Credit control method and device for data message service and medium

Abstract

The invention discloses a credit control method and device for a data message service and a medium, and relates to the technical field of intelligent scheduling, and the method comprises the following steps: when it is detected that the current network congestion level reaches a congestion deterioration threshold in the execution process of an authorization overdraft sending instruction, automatically skipping to a prepaid instruction execution process; when the prepaid instruction is required to identify that a semantic priority label is a high-priority emergency message in a payment verification process, immediately terminating a payment verification process, switching to an authorization overdraft sending instruction execution process, starting a rapid transmission channel, and applying a delayed charging mode to complete data message forwarding; and when the temporary storage message of refusing to send the instruction monitors that the current network congestion level is recovered to a smooth state, reactivating any one of the authorized overdraft sending instruction and the required prepayment instruction. According to the invention, through the established instruction execution state pool, the dynamic state conversion among the authorized overdraft sending instruction, the prepayment requesting instruction and the sending refusal instruction is realized.

Description

Technical Field

[0001] This invention relates to the field of intelligent scheduling technology, and in particular to a credit control method, device and medium for data messaging services. Background Technology

[0002] In the field of mobile communications, data messaging services, as one of the core applications, have been widely used for real-time information transmission and user interaction. Its credit control method ensures the economic sustainability of the message delivery process and the rational allocation of network resources by assessing user creditworthiness and business risks. With the popularization of 5G and edge computing technologies, credit control mechanisms are increasingly integrated into intelligent decision-making frameworks, which can dynamically adjust authorization policies based on message content semantics and real-time network status, thereby improving service response speed and user experience. This method not only optimizes tariff management but also enhances the processing efficiency of emergency messages through a priority classification mechanism, driving the evolution of data messaging services towards intelligence.

[0003] Existing credit control methods for data messaging services typically rely on static balance checks and preset trigger conditions to achieve basic risk prevention and control. However, in high-load network environments, these methods struggle to adapt to the diversity of message content and congestion fluctuations in real time, leading to delays or loss of some high-value messages. Existing technologies address the lack of dynamic adaptation in credit policies by employing extended interface interactions or post-processing call detail records, but these methods often fail to adequately address the real-time limitations. Summary of the Invention

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

[0005] Therefore, this invention provides a credit control method for data messaging services to solve the problem of lack of dynamic adaptation to credit policies.

[0006] To solve the above-mentioned technical problems, the present invention provides the following technical solution: In a first aspect, the present invention provides a credit control method for data messaging services, comprising, The system receives data messages to be sent and transmits them to the semantic analysis center for real-time processing. The semantic analysis center performs feature extraction and pattern recognition on the message text of the data messages based on a pre-set semantic core lexicon and classification rule library, and generates semantic priority tags. Collect throughput, transmission delay, and packet loss rate metrics from network nodes; input these metrics into a congestion assessment algorithm for multi-parameter fusion analysis and network congestion status determination, and output the current network congestion level. The credit decision center receives semantic priority tags and the current network congestion level, queries and matches a preset policy decision matrix, and generates credit control instructions based on the matching results of semantic priority tags and the current network congestion level; the credit control instructions include authorized overdraft sending instructions, rejected sending instructions, and prepayment request instructions. Based on credit control instructions, an instruction execution state pool is established, allowing different credit control instructions to dynamically switch according to the current network congestion level and semantic priority label: When the execution of the authorized overdraft sending instruction detects that the current network congestion level has reached the congestion deterioration threshold, it automatically jumps to the prepayment request instruction execution process, triggering the payment verification mechanism; when the prepayment request instruction identifies a high-priority urgent message as its semantic priority label during the payment verification process, it immediately terminates the payment verification process and switches to the authorized overdraft sending instruction execution process, enabling the fast transmission channel and applying the delayed billing mode to complete the data message forwarding; when the temporary message of the rejected sending instruction detects that the current network congestion level has recovered to a smooth state, it reactivates any available instruction among the authorized overdraft sending instruction and the prepayment request instruction, executing the corresponding instruction's credit control.

