Self-service aero-change service method, device and equipment and storage medium
By integrating multiple channels to collect flight change information, integrating and converting data, and distributing notifications by using message queues, the problems of delayed flight change notification and inefficient manual maintenance in the existing technology are solved, and the timely and accurate transmission of flight change information and the improvement of business operation efficiency are achieved.
Patent Information
- Application Number
- CN202411811953.6
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-12-10
- Publication Date
- 2025-05-13
AI Technical Summary
The existing flight change notification methods are delayed, and the recording of notifications and related data is inefficient in manual maintenance, resulting in an increased risk of passengers missing flights.
Collect flight change information through preset channels (email, API, SMS), perform data integration and conversion, determine flight change notification tasks, and distribute them to target users through message queues. At the same time, integrated flight data management methods are adopted to automate data recording and build knowledge graphs for data analysis and prediction.
Ensure that flight change information is transmitted to passengers in a timely and accurate manner, reduce the risks and costs of manual operations, and improve overall business operation efficiency.
Smart Images
Figure CN119991391A_ABST
Abstract
Description
Technical Field
[0001] The present application relates to the field of self-service flight change services, and in particular to a self-service flight change service method, device, equipment and storage medium. Background Art
[0002] In the current aviation business, flight changes (such as adjustments to flight time, cabin space, boarding gate, etc.) may cause inconvenience to passengers. Existing flight change notification methods involve multiple channels, such as email, SMS, mobile device notifications, API interface push, etc. These channels are mixed and redundant, resulting in inefficient manual processing, prone to data errors or notification delays, and thus increasing the risk of passengers missing their flights.
[0003] Therefore, it is necessary to provide a self-service flight change service method that integrates multiple push channels and relies on the real-time data synchronization mechanism within the airline to ensure that the change information can be delivered to passengers in a timely, accurate and correct manner. Summary of the invention
[0004] The present application provides a self-service flight change service method, device and storage medium to solve the problems of delay in existing flight change notification methods and low efficiency in manual maintenance of notifications and related data records.
[0005] In a first aspect, the present application provides a self-service flight change service method, the method comprising:
[0006] Collect flight change information through preset channels; wherein the preset channels include email data collection, API data collection and SMS data collection;
[0007] Performing data integration and conversion on the collected flight change information to obtain flight change data; wherein, for the collected email data, using an email data parser to parse the email data, extracting and converting the email content into a standardized format; for the collected API data, converting the flight change data in the API data into a preset universal format; for the collected SMS data, using a natural language processing model to parse the SMS flight change data and convert it into the standardized format;
[0008] Based on the flight change data, determining a flight change notification task;
[0009] Based on the task type of the flight change notification task, a corresponding message queue is used to distribute the message corresponding to the flight change notification task to the target user.
[0010] In some embodiments, the method further comprises:
[0011] Storing information corresponding to the distributed flight change notification task;
[0012] Perform data analysis based on the stored information and determine the data analysis results;
[0013] Based on the data analysis results, the creation and scheduling of the flight change notification task are optimized.
[0014] In some embodiments, recording the information corresponding to the distributed flight change notification task includes:
[0015] Determine whether the data volume of the information corresponding to the flight change notification task exceeds a preset threshold;
[0016] In response to the data volume exceeding a preset threshold, using a distributed database to store information corresponding to the flight change notification task;
[0017] In response to the data volume not exceeding a preset threshold, information corresponding to the flight change notification task is stored using a time series database.
[0018] In some embodiments, the method further comprises:
[0019] Based on the flight change data, obtain flight change related data; wherein the related data includes flight information, airline information, weather data, airport information, social event information and aviation management regulations;
[0020] Building a knowledge graph based on the flight change related data;
[0021] Based on the knowledge graph, obtaining a graph embedding representation corresponding to each flight change information;
[0022] Based on the graph embedding representation, predicting subsequent changes of the flight corresponding to the flight change information;
[0023] Subsequent changes to the flight are distributed to the target users.
