Advertisement player information processing method and device, electronic equipment and storage medium

Abstract

The invention provides an advertisement player information processing method and device, electronic equipment and a storage medium, and the method comprises the steps: responding to at least one of first demand information and second demand information, and processing at least one of the two pieces of demand information and third demand information based on a first big language model, first service category advertisement identification items are obtained, the first demand information is generated based on interaction information sent by the advertisement player terminal, the second demand information is generated based on real-time recording information sent by the vehicle-mounted application terminal, and the third demand information is generated based on analysis of associated historical travel orders by a second big language model; obtaining a first advertisement page corresponding to the advertisement identification item of the first service category and pushing the first advertisement page to the advertisement player terminal; and in response to the acquired push request indicating the advertisement identification item of the second media resource, pushing the second advertisement page to the advertisement player terminal. According to the invention, targeted advertisement pushing can be carried out based on different channels, and the advertisement browsing experience of the vehicle-mounted user is improved.

Description

Technical Field

[0001] The present application relates to the field of computer technology, and in particular to an advertising machine information processing method, device, electronic device and storage medium. Background Art

[0002] At present, more and more businesses use advertising machines to push advertisements. Advertising machines are usually set up in public places with large traffic. After the business chooses the time and location for advertising, the advertisements are placed to the public in the form of advertising machines to attract users' attention to the advertising content.

[0003] Advertisements are placed using advertising machines, and the target audience is people in public places. The coverage is limited, and when people browse advertisements, they will selectively pay attention to advertisements corresponding to the goods and services they are interested in. Therefore, this method of placement cannot place advertisements of interest to different groups of people based on their actual needs. There are problems such as being unable to accurately provide advertisements for different groups of people and having a low conversion rate for advertisement placement. Summary of the invention

[0004] In view of the above problems, the embodiments of the present application provide an advertising machine information processing method, device, electronic device and storage medium that overcome the above problems or at least partially solve the above problems.

[0005] In a first aspect, an embodiment of the present application provides an advertising machine information processing method, which is applied to an in-vehicle application server, wherein the in-vehicle application server communicates with an advertising machine terminal and an in-vehicle application terminal installed in an operating vehicle, and the method includes:

[0006] In response to obtaining at least one of the first demand information and the second demand information, at least one of the first demand information and the second demand information and the third demand information are processed based on the first large language model to obtain a first service category advertisement identification item, wherein the first demand information is generated based on the interaction information of the vehicle user in the interaction interface of the advertising machine terminal in the current itinerary order sent by the advertising machine terminal, the second demand information is generated based on the real-time recording information in the current itinerary order sent by the vehicle application terminal, and the third demand information is generated based on the second large language model by analyzing the historical itinerary order information associated with the target account of the current itinerary order;

[0007] Acquire a first advertisement page matching the in-vehicle user according to the first service category advertisement identification item;

[0008] Pushing the first advertisement page to the advertisement machine terminal so as to display the first advertisement page in a first display area of ​​an interactive interface of the advertisement machine terminal;

[0009] In response to obtaining a push request indicating a second media resource advertising identification item, a second advertising page corresponding to the second media resource advertising identification item is pushed to the advertising machine terminal to display the second advertising page in a second display area of ​​the interactive interface of the advertising machine terminal, wherein the push request indicating the second media resource advertising identification item is directly triggered by the in-vehicle application terminal or indirectly triggered by the in-vehicle application terminal through an artificial intelligence assistant, and the second media resource advertising identification item belongs to an element in a set of service advertising identification items allocated to the in-vehicle application terminal.

[0010] In a second aspect, an embodiment of the present application provides an advertising machine information processing device, which is applied to a vehicle-mounted application server, wherein the vehicle-mounted application server communicates with an advertising machine terminal and a vehicle-mounted application terminal installed in an operating vehicle, and the device includes:

[0011] a processing and obtaining module, for, in response to obtaining at least one of the first demand information and the second demand information, processing at least one of the first demand information and the second demand information and the third demand information based on a first large language model to obtain a first service category advertisement identification item, wherein the first demand information is generated based on the interaction information of the vehicle user in the interaction interface of the advertising machine terminal in the current itinerary order sent by the advertising machine terminal, the second demand information is generated based on the real-time recording information in the current itinerary order sent by the vehicle application terminal, and the third demand information is generated based on the second large language model by analyzing the historical itinerary order information associated with the target account of the current itinerary order;

[0012] A first acquisition module, configured to acquire a first advertisement page matching the in-vehicle user according to the first service category advertisement identification item;

[0013] A first push module, configured to push the first advertisement page to the advertisement machine terminal, so as to display the first advertisement page in a first display area of ​​an interactive interface of the advertisement machine terminal;

[0014] A second push module is used to push a second advertising page corresponding to the second media resource advertising identification item to the advertising machine terminal in response to obtaining a push request indicating a second media resource advertising identification item, so as to display the second advertising page in a second display area of ​​the interactive interface of the advertising machine terminal, wherein the push request indicating the second media resource advertising identification item is directly triggered by the in-vehicle application terminal or indirectly triggered by the in-vehicle application terminal through an artificial intelligence assistant, and the second media resource advertising identification item belongs to an element in a service advertising identification item set allocated to the in-vehicle application terminal.

[0015] In a third aspect, an embodiment of the present application provides an electronic device, comprising a processor, a memory, and a computer program stored in the memory and executable on the processor, wherein when the computer program is executed by the processor, the steps of the advertising machine information processing method described in the first aspect are implemented.

[0016] In a fourth aspect, an embodiment of the present application provides a computer-readable storage medium, on which a computer program is stored. When the computer program is executed by a processor, the steps of the advertising machine information processing method described in the first aspect are implemented.

[0017] The technical solution of the embodiment of the present application, when at least one of the first demand information associated with the interactive information submitted by the advertising machine terminal and the second demand information associated with the real-time recording information submitted by the vehicle-mounted application terminal is obtained, at least one of the first demand information and the second demand information and the third demand information analyzed based on the second large language model are processed by using the first large language model, so that the first large language model is used to analyze the historical order data while combining the real-time recording and / or user interaction needs to comprehensively analyze the at least two types of demand information, consider the needs of the vehicle-mounted user from at least two levels, obtain relatively comprehensive demand information, and then select an adapted service category advertising identification item for the vehicle-mounted user, after determining the first service category advertising identification item, obtain a first advertising page matching the vehicle-mounted user according to the first service category advertising identification item, push the first advertising page determined by the large language model reasoning to the advertising machine terminal, display the first advertising page in the first display area of ​​the interactive interface of the advertising machine terminal, and the vehicle-mounted application server pushes the second advertising page corresponding to the second media resource advertising identification item to the advertising machine terminal in response to obtaining the push request indicating the second media resource advertising identification item, so as to display the second advertising page in the second display area of ​​the interactive interface of the advertising machine terminal. It is possible to realize automatic configuration of service category information based on large language model analysis to provide in-vehicle users with advertising content that meets at least one of the interaction needs, conversation needs and historical riding preference needs through the first advertising configuration channel, and provide in-vehicle application terminal users with a second advertising configuration channel for self-configured push of second media resource advertising identification items. While realizing targeted service advertising push for in-vehicle users, it also displays media resource information, thereby improving the in-vehicle users' advertising browsing experience and advertising conversion rate, and also improving the car owner's advertising push experience. BRIEF DESCRIPTION OF THE DRAWINGS

[0018] Figure 1 A schematic diagram showing that the interactive interface of the advertising machine terminal provided in the embodiment of the present application includes different display areas;

[0019] Figure 2A A schematic diagram showing the working process of the first language model provided in an embodiment of the present application;

[0020] Figure 2B A schematic diagram showing the working process of the second largest language model provided in an embodiment of the present application;

[0021] Figure 3 An interactive flow chart showing an advertising machine information processing method provided by an embodiment of the present application;

[0022] Figure 4 A schematic diagram showing an information processing device for an advertising machine provided in an embodiment of the present application;

[0023] Figure 5 A schematic diagram of the structure of an electronic device provided in an embodiment of the present application is shown. DETAILED DESCRIPTION

[0024] The following will be combined with the drawings in the embodiments of the present application to clearly and completely describe the technical solutions 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 creative work are within the scope of protection of this application.

[0025] It should be understood that the "one embodiment" or "an embodiment" mentioned throughout the specification means that the specific features, structures or characteristics related to the embodiment are included in at least one embodiment of the present application. Therefore, "in one embodiment" or "in an embodiment" appearing throughout the specification does not necessarily refer to the same embodiment. In addition, these specific features, structures or characteristics can be combined in one or more embodiments in any suitable manner. The multiple in the embodiments of the present application can include two and more than two.

[0026] In the various embodiments of the present application, it should be understood that the size of the serial numbers of the following processes does not mean the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of the present application.

