Air ticket searching method and device, computer readable storage medium and electronic equipment

Abstract

The invention relates to an air ticket searching method and device. The method comprises the steps that man-machine interaction is carried out, a user inputs a search requirement in a voice or character mode, the system converts the voice into text information, semantic understanding and analysis are carried out on the input information through a civil aviation large model, and the input information is converted into corresponding search parameters; generating a complete airline ticket search request according to the search parameters and the airline company search request template, and accessing a search engine to obtain an airline ticket vertical search result; accessing a civil aviation large model according to destination and flight information in the air ticket vertical search result, obtaining travel knowledge related to the destination and the flight, and integrating the obtained travel knowledge with the air ticket vertical search result to obtain a search result; and screening and displaying the search results. According to the method, the problems of redundancy and inaccuracy of user search results, difficulty in travel destination and flight selection, difficulty in mass information screening and the like are solved, and the air ticket search efficiency and accuracy are improved.

Description

Technical Field

[0001] The present application belongs to the technical field of air ticket search, and in particular, relates to an air ticket search method, device, computer-readable storage medium, and electronic device. Background Art

[0002] With the continuous development of the global economy and the improvement of people's living standards, activities such as tourism and business travel are becoming more frequent. People's requirements for air ticket searches are no longer limited to simple price comparisons and flight inquiries. Instead, they expect to obtain more personalized and accurate search results, as well as a more efficient and convenient search experience.

[0003] Traditional air ticket search methods often rely on users to manually enter specific information such as departure place, destination and date. The search results are relatively simple and cannot meet the personalized needs of users. The large language model can continuously optimize the search results through interaction with users. When users are not satisfied with the search results, they can further propose modifications, such as "I want to take a flight between 9 am and 11 am" or "I want a direct flight". The large language model can quickly adjust the search strategy to provide users with air ticket search results that better meet their needs.

[0004] The large language model has powerful language understanding and generation capabilities. Whether it is a precise expression of demand or a vague travel intention, the large language model can accurately understand and parse it. If the user only has a vague travel intention, the large language model can also intelligently guide the user to complete the air ticket search and itinerary booking through heuristic dialogue. For example, after the user is guided by the large model and interacts with multiple dialogues, he enters "I want to depart from Beijing during the National Day and go to a warm coastal city for vacation. I will stay there for about 3-5 days, and I hope the air ticket price will not exceed 3,000 yuan." The large language model can quickly identify key information such as departure place, destination, date range, budget, etc., and generate a search request based on this to search for air tickets.

[0005] In addition, the big language model also has a wealth of travel knowledge, which is what travelers need to pay attention to when traveling, and may even affect their choices and purchase decisions. By integrating the travel knowledge of the big model into the vertical search results for air tickets, it can not only provide users with basic information such as flights and prices, but also provide information such as weather conditions at the destination, recommended tourist attractions, local specialties, and transportation methods, helping travelers achieve smart and convenient travel.

[0006] In summary, empowering the air ticket search business through artificial intelligence technology and large language models, achieving organic complementarity between the two, and developing new air ticket search applications based on AI+ has become a hot topic in the industry. Summary of the invention

[0007] Based on the above background and technical requirements, this application proposes a new air ticket search method and device. This method is based on the civil aviation big model, reshapes the air ticket search business and process, and uses human-computer interaction to achieve personalized search and fuzzy search; this method integrates the travel knowledge of the big model and the vertical search results of air tickets, empowers passengers in a novel result display method, helps passengers complete air ticket search and booking efficiently and conveniently, improves purchase efficiency and experience, and meets passengers' needs for a better travel.