[0024] In some embodiments, obtaining a graph embedding representation corresponding to each flight change information based on the knowledge graph includes:
[0025] Based on the knowledge graph, feature extraction is performed on entities and relationships of the flight corresponding to each flight change information to obtain an embedding vector;
[0026] Acquire a supplementary feature vector based on external factors related to the flight corresponding to each flight change information;
[0027] The embedding vector and the supplementary feature vector are fused to determine the graph embedding representation.
[0028] In some embodiments, based on the knowledge graph, feature extraction is performed on entities and relationships of the flight corresponding to each flight change information to obtain an embedding vector, including:
[0029] Based on the knowledge graph, feature extraction is performed on the entities and relationships of the flight corresponding to each flight change information to obtain an initial embedding vector;
[0030] Based on the initial embedding vector, the timing information is enhanced to obtain the embedding vector; the timing information enhancement is performed by the following formula:
[0031] h t =LSTM(h t-1 ,x t )
[0032] Among them, h t represents the status of the flight at time t, x t It is the characteristic information of the flight change input at time t.
[0033] In some embodiments, based on the knowledge graph, feature extraction is performed on entities and relationships of the flight corresponding to each flight change information to obtain an embedding vector, including:
[0034] Based on the knowledge graph, feature extraction is performed on the entities and relationships of the flight corresponding to each flight change information to obtain an initial embedding vector;
[0035] The domain information of the flight entity is aggregated using a graph neural network to obtain the neighborhood feature vector; the aggregation process is performed using the following formula:
[0036]
[0037] in, is the embedding vector of node i at layer k, w (k) is the weight matrix of the kth layer, N(i) is the set of neighbor nodes of node i, c ij is the normalization coefficient, b (k) is the bias term, σ is the activation function;
[0038] The initial embedding vector and the neighborhood feature vector are fused to obtain the embedding vector.
[0039] In a second aspect, the present application provides a self-service flight change service device, the device comprising:
[0040] A flight change information collection module, used to collect flight change information through preset channels; wherein the preset channels include email data collection, API data collection and SMS data collection;
[0041] A data integration and conversion module is used to integrate and convert the collected flight change information to obtain flight change data; wherein, for the collected email data, the email data is parsed using an email data parser to extract and convert the email content into a standardized format; for the collected API data, the flight change data in the API data is converted into a preset universal format; for the collected SMS data, the SMS flight change data is parsed using a natural language processing model and converted into the standardized format;
[0042] A task determination module, used to determine a flight change notification task based on the flight change data;
[0043] The message distribution module is used to distribute the message corresponding to the flight change notification task to the target user using the corresponding message queue based on the task type of the flight change notification task.
[0044] In a third aspect, an electronic device is provided, comprising a processor, a communication interface, a memory and a communication bus, wherein the processor, the communication interface and the memory communicate with each other via the communication bus;
[0045] Memory, used to store computer programs;
[0046] The processor is used to implement the steps of the self-service flight change service method described in any embodiment of the first aspect when executing the program stored in the memory.
[0047] In a fourth aspect, a computer-readable storage medium is provided, on which a computer program is stored. When the computer program is executed by a processor, the steps of the self-service flight change service method as described in any embodiment of the first aspect are implemented.
[0048] The above technical solution provided by the embodiment of the present application has the following advantages over the prior art: it collects and receives flight change notification data pushed by various airlines, GDS and suppliers, ensuring the comprehensiveness and diversity of information sources, and by integrating multiple push channels and relying on the real-time data synchronization mechanism within the airline, it can ensure that the change information is delivered to passengers in a timely and accurate manner. In addition, the use of an integrated flight data management method can automatically record data, realize convenient management and traceability of flight change history, reduce the risk and cost of manual operations, and improve the overall business operation efficiency. BRIEF DESCRIPTION OF THE DRAWINGS
[0049] The accompanying drawings, which are incorporated in and constitute a part of this specification, illustrate embodiments consistent with the invention and, together with the description, serve to explain the principles of the invention.
[0050] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, for ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative labor.