[0027] The embodiment of the present application provides an advertising machine information processing method applied to the operating vehicle scenario, and the scenario corresponds to a system architecture, which includes an advertising machine terminal and a vehicle-mounted application terminal installed in the operating vehicle, and a vehicle-mounted application server communicating with the advertising machine terminal and the vehicle-mounted application terminal. Among them, the advertising machine terminal is set at a preset position that is convenient for vehicle users to browse pages, such as being set behind the driving seat or the co-pilot seat, so that vehicle users sitting in the back row can browse the content displayed by the advertising machine terminal. When the advertising machine terminal is turned on, the interactive interface of the advertising machine terminal displays the recommended advertising content, and supports vehicle users to perform interactive operations on the interactive interface to select content of interest. The vehicle-mounted application terminal is a terminal used by the driver of the operating vehicle. The driver accepts orders through the vehicle-mounted application terminal and transports the vehicle-mounted user corresponding to the travel order to the destination. As the background server of the vehicle-mounted application terminal, the vehicle-mounted application server establishes a communication connection with the advertising machine terminal and the vehicle-mounted application terminal, receives information provided by the advertising machine terminal and the vehicle-mounted application terminal, provides recommended advertising content for the advertising machine terminal, and dispatches orders for the vehicle-mounted application terminal.

[0028] The following is an introduction to an advertising machine information processing method applied to a vehicle-mounted application server provided in an embodiment of the present application. The method comprises the following steps:

[0029] Step 201, in response to obtaining at least one of the first demand information and the second demand information, at least one of the first demand information and the second demand information and the third demand information are processed based on the first large language model to obtain a first service category advertising identification item, wherein the first demand information is generated based on the interaction information of the in-vehicle user in the interactive interface of the advertising machine terminal in the current itinerary order sent by the advertising machine terminal, the second demand information is generated based on the real-time recording information in the current itinerary order sent by the in-vehicle application terminal, and the third demand information is generated based on the second large language model to analyze the historical itinerary order information associated with the target account of the current itinerary order.

[0030] The interactive interface of the advertising machine terminal displays the advertising page recommended by the vehicle application server. The vehicle users in the current trip order can interact on the interactive interface of the advertising machine terminal based on their own needs and select the content they are interested in. For example, the content displayed on the interactive interface of the advertising machine terminal includes multiple advertising categories. The vehicle users can select the category they are interested in from the multiple advertising categories, and then replace the currently played advertising page with the advertising page corresponding to the advertising category that the vehicle users are interested in. By controlling the advertising machine terminal to play the content that the vehicle users are interested in based on the interactive operations of the vehicle users on the interactive interface, the riding experience of the vehicle users can be improved, and the advertising conversion rate can be improved.

[0031] The advertising machine terminal responds to the interactive operation of the vehicle user on the interactive interface, collects the interactive information corresponding to the interactive operation, and sends the interactive information to the vehicle application server. The interactive information sent by the advertising machine terminal is used to characterize the interactive characteristics of the vehicle user. After obtaining the interactive information corresponding to the vehicle user in the current trip order sent by the advertising machine terminal, the vehicle application server determines the current interactive demand of the vehicle user by analyzing the interactive information, and then generates the first demand information based on the interactive information. Among them, the advertising machine terminal collects and reports the interactive information in real time or collects and reports the interactive information at a set interval (such as 3 seconds), so that the vehicle application server can timely analyze the needs of the vehicle user based on the interactive information.

[0032] After the vehicle user gets on the vehicle, the vehicle application terminal can record the conversation of the vehicle user in the vehicle in the current trip order in real time, and send the real-time recording information to the vehicle application server, so that the vehicle application server can understand the real needs of the vehicle user based on the acquired real-time recording. For example, the vehicle user is a tourist user, and during the trip, he talks with the driver and asks the driver to recommend local food. The driver recommends restaurants to the vehicle user based on the needs of the vehicle user. For the vehicle application server, after receiving the real-time recording information provided by the vehicle application terminal, it analyzes the real-time recording information to determine the current needs of the vehicle user based on the conversation of the vehicle user, and then generates the second demand information based on the real-time recording information.

[0033] In this embodiment, the advertising machine terminal and the vehicle-mounted application terminal are used to provide the collected information to the vehicle-mounted application server, and the vehicle-mounted application server analyzes the information to obtain the first demand information associated with the interactive information submitted by the advertising machine terminal and the second demand information associated with the real-time recording information submitted by the vehicle-mounted application terminal, so as to determine the actual needs of the vehicle-mounted user in the interaction dimension of the vehicle-mounted user and the conversation dimension of the vehicle-mounted user. Optionally, the advertising machine terminal can also extract the interactive features after collecting the interactive information, and submit the interactive features to the vehicle-mounted application server, and the vehicle-mounted application server analyzes the interactive features submitted by the advertising machine terminal to determine the first demand information. Correspondingly, the vehicle-mounted application terminal can also extract the key conversation content after collecting the recording information in real time, and submit the key conversation content to the vehicle-mounted application server, and the vehicle-mounted application server analyzes the key conversation content submitted by the vehicle-mounted application terminal to determine the second demand information.

[0034] Among them, the in-vehicle application server provided in the embodiment of the present application is deployed with a first large language model and a second large language model. The two large language models are mature neural network models containing ultra-large-scale parameters and trained with a large amount of data, and the neural network model is an AI (Artificial Intelligence) large model. The in-vehicle application server uses the second large language model to analyze the historical travel order information associated with the target account of the current travel order, so as to identify the riding habits of the user corresponding to the target account by analyzing the historical travel order situation of the associated target account, and then generate the third demand information. And in the process of generating the third demand information, the large language model is used to perform extended questions and answers to analyze the potential needs of in-vehicle users based on the historical travel order information. When using the large language model for extended questions and answers, the large language model is used to play the identity role of the user, generate relevant demand questions, and match the service demand content according to the relevant demand questions to determine the third demand information. Specifically, the third sub-model of the second largest language model can be used to derive a character portrait of the vehicle user based on historical trip order information, and a matching prompt template can be determined based on the character portrait, and the prompt template includes prompt information related to service demand; the prompt information is input into the fourth sub-model of the second largest language model to obtain demand simulation information; illustratively, the following content can be input into the fourth sub-model: If you are this vehicle user, with a character portrait of... (the character portrait is analyzed by the third sub-model based on historical riding conditions), please give an evaluation of the service consumption demand. The second largest language model generates a series of related demand questions based on the prompt information, and then obtains the service demand content and determines the third demand information, which includes the consumption level of the vehicle user, service category and other information.

[0035] In response to obtaining at least one of the first demand information and the second demand information, the in-vehicle application server processes at least one of the first demand information and the second demand information and the third demand information based on the first large language model to obtain a first service category advertising identification item that matches the in-vehicle user in the current trip order.

[0036] In the process of the vehicle application server using the second language model to generate the third demand information, the vehicle application server extracts the historical trip order information associated with the target account of the current trip order from the historical trip order information, analyzes the extracted historical trip order information, and determines the vehicle user's riding preference and potential demand by analyzing the historical trip, thereby generating the third demand information. The historical trip order information at least includes the vehicle category, the frequency of trip orders, the demand items input during the ride, the pushed historical advertisement identification items, the user's adoption of the pushed advertisement page, the user's browsing time of the advertisement page, etc. And the extracted historical trip order information associated with the target account of the current trip order may include the historical trip order information corresponding to the target account, and the historical trip order information corresponding to other accounts with the same user attributes as the target account. For example, after identifying the user attributes corresponding to the target account, determine other accounts with the same user attributes as the target account, and then extract the corresponding historical trip order information. The user attributes here can be understood as the user's riding habits, user riding preferences, etc.

[0037] Since there is a situation where the in-vehicle user does not touch the advertising terminal and does not generate a session during the ride, the in-vehicle application server processes at least one of the first demand information and the second demand information and the third demand information based on the first large language model when obtaining at least one of the first demand information and the second demand information, so as to determine the first service category advertising identification item matching the in-vehicle user through model analysis. In the process of processing at least one of the first demand information and the second demand information and the third demand information using the first large language model, the first large language model is used to comprehensively consider the actual riding needs of the in-vehicle user at least two levels, so as to combine the real-time recording and / or the user interaction needs while considering the historical order data, and the large language model is used to expand the question and answer based on the demand information to select a suitable service category advertising identification item for the in-vehicle user. It should be noted that the in-vehicle user in the embodiment of the present application may include only the user corresponding to the target account, or may include the user corresponding to the target account and other in-vehicle users, or the in-vehicle user may not include the user corresponding to the target account, such as the scenario where the user corresponding to the target account calls a car for others. When the large language model performs extended questions and answers based on demand information, the large language model is used to play the role of the user's identity, generate demand questions related to demand information of at least two dimensions, match the service demand content according to the relevant demand questions, and then select the appropriate service category advertising identification item. For example, relevant prompt information can be obtained based on demand information of at least two dimensions: If you are this in-vehicle user, with the characteristics of..., what services do you think you will need. The large language model generates a series of relevant demand questions based on the above prompt information, and then obtains the service demand content, determines the service category advertising identification item adapted to the in-vehicle user, and then pushes the advertising content corresponding to the service category advertising identification item to the advertising machine terminal.