[0008] This application proposes a new air ticket search method and device to improve the efficiency and accuracy of air ticket search for customer groups with clear needs, guide customer groups with unclear needs to convert vague travel intentions into final ticket purchase behaviors, and enhance the stickiness and intelligence level of airline sales channels. This method can provide personalized and accurate search results for passengers with clear needs, and guide customer groups with unclear needs to complete air ticket search and booking through fuzzy search. This method reshapes the search processing flow by introducing large model technology, uses natural language to search for air tickets through human-computer interaction, and the trained and fine-tuned large model can correctly understand the user's natural language and convert it into corresponding search parameters, provide personalized search results, and return travel knowledge related to the results according to the result content, so as to help passengers better choose and book itineraries. The traditional air ticket search method requires users to manually enter specific information such as departure place, destination, date and passengers, lacks interactive experience, cannot accurately obtain the complete needs of passengers, usually returns a large set of search results for users to choose, and has problems such as difficulty in flight selection. It cannot provide personalized and accurate search results. On the other hand, it is difficult to meet the fuzzy search needs of user groups with unclear destinations. The air ticket search method based on the large language model of this application, through human-computer interaction, intelligent guidance and recommendation, and knowledge empowerment, effectively solves the problems of redundant and inaccurate user search results, difficulty in choosing travel destinations and flights, and difficulty in screening massive information, thereby greatly improving the efficiency and accuracy of air ticket search.

[0009] In order to achieve the above objectives, this application provides the following technical solutions:

[0010] A first aspect of the present application provides a flight ticket search method, the method comprising:

[0011] Step 1: Human-computer interaction: the user inputs the search requirements in the form of voice or text. The system converts the voice into text information, and uses the civil aviation big model to understand and analyze the input information semantically and convert it into corresponding search parameters.

[0012] Step 2: Generate a complete air ticket search request based on the search parameters and the airline search request template, and access the search engine to obtain the air ticket vertical search results;

[0013] Step 3: access the civil aviation big model according to the destination and flight information in the air ticket vertical search results, obtain travel knowledge related to the destination and flight, and integrate the obtained travel knowledge with the air ticket vertical search results to obtain search results;

[0014] Step 4: Filter the search results according to the needs of the airline and display the filtered results to the user.

[0015] Optionally, the human-computer interaction described in step 1 of the method of the present application includes the following steps:

[0016] The user starts a dialogue with the human-computer interaction module, which determines whether the information input by the user is air ticket query information; if not, it is transferred to other related business modules for processing; if yes, it is transferred to the search agent module for processing;

[0017] The search agent module determines whether the information input by the user is complete. If it is complete, it will use the civil aviation big model module to understand and analyze its semantics and generate corresponding search parameters. If it is incomplete, it will return to the human-computer interaction module to continue the dialogue. After the dialogue is completed, it will be transferred to the search agent module for processing again.

[0018] Optionally, step one of the method of the present application further includes: guiding the user to complete the input of the search requirement through a prompt word project.

[0019] Optionally, the search parameters described in step one of the method of the present application include: origin, destination, departure date, departure time, return date, return time, destination stay period, passenger information, preference information and budget information.

[0020] Optionally, in the method of the present application, the origin and destination in the search parameters are mandatory items; the departure date, departure time, return date, return time, destination stay period, passenger information, preference information and budget information in the search parameters are optional items;

[0021] The origin is obtained through user location information, IP information or user input;

[0022] The stated return date and duration of stay at the destination cannot occur at the same time.

[0023] Optionally, step 2 of the method of the present application also includes: the search engine determines whether it is a fuzzy search based on the air ticket search request. If it is a fuzzy search, the search results are directly obtained from the result cache; if the cache is not hit or there are only part of the results in the cache, the search engine calculates the remaining results in real time and merges them with the cached results and returns them to the search agent, and writes the real-time calculated results into the result cache.

[0024] Optionally, in step 4 of the method of the present application, the screening results are presented to the user in the form of a table, picture or text as needed.