[0007] As a preferred embodiment of the credit control method for data message services described in this invention, the method involves: receiving the data message to be sent, transmitting the data message to a semantic analysis center for real-time processing, and the semantic analysis center performing feature extraction and pattern recognition on the message text of the data message based on a pre-set semantic core lexicon and classification rule base to generate semantic priority tags. Specifically, Receive data messages to be sent, extract message text from the data messages, perform encryption and access authentication processing on the message text, and generate semantic analysis objects; Perform text cleaning operations on semantic analysis objects to remove invisible characters and redundant symbols, and generate standardized text; A segmentation method combining dictionary matching and language modeling is used to perform word segmentation on standardized text and generate word sequences. The word sequence is compared item by item with a pre-set semantic core vocabulary, the semantic core words that are successfully matched in the word sequence are identified, and the occurrence position and frequency of the semantic core words are recorded. Based on the preset semantic weights in the pre-set semantic core vocabulary, the occurrence position and frequency of semantic core words are weighted to generate a weighted feature vector of semantic core words. The word sequence, the position of the semantic core word, and the contextual adjacency relationship formed by the words before and after the semantic core word are matched with the rule entries in the classification rule base to determine whether there is a classification rule that meets the triggering conditions. The semantic core word weighted feature vector and the classification rule matching result that meets the triggering condition are input into the classification decision logic, and the message category result is generated through logical judgment; The priority type of the data message content is determined based on the message category result. The priority type is written into the identifier field of the semantic priority label to generate the semantic priority label.

[0008] As a preferred embodiment of the credit control method for data messaging services described in this invention, the semantic priority tags are high-priority emergency messages, normal-priority social messages, and low-priority commercial promotion messages.

[0009] As a preferred embodiment of the credit control method for data message services described in this invention, the classification rule base is a collection that stores classification rule entries, including semantic core word patterns, positional order, and adjacency conditions defined based on the message corpus.

[0010] As a preferred embodiment of the credit control method for data messaging services described in this invention, the step of inputting throughput, transmission delay, and packet loss rate indicators into a congestion assessment algorithm for multi-parameter fusion analysis and network congestion status determination, and outputting the current network congestion level, specifically involves... The throughput, transmission delay, and packet loss rate metrics are input into the congestion assessment algorithm, and a weighted fusion process is performed to generate a comprehensive congestion score. The overall congestion score is compared with the preset congestion level threshold range to determine the range of the congestion level threshold range to which the overall congestion score belongs. Perform state mapping on the level identifiers corresponding to the congestion level threshold range to generate the current network congestion level.

[0011] As a preferred embodiment of the credit control method for data messaging services described in this invention, the generation of credit control instructions specifically includes: The credit decision center receives the semantic priority label and the current network congestion level, and performs integrity verification on the semantic priority label and the current network congestion level; Using the verified semantic priority label and the current network congestion level as the query key, a matching search is performed on the preset policy decision matrix within the credit decision center to determine the policy entries in the policy decision matrix that correspond to the semantic priority label and the current network congestion level. Extract the instruction type from the policy entry and use it as a credit control instruction.

[0012] As a preferred embodiment of the credit control method for data message services described in this invention, the establishment of the instruction execution status pool refers to receiving a credit control instruction, creating an instruction execution status data structure, defining the fields of the instruction execution status data structure as an instruction identifier field and an execution status field, and writing the identifier and execution status field of the credit control instruction into the instruction execution status data structure.

[0013] As a preferred embodiment of the credit control method for data messaging services described in this invention, when the temporary message of the rejection sending instruction detects that the current network congestion level has recovered to a smooth state, any available instruction among the authorized overdraft sending instruction and the request prepayment instruction is reactivated, and the credit control of the corresponding instruction is executed. Specifically, When the "refuse to send" instruction is set to the "execute" state, the data message is placed in the temporary queue, and the current network congestion level is continuously obtained. Determine whether the current network congestion level has recovered to a smooth state. If it has recovered, remove the execution state of "refuse to send command" from the command execution state pool. Check whether the execution conditions for the authorized overdraft sending instruction are met. If they are met, activate the execution state of the authorized overdraft sending instruction and perform credit control for the authorized overdraft sending instruction. If not, check whether the execution conditions of the prepayment instruction are met. If they are met, activate the execution state of the prepayment instruction and execute the credit control of the prepayment instruction.

[0014] In a second aspect, the present invention provides a computer device including a memory and a processor, wherein the memory stores a computer program, wherein when the computer program is executed by the processor, it implements any step of the credit control method for data messaging services as described in the first aspect of the present invention.

[0015] Thirdly, the present invention provides a computer-readable storage medium having a computer program stored thereon, wherein: when the computer program is executed by a processor, it implements any step of the credit control method for data messaging services as described in the first aspect of the present invention.

[0016] The beneficial effects of this invention are as follows: By establishing an instruction execution state pool, dynamic state transitions between authorized overdraft sending instructions, prepayment request instructions, and rejection sending instructions are realized. This allows authorized overdraft sending, prepayment request, and rejection sending states to switch in real time based on network congestion levels and semantic priority tags, improving the adaptability of credit control strategies. When network congestion intensifies, the overdraft sending mechanism is switched to a prepayment mechanism in a timely manner to avoid resource overload. When a high-priority emergency message is identified, the payment verification process is immediately terminated, overdraft sending is resumed, and fast channels and delayed billing are enabled to ensure the timely transmission of critical information. In the rejection sending state, available instructions are reactivated as the network recovers, reducing the risk of message loss. The reliability of data message transmission in dynamic network environments is enhanced, and the coordinated optimization of credit control, semantic importance, and network status is achieved, improving the response efficiency of emergency services and the flexibility of resource utilization. Attached Figure Description

[0017] To more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings used in the description of the embodiments will be briefly introduced below. Obviously, the 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.