[0038] By generating the first demand information based on the interactive information provided by the advertising machine terminal, the actual demand of the in-vehicle user can be analyzed in the interactive dimension; by generating the second demand information based on the real-time recording information provided by the in-vehicle application terminal, the actual demand of the in-vehicle user can be analyzed in the conversation dimension; by extracting the historical travel order information associated with the target account and generating the third demand information based on the second language model, the user's riding preference and potential demand can be analyzed based on the user's historical riding habits. By using the first language model to conduct a comprehensive analysis of at least two types of demand information, the needs of the in-vehicle user can be considered from at least two levels to obtain relatively comprehensive demand information, and then select an adapted service advertisement identification item for the in-vehicle user.

[0039] In the embodiment of the present application, the advertising terminal may extract interaction features after collecting interaction information, and submit the interaction features to the vehicle-mounted application server, thereby determining the first demand information.

[0040] Exemplarily, the advertising machine terminal is equipped with input controls, and the user can perform voice input or click input through the click controls pre-set in the interactive interface of the advertising machine terminal. For example: Display on the advertising machine terminal: Do you need to store luggage after getting off the bus? The user selects yes, and then the advertising machine terminal submits the extracted interactive features to the on-board application server. The on-board application server obtains the first demand information of the on-board user, and combines the third demand information obtained above (such as, including the consumption level of the on-board user), and inputs the two demand information into the first sub-model of the first large language model to generate target demand information. Specifically, the target demand information can be generated according to a preset template. Exemplarily, the target demand can be: the user needs to store luggage at...destination (extracted according to the first demand information), and accept...consumption amount (extracted according to the third demand information). The target demand information is input into the second sub-model of the first large language model to generate service category features that match the target demand information. Specifically, following the above example, the category feature that can be generated by the large language model is: finding a storage point; based on the category feature, the service provider logo that meets the consumption needs of the in-vehicle user can be screened out in the service provider advertising category library at the destination as the first service category advertising logo item.

[0041] Correspondingly, the vehicle-mounted application terminal may also extract key conversation content after real-time collection of recording information, and submit the key conversation content to the vehicle-mounted application server. The vehicle-mounted application server analyzes the key conversation content submitted by the vehicle-mounted application terminal to determine the second demand information.

[0042] Exemplarily, the in-vehicle user agrees to start the trip recording during driving. For example, during driving, the in-vehicle user is captured saying: Master, where can I charge nearby? The in-vehicle application server obtains the user's second demand information based on the key conversation content submitted by the in-vehicle application terminal: the user needs to charge. Combined with the third demand information obtained above (such as the consumption level of the in-vehicle user), the two demand information are input into the first sub-model of the first large language model to generate target demand information. Specifically, the target demand information can be generated according to a preset template. Exemplarily, the target demand can be: the user needs to charge at the destination... (extracted according to the second demand information) and accept the consumption amount... (extracted according to the third demand information). The target demand information is input into the second sub-model of the first large language model to generate service category features that match the target demand information. Specifically, following the above example, the category features that can be generated by the large language model are: finding places that provide charging services; the service provider logo that meets the consumption needs of the in-vehicle user can be screened out in the service provider advertising category library at the destination as the first service category advertising logo item.

[0043] Exemplarily, the advertising machine terminal is equipped with an input control, and the user can perform voice input, or click input through the click control pre-set in the interactive interface of the advertising machine terminal. For example: Display on the advertising machine terminal: Do you need to store luggage after getting off the bus? The user selects yes, and the on-board application server obtains the user's first demand information based on the interactive features submitted by the advertising machine terminal; and the on-board user agrees to start the trip recording during driving, for example: During driving, the on-board user is captured saying: Master, my mobile phone is running out of power, can I borrow a charging cable? At this time, the on-board application server obtains the user's second demand information based on the key conversation content submitted by the on-board application terminal: the user needs to charge. At this time, combined with the third demand information obtained above (such as, including the consumption level of the on-board user), the three demand information are input into the first sub-model of the first language model to generate target demand information. Specifically, the target demand information can be generated according to a preset template. Exemplarily, the target demand can be: the user needs to store luggage at the destination... (extracted according to the first demand information), and the storage point provides charging services (extracted according to the second demand information), and accepts... consumption amount (extracted according to the third demand information). The target demand is input into the second sub-model of the first large language model to generate a service category feature that matches the target demand information. Specifically, following the above example, the category feature that can be generated by the large language model is: find a storage point that provides charging services; based on the category feature, the service provider logo that meets the consumption needs of the vehicle user can be screened out from the service provider advertising category library at the destination as the first service category advertising logo item.

[0044] Step 202: Obtain a first advertisement page matching the in-vehicle user according to the first service category advertisement identification item.

[0045] After comprehensively considering the vehicle user's riding needs at least two levels using the first language model and determining the first service category advertising identification item, a first advertising page matching the vehicle user is obtained according to the first service category advertising identification item.

[0046] The first service category advertisement identification item is a service category advertisement identification item selected from multiple service category advertisement identification items based on the vehicle user's riding demand. The first service category advertisement identification item may correspond to multiple advertisement pages, and the multiple advertisement pages all match the vehicle user's riding demand. When determining the first advertisement page from the multiple advertisement pages corresponding to the first service category advertisement identification item, the following methods are included but not limited to:

[0047] The advertisement page with the highest priority among the multiple advertisement pages corresponding to the advertisement identification item of the first service category is determined as the first advertisement page. The priority of the advertisement page can be preset or determined according to the content corresponding to the advertisement page.

[0048] An advertisement page with the best response effect is determined as a first advertisement page among multiple advertisement pages corresponding to the advertisement identification item of the first service category. The response effect of the advertisement page is determined based on the praise rate and conversion rate of the advertisement page.

[0049] An advertisement page that is urgently needed to be recommended is determined from among a plurality of advertisement pages corresponding to the advertisement identification item of the first service category, and the advertisement page is used as the first advertisement page.

[0050] The advertisement page most suitable for the in-vehicle user is screened out from the multiple advertisement pages corresponding to the advertisement identification item of the first service category. The advertisement page most suitable for the in-vehicle user is determined based on the matching degree between the content corresponding to the advertisement page and the riding demand of the in-vehicle user.

[0051] For example, if the first service category advertisement identification item determined based on the first language model is a service category advertisement identification item related to local attractions, then the advertisement page corresponding to the most popular attraction can be selected from the set of advertisement pages corresponding to the service category advertisement identification items related to local attractions, or the advertisement page corresponding to the attraction that is urgently needed to be recommended can be selected, or the advertisement page corresponding to the most suitable attraction can be selected based on the needs of the in-vehicle user to browse natural scenery. Of course, other methods can also be used to select the corresponding advertisement page, which will not be listed here one by one.

[0052] Step 203: Push the first advertisement page to the advertisement machine terminal, so as to display the first advertisement page in the first display area of ​​the interactive interface of the advertisement machine terminal.

[0053] After obtaining the first advertisement page based on the first service category advertisement identification item analyzed by the first language model, the first advertisement page is pushed to the advertisement machine terminal to display the first advertisement page on the interactive interface of the advertisement machine terminal. The first advertisement page is displayed in the first display area of ​​the interactive interface, which is a pre-defined area. The vehicle user can adjust the position of the first display area on the interactive interface according to browsing habits, and / or adjust the size of the first display area, so that the vehicle user can browse the content displayed in the first display area based on his / her own habits.

[0054] Since the first advertising page is an advertising page pushed based on the cooperation of the first language model and the second language model and the analysis of the demand information of the in-vehicle users in different dimensions, by displaying the first advertising page in the first display area, the in-vehicle users can be provided with advertising content that meets at least one of the interaction needs, conversation needs and historical riding preference needs, so as to achieve targeted advertising push for the in-vehicle users, enhance the user's advertising browsing experience during the ride, and improve the advertising conversion rate.

[0055] In the above implementation process of the present application, when at least one of the first demand information associated with the interactive information submitted by the advertising machine terminal and the second demand information associated with the real-time recording information submitted by the vehicle-mounted application terminal is obtained, the first large language model is used to process at least one of the first demand information and the second demand information and the third demand information analyzed based on the second large language model, so as to use the first large language model to analyze historical data while combining real-time recording and / or user interaction needs, conduct a comprehensive analysis of at least two types of demand information, consider the needs of vehicle-mounted users from at least two levels, obtain relatively comprehensive demand information, and then select an adapted service for the vehicle-mounted user Category advertising identification item, after determining the first service category advertising identification item, obtain the first advertising page matching the in-vehicle user according to the first service category advertising identification item, push the first advertising page determined by large language model reasoning to the advertising machine terminal, and display the first advertising page in the first display area of ​​the interactive interface of the advertising machine terminal. Automatic configuration of service category information based on large language model analysis can be achieved through the first advertising configuration channel to provide the in-vehicle user with advertising content that meets at least one of the interaction needs, conversation needs and historical riding preference needs, thereby improving the in-vehicle user's advertising browsing experience and advertising conversion rate while achieving targeted service advertising push for the in-vehicle user.