[0051] Figure 1 A flowchart of a self-service flight change service method provided in an embodiment of the present application;
[0052] Figure 2 A schematic diagram of a process for recording information corresponding to the distributed flight change notification task provided in an embodiment of the present application;
[0053] Figure 3 A flowchart of another self-service flight change service method provided in an embodiment of the present application;
[0054] Figure 4 A schematic diagram of the structure of an electronic device provided in an embodiment of the present application. DETAILED DESCRIPTION
[0055] In order to make the purpose, technical solution and advantages of the embodiments of the present application clearer, the technical solution in the embodiments of the present application will be clearly and completely described below in conjunction with the drawings in the embodiments of the present application. Obviously, the described embodiments are part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without making creative work are within the scope of protection of this application.
[0056] Figure 1 A flowchart of a self-service flight change service method provided in an embodiment of the present application. Figure 1 The process shown can be executed by an electronic device (such as a processor) or a self-service flight change service device. Figure 1 As shown, process 100 may include the following operations.
[0057] Step 101, collecting flight change information through a preset channel.
[0058] The preset channel refers to a pre-configured information collection channel. For example, the preset channel may include at least one of email data collection, API data collection, and SMS data collection.
[0059] Email data capture refers to the collection of flight change emails sent by airlines or suppliers through automated processes.
[0060] API data collection refers to obtaining flight change information from airlines, GDS (global distribution system) and suppliers through the API interface.
[0061] SMS data collection refers to the collection of flight change notification information from airlines or passengers via text messages.
[0062] In some embodiments, an email monitoring system may be configured to automatically receive flight change emails from airlines or other service providers.
[0063] In some embodiments, flight change information can be obtained by integrating with an API interface provided by an airline or supplier through scheduled pulling or real-time subscription.
[0064] In some embodiments, flight change text messages from passengers or airlines may be received via a text message gateway.
[0065] Step 102, integrating and converting the collected flight change information to obtain flight change data.
[0066] Data integration and conversion can include two types of operations: data integration and data conversion. Data integration refers to the merging of flight change information from different channels (email, API, SMS) to ensure the consistency and integrity of the information. Data conversion refers to converting the collected flight change data in different formats into a standardized format for subsequent processing and storage.
[0067] In some embodiments, the collected flight change information may be converted first, and then the converted data may be integrated.
[0068] In some embodiments, for the collected email data, an email data parser is used to parse the email data, extract and convert the email content into a standardized format; for the collected API data, the flight change data in the API data is converted into a preset universal format; for the collected SMS data, a natural language processing model is used to parse the SMS flight change data and convert it into the standardized format.
[0069] The standardized format can be a common format in the industry, and the present invention does not limit the specific format type. For example, for the collected email data, such as flight number, change type, time, etc., these data can be converted into a preset standardized format (such as JSON format) to ensure the uniformity of different data sources.
[0070] For flight change data collected through the API, the original API return data can be converted into a unified flight data format (such as JSON format) within the system through the API data converter. The API data converter can ensure that the data format returned by different API interfaces is consistent to correctly identify the key information of flight changes.
[0071] For SMS data, a natural language processing (NLP) model can be used to identify and parse flight change information in SMS. The NLP model can convert unstructured text data into a standardized format and extract key information such as flight number, change type, time, etc.
[0072] Step 103: Determine a flight change notification task based on the flight change data.
[0073] The flight change notification task is a flight change message notification task created based on the integrated and converted flight change data. Its purpose is to notify relevant personnel (such as passengers, airlines or partners) about flight change information.
[0074] Task type refers to different notification task types corresponding to different flight change situations (such as flight cancellation, flight delay, boarding gate change, etc.). Task types can correspond to different notification contents and methods.
[0075] In some embodiments, the specific circumstances of the flight change can be automatically identified based on the content of the flight change data (such as change type, change reason, passenger information, etc.). For example, if the flight is canceled, a "flight cancellation notification task" can be created; if the flight is delayed, a "flight delay notification task" can be created.
[0076] In some embodiments, the processor may determine the type of task based on the specific content of the flight change data. The determination of the task type may be implemented according to a pre-set rule or logical judgment, for example, if the delay exceeds 30 minutes, it is automatically marked as "flight delay notification"; if the flight is cancelled, it is marked as "flight cancellation notification".
[0077] Step 104: Based on the task type of the flight change notification task, use the corresponding message queue to distribute the message corresponding to the flight change notification task to the target user.