[0018] Figure 1 A flowchart for the credit control method for data messaging services.

[0019] Figure 2 A flowchart for encrypting files.

[0020] Figure 3 This is a flowchart for network congestion assessment.

[0021] Figure 4 A flowchart for the dynamic transition of the instruction execution state pool. Detailed Implementation

[0022] To make the above-mentioned objects, features and advantages of the present invention more apparent and understandable, the specific embodiments of the present invention will be described in detail below with reference to the accompanying drawings.

[0023] Many specific details are set forth in the following description in order to provide a full understanding of the invention. However, the invention may also be practiced in other ways different from those described herein, and those skilled in the art can make similar extensions without departing from the spirit of the invention. Therefore, the invention is not limited to the specific embodiments disclosed below.

[0024] Secondly, the term "one embodiment" or "embodiment" as used herein refers to a specific feature, structure, or characteristic that may be included in at least one implementation of the present invention. The phrase "in one embodiment" appearing in different places in this specification does not necessarily refer to the same embodiment, nor is it a single or selective embodiment that is mutually exclusive with other embodiments.

[0025] Reference Figures 1-4 As an embodiment of the present invention, this embodiment provides a credit control method for data messaging services, comprising the following steps: S1. Receive the data message to be sent and transmit the data message to the semantic analysis center for real-time processing. The semantic analysis center performs feature extraction and pattern recognition on the message text of the data message based on the pre-set semantic core lexicon and classification rule library, and generates semantic priority tags.

[0026] Receive data messages to be sent, extract message text from the data messages, perform encryption and access authentication processing on the message text, and generate semantic analysis objects; Specifically, the process involves: receiving a data message to be sent; checking whether the data message contains complete message text and sender identity credentials to confirm the integrity of the data message; separating the message text portion from the data message to be sent as the message text to be processed; applying a symmetric key encryption method and using a single key to perform an encryption transformation on the message text to convert it into encrypted message text to prevent unauthorized third-party access; comparing the user identifier or authentication token in the sender identity credentials with the permission entries in the preset permission list one by one to generate an access permission verification result to confirm that the access permission is granted; and combining the encrypted message text and the access permission verification result to generate a semantic analysis object. The process of comparing the user identifier or authentication token in the sender's identity credentials with the permission entries in the preset permission list to generate an access permission verification result is as follows: The user identifier or authentication token is extracted from the sender's identity credentials as the credential to be verified. This credential is then matched against each permission entry in the preset permission list. The system checks whether the credential to be verified completely matches the user identifier or authentication token in the permission entry. If a matching permission entry exists, an access permission verification result indicating that the access permission has been granted is generated; otherwise, an access permission verification result indicating that the access permission has failed is generated. The preset permission list is set based on user registration information and contains a set of allowed sender user identifiers and authentication tokens. Perform text cleaning operations on semantic analysis objects to remove invisible characters and redundant symbols, and generate standardized text; Specifically, the process involves: examining each character in the encrypted message text one by one, identifying and removing invisible characters, including newline characters, tab characters, and control characters, while also deleting redundant symbols, including extra spaces and repeated punctuation marks, and converting the processed encrypted message text into standardized text with a uniform format. A segmentation method combining dictionary matching and language modeling is used to perform word segmentation on standardized text and generate word sequences. Specifically, the process involves scanning a continuous character sequence starting from the beginning of the standardized text. A single character within this sequence is selected as the initial comparison object. This initial comparison object is then compared one by one with all semantic core words in a pre-defined semantic core lexicon. The initial comparison object is checked for complete match with each semantic core word. If they match, the initial comparison object is marked as a successfully matched semantic core word, and the end position of the successfully matched semantic core word is recorded as the boundary point of the initial segmentation result. Otherwise, the length of the initial comparison object is extended by two characters. This comparison process is repeated until the length of the initial comparison object reaches the maximum length of the semantic core words in the pre-defined semantic core lexicon or a match is found. The process involves identifying the semantic core words of a text. Starting from the end position of a successfully matched semantic core word, the next continuous character sequence is scanned, repeating the selection, comparison, and labeling process until all continuous character sequences of the standardized text have been scanned, resulting in a preliminary segmentation of the text containing all successfully matched semantic core words. The pre-built semantic core word library is a collection of semantic core words, containing common words extracted from a message corpus and their corresponding semantic classification labels. The message corpus is a collection of texts gathered from SMS and social media messages, including daily conversations, emergency notifications, and commercial promotions, used to train the language model and construct the pre-built semantic core word library. A language model is applied to the preliminary segmentation results. Based on the semantic combination rules preset by the language model, the boundary points in the preliminary segmentation results are checked. Boundary points that do not conform to the semantic combination rules are adjusted to eliminate ambiguous semantic core words. The preliminary segmentation results are re-divided based on the adjusted boundary points to generate adjusted segmentation results. The adjusted segmentation results are then organized into word sequences. The preset semantic combination rules are set based on the co-occurrence frequency and contextual relationship of semantic core words in the message corpus, including the collocation probability of semantic core words with preceding and following semantic core words. It should be noted that the specific training process of the language model is as follows: the language model is trained based on a message corpus, which includes daily dialogues, emergency notifications, and commercial promotion content; the co-occurrence frequency and contextual relationships of semantic core words are extracted from the message corpus to generate semantic combination rules; the semantic combination rules are stored as a semantic probability table, which records the probability of the semantic core words collocation with the preceding and following semantic core words; the semantic probability table is optimized using the maximum likelihood estimation method to ensure that the semantic combination rules reflect common language patterns, and is used to adjust the boundary points of the initial segmentation results; The optimization of the semantic probability table using the maximum likelihood estimation method involves: traversing text samples in the message corpus, identifying and extracting all semantic core words, recording the occurrence positions of semantic core words in the text and their adjacent words, forming a contextual relationship set for semantic core words; counting the joint occurrence frequency of each pair of semantic core words in the contextual relationship set to determine the co-occurrence frequency between semantic core words; constructing an initial semantic probability table based on the co-occurrence frequency and contextual relationships, where the semantic probability table records the collocation probability of each pair of semantic core words, determined by the ratio of the joint occurrence frequency to the total occurrence frequency of semantic core words; storing the generated initial semantic probability table; iteratively analyzing text samples in the message corpus using the maximum likelihood estimation method, adjusting the collocation probabilities in the semantic probability table, and generating an optimized semantic probability table that maximizes the fit of the collocation probabilities to the semantic core word collocation patterns observed in the message corpus; and storing the optimized semantic probability table to guide semantic analysis and boundary point adjustment.