[0025] A second aspect of the present application provides a flight ticket search device, the device comprising:

[0026] Human-computer interaction module: used to interact with users through voice or text, convert user voice input into text information, and guide users to complete search demand input through prompt word engineering; the user input information is semantically understood and analyzed through the civil aviation big model module, and converted into corresponding search parameters;

[0027] Search agent module: used to generate search requests, access the search engine module to obtain vertical search results for air tickets; access the civil aviation big model module to obtain travel knowledge related to the air ticket search results, and integrate the obtained travel knowledge into the vertical search results for air tickets;

[0028] Civil aviation big model module: used to understand and analyze user input and generate corresponding search parameters based on user input information; receive requests from the search agent module and return corresponding travel knowledge;

[0029] Search engine module: used to process received air ticket search requests and return corresponding air ticket search results;

[0030] Result display module: used to output the integrated air ticket vertical search results and travel knowledge to users.

[0031] The device implements the steps of the aforementioned air ticket search method when running.

[0032] A third aspect of the present application provides an electronic device, comprising: a memory and a processor;

[0033] Memory: used to store computer programs;

[0034] Processor: used to execute the computer program to implement the steps of the aforementioned air ticket search method.

[0035] A fourth aspect of the present application provides a computer-readable storage medium having a computer program stored thereon, and when the computer program is executed by a processor, the steps of the aforementioned air ticket search method are implemented.

[0036] In summary, this application proposes a new air ticket search method, which has the following advantages:

[0037] (1) Different from the traditional page input method, this application uses the power of a large language model to complete air ticket search input in a human-computer interactive manner. Through multiple rounds of dialogue and prompt word guidance, personalized air ticket search can be achieved, providing passengers with more accurate search results.

[0038] (2) This application innovatively integrates travel knowledge with vertical search results for air tickets, providing a new way to display search results, helping passengers to more efficiently select and purchase air tickets and achieve smart and convenient travel.

[0039] (3) This application uses retrieval enhancement generation (RAG) technology and a plug-in travel knowledge base to improve the accuracy, relevance and freshness of travel knowledge, shorten the time of knowledge retrieval, and effectively solve the problems of long response time and low accuracy of large models.

[0040] (4) This application uses a large model to intelligently screen destinations for passengers and provide travel-related knowledge, which can effectively solve the problems of passengers' difficulty in choosing travel destinations and flights and the difficulty in screening massive amounts of information.

[0041] Other features and advantages of the present application will be described in the following description, or understood by practicing the present application. The purpose and other advantages of the present application can be realized and obtained by the techniques indicated in the description, claims and drawings. BRIEF DESCRIPTION OF THE DRAWINGS

[0042] In order to more clearly illustrate the technical solutions of the embodiments of the present application, a brief introduction is given below to the drawings required for use in the description of the embodiments. Obviously, the drawings described below are only some embodiments of the present application. For those skilled in the art, other drawings can be obtained based on these drawings without creative work.

[0043] Figure 1 A structural diagram of a flight ticket search device provided in an embodiment of the present application.

[0044] Figure 2 A workflow diagram of an air ticket search device provided in an embodiment of the present application.

[0045] Figure 3 An overall implementation flow chart of an air ticket search method provided in an embodiment of the present application.

[0046] Figure 4 A human-computer interaction flow chart provided for an embodiment of the present application.

[0047] Figure 5 An example diagram of inputting air ticket search requirements in the form of a dialogue provided in an embodiment of the present application.

[0048] Figure 6 A travel knowledge classification diagram provided for an embodiment of the present application.

[0049] Figure 7 A schematic diagram of the destination information return result provided in an embodiment of the present application.

[0050] Figure 8 This is a human-computer interaction + personalized search effect diagram provided in an embodiment of the present application.

[0051] Fig. 9 This is a diagram of the human-computer interaction + fuzzy search effect provided in an embodiment of the present application.

[0052] Fig.10 A schematic diagram of the structure of an electronic device provided in an embodiment of the present application. DETAILED DESCRIPTION

[0053] In order to make the purpose, technical scheme and advantages of the embodiments of the present application clearer, the technical scheme 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 only 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.

[0054] As used herein, the term “including” and its variations are open inclusions, ie, “including but not limited to”; the term “based on” means “based at least in part on”; and the term “one embodiment” means “at least one embodiment”.