[0078] A message queue is an asynchronous messaging mechanism that can be used to distribute task messages to designated target users (such as passengers, airline personnel, or partners). Message queues can include RabbitMQ, Kafka, etc.
[0079] In some embodiments, each task type corresponds to a message queue. For example, a "flight cancellation notification task" is pushed through an email queue, and a "flight delay notification task" is pushed through a SMS queue.
[0080] In some embodiments, the processor may select a corresponding message queue according to the task type. For example, a flight cancellation notification may be distributed to passengers via an email queue, while an urgent flight delay notification may be distributed via a text message queue to ensure that passengers receive the information in a timely manner.
[0081] After determining the task type, the processor can put the task message into the corresponding message queue. Each message queue can push the message to the device of the predetermined target user (such as mailbox, SMS, API interface, etc.). The use of message queues can ensure smooth processing of tasks under high concurrency and reduce system load.
[0082] In some embodiments, the message queue is used asynchronously to ensure that tasks are not executed synchronously, but are delayed according to the capacity and priority of the message queue, thereby achieving the purpose of effective peak shaving and avoiding system crashes under high traffic conditions.
[0083] In this embodiment, by defining different data collection channels (such as email, API, SMS), efficient collection and conversion of flight change information is achieved. Subsequently, based on this data, the flight change notification task can be automatically determined and distributed to the target users through the message queue mechanism, ensuring that the flight change information can be delivered to the relevant personnel in a timely and accurate manner. The automatic execution of each step ensures the automation, efficiency and ability to handle high concurrency of the system.
[0084] In some embodiments, the system may also record and analyze relevant messages to optimize the creation and scheduling of notification tasks.
[0085] Exemplarily, this can be achieved through the following operations: storing the information corresponding to the distributed flight change notification task; performing data analysis based on the stored information to determine the data analysis results; and, based on the data analysis results, optimizing the creation and scheduling of the flight change notification task.
[0086] Storage refers to persistently storing data such as task information, notification content and its execution status in a data storage system for subsequent management and analysis. In some embodiments, after the flight change notification task is created, the system can store relevant information of the flight change notification task (such as task type, flight number, change details, notification status, etc.) in the system's database.
[0087] Data analysis refers to the process of analyzing the stored information of flight change notification tasks to extract useful trends, patterns and insights to guide subsequent task optimization and adjustment. Data analysis results refer to the analysis results based on the stored information, which may include task execution efficiency, user feedback, notification channel usage effect, common patterns of flight changes, etc.
[0088] In some embodiments, the stored flight change notification tasks can be counted and summarized. For example, the frequency of occurrence of different types of flight changes (delays, cancellations, etc.) can be counted, the usage of different notification channels (email, SMS, API push) can be analyzed, or the flight change trends in different time periods can be analyzed. Data analysis tools (such as SQL queries, data mining algorithms, machine learning models, etc.) can also be used to conduct in-depth analysis of the stored data. The goal of the analysis can be to discover potential operational bottlenecks, identify high-frequency flight change patterns, or evaluate the effectiveness of notification channels, etc.
[0089] Figure 2 A flow chart of recording information corresponding to the distributed flight change notification task provided in an embodiment of the present application. Figure 2 As shown, process 200 may include the following operations.
[0090] Step 201, determining whether the data volume of the information corresponding to the flight change notification task exceeds a preset threshold.
[0091] The data volume refers to the storage space or the number of data entries occupied by the flight change task information. For example, the data volume may refer to the size or quantity of the task information, which may be specifically calculated in bytes of storage space or in the number of data entries.
[0092] The preset threshold refers to the upper limit of a certain amount of data set during the system design phase. The preset threshold can be used to determine whether a more efficient storage method is needed. If the amount of data exceeds the threshold, the operation of step 202 is performed, and if the amount of data does not exceed the threshold, the operation of step 203 is performed.
[0093] Step 202: In response to the data volume exceeding a preset threshold, using a distributed database to store information corresponding to the flight change notification task.