[0027] The word sequence is compared item by item with a pre-set semantic core vocabulary, the semantic core words that are successfully matched in the word sequence are identified, and the occurrence position and frequency of the semantic core words are recorded. Based on the preset semantic weights in the semantic core lexicon, the occurrence position and frequency of semantic core words are weighted to generate a weighted feature vector of semantic core words. The preset semantic weights are set based on the semantic importance and usage frequency of semantic core words in the message corpus. For example, the weight of "rescue" in emergency notifications is higher than that of "greeting" in social messages. The word sequence, the position of the semantic core word, and the contextual adjacency relationship formed by the words before and after the semantic core word are matched with the rule entries in the classification rule base to determine whether there is a classification rule that meets the triggering conditions. The classification rule base is determined based on the semantic core word statistics and contextual patterns in the message corpus. It is a collection that stores classification rule entries and is pre-installed in the semantic analysis center. It contains semantic core word patterns, position order, and adjacency conditions defined based on the message corpus and is used for pattern matching and message classification. Specifically, the process involves: extracting each semantic core word, its position, and adjacent words from the word sequence; combining the extracted semantic core words, their positions, and adjacent words to form a contextual adjacency set; comparing each semantic core word, its position, and adjacent words in the contextual adjacency set with the rule entries in the pre-set classification rule base; checking whether the semantic core word matches the word pattern defined in the rule entry; whether the position of the semantic core word conforms to the positional order specified in the rule entry; and whether the adjacent words satisfy the adjacency conditions specified in the rule entry. If the semantic core word, its position, and adjacent words all satisfy the conditions defined in the rule entry, then it is confirmed that a classification rule that meets the triggering conditions exists. The semantic core word weighted feature vector and the matching results of classification rules that meet the trigger conditions are input into the classification decision logic. The message category result is generated through logical judgment. Logical judgment refers to comparing the weight values ​​in the semantic core word weighted feature vector with the rule entries in the matching results of classification rules that meet the trigger conditions. It verifies whether the weight values ​​in the semantic core word weighted feature vector meet the preset category judgment conditions in the rule entries and confirms the message category corresponding to the rule entries that meet the category judgment conditions as the message category result. The preset category judgment conditions are set based on the semantic importance and contextual patterns of semantic core words in the message corpus, including weight value thresholds, semantic core word combination patterns, and contextual adjacency conditions. Among them, the triggering condition refers to the complete match between the semantic core word, the position of the semantic core word and the adjacent words before and after it and the word pattern, position order and adjacency conditions defined in the rule entries in the classification rule base, confirming the existence of a classification rule that satisfies the definition of the rule entries in the classification rule base; The message category corresponding to the rule entry that meets the category determination conditions is confirmed as the message category result. This means that the weight value in the weighted feature vector of the semantic core word meets the weight value threshold, semantic core word combination pattern and context adjacency condition of the rule entry in the classification rule base. It should be noted that the category determination criteria directly reference the standards defined in the rule entries of the classification rule base, and are used to verify whether the weighted feature vector of semantic core words meets the requirements for determining the message category.