[0055] It should be noted that the modifications of "one" and "plurality" mentioned in the present application are illustrative rather than restrictive, and those skilled in the art should understand that unless the context clearly indicates otherwise, it should be understood as "one or more".

[0056] Explanation of terms

[0057] Flight Search / SHOP: Flight Search is the process of integrating flight information, available fare information, taxes, brands and additional services, baggage, etc. based on the itinerary and preference information entered by the passenger and returning the results to the requesting user.

[0058] Fuzzy search: air ticket search with unclear travel destination and relatively flexible travel date.

[0059] Large Language Model / LLM: A language model built with a deep neural network containing more than tens of billions of parameters.

[0060] Retrieval-augmented Generation / RAG: A technique or approach that combines external knowledge retrieval with the generation capabilities of Large Language Models (LLMs).

[0061] Figure 1 The figure shows the structure of a flight ticket search device proposed in this application. Figure 2 The working process of the device is shown, and the device includes:

[0062] Human-computer interaction module: supports users to interact with the system through voice or text, converts voice input into text input, and guides users to complete search requirements through prompt word engineering; understands user text input through the civil aviation big model module and converts it into corresponding search parameters.

[0063] Search agent module: Generates search requests, accesses the search engine module to obtain vertical search results for air tickets; accesses the civil aviation big model module to obtain travel knowledge related to the air ticket search results, including destination information and flight information; finally, integrates the obtained travel knowledge into the vertical search results for air tickets.

[0064] Civil aviation big model module: understands and analyzes user input and generates corresponding search parameters based on user input information; receives requests from the search agent module and returns corresponding travel knowledge.

[0065] Search engine module: processes received air ticket search requests and returns corresponding air ticket search results.

[0066] Result display module: outputs the integrated vertical search results of air tickets and travel knowledge to users; it can support output to users in the form of tables, pictures or texts.

[0067] When the above device is running, the steps of the air ticket search method disclosed in this application are implemented.

[0068] Figure 3 The overall implementation process of a flight ticket search method proposed in this application is shown, including the following steps:

[0069] Step 1: Conduct human-computer interaction. Users input their search requirements in the form of voice or text through the human-computer interaction module. The human-computer interaction module converts the voice into text information, and understands the passenger's question through the civil aviation big model module, guiding the passenger to complete the input of the demand; the civil aviation big model module performs semantic understanding and analysis on the passenger input information and converts it into corresponding search parameters.

[0070] Step 2: The search agent module generates a complete air ticket search request based on the search parameters passed in by the human-computer interaction module and the airline search request template, and accesses the search engine module to obtain the air ticket vertical search results.

[0071] Step 3: The search agent module accesses the civil aviation big model module according to the destination and flight information in the air ticket vertical search results, obtains travel knowledge related to the destination and flight, and then integrates the obtained travel knowledge with the air ticket vertical search results to obtain search results.

[0072] Step 4: The result display module filters the search results according to the needs of the airlines and displays the final results of the screening to the users. The result display module can output the results to the users in the form of tables, pictures or texts as needed.

[0073] In order to better understand the technical solution of the present application, it is further described with reference to the following embodiments.

[0074] 1. Human-computer interaction

[0075] Figure 4 The following is the human-computer interaction process provided by this application, including:

[0076] The passenger starts a conversation with the human-computer interaction module, which determines whether the information entered by the passenger is a ticket inquiry question; if not, it is transferred to other related business modules for processing; if yes, it is transferred to the search agent module for processing;

[0077] The search agent module determines whether the information entered by the passenger is complete. If it is complete, it will use the civil aviation big model module to understand its semantics and generate corresponding search parameters. If it is incomplete, it will return to the human-computer interaction module to continue the dialogue. After the dialogue is completed, it will be transferred to the search agent module for processing again.

[0078] 1.1 Passengers enter their flight search requirements in a dialogue format, such as Figure 5 shown.

