Command processing method and device of simulation passenger service system and storage medium

Abstract

The invention discloses a command processing method and device of a simulation passenger service system and a storage medium. Relates to the deep learning field, and the method comprises the following steps: receiving information input by a user, and obtaining state information recorded by a current passenger name, the information input by the user at least carrying a business instruction; determining a cue word according to the information input by the user and the state information recorded by the current passenger name; the cue word is input into a target model, and the target model is obtained after a low-rank matrix introduced into the initial model is adjusted based on a training data set; and outputting a target response result through the target model, the target response result including state information recorded by the passenger name adjusted according to the business instruction. According to the method and the device, the problem of poor response result accuracy of an analogue simulation passenger service system in related technologies is solved.

Description

Technical Field

[0001] This application relates to the field of deep learning, and more specifically, to a command processing method, apparatus, and storage medium for a simulated passenger service system. Background Technology

[0002] In today's civil aviation industry, the Passenger Service System (PSS) is a core pillar of airline operations, and its stability and reliability directly impact the travel experience of millions of passengers worldwide. However, traditional mainframe architectures, despite their superior transaction processing capabilities and high reliability, face challenges such as high maintenance costs, hardware silos, and barriers to integrating new technologies. This is particularly true in non-production environments such as development, testing, and training, where the heavy reliance on dedicated physical mainframe hardware hinders industry development. In recent years, virtualization technologies and protocol simulators have been widely explored to build low-cost, highly flexible PSS simulation environments to meet the demands of the software development cycle for realistic system behavior. However, issues such as modeling complexity, insufficient generalization ability, and difficulty in maintaining format consistency result in generally low accuracy of simulation system responses. These simulations cannot perfectly replicate the deterministic behavior of native mainframes, especially when dealing with state-sensitive business logic, such as the real-time updating and synchronization of Passenger Name Records (PNRs). They struggle to accurately respond to input commands, hindering the development of efficient development and testing toolchains in the modernization of PSS systems.

[0003] There is currently no effective solution to the problem of poor accuracy in the response results of simulated passenger service systems in related technologies. Summary of the Invention

[0004] The main objective of this application is to provide a command processing method, apparatus, and storage medium for a simulated passenger service system, in order to solve the problem of poor accuracy of response results in simulated passenger service systems in related technologies.

[0005] To achieve the above objectives, according to one aspect of this application, a command processing method for a simulated passenger service system is provided. The method includes: receiving user input information and obtaining current passenger name record status information, wherein the user input information carries at least a business instruction; determining a prompt word based on the user input information and the current passenger name record status information; inputting the prompt word into a target model, wherein the target model is a model obtained by adjusting a low-rank matrix introduced into an initial model based on a training dataset; and outputting a target response result through the target model, wherein the target response result includes the passenger name record status information adjusted according to the business instruction.

[0006] Optionally, the target model is trained through the following steps: introducing a low-rank matrix into the attention layer of the initial model; determining the training dataset, which includes the status information of the sample passenger information records, sample input commands, and sample response results; inputting the status information of the sample passenger information records and sample input commands into the initial model; outputting the predicted response result through the initial model; calculating the loss value of the predicted response result and the sample response result based on the loss function; adjusting the low-rank matrix introduced into the initial model through the loss value to obtain the adjusted initial model; repeating the steps of inputting the status information of the sample passenger information records and sample input commands into the adjusted initial model, the adjusted initial model outputting a new predicted response result, calculating a new loss value of the new predicted response result and the sample response result based on the loss function, and adjusting the low-rank matrix through the new loss value to obtain a new adjusted initial model, until the preset conditions are met to obtain the target model.

[0007] Optionally, determining the training dataset includes: collecting historical interaction logs from the passenger service system to obtain initial training data; removing user personal identification information from the initial training data and replacing it with anonymous representation to obtain first training data; standardizing the data format of the first training data to obtain second training data; extracting target fields from the second training data, wherein the target fields include the status information field of the passenger name record before the response, the input instruction field, and the status information field of the passenger information record after the response; and generating a training dataset based on the target fields.

[0008] Optionally, determining the prompt word based on the user input information and the current status information of the passenger name record includes: converting the current status information of the passenger name record into passenger information description text; parsing the business instruction by analyzing the user input information; and concatenating the passenger information description text and the business instruction into a prompt word.

[0009] Optionally, after inputting the prompt words into the target model, the method further includes: receiving and parsing the prompt words through the target model to determine the business instructions in the prompt words; and adjusting the status information of the current passenger name record through the target model based on the business instructions and passenger information description text.

[0010] Optionally, after outputting the target response result through the target model, the method further includes: inputting business instructions into the passenger service system; outputting the actual response result through the passenger service system; calculating the consistency rate between the actual response result and the target response result; and adjusting the target model if the consistency rate is less than a preset threshold.

[0011] Optionally, calculating the consistency rate between the actual response result and the target response result includes: using target matching to calculate the consistency rate between the characters in the actual response result and the characters in the target response result.

[0012] To achieve the above objectives, according to another aspect of this application, a command processing apparatus for a simulated passenger service system is provided. The apparatus includes: a receiving unit for receiving user input information and acquiring current passenger name record status information, wherein the user input information carries at least a business instruction; a first determining unit for determining a prompt word based on the user input information and the current passenger name record status information; a first input unit for inputting the prompt word into a target model, wherein the target model is a model obtained by adjusting a low-rank matrix introduced into an initial model based on a training dataset; and a first output unit for outputting a target response result through the target model, wherein the target response result includes passenger name record status information adjusted according to the business instruction.

[0013] Optionally, the device further includes: an introduction unit for introducing a low-rank matrix into the attention layer of the initial model; a second determination unit for determining a training dataset, wherein the training dataset includes the status information of the sample passenger information record, sample input instructions, and sample response results; a second input unit for inputting the status information of the sample passenger information record and the sample input instructions into the initial model; a second output unit for outputting a predicted response result through the initial model; a first calculation unit for calculating the loss value of the predicted response result and the sample response result based on a loss function; a first adjustment unit for adjusting the low-rank matrix introduced into the initial model through the loss value to obtain an adjusted initial model; and a second adjustment unit for repeatedly executing the steps of inputting the status information of the sample passenger information record and the sample input instructions into the adjusted initial model, the adjusted initial model outputting a new predicted response result, calculating a new loss value of the new predicted response result and the sample response result based on the loss function, and adjusting the low-rank matrix through the new loss value to obtain a new adjusted initial model, until a preset condition is met to obtain the target model.