[0028] The priority type of the data message content is determined based on the message category result. The priority type is written into the identifier field of the semantic priority label to generate a semantic priority label. The semantic priority label is one of the following: high priority emergency message, normal priority social message, and low priority commercial promotion message.

[0029] It should be noted that this step receives the data messages to be sent and generates semantic priority tags. By applying symmetric key encryption to the message text and verifying the sender's identity credentials, data security is ensured, effectively preventing unauthorized third-party access. Through precise word segmentation and semantic analysis, the accuracy of message classification is improved, and the efficiency of prioritizing urgent messages is optimized. By combining contextual relationships and weighted processing, the intelligence and reliability of credit control are enhanced, the risk of misjudgment under network congestion is reduced, and the overall security and resource scheduling efficiency of data message services are improved.

[0030] S2. Collect throughput, transmission delay, and packet loss rate metrics from network nodes; input these metrics into a congestion assessment algorithm for multi-parameter fusion analysis and network congestion status determination, and output the current network congestion level.

[0031] The total amount of data transmitted per unit time in a network node is collected as a throughput indicator; a network node is a collection of devices responsible for data transmission in a communication network, including network base stations, network routers, and network switches. Collect the time difference sequence of communication data packets from sending to receiving in network nodes, and perform a moving average processing on the time difference sequence to generate a transmission delay index; The process of performing a moving average on the time difference sequence is as follows: a fixed-length time window is selected from the time difference sequence of communication data packets, each time difference value in the time difference sequence is traversed one by one, the continuous time difference values ​​within the time window are averaged to generate a smoothed time difference sequence, and the latest time difference value in the smoothed time difference sequence is used as the transmission delay indicator. The ratio of the number of communication data packets that failed to reach the target node within a specified observation window to the total number of communication data packets sent is used as a data packet loss rate indicator. The target node is determined by the receiver address specified in the header information of the communication data packet in the network communication protocol, which identifies the target device or node of the communication data packet. The throughput, transmission delay, and packet loss rate metrics are input into the congestion assessment algorithm, weighted and fused to generate a comprehensive congestion score, expressed as: ; in, The overall congestion score represents the degree of network congestion. This represents the weight of the throughput metric, for example, 0.4, indicating the proportion of influence of the throughput metric on the overall congestion score. Throughput is a metric that measures the total amount of data transmitted per unit of time, expressed in megabytes per second (Mbps). This represents the weight of the transmission delay metric, for example, 0.3, indicating the proportion of influence of the transmission delay metric on the overall congestion score. Transmission delay is a metric that measures the smooth time difference between the sending and receiving of communication data packets, measured in milliseconds (ms). The packet loss rate metric measures the ratio of the number of communication packets that fail to reach the target node to the total number of communication packets sent, expressed as a percentage (%). The weight of the packet loss rate metric is 0.3, which represents the proportion of the packet loss rate metric's influence on the overall congestion score. The overall congestion score is compared with the preset congestion level threshold range to determine the range of the congestion level threshold range to which the overall congestion score belongs. Specifically, the process involves comparing the overall congestion score with preset congestion level threshold intervals. By checking whether the overall congestion score falls within the preset smooth flow threshold interval, mild congestion threshold interval, or severe congestion threshold interval, the range of the corresponding congestion level threshold interval is determined. The preset congestion level threshold intervals are set based on historical network performance data. For example, the smooth flow threshold interval is an overall congestion score less than or equal to 30, the mild congestion threshold interval is an overall congestion score greater than 30 and less than or equal to 70, and the severe congestion threshold interval is an overall congestion score greater than 70. The comparison operation starts from the smooth flow threshold interval and progressively checks the boundary values ​​between the overall congestion score and each congestion level threshold interval. After confirming the interval to which the overall congestion score belongs, the corresponding congestion level threshold interval range is output. Perform state mapping on the level identifiers corresponding to the congestion level threshold range to generate the current network congestion level; the current network congestion level includes smooth flow, mild congestion, and severe congestion.