[0056] Compared with the traditional processing method of pushing advertisements based on historical data, this embodiment uses a large language model to analyze the user's potential needs, combines real-time recordings and / or user interaction information while considering historical order data, and uses the large language model to expand questions and answers based on the demand information. It can provide a first advertisement page determined based on the large language model reasoning, and implement advertisement push through the first advertisement configuration channel that is automatically configured based on the large language model analysis.

[0057] The in-vehicle application server may also use the second advertisement configuration channel to push advertisements. For information on pushing advertisements based on the second advertisement configuration channel, see step 204 .

[0058] Step 204: in response to obtaining a push request indicating the second media resource advertisement identification item, push a second advertisement page corresponding to the second media resource advertisement identification item to the advertisement machine terminal, so as to display the second advertisement page in a second display area of ​​the interactive interface of the advertisement machine terminal.

[0059] The interactive interface of the advertising terminal includes not only a first display area but also a second display area. The first display area is used to display the advertising page pushed based on the first advertising configuration channel, and the second display area is used to display the advertising page pushed based on the second advertising configuration channel. Specifically, the first display area is used to display the advertising page pushed by the vehicle application server based on the analyzed vehicle user's riding needs, and the second display area is used to display the advertising page pushed by the vehicle application terminal. And the relative position relationship between the first display area and the second display area on the interactive interface can be adaptively adjusted by the vehicle user. Figure 1 As shown, it is an example that the interactive interface of the advertising machine terminal includes a first display area and a second display area.

[0060] The push request indicating the second media resource advertisement identification item obtained by the vehicle-mounted application server may be a push request directly triggered by the vehicle-mounted application terminal, or may be a push request triggered by an AI (Artificial Intelligence) assistant on the vehicle-mounted application server. The vehicle-mounted application server is configured with an AI assistant, which is used to assist the vehicle-mounted application terminal user to automatically configure advertisements. The second media resource advertisement identification item is an element in the service advertisement identification item set allocated to the vehicle-mounted application terminal.

[0061] The vehicle-mounted application server can allocate a service advertisement identification item set to the vehicle-mounted application terminal through interaction with the vehicle-mounted application terminal, and the contents of the service advertisement identification item sets corresponding to different vehicle-mounted application terminals can be the same or different. In case that the contents of the service advertisement identification item sets corresponding to different vehicle-mounted application terminals are different, the vehicle-mounted application server may correspond to multiple service advertisement identification item sets, and each service advertisement identification item set may be allocated by sending different service advertisement identification item sets to different vehicle-mounted application terminals; or multiple service advertisement identification item sets may correspond to multiple advertising categories, and the corresponding category of service advertisement identification item sets are allocated to the vehicle-mounted application terminal based on the selection of the user (driver) to whom the vehicle-mounted application terminal belongs.

[0062] The following first introduces the situation where the vehicle-mounted application terminal sends a push request indicating the second media resource advertisement identification item.

[0063] The interactive interface of the vehicle-mounted application terminal displays the icon corresponding to the advertising identification item included in the corresponding service advertising identification item set. When receiving the input of one of the icons by the user of the vehicle-mounted application terminal, the vehicle-mounted application terminal sends a push request to the vehicle-mounted application server, and the push request indicates the second media resource advertising identification item, and the second media resource advertising identification item is the advertising identification item corresponding to the touched icon. In response to the acquired push request, the vehicle-mounted application server pushes the second advertising page corresponding to the second media resource advertising identification item to the advertising machine terminal, and displays the second advertising page in the second display area of ​​the interactive interface of the advertising machine terminal, so as to realize the display of the advertising page pushed by the vehicle-mounted application terminal in the second display area.

[0064] Among them, the advertisement identification item assigned by the vehicle application server to the vehicle application terminal can be an advertisement identification item with a high recommendation level, and the recommendation level can be related to the urgency of the recommendation and the recommendation category. For example, if a certain advertisement identification item is urgently needed, the corresponding recommendation level is high, and if the audience of a certain category of advertisements is small, the corresponding recommendation level is high. The advertisement identification item assigned by the vehicle application server to the vehicle application terminal can also be an advertisement identification item with a high audience, an advertisement identification item with a high commission return, or an advertisement identification item with a high priority, etc., and the priority of the advertisement identification item can be predetermined or determined based on specific parameters.

[0065] For the user (driver) of the vehicle application terminal, the advertising identification item corresponding to the icon selected in the interactive interface of the vehicle application terminal can be an advertising identification item that matches the current needs of the vehicle user, or an advertising identification item that is urgently recommended, or an advertising identification item with a high commission. As an example, the vehicle user generates a conversation during the ride, which can be a conversation between the vehicle user and the driver, a conversation between multiple vehicle users, or a conversation between the vehicle user and his communication friend. The driver identifies the needs of the vehicle user based on the content of the vehicle user's conversation, and selects a suitable icon on the interactive interface of the vehicle application terminal. The advertising identification item A corresponding to the selected icon A is adapted to the vehicle user's riding needs. The vehicle application terminal sends a push request carrying the advertising identification item A to the vehicle application server, so that the vehicle application server can push the second advertising page corresponding to the advertising identification item A to the advertising machine terminal, so as to realize the push of the adapted advertising content by the vehicle application terminal based on the vehicle user's conversation situation. Here, the advertising identification item A is the second media resource advertising identification item.

[0066] As another example, during the driving process corresponding to the current travel order, the driver selects the icon corresponding to the advertising identification item that needs to be recommended urgently on the interactive interface of the on-board application terminal. The on-board application terminal sends a push request to the on-board application server based on the driver's operation. The on-board application server pushes an adapted second advertising page to the advertising machine terminal in response to the push request, thereby pushing the advertising content corresponding to the advertising identification item that needs to be recommended urgently (the second media resource advertising identification item) to the on-board user based on the driver's operation, so that the driver can complete the recommendation task issued by the on-board application server.

[0067] In another example, the driver selects the icon corresponding to the advertising identification item with a high commission return on the interactive interface of the vehicle-mounted application terminal. The vehicle-mounted application terminal sends a push request to the vehicle-mounted application server based on the driver's operation. The vehicle-mounted application server pushes an adapted second advertising page to the advertising machine terminal in response to the push request. When the playback of the second advertising page meets the conditions (such as the playback time reaches a preset time, the number of playbacks meets a preset number of times), or when the second advertising page is converted based on the operation of the vehicle-mounted user (such as the vehicle-mounted user collects or purchases the product introduced on the second advertising page), the vehicle-mounted application server returns the commission to the user account to which the vehicle-mounted application terminal belongs.

[0068] The following describes the situation where the AI ​​assistant on the in-vehicle application server triggers a push request. In response to obtaining a push request indicating a second media resource advertisement identification item, when a second advertisement page corresponding to the second media resource advertisement identification item is pushed to the advertisement machine terminal, it includes:

[0069] In response to obtaining the request information for starting the AI ​​assistant sent by the in-vehicle application terminal, after obtaining the second indication information that the current travel order has started, obtain the second media resource advertising identification item according to the preset configuration rules; generate a push request indicating the second media resource advertising identification item, and push the second advertising page corresponding to the second media resource advertising identification item to the advertising machine terminal according to the push request.

[0070] For the in-vehicle application server, it can count the historical data of transactions made by in-vehicle users based on the short dramas that bring goods, and analyze the preferences of in-vehicle users based on the counted historical data. Furthermore, the in-vehicle application server can be configured with an AI assistant, which is used to assist the in-vehicle application terminal users to automatically configure advertisements. After obtaining the request information for starting the AI ​​assistant sent by the in-vehicle application terminal, the in-vehicle application server turns on the AI ​​assistant configured by the in-vehicle application server for the in-vehicle application terminal. In response to obtaining that the current trip order corresponding to the in-vehicle user has started, the in-vehicle application server obtains the second media resource advertising identification item according to the preset configuration rules, and generates a push request indicating the second media resource advertising identification item, and pushes the second advertising page corresponding to the second media resource advertising identification item to the advertising machine terminal based on the push request, thereby realizing advertising push for the in-vehicle user based on the second advertising configuration channel, and realizing the display of the second advertising page in the second display area. Among them, the preset configuration rules can be configuration rules pre-set by the in-vehicle application terminal. For example, filtering conditions can be set according to the popularity of the short drama, commission price, etc. The in-vehicle application server considers the configuration rules and uses the AI ​​assistant to analyze the historical short drama subscription data of the in-vehicle user to filter out the short drama content suitable for the in-vehicle user, and then pushes the short drama content suitable for the in-vehicle user to the advertising machine terminal in the form of a second advertising page for display in the second display area of ​​the interactive interface.