[0094] A distributed database is a database system consisting of multiple independent database nodes. Data can be distributed to multiple physical locations to enable horizontal expansion and large-scale data storage and query. Distributed databases can include Cassandra, MongoDB, HBase, etc.
[0095] In some embodiments, the processor may distribute the flight change task information to multiple distributed database nodes. The distributed database may store data in different physical storage units according to certain sharding rules (such as hash sharding, range sharding, etc.).
[0096] In some embodiments, the flight change task information may be written into a distributed database in the form of structured data (such as JSON, XML, etc.).
[0097] Step 203: In response to the data volume not exceeding a preset threshold, the information corresponding to the flight change notification task is stored using a time series database.
[0098] A time series database refers to a database system specifically used to store and query time series data. Time series data refers to data points arranged in chronological order, usually used to record the occurrence of events, and is suitable for processing high-frequency event data, such as sensor data, log information, etc. Time series databases can include InfluxDB, Prometheus, etc.
[0099] In some embodiments, when it is determined that the data volume does not exceed a preset threshold, the information of the flight change task can be written into the time series database in chronological order. The written data can be indexed by the timestamp.
[0100] Figure 3 A flowchart of another self-service flight change service method provided in an embodiment of the present application. Figure 3 The process shown can be executed by an electronic device (such as a processor) or a self-service flight change service device. Figure 3 As shown, process 300 may include the following operations.
[0101] Step 301: Acquire flight change related data based on the flight change data.
[0102] Flight change data refers to information related to changes in flight operations such as flight time, cabin space, boarding gate, flight cancellations, etc.
[0103] Flight change related data refers to external information directly related to flight change. In some embodiments, flight change related data includes flight information, airline information, weather data, airport information, social event information and aviation management regulations.
[0104] Flight information includes flight number, take-off and landing time, take-off and landing location, flight type, etc. Airline information refers to the name of the airline, contact information, policies, etc. Weather data includes weather forecasts, meteorological conditions in the flight area, and sudden weather events (such as storms, snowstorms, etc.). Airport information includes airport name, airport facilities, and operational capacity for flight take-off and landing. Social event information refers to external social events that affect flight operations, such as large-scale events, traffic jams, etc. Aviation management regulations refer to the specifications, regulations, laws and regulations of the aviation industry, involving flights, airspace, flight scheduling, etc.
[0105] In some embodiments, flight change related data may be collected from multiple sources (eg, airlines, weather service providers, airport authorities, etc.).
[0106] Step 302: construct a knowledge graph based on the flight change related data.
[0107] A knowledge graph is a graphical data structure that represents different entities and their relationships in the form of nodes and edges. Through a knowledge graph, we can better understand the connections and dependencies between entities.
[0108] In some embodiments, the nodes in the knowledge graph can be defined first. For example, flight information, airlines, airports, weather events, social events, etc. can all be used as nodes of the knowledge graph. Afterwards, relationships can be built between nodes. For example, there is an influence relationship between "Flight A" and "Weather Event X"; there is an ownership relationship between "Flight B" and "Airline Y"; there is a take-off and landing relationship between "Airport Z" and "Flight A", etc. Using a graph database (such as Neo4j, ArangoDB, etc.) to store nodes and relationships in a database, the knowledge graph can be obtained.
[0109] Step 303: Based on the knowledge graph, obtain a graph embedding representation corresponding to each flight change information.
[0110] Graph embedding is a numerical vector obtained by transforming the nodes and edges in the graph through an algorithm. Graph embedding can map the structural information in the graph to a continuous vector space.
[0111] In some embodiments, an embedding vector may be generated by calculating the adjacency relationship of nodes in a graph using a graph embedding algorithm, such as Node2Vec, GraphSAGE, etc. For example, a selected graph embedding algorithm may be used to convert flight change information and various types of data associated therewith into a high-dimensional vector representation.
[0112] In some embodiments, obtaining a graph embedding representation corresponding to each flight change information based on the knowledge graph may include the following operations.
[0113] S1. Based on the knowledge graph, feature extraction is performed on the entities and relationships of the flight corresponding to each flight change information to obtain an embedding vector.
[0114] Feature extraction refers to the process of extracting meaningful information from entities and their relationships. The extracted features can include the attribute information of entities and the relationships between entities.