[0032] S3. The credit decision center receives the semantic priority label and the current network congestion level, queries and matches the preset policy decision matrix, and generates credit control instructions based on the matching results of the semantic priority label and the current network congestion level. The credit control instructions include authorization overdraft sending instructions, rejection sending instructions, and prepayment request instructions.

[0033] The credit decision center receives the semantic priority label and the current network congestion level, and performs integrity verification on the semantic priority label and the current network congestion level; Specifically, the process involves: checking whether the semantic priority label contains a complete priority type identifier field and a priority type value; checking whether the current network congestion level contains a complete level identifier and a level value; and verifying whether the priority type value of the semantic priority label and the level value of the current network congestion level match a preset format template. If both the semantic priority label and the current network congestion level match the preset format template, the integrity verification is passed. The preset format template is determined based on the data message processing standard and includes the expected structure of the priority type identifier field and priority type value of the semantic priority label, as well as the level identifier and level value of the current network congestion level. Using the verified semantic priority label and the current network congestion level as the query key, a matching search operation is performed on the preset policy decision matrix in the credit decision center to determine the policy entries in the policy decision matrix that correspond to the semantic priority label and the current network congestion level. Specifically, the priority type value of the verified semantic priority tag is used as the row index, and the level value of the current network congestion level is used as the column index to locate the intersection position in the preset policy decision matrix. The policy entry content corresponding to the intersection position is extracted, and the policy entry content is determined to be the policy entry corresponding to the semantic priority tag and the current network congestion level. The preset policy decision matrix is ​​a two-dimensional table based on the message corpus and historical network performance data. It stores the policy entries corresponding to the credit control instructions with the priority type value of the semantic priority tag as the row and the level value of the current network congestion level as the column. Extract the instruction type from the policy entry as a credit control instruction; the credit control instruction is one of the following: authorized overdraft sending instruction, rejection sending instruction, and prepayment request instruction.

[0034] S4. Based on credit control instructions, establish an instruction execution state pool, allowing different credit control instructions to dynamically switch according to the current network congestion level and semantic priority label: When the current network congestion level is detected to have reached the congestion deterioration threshold during the execution of the authorized overdraft sending instruction, it automatically jumps to the prepayment request instruction execution process, triggering the payment verification mechanism; when the prepayment request instruction identifies a high-priority urgent message as its semantic priority label during the payment verification process, it immediately terminates the payment verification process and switches to the authorized overdraft sending instruction execution process, enabling the fast transmission channel and applying the delayed billing mode to complete the data message forwarding; when the temporary message of the rejected sending instruction detects that the current network congestion level has recovered to a smooth state, it reactivates any available instruction among the authorized overdraft sending instruction and the prepayment request instruction, executing the corresponding instruction's credit control.

[0035] Receive credit control instructions, create an instruction execution status data structure, define the fields of the instruction execution status data structure including the instruction identifier field and the execution status field, write the identifier of the credit control instruction and the initial execution status into the instruction execution status data structure, and generate an instruction execution status pool; When the authorized overdraft sending instruction is in the execution state, the current network congestion level is continuously acquired to determine whether the current network congestion level has reached the preset congestion deterioration threshold. If the preset congestion deterioration threshold is reached, the execution state of the authorized overdraft sending instruction is removed from the instruction execution state pool, and the execution state of the prepaid instruction is written, triggering the payment verification mechanism. The payment verification mechanism is the process of checking whether the user's account balance is sufficient to pay for data message forwarding fees and confirming payment authorization. The preset congestion deterioration threshold is set based on the distribution of the comprehensive congestion score in the historical network performance data. For example, the value range is that a comprehensive congestion score greater than 70 indicates a severe congestion state. When a prepaid instruction is in the payment verification state, the semantic priority tag value is read from the semantic priority tag associated with the data message. It is then determined whether the semantic priority tag value is a high-priority urgent message. If it is a high-priority urgent message, the payment verification state is terminated, the execution state of the prepaid instruction is removed from the instruction execution state pool, and the execution state of the authorized overdraft sending instruction is switched back. The fast transmission channel is enabled and the delayed billing mode is applied to complete the data message forwarding. Enabling the fast transmission channel and applying the delayed billing mode means allocating a high-bandwidth network path to the data message to accelerate forwarding, prioritizing it over other data message transmissions, while recording the cost incurred in forwarding the data message. The cost will be deducted after the high-priority urgent message is confirmed to have been delivered. When a "Reject Sending" instruction is set to the "Execute" state, the data message is placed in a temporary queue, and the current network congestion level is continuously acquired. It is then determined whether the current network congestion level has recovered to a smooth state. If it has recovered to a smooth state, the "Reject Sending" instruction's "Execute" state is removed from the instruction execution state pool. The execution conditions for an "Authorize Overdraft Sending" instruction are then checked. If they are met, the "Authorize Overdraft Sending" instruction's "Execute" state is activated, and credit control for the "Authorize Overdraft Sending" instruction is executed. If not, the execution conditions for a "Request Prepayment" instruction are checked. If they are met, the "Request Prepayment" instruction's "Execute" state is activated, and credit control for the "Request Prepayment" instruction is executed.