[0071] For the in-vehicle application terminal, it starts the AI ​​assistant based on the driver's operation, or automatically sends a request message to start the AI ​​assistant based on specific trigger conditions. After starting the AI ​​assistant, the driver does not need to perform any operation. The AI ​​assistant configured on the in-vehicle application server side automatically configures the short drama content suitable for the current in-vehicle user, so as to enable the in-vehicle application terminal to autonomously push advertisements to the advertising machine terminal based on the started AI assistant.

[0072] In the above implementation scheme of the present application, the on-board application terminal sends a push request indicating the second media resource advertising identification item to the on-board application server, or the push request is triggered by the AI ​​assistant. The on-board application server pushes an adapted advertising page to the advertising machine terminal based on the push request. The driver can push advertising content to the on-board user. On the basis of analyzing the user's potential needs based on the large language model and actively pushing the first advertising page displayed in the first display area, a second advertising page pushed by the driver is configured at the same time, which increases the advertising configuration channel of the advertising machine terminal, realizes the push of advertising content from another angle, and enriches the display content of the advertising machine terminal.

[0073] In another optional embodiment of the present application, the first advertising page and the second advertising page are configured with a service QR code corresponding to the merchant end that provides the target service item; a scanning record sent by the advertising machine terminal is received after the target customer terminal corresponding to the in-vehicle user scans the service QR code to establish communication with the merchant end; in response to the scanning record, virtual resources matching the target service item are allocated to the account corresponding to the in-vehicle application terminal.

[0074] The first and second advertising pages displayed on the interactive interface of the advertising machine terminal are both configured with service QR codes corresponding to the merchant end that provides the target service items, and the corresponding target service items are different for different advertising pages. For example, the first advertising page is a product recommendation page, and the service QR code displayed on the first advertising page is the QR code corresponding to the merchant end that provides the product, and the product is the target service item; the second advertising page is a hotel recommendation page, and the service QR code displayed on the second advertising page is the QR code corresponding to the merchant end that provides hotel services.

[0075] After the target client terminal corresponding to the in-vehicle user scans the service QR code on the advertising page and the target client terminal establishes communication with the merchant, the advertising machine terminal sends the scanning record to the in-vehicle application server. The in-vehicle application server determines, based on the acquired scanning record, that the in-vehicle user has visited the matching merchant based on the advertising content pushed by the advertising machine terminal, that is, the advertising content pushed by the advertising machine terminal has achieved conversion. At this time, the in-vehicle application server allocates virtual resources matching the target service item to the account corresponding to the in-vehicle application terminal, so as to reward the driver based on the advertising conversion achieved on the current operating vehicle.

[0076] Among them, the target customer terminal can be the terminal corresponding to the target account that placed the order for the current itinerary order, or it can be the terminal used by other on-board users in the same trip, or it can be the user corresponding to the target account who is the ordering party but did not participate in the ride, and the target customer terminal is the terminal used by the passenger in the current itinerary order.

[0077] For in-vehicle users, they scan the service QR code and visit the corresponding merchant through the terminal they use (target client terminal) during the ride, establish a communication connection with the merchant through the target client terminal, and then realize the conversion of the advertising content pushed by the advertising machine terminal during the ride. For the advertising machine terminal, based on the scanning action of the target client terminal, it sends a scanning record to report the notification of advertising content conversion to the in-vehicle application server. The in-vehicle application server determines that the advertisement displayed on the advertising machine terminal has been converted based on the scanning record, and allocates virtual resources (such as electronic currency) to the account corresponding to the in-vehicle application terminal, so that the driver can obtain corresponding rewards based on the advertising conversion on the current vehicle.

[0078] The above implementation process configures the service QR code on the advertising page, determines the conversion of the advertisement played by the advertising terminal based on the scanning situation of the target customer terminal, and allocates virtual resources to the account corresponding to the in-vehicle application terminal. On the basis of using the operating vehicle as an advertising delivery platform, it can provide rewards to the driver to enhance the driver's enthusiasm for participating in advertising push.

[0079] The following introduces the process in which the vehicle-mounted application server sends an interactive page to the advertising terminal and the vehicle-mounted application server obtains the first demand information and the second demand information.

[0080] In response to obtaining that the position of the in-vehicle application terminal is within the starting position range of the current trip order, the in-vehicle application server sends an interactive page to the advertising machine terminal to display the interactive page on the interactive interface of the advertising machine terminal, so that the interactive information in the current trip order and input by the in-vehicle user on the interactive interface can be obtained through the content displayed by the advertising machine terminal.

[0081] For the vehicle application server, it communicates with the vehicle application terminal and obtains the location of the vehicle application terminal. Since the starting and ending locations of the current trip order are known information, the vehicle application server sends an interactive page to the advertising machine terminal when it determines that the vehicle application terminal is within the starting location range of the current trip order. The interactive page sent at this time is usually the default page. The interactive interface of the advertising machine terminal displays the interactive page provided by the vehicle application server, so that the vehicle user can browse the page and perform interactive operations on the interactive interface.

[0082] By sending the interactive page to the advertising machine terminal in advance, the advertising machine terminal can display the page content provided by the vehicle application server on the interactive interface before the vehicle user gets on the vehicle, so that the vehicle user can browse the page after getting on the vehicle, without the need to manually trigger the interactive interface to display the content, thereby improving the riding experience of the vehicle user. And when a trip order is generated, the advertising machine terminal is controlled to display the interactive page on the interactive interface, which can reasonably utilize the display resources and avoid the waste of resources caused by playing the advertising content through the advertising machine terminal when there is no vehicle user.

[0083] When the advertising terminal displays information based on the interactive interface, the in-vehicle user performs operations on the interactive interface and selects content of interest to update the interactive interface. The advertising terminal collects the corresponding interactive information of the in-vehicle user when operating the interactive interface, and reports the interactive information to the in-vehicle application server. After obtaining the interactive information corresponding to the in-vehicle user in the current trip order sent by the advertising terminal, the in-vehicle application server analyzes the interactive information, extracts the interactive features, identifies the interactive needs of the in-vehicle user based on the interactive features, and then obtains the first demand information.

[0084] For the vehicle-mounted application server, it obtains the real-time recording information provided by the vehicle-mounted application terminal, and obtains the second demand information by analyzing the real-time recording information. Among them, when the vehicle-mounted application server receives the first indication information allowing recording sent by the target client terminal corresponding to the target account, it sends a recording instruction to the vehicle-mounted application terminal to instruct the vehicle-mounted application terminal to perform real-time recording during the current trip. Since the ordering user performs the ordering action based on the target account, the first indication information is sent by the target client terminal corresponding to the target account used by the ordering user. After obtaining the recording instruction sent by the vehicle-mounted application server, the vehicle-mounted application terminal performs real-time recording of the current trip based on the recording instruction and provides the real-time recording information to the vehicle-mounted application server. The vehicle-mounted application server processes the obtained real-time recording information, extracts key information, and identifies the actual needs of the vehicle-mounted user during the conversation by analyzing the key information, so as to obtain the second demand information in the conversation dimension of the vehicle-mounted user.

[0085] In an embodiment of the present application, for the in-vehicle application server, in response to at least one of the first demand information and the second demand information obtained, it uses a first large language model to perform demand information analysis at at least two levels to determine a service category advertising identification item that is adapted to the in-vehicle user.

[0086] Wherein, after obtaining the second indication information that the current trip order has started, the in-vehicle application server processes the first demand information and the third demand information based on the first language model in response to obtaining the interactive information sent by the advertising machine terminal within the preset time period and not obtaining the real-time recording information sent by the in-vehicle application terminal, and obtains the first service category advertisement identification item; or

[0087] After obtaining the second indication information that the current trip order has started, the in-vehicle application server processes the second demand information and the third demand information based on the first language model to obtain the first service category advertisement identification item in response to not obtaining the interaction information sent by the advertising machine terminal within the preset time period and obtaining the real-time recording information sent by the in-vehicle application terminal; or

[0088] After obtaining the second indication information that the current travel order has started, the in-vehicle application server responds to the interactive information sent by the advertising machine terminal within a preset time period and the real-time recording information sent by the in-vehicle application terminal, and processes the first demand information, the second demand information and the third demand information based on the first language model to obtain the first service category advertising identification item.

[0089] When the in-vehicle user has boarded the vehicle, the driver operates the in-vehicle application terminal, and the in-vehicle application terminal sends a second indication message to the in-vehicle application server that the current trip order has started. The in-vehicle application server responds to the second indication message and monitors whether the interactive information sent by the advertising machine terminal and the real-time recording information sent by the in-vehicle application terminal are obtained within the preset time. If the interactive information sent by the advertising machine terminal is obtained within the preset time, the first demand information and the third demand information are processed based on the first large language model to analyze the needs of the in-vehicle user at the interactive level and the historical riding habit level, and determine the matching first service category advertising identification item. If the real-time recording information sent by the in-vehicle application terminal is obtained within the preset time, the second demand information and the third demand information are processed based on the first large language model to analyze the needs of the in-vehicle user at the recording level and the historical riding habit level, and determine the matching first service category advertising identification item. If the interactive information sent by the advertising machine terminal and the real-time recording information sent by the vehicle-mounted application terminal are obtained within the preset time length, the first demand information, the second demand information and the third demand information are processed based on the first language model to analyze the needs of the vehicle-mounted users at the interaction level, the recording level and the historical riding habit level, and determine the matching first service category advertising identification item.