[0115] The embedding vector is a numerical representation (vector) of the features of the flight change information.
[0116] In some embodiments, the processor can extract all entities related to flight change information (such as flights, airlines, airports, etc.) and the relationships between them (such as the take-off and landing relationship between flights and airports, the impact relationship between flights and weather, etc.) from the knowledge graph, and extract features from the entities and relationships. For example, for flight entities, features such as flight time, date, and cabin space can be extracted; for weather events, features such as temperature, wind speed, and precipitation can be extracted. Relationship features include the connection between flights and airports, and flights and airlines.
[0117] In some embodiments, a graph embedding algorithm (such as Node2Vec, GraphSAGE) may be used to convert entities and their relationships into embedding vectors.
[0118] In some embodiments, based on the knowledge graph, feature extraction is performed on the entities and relationships of the flight corresponding to each flight change information to obtain an embedding vector, including: based on the knowledge graph, feature extraction is performed on the entities and relationships of the flight corresponding to each flight change information to obtain an initial embedding vector; based on the initial embedding vector, time series information enhancement is performed to obtain the embedding vector.
[0119] In some embodiments, the timing information enhancement is performed by the following formula (1):
[0120] ht=LSTM(ht-1,xt) (1)
[0121] Among them, h t represents the status of the flight at time t (embedded vector), h t-1 represents the flight status at time t-1 (initial embedding vector), that is, the embedding vector at the previous moment, which is used as one of the inputs of the model, x t It is the characteristic information of flight changes input at time t, such as flight delay time, weather conditions, etc. LSTM(*) represents the mapping function of the time series model (it can also be GRU or other recursive neural networks), which is used to calculate the current state based on the state at the previous moment and the characteristics of the current input.
[0122] The initial embedding vector refers to the initial vector obtained by feature extraction based on the knowledge graph.
[0123] Flight change information usually contains the time dimension (such as flight departure time, delay time, etc.). The purpose of enhancing the temporal information is to take the time factor into account and improve the adaptability of the embedding vector to time changes, so as to more accurately reflect the timeliness of flight changes.
[0124] In some embodiments, the extracting features of entities and relationships of the flight corresponding to each flight change information based on the knowledge graph to obtain an embedding vector may include the following operations.
[0125] S11, based on the knowledge graph, feature extraction is performed on the entities and relationships of the flight corresponding to each flight change information to obtain an initial embedding vector.
[0126] S12, utilizing a graph neural network to aggregate the domain information of the entities of the flight to obtain the neighborhood feature vector.
[0127] In some embodiments, the aggregation process can be performed by the following formula (2):
[0128]
[0129] in, is the embedding vector of node i at layer k, w (k) is the weight matrix of the kth layer, N(i) is the set of neighbor nodes of node i, c ij is the normalization coefficient, b (k) is the bias term, and σ is the activation function, for example, the ReLU function.
[0130] Graph neural network (GNN) is a neural network model that learns node representation by aggregating information from neighborhood nodes and is capable of processing graph-structured data.
[0131] Domain information refers to the adjacent entity information related to flights in the graph, such as airlines, airports, etc.
[0132] The neighborhood feature vector is a vector obtained by aggregating the neighborhood information of flight entities through GNN.
[0133] In some embodiments, a graph neural network model can be constructed based on flights and other related entities (such as airlines, airports, etc.), taking entity nodes and relationship edges as input. Using the aggregation operation of GNN (such as graph convolution layer), the information of neighbor nodes is weightedly aggregated to calculate the neighborhood feature vector of each flight.
[0134] S13, fusing the initial embedding vector and the neighborhood feature vector to obtain the embedding vector.
[0135] S2, acquiring a supplementary feature vector based on external factors related to the flight corresponding to each flight change information;
[0136] External factors refer to external variables that may affect flight changes or decisions during flight changes, such as weather conditions, social events, aviation policies, airport conditions, etc.
[0137] Supplementary feature vectors refer to additional features extracted from external factors in addition to the features extracted from the knowledge graph, which are used to supplement the performance of the embedding vector to improve the prediction accuracy of the model.