[0036] This embodiment also provides a computer device applicable to a credit control method for data messaging services, comprising: a memory and a processor; the memory is used to store computer-executable instructions, and the processor is used to execute the computer-executable instructions to implement the credit control method for data messaging services as proposed in the above embodiment.

[0037] The computer device can be a terminal, comprising a processor, memory, communication interface, display screen, and input devices connected via a system bus. The processor provides computing and control capabilities. The memory includes non-volatile storage media and internal memory. The non-volatile storage media stores the operating system and computer programs. The internal memory provides an environment for the operation of the operating system and computer programs stored in the non-volatile storage media. The communication interface is used for wired or wireless communication with external terminals; wireless communication can be achieved through Wi-Fi, carrier networks, NFC (Near Field Communication), or other technologies. The display screen can be an LCD screen or an e-ink screen. The input devices can be a touch layer covering the display screen, buttons, a trackball, or a touchpad on the computer device's casing, or an external keyboard, touchpad, or mouse.

[0038] This embodiment also provides a storage medium storing a computer program, which, when executed by a processor, implements the credit control method for implementing data message services as proposed in the above embodiments. The storage medium can be implemented by any type of volatile or non-volatile storage device or a combination thereof, such as Static Random Access Memory (SRAM), Electrically Erasable Programmable Read-Only Memory (EEPROM), Erasable Programmable Read Only Memory (EPROM), Programmable Red-Only Memory (PROM), Read-Only Memory (ROM), magnetic storage, flash memory, magnetic disk, or optical disk.

[0039] In summary, this invention achieves dynamic state transitions between authorized overdraft sending instructions, prepayment request instructions, and rejection sending instructions through an established instruction execution state pool. This allows the authorized overdraft sending, prepayment request, and rejection sending states to switch in real time based on network congestion levels and semantic priority tags, improving the adaptability of credit control strategies. When network congestion worsens, it promptly switches from interrupting overdraft sending to a prepayment mechanism to avoid resource overload. When a high-priority urgent message is identified, it immediately terminates the payment verification process, resumes overdraft sending, and activates fast channels and delayed billing to ensure timely transmission of critical information. In the rejection sending state, available instructions are reactivated as the network recovers, reducing the risk of message loss. This enhances the reliability of data message transmission in dynamic network environments, achieves coordinated optimization of credit control, semantic importance, and network status, and improves the response efficiency and resource utilization flexibility of urgent services.

[0040] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit it. Although the present invention has been described in detail with reference to preferred embodiments, those skilled in the art should understand that modifications or equivalent substitutions can be made to the technical solutions of the present invention without departing from the spirit and scope of the technical solutions of the present invention, and all such modifications or substitutions should be covered within the scope of the claims of the present invention.

Claims

1. A credit control method for data messaging services, characterized in that: include, The system receives data messages to be sent and transmits them to the semantic analysis center for real-time processing. The semantic analysis center performs feature extraction and pattern recognition on the message text of the data messages based on a pre-set semantic core lexicon and classification rule library, and generates semantic priority tags. Collect throughput, transmission delay, and packet loss rate metrics from network nodes; input these metrics into a congestion assessment algorithm for multi-parameter fusion analysis and network congestion status determination, and output the current network congestion level. The credit decision center receives semantic priority tags and the current network congestion level, queries and matches a preset policy decision matrix, and generates credit control instructions based on the matching results of semantic priority tags and the current network congestion level; the credit control instructions include authorized overdraft sending instructions, rejected sending instructions, and prepayment request instructions. Based on credit control instructions, an instruction execution state pool is established, allowing different credit control instructions to dynamically switch according to the current network congestion level and semantic priority label: When the execution of the authorized overdraft sending instruction detects that the current network congestion level has reached the congestion deterioration threshold, it automatically jumps to the prepayment request instruction execution process, triggering the payment verification mechanism; when the prepayment request instruction identifies a high-priority urgent message as its semantic priority label during the payment verification process, it immediately terminates the payment verification process and switches to the authorized overdraft sending instruction execution process, enabling the fast transmission channel and applying the delayed billing mode to complete the data message forwarding; when the temporary message of the rejected sending instruction detects that the current network congestion level has recovered to a smooth state, it reactivates any available instruction among the authorized overdraft sending instruction and the prepayment request instruction, executing the corresponding instruction's credit control.