[0014] Optionally, the second determining unit includes: a collection module for collecting historical interaction logs of the passenger service system to obtain initial training data; a removal module for removing user personal identification information from the initial training data and replacing it with anonymous representation to obtain first training data; a unified format module for unifying the data format of the first training data to obtain second training data; an extraction module for extracting target fields from the second training data, wherein the target fields include the status information field of the passenger name record before the response, the input instruction field, and the status information field of the passenger information record after the response; and a generation module for generating a training dataset based on the target fields.

[0015] Optionally, the first determining unit includes: a conversion module for converting the current passenger name record status information into passenger information description text; a parsing module for parsing out business instructions by analyzing user input information; and a splicing module for splicing the passenger information description text and business instructions into prompt words.

[0016] Optionally, the device further includes: a third determining unit, used to receive and parse the prompt words through the target model after the prompt words are input into the target model, and determine the business instructions in the prompt words; and a third adjusting unit, used to adjust the status information of the current passenger name record through the target model according to the business instructions and passenger information description text.

[0017] Optionally, the device further includes: a third input unit for inputting business instructions into the passenger service system after outputting the target response result through the target model; a third output unit for outputting the actual response result through the passenger service system; a second calculation unit for calculating the consistency rate between the actual response result and the target response result; and a fourth adjustment unit for adjusting the target model if the consistency rate is less than a preset threshold.

[0018] Optionally, the second calculation unit includes: a calculation module for calculating the consistency rate between the characters in the actual response result and the characters in the target response result using a target matching method.

[0019] According to another aspect of this application, a computer-readable storage medium is provided, which includes a stored program, wherein, when the program is executed, it controls the device where the computer-readable storage medium is located to execute any command processing method of a simulated passenger service system.

[0020] According to another aspect of this application, an electronic device is provided, comprising: one or more processors, a memory, and one or more programs, wherein the one or more programs are stored in the memory and configured to be executed by the one or more processors, and the one or more programs include a command processing method for executing any simulated passenger service system.

[0021] According to another aspect of this application, a computer program product is provided, including computer instructions, which, when executed by a processor, implement the steps of a command processing method for a simulated passenger service system as described above.

[0022] In this embodiment, by receiving user input information and obtaining the current status information of passenger name records, wherein the user input information carries at least a business instruction; determining prompt words based on the user input information and the current status information of passenger name records; inputting the prompt words into a target model, wherein the target model is a model obtained by adjusting the low-rank matrix introduced into the initial model based on the training dataset; and outputting a target response result through the target model, wherein the target response result includes the status information of passenger name records adjusted according to the business instruction, thereby solving the technical problem of poor accuracy of the response results of the simulated passenger service system. In this application, prompt words are determined based on the received user input information and the obtained current status information of passenger name records. The user input information carries a business instruction. The prompt words are input into the trained target model, and the target model outputs a target response result, which includes the status information of passenger name records adjusted according to the business instruction, thereby achieving the technical effect of improving the accuracy of the response results of the simulated passenger service system. Attached Figure Description

[0023] The accompanying drawings, which form part of this application, are used to provide a further understanding of this application. The illustrative embodiments and descriptions of this application are used to explain this application and do not constitute an undue limitation of this application. In the drawings:

[0024] Figure 1 A hardware block diagram of a computer terminal for implementing a command processing method of a simulated passenger service system is shown.

[0025] Figure 2 This is a flowchart of a command processing method for a simulated passenger service system provided according to an embodiment of this application;

[0026] Figure 3 This is a schematic diagram of a command processing device for a simulated passenger service system provided according to an embodiment of this application;

[0027] Figure 4 This is a structural block diagram of an electronic device according to an embodiment of this application. Detailed Implementation

[0028] To enable those skilled in the art to better understand the present application, the technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present application, and not all embodiments. Based on the embodiments in the present application, all other embodiments obtained by those of ordinary skill in the art without creative effort should fall within the scope of protection of the present application.

[0029] It should be noted that the terms "first," "second," etc., in the specification, claims, and accompanying drawings of this application are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such data can be interchanged where appropriate so that the embodiments of this application described herein can be implemented in orders other than those illustrated or described herein. Furthermore, the terms "comprising" and "having," and any variations thereof, are intended to cover non-exclusive inclusion; for example, a process, method, system, product, or apparatus that comprises a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or units not explicitly listed or inherent to such processes, methods, products, or apparatus.

[0030] Example 1

[0031] According to an embodiment of this application, a method embodiment for command processing of a simulated passenger service system is also provided. It should be noted that the steps shown in the flowchart in the accompanying drawings can be executed in a computer system such as a set of computer-executable instructions. Furthermore, although a logical order is shown in the flowchart, in some cases, the steps shown or described may be executed in a different order than that shown here.

[0032] The method embodiment provided in Embodiment 1 of this application can be executed on a mobile terminal, computer terminal, or similar computing device. Figure 1 A hardware block diagram of a computer terminal (or mobile device) for implementing a command processing method in a simulated passenger service system is shown. Figure 1 As shown, the computer terminal 10 (or mobile device) may include one or more processors 102 (shown as 102a, 102b, ..., 102n in the figure) 102 (processor 102 may include, but is not limited to, a microprocessor MCU or a programmable logic device FPGA, etc.), a memory 104 for storing data, and a transmission device 106 for communication functions. In addition, it may also include: a display, an input / output interface (I / O interface), a Universal Serial Bus (USB) port (which may be included as one of the ports of a BUS bus), a network interface, a power supply, and / or a camera. Those skilled in the art will understand that... Figure 1 The structure shown is for illustrative purposes only and does not limit the structure of the aforementioned electronic device. For example, computer terminal 10 may also include... Figure 1 The more or fewer components shown, or having the same Figure 1 The different configurations shown.