[0079] 1.2 The Civil Aviation Model Module converts passenger input into corresponding search parameters

[0080] The human-computer interaction module uses the civil aviation big model module to semantically understand and analyze the information input by the passenger (which can be input all at once or in sequence through multiple rounds of dialogue), and divides the passenger's search conditions into corresponding fields and converts them into corresponding search parameters in the following format.

[0081] KEY (origin-destination-departure date-departure time-return date-return time-destination stay period-passenger information-preference information-budget information);

[0082] Field Description:

[0083] Origin: Required, such as Beijing. The origin information is obtained through user location information, IP information or user input. User input is used first. If there is no user input, the user's city information obtained by converting the location information is used first. In this example, the user has no input, and the origin obtained by using the location information is Beijing, that is, originCity = BJS;

[0084] Destination: Required, such as Shanghai, southern cities, islands, domestic hot spots, etc. In this example, the destination is hiking. Since the destination is not a specific city or airport, the civil aviation model module intelligently recommends destinations for users based on the hiking theme. The system can limit the number of intelligently recommended destinations. In this example, the civil aviation model module recommends Huashan, Taishan and Huangshan as hiking destinations for users. The corresponding cities are Xi'an, Jinan and Huangshan, that is, destCity = {XIY, TNA, TXN}. Since different airlines operate different routes, for the same theme, such as islands, the civil aviation model module can recommend different destinations for different airlines, thereby achieving differentiated recommendations;

[0085] Departure date: optional, such as August 18, next week, late August, etc. In this case, it is October 1, that is, departureDate = 2024-10-01;

[0086] Departure time: optional, such as 8:00 a.m. to 10:00 a.m., morning, etc. In this example, no input is made and no corresponding parameters are generated;

[0087] Return date: optional, such as August 28, the week after next, early September, etc. In this example, no input is made and no corresponding parameters are generated;

[0088] Return time: optional, such as 3pm to 5pm, afternoon, etc. In this example, no input is made and no corresponding parameters are generated;

[0089] Destination stay period: optional, such as 3 days, 3 to 5 days, etc. The return date and stay period cannot appear at the same time. In this case, it is 5 to 7 days, that is, minStay = 5, maxStay = 7;

[0090] Passenger information: optional, such as 2 adults and 1 child. If no input is made, it will be treated as 1 adult by default. In this case, it is 1 adult, that is, passengerType = ADT, number = 1;

[0091] Preference information: optional, including direct / transfer, cabin class, etc. In this example, no input is made and no corresponding parameters are generated;

[0092] Budget information: optional. No input means no limit. In this case, it is within three thousand, that is, budget = 3000.

[0093] 2. Get vertical search results for air tickets

[0094] 2.1 Search agent module generates complete air ticket search request

[0095] The search proxy module replaces the corresponding variable parameter fields in the airline search request template according to the search request parameters generated by the civil aviation big model module to generate a complete air ticket search request. This system supports airlines to define different search request templates according to their own business needs. The request parameters contained in the search request template can be roughly divided into fixed parameters and variable parameters. Fixed parameters are usually airline identity parameters, system parameters, and search mode parameters; variable parameters include passenger itinerary related parameters, such as origin, destination, departure date, departure time, return date, return time, etc.; passenger information corresponding parameters, such as passenger type and quantity; passenger preference related parameters, such as whether it is direct, cabin class, budget limit, etc.

[0096] 2.2 Access the search engine module to obtain flight ticket vertical search results

[0097] The search proxy module sends the generated search request to the search engine module, and the search engine module determines whether it is a fuzzy search based on the search request. If it is a fuzzy search, the search results are directly obtained from the result cache. If the cache is not hit or there are only some results in the cache (partial hits), the search engine module calculates the remaining results in real time and merges them with the cached results and returns them to the search proxy module, and writes the real-time calculated results into the result cache.