[0090] The above preset duration is usually the default duration, and can also be adaptively adjusted according to orders of different mileages, such as 50 seconds for orders of travel less than 5 kilometers and 90 seconds for orders of travel greater than 5 kilometers. If the in-vehicle application server does not obtain the interactive information sent by the advertising terminal within the preset duration, or does not obtain the real-time recording information sent by the in-vehicle application terminal, it directly analyzes the needs of the in-vehicle user based on the existing information, promptly determines the adapted service advertisement identification item for the in-vehicle user, and then pushes the in-vehicle user the advertisement content that meets the needs.

[0091] When at least one of the first demand information and the second demand information is obtained within the preset time period, a comprehensive analysis is performed on the currently obtained demand information and the third demand information, so that the needs of the in-vehicle users can be analyzed in a timely manner based on the existing demand information, and then the in-vehicle users can be provided with targeted and adapted advertising content to ensure the in-vehicle users' advertising browsing experience during the ride.

[0092] Wherein, when at least one of the first demand information and the second demand information and the third demand information are processed based on the first language model to obtain the first service category advertisement identification item, it includes:

[0093] Inputting at least one of the first demand information and the second demand information and the third demand information into a first sub-model of the first large language model to generate target demand information;

[0094] Inputting the target demand information into the second sub-model of the first language model to generate service category features matching the target demand information;

[0095] The first service category advertisement identification item is filtered in the advertisement information library according to the service category characteristics.

[0096] like Figure 2A As shown, the first large language model deployed by the vehicle-mounted application server includes a first sub-model and a second sub-model. Among them, the first sub-model is used to analyze demand information at multiple levels and output comprehensive demand. Specifically, in response to obtaining at least one of the first demand information and the second demand information, the vehicle-mounted application server inputs at least one of the first demand information and the second demand information and the third demand information into the first sub-model, and the first sub-model analyzes at least two types of demand information and outputs target demand information representing comprehensive demands of at least two dimensions.

[0097] The second sub-model included in the first large language model is used to analyze the comprehensive demand output by the first sub-model, and output the service category characteristics that match the target demand information, so as to generate the service category characteristics through the cooperation of the two sub-models. After generating the service category characteristics based on the first large language model, the in-vehicle application server screens the first service category advertisement identification item that is adapted to the service category characteristics in the advertisement information library to determine the service category advertisement identification item that matches the in-vehicle user, and then pushes the in-vehicle user the advertisement content that meets his needs.

[0098] Since the service category feature is a feature that matches the target demand information by comprehensively considering the demand information of the vehicle users at different levels, the feature can reflect the needs of the vehicle users at least two levels. The first service category advertising identification item selected in the advertising information library based on the service category feature can be a service category advertising identification item of a specific category. The service category advertising identification item corresponds to multiple advertising pages. The vehicle application server can randomly select an advertising page from the corresponding multiple advertising pages or select an advertising page according to a specific strategy and push it to the advertising machine terminal. The first service category advertising identification item selected in the advertising information library based on the service category feature can also be a service category advertising identification item corresponding to a specific advertisement. The vehicle application server can directly push the corresponding advertising page to the advertising machine terminal.

[0099] By inputting at least one of the first demand information and the second demand information and the third demand information into the first sub-model, using the first sub-model to comprehensively analyze the demand information of different levels of vehicle users to output target demand information, using the second sub-model to process the target demand information to generate service category features, based on the service category features, screening corresponding service category advertising identification items in the advertising information library, and then pushing suitable advertising content to the advertising machine terminal, it is possible to achieve comprehensive reasoning and analysis using a large language model to provide vehicle users with targeted and adapted advertising content.

[0100] The third demand information in the embodiment of the present application is generated by the vehicle-mounted application server based on the deployed second largest language model to analyze the historical travel order information associated with the target account. The process of generating the third demand information is introduced below.

[0101] The in-vehicle application server extracts record information related to guiding service needs from the historical travel order information associated with the target account; inputs the record information into the third sub-model of the second largest language model to generate a character portrait corresponding to the target account; determines a matching prompt template based on the character portrait, and the prompt template includes prompt information related to the service needs; inputs the prompt information into the fourth sub-model of the second largest language model to obtain demand simulation information; extracts key information from the demand simulation information to obtain third demand information.

[0102] The historical trip order information associated with the target account counted by the vehicle-mounted application server includes the historical trip order information corresponding to the target account, and may also include the historical trip order information corresponding to other accounts with the same user attributes as the target account. The user attributes can be understood as the user's riding habits, user riding preferences, etc. After counting the relevant historical trip order information, the vehicle-mounted application server extracts the record information related to the historical trip order information for guiding the service needs. For example, the extracted record information includes the user's taxi frequency, the selected car model (special car, express car, etc.), the order mileage, the advertising content concerned during the trip (such as the preferred advertising category, the advertisements that have been converted in the pushed advertisements, etc.), the taxi time, the starting and ending locations, etc. After extracting the record information related to the service needs from the relevant historical trip order information, the extracted record information is input into the second largest language model.

[0103] like Figure 2BAs shown, the second largest language model includes a third sub-model and a fourth sub-model. The recorded information is input into the third sub-model of the second largest language model, and the third sub-model processes the recorded information used to guide the service needs and generates a character portrait corresponding to the target account. For example, the extracted recorded information includes the following content: the user usually rides a private car, takes a taxi frequently, and the advertisements he pays attention to involve luxury goods. The character portrait output based on the third sub-model indicates that the user belongs to the high-spending group, so as to locate the user's consumption ability.

[0104] After generating the character portrait corresponding to the target account based on the third sub-model, a matching prompt template is determined based on the character portrait, and the prompt template includes prompt information related to the service demand. Among them, determining the matching prompt template based on the character portrait is to select the template that best matches the current user from multiple preset templates. After determining the matching prompt template based on the character portrait, the prompt information in the prompt template is obtained, and the prompt information is input into the fourth sub-model of the second largest language model to obtain demand simulation information. At this stage, the fourth sub-model performs extended questions and answers based on the prompt information, and then analyzes the potential needs of the vehicle user. When using the fourth sub-model for extended questions and answers, the fourth sub-model is used to play the role of the user's identity, generate a series of related demand questions based on the prompt information, and match the service demand content according to the related demand questions to obtain demand simulation information. Exemplarily, the prompt information is: If you are this vehicle user, with a character portrait of..., what services do you think you will need. The fourth sub-model generates a series of related demand questions based on the prompt information, and then obtains demand simulation information. The demand simulation information provided by the fourth sub-model is demand reference information, and the vehicle application server selects key information from these demand reference information as the third demand information. For example, the character portrait has the characteristic of relaxing at the end of the trip. The demand simulation information output by the fourth sub-model is related to the entertainment venues near the end of the trip. The demand simulation information includes the screened entertainment venues near the end of the trip: venue A, venue B and venue C. The on-board application server determines based on the collected historical trip order information that the user corresponding to the target account has taken a large number of rides near venue C, and further screens out venue C near the end of the trip as the third demand information.

[0105] In the above implementation scheme, the historical travel order information associated with the target account is determined, the historical travel order information is analyzed to obtain record information related to guiding service needs, the record information is analyzed using a large language model, and finally demand simulation information is output to provide the vehicle user's demand information based on the large language model, thereby analyzing the potential needs of the vehicle user from the dimension of historical riding preferences.

[0106] The following is an explanation of the advertising machine information processing method provided by the embodiment of the present application through a specific implementation process. Figure 3As shown:

[0107] In response to obtaining that the position of the in-vehicle application terminal is within the starting position range of the current trip order, the in-vehicle application server sends an interactive page to the advertising machine terminal so that the interactive interface of the advertising machine terminal displays the interactive content provided by the in-vehicle application server.

[0108] After receiving the first indication information of allowing recording sent by the target client terminal corresponding to the target account, the vehicle application server sends a recording instruction to the vehicle application terminal, and the vehicle application terminal records the current trip in real time based on the recording instruction.

[0109] The vehicle-mounted application server receives the interactive information submitted by the advertising machine terminal and the real-time recording information submitted by the vehicle-mounted application terminal, generates the first demand information according to the interactive information, and generates the second demand information according to the real-time recording information.

[0110] The on-board application server uses the first largest language model to perform a comprehensive analysis on the first demand information, the second demand information and the third demand information, obtains the first service category advertising identification item, pushes the first advertising page corresponding to the first service category advertising identification item to the advertising machine terminal, and displays the first advertising page on the advertising machine terminal; among which, the third demand information is obtained by analyzing historical travel orders using the second largest language model.