[0033] It should be noted that the aforementioned one or more processors 102 and / or other data processing circuits are generally referred to herein as "data processing circuits". These data processing circuits may be embodied, in whole or in part, in software, hardware, firmware, or any other combination thereof. Furthermore, the data processing circuits may be a single, independent processing module, or may be integrated, in whole or in part, into any other element within the computer terminal 10 (or mobile device). As involved in the embodiments of this application, the data processing circuits serve as a processor control mechanism (e.g., selection of a variable resistor termination path connected to an interface).

[0034] The memory 104 can be used to store software programs and modules of application software, such as the program instructions / data storage device corresponding to the command processing method of the simulated passenger service system in this embodiment of the application. The processor 102 executes various functional applications and data processing by running the software programs and modules stored in the memory 104, thereby realizing the command processing method of the simulated passenger service system described above. The memory 104 may include high-speed random access memory, and may also include non-volatile memory, such as one or more magnetic storage devices, flash memory, or other non-volatile solid-state memory. In some instances, the memory 104 may further include memory remotely located relative to the processor 102, and these remote memories can be connected to the computer terminal 10 via a network. Examples of such networks include, but are not limited to, the Internet, corporate intranets, local area networks, mobile communication networks, and combinations thereof.

[0035] The transmission device 106 is used to receive or send data via a network. Specific examples of the network described above may include a wireless network provided by the communication provider of the computer terminal 10. In one example, the transmission device 106 includes a Network Interface Controller (NIC), which can connect to other network devices via a base station to communicate with the Internet. In another example, the transmission device 106 may be a Radio Frequency (RF) module, used for wireless communication with the Internet.

[0036] The display can be, for example, a touchscreen liquid crystal display (LCD), which allows the user to interact with the user interface of the computer terminal 10 (or mobile device).

[0037] Under the aforementioned operating environment, this application provides the following: Figure 2 The command processing method of the simulated passenger service system is shown. Figure 2 This is a flowchart of the command processing method of the simulated passenger service system according to Embodiment 1 of this application.

[0038] Step S201: Receive information input by the user and obtain the status information of the current passenger name record, wherein the information input by the user carries at least a business instruction.

[0039] Optionally, user-inputted information can be presented in natural language or a specific command format, containing at least one business instruction, such as "NM" (Name Entry), "TK" (Ticket Issuance), "OSI" (Special Service Information), or "CT" (Check-in Transaction). Simultaneously, the system needs to obtain the current status information of passenger name records. Passenger name records are the core data structure of the passenger service system, containing all information about passenger itineraries, such as flight reservations, passenger details, and special service requests. The system can read the latest passenger name record status information through interaction with databases or other data sources, ensuring that subsequent processing is based on the most up-to-date information.

[0040] Step S202: Determine the prompt word based on the information entered by the user and the status information of the current passenger name record.

[0041] Optionally, prompt words serve as guidance for the large language model in processing the input content, containing natural language representations of the user's input information and the current status information of the passenger name record. By constructing more accurate prompt words, complete contextual information can be effectively conveyed to the target model, including the intent of the business instruction and the specific content of the current passenger name record's status information. This helps the target model generate correct response results based on existing experience and knowledge.

[0042] Step S203: Input the prompt word into the target model, where the target model is a model obtained by adjusting the low-rank matrix introduced into the initial model based on the training dataset.

[0043] Optionally, the constructed prompt words are input into the target model, which is obtained by adjusting the low-rank matrix introduced into the initial model. The initial model is typically a large language model that has been pre-trained using extensive source code and technical documentation. Once the prompt words are input into the target model, the target model can use its internal sequence learning mechanism, referencing the existing training dataset, to generate response results that conform to the business instructions.

[0044] Optionally, adjusting the low-rank matrix introduced into the initial model can improve the processing performance of the target model in specific business domains and reduce training resource requirements without changing the core of the model, by adjusting only some parameters of the attention mechanism.

[0045] Step S204: Output the target response result through the target model, wherein the target response result includes the status information of the passenger name record after adjustment according to the business instruction.

[0046] Optionally, the target model generates a target response result based on the input prompt words. The target response result includes the status information of the updated passenger name record after the execution of the business instruction, and the target response result conforms to the passenger service specifications.

[0047] The command processing method for a simulated passenger service system provided in this application receives user input information and obtains the current status information of passenger name records, wherein the user input information carries at least a business instruction; based on the user input information and the current status information of passenger name records, a prompt word is determined; the prompt word is input into a target model, wherein the target model is a model obtained by adjusting the low-rank matrix introduced into the initial model based on the training dataset; and the target model outputs a target response result, wherein the target response result includes the status information of passenger name records adjusted according to the business instruction, thus solving the technical problem of poor accuracy of the response results of the simulated passenger service system. In this application, the prompt word is determined based on the received user input information and the obtained current status information of passenger name records. The user input information carries a business instruction. The prompt word is input into the trained target model, and the target model outputs a target response result, which includes the status information of passenger name records adjusted according to the business instruction, thereby achieving the technical effect of improving the accuracy of the response results of the simulated passenger service system.

[0048] To make the response results of the target model more accurate, optionally, in the command processing method of the simulated passenger service system provided in this application embodiment, the target model is trained through the following steps:

[0049] The first step is to introduce a low-rank matrix into the attention layer of the initial model.

[0050] Optionally, introducing a low-rank matrix into the attention layer of the initial model is an efficient parameter tuning mechanism that minimizes the increase in model parameters while ensuring that the target model can capture key information in the new learning task. In the initial model based on the Transformer architecture, parameters are distributed across various layers, including but not limited to the embedding layer, multi-head attention layer, feedforward neural network layer, and the final output layer. During fine-tuning, most parameters of the initial model remain unchanged, and only a small subset of newly added low-rank matrix parameters in the attention layer are adjusted.

[0051] The second step is to determine the training dataset, which includes the status information of the sample passenger information records, sample input commands, and sample response results.

