on the computer device 3. Furthermore, the memory 302 may include both internal and external storage units of the computer device 3. The memory 302 is used to store the operating system, applications, bootloader, data, and other programs, such as the program code of the computer program. The memory 302 can also be used to temporarily store data that has been output or will be output.

[0111] Example 5

[0112] This invention also provides a computer-readable storage medium storing a computer program thereon, which, when executed by a processor, implements a data transmission method based on artificial intelligence services as described in any of the above methods.

[0113] In this embodiment, if the integrated unit is implemented as a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, all or part of the processes in the methods of the above embodiments can be implemented by a computer program instructing related hardware. The computer program can be stored in a computer-readable storage medium, and when executed by a processor, it can implement the steps of the various method embodiments described above. The computer program includes computer program code, which can be in the form of source code, object code, executable files, or certain intermediate forms. The computer-readable medium can include at least: any entity or device capable of carrying computer program code to a photographing device / terminal device, a recording medium, a computer memory, a read-only memory (ROM), a random access memory (RAM), an electrical carrier signal, a telecommunication signal, and a software distribution medium. Examples include USB flash drives, portable hard drives, magnetic disks, or optical disks. In some jurisdictions, according to legislation and patent practice, computer-readable media cannot be electrical carrier signals or telecommunication signals.

[0114] In the above embodiments, the descriptions of each embodiment have different focuses. For parts that are not described in detail or recorded in a certain embodiment, please refer to the relevant descriptions of other embodiments.

[0115] Those skilled in the art will recognize that the units and algorithm steps of the various examples described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are implemented in hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of this application.

[0116] In the embodiments disclosed in this application, it should be understood that the disclosed devices / terminal equipment and methods can be implemented in other ways. For example, the device / terminal equipment embodiments described above are merely illustrative. For instance, the division of modules or units is only a logical functional division, and in actual implementation, there may be other division methods. For example, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the displayed or discussed mutual coupling or direct coupling or communication connection may be through some interfaces; the indirect coupling or communication connection between devices or units may be electrical, mechanical, or other forms.

[0117] The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the units can be selected to achieve the purpose of this embodiment according to actual needs.

[0118] The above formulas are all dimensionless calculations. The formulas are derived from software simulations based on a large amount of collected data to obtain the most recent real-world results. The preset parameters in the formulas are set by those skilled in the art according to the actual situation.

[0119] The foregoing has provided a detailed description of one embodiment of the present invention, but this description is merely a preferred embodiment and should not be construed as limiting the scope of the invention. All equivalent variations and modifications made within the scope of the claims of this invention should still fall within the patent coverage of this invention.

Claims

1. A data transmission system based on artificial intelligence services, characterized in that, include: Data acquisition module: Acquires data packets from artificial intelligence services, performs word segmentation on the text in the data packets, constructs word segmentation data groups, and organizes the word segmentation data groups of the data packets; Lexicon building module: Based on AI service scenarios, build key lexicons; Word segmentation comparison module: Based on word segmentation data groups and key word libraries, numerical analysis is performed. Specifically, based on the word frequency and weight of word groups in the key word library, the priority value Cy of the word group is calculated. Based on the priority value Cy of the word group, a data sending sequence is constructed. Transmission Analysis Module: Data packets with the same priority value Cy in the data transmission sequence are marked as equivalent data packets. The byte values ​​of all equivalent data packets are obtained and compared with the maximum byte value sent simultaneously by the current transmission network to determine whether conflict handling of equivalent data packets is required. Conflict Analysis Module: If conflict handling is required for equivalent data packets, construct the word frequency vector A and weight vector B of the equivalent data packets and perform numerical analysis to construct a transmission optimization sequence; Transmission optimization module: Based on the transmission optimization sequence, invalid word frequency groups are removed from the data packets of the transmission optimization sequence, and sentence analysis is performed on the word groups of the data packets of the processed transmission optimization sequence to obtain optimized data packets, and the optimized data packets are sent. The method for constructing the transmission optimization sequence is as follows: Calculate the high-frequency value Gp of a word group based on the continuous invalid word frequency Wx and the dot product of word group vectors Dj; Through the formula: Get the high frequency value Gp of the phrase, where c and d are preset ratio coefficients; Obtain the high-frequency value Gp of all equivalent data packets, and construct a transmission optimization sequence according to the high-frequency value Gp of the packets in descending order; The dot product Dj of the phrase vector is obtained as follows: Obtain the term frequency (TF) of each word group in the segmented data group corresponding to the equivalent data package. i Construct the word frequency vector A=[TF1,TF2,...,TF] i ]; Get the weight Qz of each word group in the key vocabulary in the word segmentation data group corresponding to the equivalent data package. j Construct a weight vector B=[Qz1,Qz2,...,Qz j ]; Based on the word frequency vector A=[TF1,TF2,...,TF] i The weight vector B = [Qz1, Qz2, ..., Qz] j Calculate the dot product of the vectors to obtain the dot product of the word vectors, Dj; Through the formula: Obtain the dot product of the word vectors, Dj; The invalid word frequency group is obtained as follows: Obtain the word frequency vector A=[TF1,TF2,...,TF] from each segmented data group. i [Chinese word frequency TF] i Phrases with a frequency of 0 are marked as invalid word frequency groups. The sum of the occurrences of invalid word frequency groups is marked as continuous invalid word frequency Wx.