[0111] In response to obtaining the push request indicating the second media resource advertisement identification item, the in-vehicle application server pushes the second advertisement page corresponding to the second media resource advertisement identification item to the advertisement machine terminal, so as to display the second advertisement page on the interactive interface of the advertisement machine terminal.

[0112] When the first advertising page and / or the second advertising page displays the service QR code corresponding to the merchant side that provides the target service item, after the target customer terminal scans the service QR code to establish communication with the merchant side, the advertising machine terminal sends a scanning record, and the on-board application server allocates corresponding virtual resources to the account corresponding to the on-board application terminal based on the scanning record.

[0113] In the above interaction process, the vehicle-mounted application server interacts with the vehicle-mounted application terminal and the advertising machine terminal to push advertising content to the advertising machine terminal. Targeted advertising push can be performed based on the needs of vehicle-mounted users, thereby improving the information browsing experience of vehicle-mounted users during the ride; the vehicle-mounted application server pushes adapted advertising content to the advertising machine terminal in response to the push request, thereby implementing advertising recommendations for the advertising machine terminal based on the second advertising configuration channel associated with the vehicle-mounted application terminal; by allocating virtual resources to the account corresponding to the vehicle-mounted application terminal, rewards can be provided to drivers on the basis of using operating vehicles as advertising delivery platforms to enhance the driver's enthusiasm for participating in advertising push.

[0114] The embodiment of the present application also provides an advertising machine information processing device, which is applied to a vehicle-mounted application server, and the vehicle-mounted application server communicates with an advertising machine terminal and a vehicle-mounted application terminal installed in an operating vehicle, such as Figure 4 As shown, the device comprises:

[0115] The processing and obtaining module 601 is used for, in response to obtaining at least one of the first demand information and the second demand information, processing at least one of the first demand information and the second demand information and the third demand information based on the first large language model to obtain a first service category advertisement identification item, wherein the first demand information is generated based on the interaction information of the vehicle user in the interaction interface of the advertising machine terminal in the current itinerary order sent by the advertising machine terminal, the second demand information is generated based on the real-time recording information in the current itinerary order sent by the vehicle application terminal, and the third demand information is generated based on the second large language model by analyzing the historical itinerary order information associated with the target account of the current itinerary order;

[0116] A first acquisition module 602 is used to acquire a first advertisement page matching the in-vehicle user according to the first service category advertisement identification item;

[0117] A first push module 603 is used to push the first advertisement page to the advertisement machine terminal so as to display the first advertisement page in a first display area of ​​an interactive interface of the advertisement machine terminal;

[0118] The second push module 604 is used to push the second advertising page corresponding to the second media resource advertising identification item to the advertising machine terminal in response to obtaining a push request indicating the second media resource advertising identification item, so as to display the second advertising page in the second display area of ​​the interactive interface of the advertising machine terminal, wherein the push request indicating the second media resource advertising identification item is directly triggered by the in-vehicle application terminal or indirectly triggered by the in-vehicle application terminal through an artificial intelligence assistant, and the second media resource advertising identification item belongs to an element in the service advertising identification item set allocated to the in-vehicle application terminal.

[0119] Optionally, the in-vehicle application server is configured with an artificial intelligence AI assistant, and the AI ​​assistant is used to assist the in-vehicle application terminal user to automatically configure advertisements;

[0120] The second push module is further used for:

[0121] In response to obtaining the request information for starting the AI ​​assistant sent by the in-vehicle application terminal, after obtaining the second indication information that the current trip order has started, obtaining the second media resource advertisement identification item according to a preset configuration rule;

[0122] A push request indicating the second media resource advertisement identification item is generated, and a second advertisement page corresponding to the second media resource advertisement identification item is pushed to the advertisement machine terminal according to the push request.

[0123] Optionally, the first advertising page and the second advertising page are configured with a service QR code corresponding to a merchant end providing the target service item, and the device further includes:

[0124] A first receiving module is used to receive a scanning record sent by the advertising machine terminal after the target client terminal corresponding to the vehicle-mounted user scans the service QR code and establishes communication with the merchant terminal;

[0125] An allocation module is used to allocate virtual resources matching the target service item to the account corresponding to the in-vehicle application terminal in response to the scanning record.

[0126] Optionally, the device further comprises:

[0127] a sending module, configured to send a recording instruction to the in-vehicle application terminal in response to first indication information of allowing recording sent by a target client terminal corresponding to the target account;

[0128] The second receiving module is used to receive the real-time recording information provided by the vehicle-mounted application terminal for real-time recording of the current trip based on the recording instruction;

[0129] The second acquisition module is used to process the real-time recording information to obtain the second demand information.

[0130] Optionally, the processing acquisition module is further used for:

[0131] After obtaining the second indication information indicating that the current trip order has started, in response to obtaining the interactive information sent by the advertising terminal within a preset time period and not obtaining the real-time recording information sent by the vehicle-mounted application terminal, the first demand information and the third demand information are processed based on the first language model to obtain the first service category advertisement identification item;

[0132] After obtaining the second indication information that the current trip order has started, in response to not obtaining the interactive information sent by the advertising terminal within a preset time period, obtaining the real-time recording information sent by the vehicle-mounted application terminal, processing the second demand information and the third demand information based on the first language model to obtain the first service category advertisement identification item;

[0133] After obtaining the second indication information that the current travel order has started, in response to obtaining the interactive information sent by the advertising machine terminal within a preset time period and obtaining the real-time recording information sent by the vehicle-mounted application terminal, the first demand information, the second demand information and the third demand information are processed based on the first large language model to obtain the first service category advertising identification item.

[0134] Optionally, the processing acquisition module includes:

[0135] A first generating submodule, configured to input at least one of the first demand information and the second demand information and the third demand information into a first submodel of the first large language model to generate target demand information;

[0136] A second generating submodule, configured to input the target demand information into a second submodel of the first large language model, and generate service category features matching the target demand information;

[0137] The screening submodule is used to screen the first service category advertisement identification items in the advertisement information library according to the service category characteristics.

[0138] Optionally, the device further comprises:

[0139] An extraction module, used to extract record information related to the service demand from the historical itinerary order information associated with the target account;

[0140] A generating module, configured to input the recorded information into a third sub-model of the second largest language model to generate a character portrait corresponding to the target account;

[0141] A determination module, used to determine a matching prompt template according to the character portrait, wherein the prompt template includes prompt information related to the service demand;

[0142] A third acquisition module, configured to input the prompt information into a fourth sub-model of the second largest language model to acquire demand simulation information;

[0143] The fourth acquisition module is used to extract key information from the demand simulation information to obtain the third demand information.

[0144] As for the device embodiment, since it is basically similar to the method embodiment, the description is relatively simple, and the relevant parts can be referred to the partial description of the method embodiment.

[0145] The embodiment of the present application also provides an electronic device, including: a processor, a memory, and a computer program stored in the memory and executable on the processor. When the computer program is executed by the processor, the various processes of the above-mentioned advertising machine information processing method embodiment are implemented, and the same technical effect can be achieved. To avoid repetition, it will not be described here.

[0146] For example, Figure 5 FIG. 1 shows a schematic diagram of the physical structure of an electronic device. Figure 5 As shown, the electronic device may include: a processor 710, a communication interface 720, a memory 730 and a communication bus 740, wherein the processor 710, the communication interface 720 and the memory 730 communicate with each other via the communication bus 740. The processor 710 may call the logic instructions in the memory 730, and the processor 710 is used to execute each process of the information processing method of the advertising machine in the embodiment of the present application, which will not be described one by one here.

[0147] In addition, the logic instructions in the above-mentioned memory 730 can be implemented in the form of software functional units and can be stored in a computer-readable storage medium when sold or used as an independent product. Based on this understanding, the technical solution of the present application, or the part that contributes to the prior art, or the part of the technical solution, can be embodied in the form of a software product, which is stored in a storage medium and includes a number of instructions for a computer device (which can be a personal computer, a server, or a network device, etc.) to perform all or part of the steps of the method described in each embodiment of the present application.

[0148] The embodiment of the present application also provides a computer-readable storage medium, on which a computer program is stored. When the computer program is executed by a processor, each process of the above-mentioned advertising machine information processing method embodiment is implemented, and the same technical effect can be achieved. To avoid repetition, it is not repeated here. The computer-readable storage medium is, for example, a read-only memory (ROM), a random access memory (RAM), a magnetic disk or an optical disk.

[0149] The various embodiments in this specification are described in a progressive manner, and each embodiment focuses on the differences from other embodiments. The same or similar parts between the various embodiments can be referenced to each other.