[0052] Optionally, the training dataset includes the status information of the sample passenger information records, sample input commands, and their corresponding sample response results. The status information of the sample passenger information records reflects the details of the passenger's itinerary in the passenger service system, including flight reservations, passenger information, etc.; the sample input commands are the instructions issued by the user or the system in the passenger service system, such as name entry (NM), ticket issuance (TK), etc.; and the sample response results are the output generated by the system after executing the command, including the updated PNR status and other relevant information.

[0053] The third step is to input the status information of the sample passenger information record and the sample input instructions into the initial model.

[0054] Optionally, the status information of the sample passenger information record and the sample input instructions are input into the initial model, providing the initial model with specific business scenarios and instructions to be processed.

[0055] The fourth step is to output the predicted response results using the initial model.

[0056] Optionally, the predicted response output by the initial model is the initial prediction of the sample input command based on its current knowledge and capabilities, reflecting the model's preliminary understanding of the sample input command and the expected system behavior.

[0057] The fifth step is to calculate the loss value of the predicted response and the sample response based on the loss function.

[0058] Optionally, the loss value between the predicted response and the sample response is calculated based on the loss function. The loss function can be of various types, such as cross-entropy loss and mean squared error. The purpose is to quantify the difference between the model output and the expected output, and to provide a clear direction for adjusting the model.

[0059] The sixth step is to adjust the low-rank matrix introduced into the initial model by adjusting the loss value, thus obtaining the adjusted initial model.

[0060] Optionally, adjusting the low-rank matrix introduced into the initial model using the loss value can narrow the gap between the predicted output and the true output. The adjustment process is driven by the backpropagation algorithm, and the parameters of the low-rank matrix introduced into the initial model can be updated by calculating the gradient and applying the optimizer.

[0061] Step 7: Repeat the process of inputting the status information recorded by the sample passenger information and the sample input command into the adjusted initial model. The adjusted initial model outputs a new predicted response result. Based on the loss function, calculate the new loss value of the new predicted response result and the sample response result. Adjust the low-rank matrix with the new loss value to obtain the new adjusted initial model. Continue until the preset conditions are met to obtain the target model.

[0062] Optionally, the process of inputting sample data, predicting response results, and adjusting model parameters is repeated. This iterative process continues until preset conditions are met. Preset conditions include, but are not limited to, the accuracy of the model output reaching a certain threshold, the completion of the training cycle, or the minimum loss on the validation set.

[0063] In summary, through the above steps, the entire training process, from the introduction of the low-rank matrix to the determination of the training dataset, and then to the iterative training process of model prediction and parameter adjustment, constructs a highly accurate target model that can effectively mimic the passenger service system. The target model has high accuracy in responding to business instructions.

[0064] In order to train a target model that can effectively simulate a passenger service system, optionally, in the command processing method for simulating a passenger service system provided in this application embodiment, the training dataset includes:

[0065] The first step is to collect historical interaction logs from the passenger service system to obtain initial training data.

[0066] Optionally, the historical interaction log is a record of past interactions between the passenger service system and the user, including various commands entered by the user, the system's response results after executing these commands, and system status information during the execution of the commands.

[0067] The second step is to remove the user's personal identification information from the initial training data and replace it with anonymous representation to obtain the first training data.

[0068] Optionally, data cleaning and privacy protection algorithms can be used to identify and remove personally identifiable information in logs, such as real names, ID card numbers, and contact numbers, and replace them with uniform anonymous identifiers, such as "User_A" and "User_B".

[0069] The third step is to standardize the data format of the first training data to obtain the second training data.

[0070] Optionally, the first training data can be standardized to unify the format of fields such as date representation, flight number, and service code, making the data easier for the model to understand and process. For example, the date field in the historical interaction log can be converted to a unified ISO 8601 format (such as YYYY-MM-DD), the flight number prefix can be unified, such as converting "CA" to "CCA", and other fields can be standardized.

[0071] The fourth step is to extract the target fields from the second training data. The target fields include the status information field of the passenger name record before the response, the input instruction field, and the status information field of the passenger information record after the response.

[0072] Optionally, key fields can be extracted from the second training data using regular expressions, natural language processing algorithms, or specialized data parsing tools. These fields include the status information field of the passenger name record before the response, the input instruction field, and the status information field of the passenger name record after the response. These fields are the core components for constructing the triples in the training dataset.

[0073] The fifth step is to generate a training dataset based on the target fields.

[0074] Optionally, based on the target fields extracted above, the final training dataset can be constructed. Each sample consists of a triple: (instruction, input, output), where instruction is the status information field of the passenger name record before the response, input is the input instruction field, and output is the status information field of the passenger name record after the response.

[0075] In summary, through the above steps, from log collection to data preprocessing and format standardization, to key field extraction and training dataset construction, a high-quality training dataset is obtained, ensuring that the target model can accurately learn the behavior patterns and rules of the passenger service system and provide high-fidelity command processing and response generation services.

[0076] To ensure the target model outputs a more accurate response, optionally, in the command processing method of the simulated passenger service system provided in this application embodiment, the prompt words are determined based on the user input information and the current status information of the passenger name record, including:

[0077] The first step is to convert the current passenger name record status information into passenger information description text.

[0078] Optionally, the status information of the current passenger name record can be converted into passenger information description text. The purpose is to convert the complex status and details contained in the passenger name record into a machine-readable form so as to serve as contextual information to guide the processing of the target model.

[0079] The second step is to parse out the business instructions by analyzing the information input by the user.

[0080] Optionally, the information entered by the user may contain multiple parts, such as business instructions, parameters, and operators. Business instructions refer to the specific operations that the user wants to perform in the passenger service system, such as "NM" (name entry), "TK" (ticket issuance), "OSI" (special service information addition), or "CT" (check-in confirmation). By parsing the information entered by the user, the business instructions can be identified.

[0081] The third step is to combine the passenger information description text and business instructions into a prompt word.

[0082] Optionally, concatenating passenger information description text and business instructions into prompts using a specific format can help the target model better understand the context and generate responses that conform to actual business logic. For example, a specific template, such as "Current PNR Information: ...; User Instruction: ...", can be designed to incorporate the information obtained in the first and second steps into a complete prompt.