2. The data transmission system based on artificial intelligence services according to claim 1, characterized in that, The method for obtaining the phrase priority value Cy is as follows: Based on the word frequency ratio TFb and the phrase weight ratio Qzb, the phrase priority value Cy is calculated; Through the formula: Get the phrase priority value Cy, where a and b are preset ratio coefficients; The data transmission sequence is constructed as follows; Obtain the phrase priority value Cy for each data packet, and sort the phrase priority values ​​Cy of each data packet in descending order to obtain the data transmission sequence.

3. A data transmission system based on artificial intelligence services according to claim 2, characterized in that, The term frequency ratio (TFb) is obtained as follows: Based on the segmented data set and the key vocabulary, the frequency of word phrases in the segmented data set appearing in the key vocabulary is calculated, i.e., the term frequency (TF). i , where i is the word group number in the word segmentation data group; The word segmentation data group is labeled Cz, and each word in the word segmentation data group is labeled Cz. i ; Through formula Obtain the term frequency (TF) of a single word phrase in the key vocabulary within the segmented data set. i , where n i This represents the word group Cz in the segmented data set. i The number of times it appears in the key vocabulary, n j Indicates the number of all phrases in the key vocabulary; Where j is the number of the phrase in the key vocabulary; Each word frequency TF in the segmented data group i Perform summation to obtain the word frequency sum; Obtain the number of word groups in the segmented data set, calculate the ratio of the sum of word frequencies to the number of word groups in the segmented data set, and label the word frequency ratio as TFb; The phrase weight ratio Qzb is obtained as follows: If a word phrase in the segmented data set exists in the key vocabulary, the weight Qz of that word phrase in the segmented data set in the key vocabulary will be increased. j Perform summation to obtain the sum of word segmentation weights; The word segmentation weight is calculated as a ratio to the number of word segments in the segmentation data set, resulting in the word segmentation weight ratio, which is then labeled as Qzb.

4. The data transmission system based on artificial intelligence services according to claim 1, characterized in that, The method for determining whether conflict resolution is needed for equal data packets is as follows: Based on the data transmission sequence, data packets with the same phrase priority value Cy in the data transmission sequence are marked as equivalent data packets; Obtain the byte values ​​of all equivalent data packets, sum the byte values ​​of all equivalent data packets, and obtain the total data byte value; Get the maximum number of bytes that can be sent simultaneously on the current network; Compare the total number of data bytes with the maximum number of bytes to determine whether conflict resolution is needed for data packets of equal value. If the total value of data bytes is greater than or equal to the maximum value of bytes, conflict resolution is required for data packets of equal value.

5. A data transmission system based on artificial intelligence services according to claim 1, characterized in that, The method for obtaining the continuous invalid word frequency Wx is as follows: Summing the run counts r in the run data set yields the run sum, which is the sum of the consecutive occurrences of invalid word frequency groups. The sum of the occurrences of invalid word frequency groups is denoted as the continuous invalid word frequency Wx.

6. A data transmission system based on artificial intelligence services according to claim 5, characterized in that, The method for constructing the run-length data set is as follows: A word frequency run analysis model is constructed using the run test method to obtain the number of runs r of invalid word frequency groups in word frequency vector A; The construction process of the word frequency runs analysis model is as follows: S1. Initialize the number of runs r=0 and locate the run to the initial word frequency TF1 of the word frequency vector A; S2. Starting from the second term frequency TF2 in term frequency vector A, traverse term frequency vector A; If the term frequency TF2 is an invalid term frequency group, the run count r is incremented by 1, and the next term frequency TF3 of the term frequency vector A is traversed. If the term frequency TF2 is a valid term frequency group, store the run number r in the run data group, set the run number r to 0, and continue to traverse the next term frequency TF3 of the term frequency vector A; S3. Repeat steps S1 and S2 until the last term frequency TF of the term frequency vector A is obtained. i This yields the run data set.

7. A data transmission system based on artificial intelligence services according to claim 1, characterized in that, The optimized data packet is obtained in the following way: According to the order of the transmission optimization sequence, the data packets of the transmission optimization sequence are obtained sequentially, and invalid word frequency groups are removed from the data packets of the transmission optimization sequence to obtain the data packets to be optimized; The word segmentation data groups in the data package to be optimized are used to perform word entity recognition using a natural language processing model to identify the part of speech and grammatical relationships of the words. The word groups in the segmented data are reorganized according to their parts of speech and grammatical relationships to obtain the segmented data groups with the sentences sorted out. The data packets corresponding to the segmented data groups with the sentences sorted out are marked as optimized data packets, i.e. optimized data packets, and then sent.

Images