2. The credit control method for data messaging services as described in claim 1, characterized in that: The process involves receiving the data message to be sent, transmitting the data message to the semantic analysis center for real-time processing, and the semantic analysis center performing feature extraction and pattern recognition on the message text based on a pre-set semantic core lexicon and classification rule base to generate semantic priority tags. Specifically, Receive data messages to be sent, extract message text from the data messages, perform encryption and access authentication processing on the message text, and generate semantic analysis objects; Perform text cleaning operations on semantic analysis objects to remove invisible characters and redundant symbols, and generate standardized text; A segmentation method combining dictionary matching and language modeling is used to perform word segmentation on standardized text and generate word sequences. The word sequence is compared item by item with a pre-set semantic core vocabulary, the semantic core words that are successfully matched in the word sequence are identified, and the occurrence position and frequency of the semantic core words are recorded. Based on the preset semantic weights in the pre-set semantic core vocabulary, the occurrence position and frequency of semantic core words are weighted to generate a weighted feature vector of semantic core words. The word sequence, the position of the semantic core word, and the contextual adjacency relationship formed by the words before and after the semantic core word are matched with the rule entries in the classification rule base to determine whether there is a classification rule that meets the triggering conditions. The semantic core word weighted feature vector and the classification rule matching result that meets the triggering condition are input into the classification decision logic, and the message category result is generated through logical judgment; The priority type of the data message content is determined based on the message category result. The priority type is written into the identifier field of the semantic priority label to generate the semantic priority label.

3. The credit control method for data messaging services as described in claim 2, characterized in that: The semantic priority tags are high-priority urgent messages, normal-priority social messages, and low-priority commercial promotion messages.

4. The credit control method for data messaging services as described in claim 2, characterized in that: The classification rule base is a collection that stores classification rule entries, including semantic core word patterns, positional order, and adjacency conditions defined based on the message corpus.

5. The credit control method for data messaging services as described in claim 1, characterized in that: The process involves inputting throughput, transmission delay, and packet loss rate metrics into a congestion assessment algorithm for multi-parameter fusion analysis and network congestion status determination, outputting the current network congestion level. Specifically, this involves... The throughput, transmission delay, and packet loss rate metrics are input into the congestion assessment algorithm, and a weighted fusion process is performed to generate a comprehensive congestion score. The overall congestion score is compared with the preset congestion level threshold range to determine the range of the congestion level threshold range to which the overall congestion score belongs. Perform state mapping on the level identifiers corresponding to the congestion level threshold range to generate the current network congestion level.

6. The credit control method for data messaging services as described in claim 1, characterized in that: The generation of credit control instructions is specifically as follows: The credit decision center receives the semantic priority label and the current network congestion level, and performs integrity verification on the semantic priority label and the current network congestion level; Using the verified semantic priority label and the current network congestion level as the query key, a matching search is performed on the preset policy decision matrix within the credit decision center to determine the policy entries in the policy decision matrix that correspond to the semantic priority label and the current network congestion level. Extract the instruction type from the policy entry and use it as a credit control instruction.

7. The credit control method for data messaging services as described in claim 1, characterized in that: The establishment of the instruction execution status pool refers to receiving credit control instructions, creating an instruction execution status data structure, defining the fields of the instruction execution status data structure as instruction identifier field and execution status field, and writing the identifier and execution status field of the credit control instruction into the instruction execution status data structure.

8. The credit control method for data messaging services as described in claim 1, characterized in that: When the temporary message for the rejected sending instruction detects that the current network congestion level has returned to a smooth state, any available instruction among the authorized overdraft sending instruction and the request prepayment instruction is reactivated, and the corresponding instruction's credit control is executed. Specifically, When the "refuse to send" instruction is set to the "execute" state, the data message is placed in the temporary queue, and the current network congestion level is continuously obtained. Determine whether the current network congestion level has recovered to a smooth state. If it has recovered, remove the execution state of "refuse to send command" from the command execution state pool. Check whether the execution conditions for the authorized overdraft sending instruction are met. If they are met, activate the execution state of the authorized overdraft sending instruction and perform credit control for the authorized overdraft sending instruction. If not, check whether the execution conditions of the prepayment instruction are met. If they are met, activate the execution state of the prepayment instruction and execute the credit control of the prepayment instruction.

9. A computer device comprising a memory and a processor, wherein the memory stores a computer program, characterized in that: When the processor executes the computer program, it implements the steps of the credit control method for data message services according to any one of claims 1 to 8.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by the processor, it implements the steps of the credit control method for data message services according to any one of claims 1 to 8.

Images