[0083] In summary, the above steps provide clear prompts for the target model, enabling it to output more accurate target response results.

[0084] To enable the target model to more accurately adjust the status information of passenger name records according to business instructions, optionally, in the command processing method of the simulated passenger service system provided in this application embodiment, after inputting the prompt word into the target model, the method further includes:

[0085] The first step is to receive and parse the prompt words through the target model to determine the business instructions in the prompt words.

[0086] Optionally, after the prompt word is input into the target model, the target model can receive and parse the prompt word through its internal semantic parsing mechanism to determine the business instructions contained therein.

[0087] The second step is to adjust the status information of the current passenger name record based on the business instructions and passenger information description text, using the target model.

[0088] Optionally, the target model adjusts the current status information of the passenger name record based on the meaning of the business instruction and the passenger information description text. The target model can combine its learned knowledge and rules to generate updated status information for the passenger name record. The target model can perform sequence reasoning based on the business instruction, considering the current status information of the passenger name record. If the business instruction is "NM" (name entry), the target model will generate a new passenger name record entry to reflect the addition of the name based on the information in the prompt word; if it is "CT" (check-in confirmation), the target model will generate status information for the passenger name record that includes the check-in confirmation status.

[0089] In summary, through the above steps, the target model can respond to business instructions more accurately, from parsing the business instructions in the prompt words to adjusting the status information of the passenger name record according to the instructions.

[0090] To further adjust the target model based on the target response result output by the target model, optionally, in the command processing method of the simulated passenger service system provided in this application embodiment, after outputting the target response result through the target model, the method further includes:

[0091] The first step is to input the business instructions into the passenger service system.

[0092] Optionally, business instructions can be directly input into a real passenger service system. A real passenger service system refers to an actual operating passenger service system based on a mainframe or legacy system architecture, which serves as a reference standard for evaluating the accuracy of the target model simulation.

[0093] The second step is to output the actual response results through the passenger service system.

[0094] Optionally, upon receiving a business instruction, the actual passenger service system will generate a realistic response based on its internal rules and status. This realistic response can help evaluate the accuracy of the target response generated by the target model, including aspects such as format and content details.

[0095] The third step is to calculate the consistency rate between the actual response results and the target response results.

[0096] Optionally, the consistency rate between the actual response and the target response can be calculated. Consistency verification can be based on the Exact Match (EM) metric, which requires that the two responses be identical at the character level, including all fields, delimiters, spaces, line numbers, and status codes. A string comparison algorithm can be used to compare the actual and target responses character by character, and the proportion of perfectly matching characters to the total number of characters is used as the consistency rate.

[0097] The fourth step is to adjust the target model if the consistency rate is less than the preset threshold.

[0098] Optionally, if the consistency rate is lower than a preset threshold, it indicates that the output of the target model differs significantly from the real system, and the target model needs to be adjusted. The accuracy of the target model's generated results can be improved by retraining or fine-tuning the target model.

[0099] In summary, by implementing a closed-loop feedback mechanism through the above steps, the target model can be continuously optimized, its performance in the simulation environment can be improved, and high-fidelity simulation output can ultimately be achieved.

[0100] To accurately assess the consistency between the target response result and the actual response result, optionally, in the command processing method of the simulated passenger service system provided in this application embodiment, calculating the consistency rate between the actual response result and the target response result includes: using a target matching method to calculate the consistency rate between the characters of the actual response result and the characters of the target response result.

[0101] Optionally, target matching can be based on the Exact Match (EM) metric, which compares the target response generated by the target model with the actual response one-to-one, ensuring that every character, including spaces, separators, numbers, letters, and even newlines, is completely identical. The total number of characters matching the actual response in the target response is counted, and divided by the total number of characters in the actual response to obtain the consistency rate. A higher consistency rate indicates that the target response generated by the target model is closer to the actual system output, and the better the fidelity in format and content. Using target matching to calculate the consistency rate ensures that the response generated by the target model achieves character-level precision consistency with the actual passenger service system output.

[0102] It should be noted that the steps shown in the flowchart in the accompanying drawings can be executed in a computer system such as a set of computer-executable instructions, and although a logical order is shown in the flowchart, in some cases the steps shown or described may be executed in a different order than that shown here.

[0103] Example 2

[0104] This application also provides a command processing device for a simulated passenger service system. It should be noted that the command processing device for the simulated passenger service system in this application can be used to execute the command processing method for the simulated passenger service system provided in this application. The command processing device for the simulated passenger service system provided in this application is described below.

[0105] According to embodiments of this application, an apparatus for implementing the command processing method of the above-described simulated passenger service system is also provided, such as... Figure 3 As shown, the device includes: a receiving unit 301, a first determining unit 302, a first input unit 303, and a first output unit 304.

[0106] Specifically, the receiving unit 301 is used to receive information input by the user and obtain the status information of the current passenger name record, wherein the information input by the user carries at least a business instruction.

[0107] The first determining unit 302 is used to determine the prompt word based on the information input by the user and the status information of the current passenger name record;

[0108] The first input unit 303 is used to input the prompt word into the target model, wherein the target model is a model obtained by adjusting the low-rank matrix introduced into the initial model based on the training dataset.

[0109] The first output unit 304 is used to output the target response result through the target model, wherein the target response result includes the status information of the passenger name record adjusted according to the business instruction.

[0110] The command processing device for the simulated passenger service system provided in this application embodiment receives user input information through a receiving unit 301 and obtains the current status information of the passenger name record. The user input information carries at least a business instruction. A first determining unit 302 determines a prompt word based on the user input information and the current status information of the passenger name record. A first input unit 303 inputs the prompt word into a target model, where the target model is a model obtained by adjusting the low-rank matrix introduced into the initial model based on the training dataset. A first output unit 304 outputs a target response result through the target model, where the target response result includes the status information of the passenger name record adjusted according to the business instruction. This solves the problem of poor accuracy of the response results in simulated passenger service systems in related technologies, achieving the effect of improving the accuracy of the response results in simulated passenger service systems.