[0150] Those skilled in the art will appreciate that the embodiments of the present application may be provided as methods, devices, electronic devices and storage media. Therefore, the embodiments of the present application may adopt the form of a complete hardware embodiment, a complete software embodiment, or an embodiment in combination with software and hardware. Moreover, the embodiments of the present application may adopt the form of a computer program product implemented on one or more computer-readable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0151] The embodiments of the present application are described with reference to the flowcharts and / or block diagrams of the methods, devices, electronic devices, and storage media according to the embodiments of the present application. It should be understood that each process and / or box in the flowchart and / or block diagram, as well as the combination of the processes and / or boxes in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing terminal device to generate a machine, so that the instructions executed by the processor of the computer or other programmable data processing terminal device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 A process or multiple processes and / or boxes Figure 1 These computer program instructions can also be stored in a computer-readable memory that can guide a computer or other programmable data processing terminal device to work in a specific way, so that the instructions stored in the computer-readable memory produce a product including an instruction device, which implements the functions specified in the process. Figure 1 A process or multiple processes and / or boxes Figure 1 These computer program instructions can also be loaded into a computer or other programmable data processing terminal device, so that a series of operation steps are executed on the computer or other programmable terminal device to produce a computer-implemented process, so that the instructions executed on the computer or other programmable terminal device provide for implementing the process in the process. Figure 1 A process or multiple processes and / or boxes Figure 1 The steps for the functions specified in one or more boxes.

[0152] Although the preferred embodiments of the present application have been described, those skilled in the art may make additional changes and modifications to these embodiments once they have learned the basic creative concept. Therefore, the appended claims are intended to be interpreted as including the preferred embodiments and all changes and modifications that fall within the scope of the embodiments of the present application.

[0153] Finally, it should be noted that, in this article, relational terms such as first and second, etc. 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 term "comprise" or any other variant thereof is intended to cover non-exclusive inclusion, so that a process, method, article or terminal 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 terminal device. In the absence of further restrictions, the elements defined by the sentence "comprise..." do not exclude the existence of other identical elements in the process, method, article or terminal device including the elements.

[0154] The above is only a specific implementation of the present application, but the protection scope of the present application is not limited thereto. Any person skilled in the art who is familiar with the present technical field can easily think of changes or substitutions within the technical scope disclosed in the present application, which should be included in the protection scope of the present application. Therefore, the protection scope of the present application should be based on the protection scope of the claims.

Claims

1. An advertising machine information processing method, applied to a vehicle-mounted application server, wherein the vehicle-mounted application server communicates with an advertising machine terminal and a vehicle-mounted application terminal installed in an operating vehicle, characterized in that: The method comprises: In response to obtaining at least one of the first demand information and the second demand information, at least one of the first demand information and the second demand information and the third demand information are processed based on the first large language model to obtain a first service category advertisement identification item, wherein the first demand information is generated based on the interaction information of the vehicle user in the interaction interface of the advertising machine terminal in the current itinerary order sent by the advertising machine terminal, the second demand information is generated based on the real-time recording information in the current itinerary order sent by the vehicle application terminal, and the third demand information is generated based on the second large language model by analyzing the historical itinerary order information associated with the target account of the current itinerary order; Acquire a first advertisement page matching the in-vehicle user according to the first service category advertisement identification item; Pushing the first advertisement page to the advertisement machine terminal so as to display the first advertisement page in a first display area of ​​an interactive interface of the advertisement machine terminal; In response to obtaining a push request indicating a second media resource advertising identification item, a second advertising page corresponding to the second media resource advertising identification item is pushed to the advertising machine terminal to display the second advertising page in a second display area of ​​the interactive interface of the advertising machine terminal, wherein the push request indicating the second media resource advertising identification item is directly triggered by the in-vehicle application terminal or indirectly triggered by the in-vehicle application terminal through an artificial intelligence assistant, and the second media resource advertising identification item belongs to an element in a set of service advertising identification items allocated to the in-vehicle application terminal.

2. The advertising machine information processing method according to claim 1, characterized in that: The in-vehicle application server is configured with an artificial intelligence assistant, and the artificial intelligence assistant is used to assist the in-vehicle application terminal user to automatically configure advertisements; The method of, in response to obtaining a push request indicating a second media resource advertisement identification item, pushing a second advertisement page corresponding to the second media resource advertisement identification item to the advertisement machine terminal includes: In response to obtaining the request information for starting the artificial intelligence assistant sent by the in-vehicle application terminal, after obtaining the second indication information that the current trip order has started, obtaining the second media resource advertisement identification item according to a preset configuration rule; A push request indicating the second media resource advertisement identification item is generated, and a second advertisement page corresponding to the second media resource advertisement identification item is pushed to the advertisement machine terminal according to the push request.

3. The advertising machine information processing method according to claim 1, characterized in that: The first advertising page and the second advertising page are configured with a service QR code corresponding to the merchant end providing the target service item; Receiving a scanning record sent by the advertising machine terminal after the target client terminal corresponding to the in-vehicle user scans the service QR code and establishes communication with the merchant terminal; In response to the scanning record, a virtual resource matching the target service item is allocated to an account corresponding to the in-vehicle application terminal.

4. The advertising machine information processing method according to claim 1, characterized in that: Also includes: In response to first indication information of allowing recording sent by a target client terminal corresponding to the target account, sending a recording instruction to the in-vehicle application terminal; Receiving real-time recording information provided by the vehicle-mounted application terminal for real-time recording of the current trip based on the recording instruction; The real-time recording information is processed to obtain the second demand information.

5. The advertising machine information processing method according to claim 1 or 4, characterized in that: After obtaining the second indication information indicating that the current trip order has started, in response to obtaining the interactive information sent by the advertising terminal within a preset time period and not obtaining the real-time recording information sent by the vehicle-mounted application terminal, the first demand information and the third demand information are processed based on the first language model to obtain the first service category advertisement identification item; After obtaining the second indication information that the current trip order has started, in response to not obtaining the interactive information sent by the advertising terminal within a preset time period, obtaining the real-time recording information sent by the vehicle-mounted application terminal, processing the second demand information and the third demand information based on the first language model to obtain the first service category advertisement identification item; After obtaining the second indication information that the current travel order has started, in response to obtaining the interactive information sent by the advertising machine terminal within a preset time period and obtaining the real-time recording information sent by the vehicle-mounted application terminal, the first demand information, the second demand information and the third demand information are processed based on the first large language model to obtain the first service category advertising identification item.

6. The advertising machine information processing method according to claim 1, characterized in that: The step of processing at least one of the first demand information and the second demand information and the third demand information based on the first large language model to obtain the first service category advertisement identification item includes: Inputting at least one of the first demand information and the second demand information and the third demand information into the first sub-model of the first large language model to generate target demand information; Inputting the target demand information into the second sub-model of the first large language model to generate service category features matching the target demand information; The first service category advertisement identification item is filtered in the advertisement information library according to the service category characteristics.

7. The advertising machine information processing method according to claim 1, characterized in that: The method further comprises: Extracting record information related to the service demand from the historical itinerary order information associated with the target account; Inputting the recorded information into the third sub-model of the second largest language model to generate a character portrait corresponding to the target account; Determining a matching prompt template according to the character portrait, wherein the prompt template includes prompt information related to the service demand; Inputting the prompt information into the fourth sub-model of the second largest language model to obtain demand simulation information; Key information is extracted from the demand simulation information to obtain the third demand information.

8. An advertising machine information processing device, applied to a vehicle-mounted application server, wherein the vehicle-mounted application server communicates with an advertising machine terminal and a vehicle-mounted application terminal installed in an operating vehicle, characterized in that: The device comprises: a processing and obtaining module, for, in response to obtaining at least one of the first demand information and the second demand information, processing at least one of the first demand information and the second demand information and the third demand information based on a first large language model to obtain a first service category advertisement identification item, wherein the first demand information is generated based on the interaction information of the vehicle user in the interaction interface of the advertising machine terminal in the current itinerary order sent by the advertising machine terminal, the second demand information is generated based on the real-time recording information in the current itinerary order sent by the vehicle application terminal, and the third demand information is generated based on the second large language model by analyzing the historical itinerary order information associated with the target account of the current itinerary order; A first acquisition module, configured to acquire a first advertisement page matching the in-vehicle user according to the first service category advertisement identification item; A first push module, configured to push the first advertisement page to the advertisement machine terminal, so as to display the first advertisement page in a first display area of ​​an interactive interface of the advertisement machine terminal; A second push module is used to push a second advertising page corresponding to the second media resource advertising identification item to the advertising machine terminal in response to obtaining a push request indicating a second media resource advertising identification item, so as to display the second advertising page in a second display area of ​​the interactive interface of the advertising machine terminal, wherein the push request indicating the second media resource advertising identification item is directly triggered by the in-vehicle application terminal or indirectly triggered by the in-vehicle application terminal through an artificial intelligence assistant, and the second media resource advertising identification item belongs to an element in a service advertising identification item set allocated to the in-vehicle application terminal.

9. An electronic device, characterized in that: The invention comprises a processor, a memory and a computer program stored in the memory and executable on the processor, wherein when the computer program is executed by the processor, the steps of the advertising machine information processing method as claimed in any one of claims 1 to 7 are implemented.

10. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the steps of the advertising machine information processing method according to any one of claims 1 to 7 are implemented.

Images