[0098] 3. Acquire travel knowledge and integrate with flight vertical search results

[0099] 3.1 Acquire travel knowledge

[0100] The search agent module extracts the destination and flight information from the air ticket search results, and uses the retrieval enhancement generation (RAG) technology to access the travel knowledge base in the civil aviation big model module to obtain travel knowledge related to the destination and flight. While ensuring the accuracy, relevance and freshness of the knowledge, it can also greatly shorten the time of knowledge retrieval and solve the problem of long response time and low accuracy of the big model. The travel knowledge classification related to the destination and flight is as follows: Figure 6 As shown, the system can return all travel knowledge at once, or it can return specified types of travel knowledge according to the business needs of the airline.

[0101] Various types of knowledge in the travel knowledge base will be captured and updated regularly. For volatile data such as weather and currency exchange, a higher update frequency is set to ensure the accuracy and freshness of the data. Figure 7 The following is a schematic diagram of the destination information structure returned by the interface.

[0102] 3.2 Integrate flight search results

[0103] The search agent module integrates the retrieved travel knowledge and air ticket vertical search results, generates the final search results and sends them to the result display module.

[0104] 4. Display search results

[0105] The result display module filters the results according to the display channel and airline requirements (if required by the airline), and then outputs the search results to passengers in the form of tables, pictures or text. Figure 8 and Fig. 9 Shown is a graph showing the search results.

[0106] The flow chart and block diagram in the accompanying drawings illustrate the possible architecture, function and operation of the device, method and computer program product according to various embodiments of the present application. In this regard, each square box in the flow chart or block diagram can represent a module, program segment or a part of code, and the module, program segment or a part of code contains one or more executable instructions for realizing the specified logical function. It should also be noted that each square box in the block diagram and / or flow chart, and the combination of the square boxes in the block diagram and / or flow chart can be realized by a special hardware-based system that performs the specified function or operation, or can be realized by a combination of special hardware and computer instructions.

[0107] like Fig.10 As shown, the embodiment of the present application also discloses an electronic device, including: a processor 310, a communication interface 320, a memory 330 for storing a computer program executable by the processor, and a communication bus 340. The processor 310, the communication interface 320, and the memory 330 communicate with each other through the communication bus 340. The processor 310 implements the steps of the above-mentioned air ticket search method by running the executable computer program.

[0108] It is understandable that, in addition to the memory and the processor, the electronic device may also include an input device such as a keyboard, an output device such as a display, and other communication modules. The input device, the output device, and other communication modules communicate with the processor via an I / O interface (i.e., an input / output interface).

[0109] Computer program code for performing the operations of the present application may be written in one or more programming languages ​​or a combination thereof, including, but not limited to, object-oriented programming languages, such as Java, Smalltalk, C++, and conventional procedural programming languages, such as "C" or similar programming languages. The program code may be executed entirely on the user's computer, partially on the user's computer, as a separate software package, partially on the user's computer and partially on a remote computer, or entirely on a remote computer or server. In cases involving a remote computer, the remote computer may be connected to the user's computer via any type of network, including a local area network (LAN) or a wide area network (WAN), or may be connected to an external computer (e.g., via the Internet using an Internet service provider).

[0110] Furthermore, the present application also discloses a computer-readable storage medium. When the instructions in the computer-readable storage medium are executed by a processor of an electronic device, the electronic device can execute each step of the air ticket search method disclosed in the present application.

[0111] In the context of the present application, a computer-readable storage medium may be a tangible medium, more specific examples of which would include a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the foregoing.

[0112] In particular, according to an embodiment of the present application, the process described in the flowchart can be implemented as a computer software program. For example, an embodiment of the present application includes a computer program product, which includes a computer program carried on a non-transitory computer-readable medium, and the computer program contains program code for executing the air ticket search method disclosed in the present application. When the computer program is executed by a processing device, the above functions defined in the method of the embodiment of the present application are executed.