[0138] In some embodiments, external factors that may affect the flight can be identified based on the background information of the flight change. External factors may include weather data (such as storms, snowstorms), social events (such as traffic accidents, large-scale events), airline policies, airport operating conditions, etc.
[0139] The extracted external factor features are converted into numerical vectors to obtain the supplementary feature vectors.
[0140] S3, fusing the embedding vector and the supplementary feature vector to determine the graph embedding representation.
[0141] Fusion refers to combining the embedding vector with the supplementary feature vector to form a composite vector containing all the key information.
[0142] In some embodiments, a suitable fusion method, such as simple concatenation, weighted averaging, neural network fusion in deep learning, etc., can be selected to combine the embedding vector and the supplementary feature vector into the graph embedding representation.
[0143] In this embodiment, the graph embedding representation finally obtained includes more information that is helpful for prediction, which can effectively improve the accuracy of flight change prediction and thus support more efficient flight change notification and management.
[0144] Step 304: predicting subsequent changes of the flight corresponding to the flight change information based on the graph embedding representation.
[0145] Subsequent changes to a flight refer to possible changes to the flight in the subsequent period, such as further delays, cancellations, readjustment of boarding gates, etc.
[0146] Forecasting refers to predicting possible future events or changes based on historical data and current status using machine learning or data mining techniques.
[0147] In some embodiments, a prediction model can be trained using graph embedding representations and historical flight change data. Common prediction models include regression models, decision trees, neural networks, etc. The trained model is used to input the graph embedding representation of the current flight to predict subsequent changes to the flight. For example, the model can predict whether a flight is likely to be delayed or canceled, or whether certain external factors (such as weather or social events) will cause flight adjustments.
[0148] In some embodiments, during the prediction process, multiple factors may be combined, such as the airline to which the flight belongs, the busyness of the take-off and landing airports, weather forecasts, etc., to conduct multi-dimensional prediction analysis to improve the accuracy of the prediction.
[0149] Step 305: Distribute the subsequent changes of the flight to the target users.
[0150] In some embodiments, the appropriate notification channel can be selected based on the target user's preferences or historical data. For example, a passenger can receive notifications via SMS or mobile application, while another can receive notifications via internal API or email.
[0151] The notification content can be automatically generated based on the subsequent changes of the flight. The notification content includes the flight number, change type, change time, etc.
[0152] like Figure 4 As shown, an embodiment of the present application provides an electronic device, including a processor 111, a communication interface 112, a memory 113 and a communication bus 114, wherein the processor 111, the communication interface 112, and the memory 113 communicate with each other through the communication bus 114.
[0153] Memory 113, used for storing computer programs;
[0154] In one embodiment of the present application, the processor 111 is used to execute the program stored in the memory 113 to implement the method of the self-service flight change service provided by any of the above method embodiments, including:
[0155] An embodiment of the present application further provides a computer-readable storage medium having a computer program stored thereon. When the computer program is executed by a processor, the steps of the self-service flight change service method provided by any of the aforementioned method embodiments are implemented.
[0156] It should be noted that, in this article, relational terms such as "first" and "second" are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Moreover, the terms "include", "comprise" or any other variants thereof are intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or device. In the absence of further restrictions, the elements defined by the sentence "comprise a ..." do not exclude the existence of other identical elements in the process, method, article or device including the elements.
[0157] The foregoing is merely a specific embodiment of the present invention, which enables those skilled in the art to understand or implement the present invention. Various modifications to these embodiments will be apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the present invention. Therefore, the present invention will not be limited to the embodiments shown herein, but rather to the widest scope consistent with the principles and novel features claimed herein.
Claims
1. A self-service flight change service method, characterized in that: The method comprises: Collect flight change information through preset channels; wherein the preset channels include email data collection, API data collection and SMS data collection; Performing data integration and conversion on the collected flight change information to obtain flight change data; wherein, for the collected email data, using an email data parser to parse the email data, extracting and converting the email content into a standardized format; for the collected API data, converting the flight change data in the API data into a preset universal format; for the collected SMS data, using a natural language processing model to parse the SMS flight change data and convert it into the standardized format; Based on the flight change data, determining a flight change notification task; Based on the task type of the flight change notification task, a corresponding message queue is used to distribute the message corresponding to the flight change notification task to the target user.