[0111] Optionally, in the command processing device of the simulated passenger service system provided in this application embodiment, an introduction unit is used to introduce a low-rank matrix into the attention layer of the initial model; a second determination unit is used to determine the training dataset, wherein the training dataset includes the status information of the sample passenger information record, the sample input command, and the sample response result; a second input unit is used to input the status information of the sample passenger information record and the sample input command into the initial model; a second output unit is used to output the predicted response result through the initial model; a first calculation unit is used to calculate the loss value of the predicted response result and the sample response result based on the loss function; a first adjustment unit is used to adjust the low-rank matrix introduced into the initial model through the loss value to obtain the adjusted initial model; a second adjustment unit is used to repeatedly execute the steps of inputting the status information of the sample passenger information record and the sample input command into the adjusted initial model, the adjusted initial model outputting a new predicted response result, calculating a new loss value of the new predicted response result and the sample response result based on the loss function, and adjusting the low-rank matrix through the new loss value to obtain a new adjusted initial model, until a preset condition is reached to obtain the target model.

[0112] Optionally, in the command processing device of the simulated passenger service system provided in this application embodiment, the second determining unit includes: a collection module, used to collect historical interaction logs of the passenger service system to obtain initial training data; a removal module, used to remove user personal identity information from the initial training data and replace the user personal identity information in the initial training data with anonymous representation to obtain first training data; a unified format module, used to unify the data format of the first training data to obtain second training data; an extraction module, used to extract target fields from the second training data, wherein the target fields include the status information field of the passenger name record before the response, the input instruction field, and the status information field of the passenger information record after the response; and a generation module, used to generate a training dataset based on the target fields.

[0113] Optionally, in the command processing device of the simulated passenger service system provided in this application embodiment, the first determining unit 302 includes: a conversion module, used to convert the status information of the current passenger name record into passenger information description text; a parsing module, used to parse out the business instruction by analyzing the information input by the user; and a splicing module, used to splice the passenger information description text and the business instruction into a prompt word.

[0114] Optionally, in the command processing device of the simulated passenger service system provided in the embodiments of this application, the device further includes: a third determining unit, used to receive and parse the prompt word through the target model after the prompt word is input into the target model, and determine the business instruction in the prompt word; and a third adjusting unit, used to adjust the status information of the current passenger name record through the target model according to the business instruction and passenger information description text.

[0115] Optionally, in the command processing device of the simulated passenger service system provided in the embodiments of this application, the device further includes: a third input unit, used to input business instructions into the passenger service system after outputting the target response result through the target model; a third output unit, used to output the real response result through the passenger service system; a second calculation unit, used to calculate the consistency rate between the real response result and the target response result; and a fourth adjustment unit, used to adjust the target model when the consistency rate is less than a preset threshold.

[0116] Optionally, in the command processing device of the simulated passenger service system provided in this application embodiment, the second calculation unit includes: a calculation module, used to calculate the consistency rate between the characters of the real response result and the characters of the target response result by means of target matching.

[0117] It should be noted that the receiving unit 301, the first determining unit 302, the first input unit 303, and the first output unit 304 mentioned above correspond to steps S201 to S204 in Embodiment 1. The four units and the corresponding steps implement the same instances and application scenarios, but are not limited to the content disclosed in Embodiment 1. It should be noted that the above modules or units can be hardware or software components stored in memory (e.g., memory 104) and processed by one or more processors (e.g., processors 102a, 102b, ..., 102n). The above modules can also be part of a device and run in the computer terminal 10 provided in Embodiment 1.

[0118] Example 3

[0119] Embodiments of this application may provide a computer terminal, which may be any computer terminal device in a group of computer terminals. Optionally, in this embodiment, the aforementioned computer terminal may also be replaced with a mobile terminal or an electronic device, etc.

[0120] Optionally, in this embodiment, the computer terminal may be located in at least one of a plurality of network devices in a computer network.

[0121] In this embodiment, the computer terminal described above can execute the program code for the following steps in the command processing method of the simulated passenger service system: receiving information input by the user and obtaining the current status information of the passenger name record, wherein the information input by the user carries at least a business instruction; determining a prompt word based on the information input by the user and the current status information of the passenger name record; inputting the prompt word into the target model, wherein the target model is a model obtained by adjusting the low-rank matrix introduced into the initial model based on the training dataset; and outputting a target response result through the target model, wherein the target response result includes the status information of the passenger name record adjusted according to the business instruction.

[0122] Optionally, the aforementioned computer terminal can execute the program code for the following steps in the command processing method of the simulated passenger service system: The target model is trained through the following steps: introducing a low-rank matrix into the attention layer of the initial model; determining the training dataset, wherein the training dataset includes the status information of the sample passenger information record, sample input commands, and sample response results; inputting the status information of the sample passenger information record and the sample input commands into the initial model; outputting the predicted response result through the initial model; calculating the loss value of the predicted response result and the sample response result based on the loss function; adjusting the low-rank matrix introduced into the initial model through the loss value to obtain the adjusted initial model; repeatedly executing the steps of inputting the status information of the sample passenger information record and the sample input commands into the adjusted initial model, the adjusted initial model outputting a new predicted response result, calculating a new loss value of the new predicted response result and the sample response result based on the loss function, adjusting the low-rank matrix through the new loss value to obtain a new adjusted initial model, until the preset conditions are met to obtain the target model.

[0123] Optionally, the aforementioned computer terminal can execute the program code for the following steps in the command processing method of the simulated passenger service system: determining the training dataset includes: collecting historical interaction logs of the passenger service system to obtain initial training data; removing user personal identification information from the initial training data and replacing it with anonymous representation to obtain first training data; unifying the data format of the first training data to obtain second training data; extracting target fields from the second training data, wherein the target fields include the status information field of the passenger name record before the response, the input instruction field, and the status information field of the passenger information record after the response; and generating a training dataset based on the target fields.

[0124] Optionally, the aforementioned computer terminal can execute the program code for the following steps in the command processing method of the simulated passenger service system: determining the prompt word based on the information input by the user and the status information of the current passenger name record, including: converting the status information of the current passenger name record into passenger information description text; parsing out the business instruction by analyzing the information input by the user; and concatenating the passenger information description text and the business instruction into a prompt word.