[0113] Although several specific implementation details are included in the above discussion, these should not be interpreted as limiting the scope of the present application. The above description is only a preferred embodiment of the present application and an explanation of the technical principles used. Those skilled in the art should understand that the scope of disclosure involved in the present application is not limited to the technical solutions formed by a specific combination of the above technical features, but also should cover other technical solutions formed by any combination of the above technical features or their equivalent features without departing from the above public concept.

[0114] Those skilled in the art should also understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some of the technical features therein; and these modifications or replacements do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of this application.

Claims

1. A method for searching air tickets, characterized in that: The method comprises: Step 1: Human-computer interaction: the user inputs the search requirements in the form of voice or text. The system converts the voice into text information, and uses the civil aviation big model to understand and analyze the input information semantically and convert it into corresponding search parameters. Step 2: Generate a complete air ticket search request based on the search parameters and the airline search request template, and access the search engine to obtain the air ticket vertical search results; Step 3: access the civil aviation big model according to the destination and flight information in the air ticket vertical search results, obtain travel knowledge related to the destination and flight, and integrate the obtained travel knowledge with the air ticket vertical search results to obtain search results; Step 4: Filter the search results according to the needs of the airline and display the filtered results to the user.

2. The method according to claim 1, characterized in that The human-computer interaction described in step 1 includes the following steps: The user starts a dialogue with the human-computer interaction module, which determines whether the information input by the user is air ticket query information; if not, it is transferred to other related business modules for processing; if yes, it is transferred to the search agent module for processing; The search agent module determines whether the information input by the user is complete. If it is complete, it will use the civil aviation big model module to understand and analyze its semantics and generate corresponding search parameters. If it is incomplete, it will return to the human-computer interaction module to continue the dialogue. After the dialogue is completed, it will be transferred to the search agent module for processing again.

3. The method according to claim 1, characterized in that Step one also includes: guiding the user to complete the input of the search requirement through the prompt word project.

4. The method according to claim 1, characterized in that: The search parameters described in step 1 include: origin, destination, departure date, departure time, return date, return time, destination stay period, passenger information, preference information and budget information.

5. The method according to claim 4, characterized in that The origin and destination in the search parameters are mandatory items; the departure date, departure time, return date, return time, destination stay period, passenger information, preference information and budget information in the search parameters are optional items; The origin is obtained through user location information, IP information or user input; The stated return date and duration of stay at the destination cannot occur at the same time.

6. The method according to claim 1, characterized in that Step 2 also includes: the search engine determines whether it is a fuzzy search based on the air ticket search request. If it is a fuzzy search, the search results are directly obtained from the result cache; if the cache is not hit or there are only part of the results in the cache, the search engine calculates the remaining results in real time and merges them with the cached results and returns them to the search agent, and writes the real-time calculated results into the result cache.

7. The method according to claim 1, characterized in that In step 4, the screening results are displayed to the user in the form of tables, pictures or texts as needed.

8. An air ticket search device, characterized in that: The device comprises: Human-computer interaction module: used to interact with users through voice or text, convert user voice input into text information, and guide users to complete search demand input through prompt word engineering; the user input information is semantically understood and analyzed through the civil aviation big model module, and converted into corresponding search parameters; Search agent module: used to generate search requests, access the search engine module to obtain vertical search results for air tickets; access the civil aviation big model module to obtain travel knowledge related to the air ticket search results, and integrate the obtained travel knowledge into the vertical search results for air tickets; Civil aviation big model module: used to understand and analyze user input and generate corresponding search parameters based on user input information; receive requests from the search agent module and return corresponding travel knowledge; Search engine module: used to process received air ticket search requests and return corresponding air ticket search results; Result display module: used to output the integrated air ticket vertical search results and travel knowledge to users.

9. 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 air ticket search method according to any one of claims 1 to 7 are implemented.

10. An electronic device, characterized in that: include: Memory and processor; Memory: used to store computer programs; Processor: used to execute the computer program to implement the steps of the air ticket search method according to any one of claims 1 to 7.

Images