2. The method according to claim 1, characterized in that The method further comprises: Storing information corresponding to the distributed flight change notification task; Perform data analysis based on the stored information and determine the data analysis results; Based on the data analysis results, the creation and scheduling of the flight change notification task are optimized.
3. The method according to claim 2, characterized in that The recording of information corresponding to the distributed flight change notification task includes: Determine whether the data volume of the information corresponding to the flight change notification task exceeds a preset threshold; In response to the data volume exceeding a preset threshold, using a distributed database to store information corresponding to the flight change notification task; In response to the data volume not exceeding a preset threshold, information corresponding to the flight change notification task is stored using a time series database.
4. The method according to claim 1, characterized in that: The method further comprises: Based on the flight change data, obtain flight change related data; wherein the related data includes flight information, airline information, weather data, airport information, social event information and aviation management regulations; Building a knowledge graph based on the flight change related data; Based on the knowledge graph, obtaining a graph embedding representation corresponding to each flight change information; Based on the graph embedding representation, predicting subsequent changes of the flight corresponding to the flight change information; Subsequent changes to the flight are distributed to the target users.
5. The method according to claim 4, characterized in that The step of obtaining a graph embedding representation corresponding to each flight change information based on the knowledge graph includes: Based on the knowledge graph, feature extraction is performed on entities and relationships of the flight corresponding to each flight change information to obtain an embedding vector; Acquire a supplementary feature vector based on external factors related to the flight corresponding to each flight change information; The embedding vector and the supplementary feature vector are fused to determine the graph embedding representation.
6. The method according to claim 5, characterized in that The step of extracting features of entities and relationships of flights corresponding to each flight change information based on the knowledge graph to obtain an embedding vector includes: Based on the knowledge graph, feature extraction is performed on entities and relationships of the flight corresponding to each flight change information to obtain an initial embedding vector; Based on the initial embedding vector, the timing information is enhanced to obtain the embedding vector; the timing information enhancement is performed by the following formula: h t =LSTM(h t-1 ,x t ) Among them, h t represents the status of the flight at time t, x t It is the characteristic information of the flight change input at time t.
7. The method according to claim 5, characterized in that The step of extracting features of entities and relationships of flights corresponding to each flight change information based on the knowledge graph to obtain an embedding vector includes: Based on the knowledge graph, feature extraction is performed on entities and relationships of the flight corresponding to each flight change information to obtain an initial embedding vector; The domain information of the flight entity is aggregated using a graph neural network to obtain the neighborhood feature vector; the aggregation process is performed using the following formula: in, is the embedding vector of node i at layer k, w (k) is the weight matrix of the kth layer, N(i) is the set of neighbor nodes of node i, c ij is the normalization coefficient, b (k) is the bias term, σ is the activation function; The initial embedding vector and the neighborhood feature vector are fused to obtain the embedding vector.
8. A self-service flight change service device, characterized in that: The device comprises: A flight change information collection module, used to collect flight change information through preset channels; wherein the preset channels include email data collection, API data collection and SMS data collection; A data integration and conversion module is used to integrate and convert the collected flight change information to obtain flight change data; wherein, for the collected email data, the email data is parsed using an email data parser to extract and convert the email content into a standardized format; for the collected API data, the flight change data in the API data is converted into a preset universal format; for the collected SMS data, the SMS flight change data is parsed using a natural language processing model and converted into the standardized format; A task determination module, used to determine a flight change notification task based on the flight change data; The message distribution module is used to distribute the message corresponding to the flight change notification task to the target user using the corresponding message queue based on the task type of the flight change notification task.
9. An electronic device, characterized in that: It includes a processor, a communication interface, a memory and a communication bus, wherein the processor, the communication interface and the memory communicate with each other through the communication bus; Memory, used to store computer programs; The processor is used to implement the steps of the self-service flight change service method according to any one of claims 1 to 7 when executing the program stored in the memory.
10. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the self-service flight change service method according to any one of claims 1 to 7 are implemented.