[0125] Optionally, the aforementioned computer terminal can execute the program code for the following steps in the command processing method of the simulated passenger service system: after inputting the prompt word into the target model, the method further includes: receiving and parsing the prompt word through the target model to determine the business instruction in the prompt word; and adjusting the status information of the current passenger name record through the target model according to the business instruction and the passenger information description text.

[0126] Optionally, the aforementioned computer terminal can execute the program code for the following steps in the command processing method of the simulated passenger service system: after outputting the target response result through the target model, the method further includes: inputting the business instruction into the passenger service system; outputting the real response result through the passenger service system; calculating the consistency rate between the real response result and the target response result; and adjusting the target model if the consistency rate is less than a preset threshold.

[0127] Optionally, the computer terminal described above can execute program code for the following steps in the command processing method of the simulated passenger service system: calculating the consistency rate between the actual response result and the target response result includes: using a target matching method to calculate the consistency rate between the characters in the actual response result and the characters in the target response result.

[0128] Optionally, Figure 4 This is a structural block diagram of an electronic device according to an embodiment of this application. Figure 4 As shown, the electronic device may include: one or more ( Figure 4 (Only one is shown) Processor 402, memory 404, memory controller, and peripheral interface, wherein the peripheral interface is connected to the radio frequency module, audio module and display.

[0129] The memory can be used to store software programs and modules, such as the program instructions / modules corresponding to the command processing method and apparatus of the simulated passenger service system in this application embodiment. The processor executes various functional applications and data processing by running the software programs and modules stored in the memory, thereby realizing the command processing method of the simulated passenger service system described above. The memory may include high-speed random access memory, and may also include non-volatile memory, such as one or more magnetic storage devices, flash memory, or other non-volatile solid-state memory. In some instances, the memory may further include memory remotely located relative to the processor, and these remote memories can be connected to the terminal via a network. Examples of such networks include, but are not limited to, the Internet, corporate intranets, local area networks, mobile communication networks, and combinations thereof.

[0130] The processor can access the information and application programs stored in the memory via the transmission device to execute the steps described above in the command processing method of the simulated passenger service system.

[0131] This application provides a command processing scheme for a simulated passenger service system. It receives user input and obtains the current status information of passenger name records, where the user input contains at least a business instruction. Based on the user input and the current status information of passenger name records, a prompt word is determined. The prompt word is input into a target model, which is a model obtained by adjusting the low-rank matrix introduced into the initial model based on the training dataset. The target model outputs a target response result, which includes the adjusted status information of passenger name records according to the business instruction. This solution addresses the technical problem of poor accuracy in the response results of simulated passenger service systems and improves the accuracy of the response results.

[0132] Those skilled in the art will understand that Figure 4 The structure shown is for illustrative purposes only. Electronic devices can also be smartphones, tablets, handheld computers, mobile internet devices (MIDs), PADs, and other terminal devices. Figure 4 This does not limit the structure of the aforementioned electronic device. For example, electronic devices may also include components that are more... Figure 4 The more or fewer components shown (such as network interfaces, display devices, etc.), or having the same Figure 4 The different configurations shown.

[0133] Those skilled in the art will understand that all or part of the steps in the various methods of the above embodiments can be implemented by a program instructing the hardware related to the terminal device. The program can be stored in a computer-readable storage medium, which may include: flash drive, read-only memory (ROM), random access memory (RAM), disk or optical disk, etc.

[0134] Example 4

[0135] Embodiments of this application also provide a storage medium. Optionally, in this embodiment, the storage medium can be used to store the program code executed by the command processing method of the simulated passenger service system provided in Embodiment 1.

[0136] Optionally, in this embodiment, the storage medium may be located in any computer terminal in a group of computer terminals in a computer network, or in any mobile terminal in a group of mobile terminals.

[0137] Optionally, in this embodiment, the storage medium is configured to store program code for performing the following steps: receiving information input by the user and obtaining the status information of the current passenger name record, wherein the information input by the user carries at least a business instruction; determining a prompt word based on the information input by the user and the status information of the current passenger name record; inputting the prompt word into a target model, wherein the target model is a model obtained by adjusting the low-rank matrix introduced into the initial model based on the training dataset; and outputting a target response result through the target model, wherein the target response result includes the status information of the passenger name record adjusted according to the business instruction.

[0138] Optionally, the storage medium is also configured to store program code for performing the following steps: The target model is trained through the following steps: Introducing a low-rank matrix into the attention layer of the initial model; determining a training dataset, wherein the training dataset includes the status information of the sample passenger information record, sample input instructions, and sample response results; inputting the status information of the sample passenger information record and sample input instructions into the initial model; outputting a predicted response result through the initial model; calculating the loss value of the predicted response result and the sample response result based on the loss function; adjusting the low-rank matrix introduced into the initial model through the loss value to obtain an adjusted initial model; repeatedly executing the steps of inputting the status information of the sample passenger information record and sample input instructions into the adjusted initial model, the adjusted initial model outputting a new predicted response result, calculating a new loss value of the new predicted response result and the sample response result based on the loss function, adjusting the low-rank matrix through the new loss value to obtain a new adjusted initial model, until a preset condition is met to obtain the target model.

[0139] Optionally, the storage medium is also configured to store program code for performing the following steps: determining the training dataset includes: collecting historical interaction logs of the passenger service system to obtain initial training data; removing user personal identification information from the initial training data and replacing it with anonymous representation to obtain first training data; unifying the data format of the first training data to obtain second training data; extracting target fields from the second training data, wherein the target fields include the status information field of the passenger name record before the response, the input instruction field, and the status information field of the passenger information record after the response; and generating a training dataset based on the target fields.

[0140] Optionally, the storage medium is also configured to store program code for performing the following steps: determining prompt words based on user input information and the status information of the current passenger name record, including: converting the status information of the current passenger name record into passenger information description text; parsing business instructions by analyzing user input information; and concatenating the passenger information description text and business instructions into prompt words.

[0141] Optionally, the storage medium is also configured to store program code for performing the following steps: after the prompt word is input into the target model, the method further includes: receiving and parsing the prompt word through the target model to determine the business instruction in the prompt word; and adjusting the status information of the current passenger name record through the target model according to the business instruction and passenger information description text.

[0142] Optionally, the storage medium is also configured to store program code for performing the following steps: after outputting the target response result through the target model, the method further includes: inputting business instructions into the passenger service system; outputting the actual response result through the passenger service system; calculating the consistency rate between the actual response result and the target response result; and adjusting the target model if the consistency rate is less than a preset threshold.

[0143] Optionally, the storage medium is also configured to store program code for performing the following steps: calculating the consistency rate between the actual response result and the target response result, including: using a target matching method, calculating the consistency rate between the characters in the actual response result and the characters in the target response result.

[0144] This application also provides a computer program product that, when executed on a data processing device, is suitable for performing command processing method steps of a simulated passenger service system.

[0145] The sequence numbers of the embodiments in this application are for descriptive purposes only and do not represent the superiority or inferiority of the embodiments.

[0146] In the above embodiments of this application, the descriptions of each embodiment have different focuses. For parts not described in detail in a certain embodiment, please refer to the relevant descriptions of other embodiments.

[0147] In the several embodiments provided in this application, it should be understood that the disclosed technical content can be implemented in other ways. The device embodiments described above are merely illustrative; for example, the division of 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 coupling or direct coupling or communication connection shown or discussed may be through some interfaces; the indirect coupling or communication connection of units or modules may be electrical or other forms.

[0148] 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.

[0149] Furthermore, the functional units in the various embodiments of this application can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The integrated unit can be implemented in hardware or as a software functional unit.

[0150] 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, the technical solution of this application, in essence, or the part that contributes to the prior art, or all or part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods of the various embodiments of this application. The aforementioned storage medium includes various media capable of storing program code, such as a USB flash drive, read-only memory (ROM), random access memory (RAM), portable hard drive, magnetic disk, or optical disk.

[0151] The above are merely preferred embodiments of this application. It should be noted that those skilled in the art can make various improvements and modifications without departing from the principles of this application, and these improvements and modifications should also be considered within the scope of protection of this application.

Claims

1. A command processing method of an emulated passenger service system, characterized by, include: Receive information input by the user and obtain the status information of the current passenger name record, wherein the information input by the user carries at least a business instruction; Based on the information input by the user and the status information of the current passenger name record, determine the prompt word; The prompt words are input into the target model, wherein the target model is a model obtained by adjusting the low-rank matrix introduced into the initial model based on the training dataset; The target model outputs a target response result, which includes the status information of the passenger name record adjusted according to the business instruction.

2. The method of claim 1, wherein, The target model is trained through the following steps: A low-rank matrix is ​​introduced into the attention layer of the initial model; Determine the training dataset, wherein the training dataset includes status information of sample passenger information records, sample input commands, and sample response results; The status information recorded in the sample passenger information and the sample input command are input into the initial model; The initial model outputs the predicted response result; The loss value of the predicted response result and the sample response result is calculated based on the loss function; The adjusted initial model is obtained by adjusting the low-rank matrix introduced into the initial model using the loss value; The process of repeatedly inputting the status information recorded by the sample passenger information and the sample input command into the adjusted initial model, the adjusted initial model outputting a new predicted response result, calculating a new loss value based on the new predicted response result and the sample response result based on the loss function, adjusting the low-rank matrix with the new loss value to obtain a new adjusted initial model, and so on, until the preset conditions are met to obtain the target model.

3. The method of claim 2, wherein, Determining the training dataset includes: Collect historical interaction logs from the passenger service system to obtain initial training data; Remove the user's personal identification information from the initial training data and replace the user's personal identification information in the initial training data with anonymous representation to obtain the first training data; The data format of the first training data is unified to obtain the second training data; Extract the target fields from the second training data, wherein the target fields include the status information field of the passenger name record before the response, the input instruction field, and the status information field of the passenger information record after the response; The training dataset is generated based on the target field.

4. The method of claim 1, wherein, Based on the information input by the user and the status information of the current passenger name record, the prompt words are determined to include: Convert the current passenger name record status information into passenger information description text; The business instructions are parsed by analyzing the information input by the user. The passenger information description text and the business instruction are combined to form the prompt word.

5. The method of claim 4, wherein, After inputting the prompt words into the target model, the method further includes: The target model receives and parses the prompt words to determine the business instructions within the prompt words; Based on the business instruction and the passenger information description text, the status information of the current passenger name record is adjusted through the target model.

6. The method of claim 1, wherein, After outputting the target response result through the target model, the method further includes: Input the business instructions into the passenger service system; The passenger service system outputs the actual response results. Calculate the consistency rate between the actual response result and the target response result; If the consistency rate is less than a preset threshold, the target model is adjusted.

7. The method of claim 6, wherein, Calculating the consistency rate between the actual response result and the target response result includes: The matching rate between the characters in the actual response result and the characters in the target response result is calculated using a target matching method.

8. A command processing apparatus of an emulated passenger service system, characterized by comprising: include: The receiving unit is used to receive information input by the user and obtain the status information of the current passenger name record, wherein the information input by the user carries at least a business instruction; The first determining unit is used to determine the prompt word based on the information input by the user and the status information of the current passenger name record; The first input unit is used to input the prompt word into the target model, wherein the target model is a model obtained by adjusting the low-rank matrix introduced into the initial model based on the training dataset; The first output unit is used to output a target response result through the target model, wherein the target response result includes the status information of the passenger name record adjusted according to the business instruction.

9. A computer-readable storage medium, characterized in that, The computer-readable storage medium includes a stored executable program, wherein, when the executable program is executed, it controls the device containing the computer-readable storage medium to perform the command processing method of the simulated passenger service system according to any one of claims 1 to 7.

10. An electronic device, comprising: include: Memory, which stores executable programs; A processor for running the program, wherein the program, when running, executes the command processing method of the simulated passenger service system according to any one of claims 1 to 7.

11. A computer program product comprising computer instructions, characterized in that, When the computer instructions are executed by the processor, they implement the steps of the command processing method of the simulated passenger service system according to any one of claims 1 to 7.

Images