Ticket business recommendation method and system based on artificial intelligence

Abstract

The invention belongs to the technical field of ticket business recommendation. The invention provides a ticket business recommendation method and system based on artificial intelligence. The method comprises the steps of obtaining ticket business purchase big data of multiple types of users, constructing a first training data set according to the ticket business purchase big data, and training a first ticket business recommendation model by using the first training data set; acquiring first ticket business purchase historical data of the target user, analyzing the first ticket business purchase historical data by using a first ticket business recommendation model, and predicting to obtain multiple pieces of first ticket business recommendation information; acquiring second ticket purchase historical data of the target user in a recent first time period, analyzing the second ticket purchase historical data by using a second ticket recommendation model, and predicting to obtain multiple pieces of second ticket recommendation information; and integrating the first ticket business recommendation information and the second ticket business recommendation information to obtain third ticket business recommendation information, and pushing the third ticket business recommendation information to the target user. According to the scheme provided by the invention, more suitable ticket recommendation information is determined by integrating ticket purchase rule hobbies of multiple types of users and individual conditions of individual users.

Description

Technical Field

[0001] The present invention relates to the technical field of ticket recommendation, and in particular to an artificial intelligence-based ticket recommendation method and system. Background Art

[0002] The theater ticketing system's intelligent recommendation function can recommend the most suitable performances to the audience based on their preferences and historical records, so that every performance can meet the audience's expectations.

[0003] The existing ticket recommendation system is based on the analysis of the big data of ticket purchase and viewing of multiple types of users to obtain the viewing habits of different types of users, and then recommend tickets for the corresponding types of users. However, this method does not take into account the personalized needs of individual users, as well as the real and immediate needs of individual users in certain special scenes and locations, resulting in the recommended ticket information not meeting the real needs of individual users, and poor user experience.

[0004] Therefore, how to further optimize the ticket information recommendation method is a technical problem that urgently needs to be solved. Summary of the invention

[0005] In this regard, the present invention provides a ticket recommendation method, system, electronic device, computer storage medium and computer program product based on artificial intelligence to solve the above technical problems.

[0006] The present invention also provides a ticket recommendation method based on artificial intelligence, which includes the following steps: obtaining ticket purchase big data of multiple types of users, constructing a first training data set based on the ticket purchase big data, and using the first training data set to train a first ticket recommendation model; obtaining first ticket purchase history data of a target user, using the first ticket recommendation model to analyze the first ticket purchase history data, and predicting multiple first ticket recommendation information; obtaining second ticket purchase history data of the target user in a recent first time period, using a second ticket recommendation model to analyze the second ticket purchase history data, and predicting multiple second ticket recommendation information; wherein the second ticket recommendation model is trained based on the second training data set, and the second training data set is constructed based on the first ticket purchase history data; integrating the first ticket recommendation information and the second ticket recommendation information to obtain third ticket recommendation information, and pushing the third ticket recommendation information to the target user.

[0007] In some embodiments, the first training data set is constructed based on the ticket purchase big data, and the first ticket recommendation model is trained using the first training data set, including: classifying the acquired ticket purchase big data according to user type, and constructing multiple first training data sets based on part of the classified ticket purchase big data; screening multiple collaborative nodes corresponding to the number of classifications of the ticket purchase big data in the blockchain network, sending the first ticket recommendation model and the corresponding first training data set to the corresponding collaborative nodes respectively, and each collaborative node uses the received first training data set to train the first ticket recommendation model, and when the convergence conditions corresponding to each collaborative node are met, the training of the first ticket recommendation model is stopped, and the key model parameters of the first ticket recommendation model are fed back to the local node; the local node integrates the fedback groups of the key model parameters to obtain target key model parameters, and inputs the target key model parameters into the first ticket recommendation model of the local node; randomly selects part of the training data from each of the first training data sets to form a third training data set, and uses the third training data set to train the first ticket recommendation model with the target key model parameters input until the convergence conditions of the local node are met.

[0008] In some embodiments, the first ticket recommendation model is constructed based on a convolutional neural network, and the second ticket recommendation model is constructed based on a large language model.

[0009] In some embodiments, the obtaining of the second ticket purchase history data of the target user within a recent first time period includes: obtaining the activity venue information of the target user within a recent preset time period, performing semantic analysis and semantic clustering on the activity venue information to obtain several activity venue types, and calculating the total number of each activity venue type; determining a duration adjustment coefficient based on the total number of the activity venue types, and using the duration adjustment coefficient to adjust the preset time period to the first time period; obtaining the second ticket purchase history data of the target user within the recent first time period.

[0010] In some embodiments, the using of the second ticket recommendation model to analyze the second ticket purchase history data to predict multiple pieces of second ticket recommendation information includes: using a semantic analysis component to perform semantic transformation and clustering on the second ticket purchase history data to obtain simplified third ticket purchase history data, inputting the third ticket purchase history data into the second ticket recommendation model to obtain predicted multiple pieces of second ticket recommendation information.

[0011] In some embodiments, the first ticket recommendation information and the second ticket recommendation information are integrated to obtain the third ticket recommendation information, including: determining the integration ratio according to the duration adjustment coefficient, determining the first integrated quantity of the second ticket recommendation information according to the preset total number of ticket recommendation information and the integration ratio, and calculating the second integrated quantity of the first ticket recommendation information according to the total number and the first integrated quantity; sorting the second ticket recommendation information from large to small according to the recommendation degree value of each piece of the second ticket recommendation information, and selecting the first integrated quantity of the second ticket recommendation information at the head of the sorting queue; similarly, screening the first ticket recommendation information from each piece of the first ticket recommendation information to obtain the second integrated quantity; sorting the selected first ticket recommendation information and second ticket recommendation information from large to small according to the recommendation degree value, so as to obtain the third ticket recommendation information.

[0012] The present invention also provides a ticket recommendation system based on artificial intelligence, which includes a data processor and a data storage device; the data processor calls and runs the computer program code in the data storage device to implement the following steps: obtaining ticket purchase big data of multiple types of users, constructing a first training data set based on the ticket purchase big data, and using the first training data set to train a first ticket recommendation model; obtaining first ticket purchase history data of a target user, using the first ticket recommendation model to analyze the first ticket purchase history data, and predicting multiple first ticket recommendation information; obtaining second ticket purchase history data of the target user in a recent first time period, using the second ticket recommendation model to analyze the second ticket purchase history data, and predicting multiple second ticket recommendation information; wherein the second ticket recommendation model is trained based on the second training data set, and the second training data set is constructed based on the first ticket purchase history data; integrating the first ticket recommendation information and the second ticket recommendation information to obtain third ticket recommendation information, and pushing the third ticket recommendation information to the target user.

[0013] The present invention also provides an electronic device, comprising: at least one processor, a memory, and a computer program stored in the memory and executable on the at least one processor, wherein the processor executes the computer program to implement any of the methods described above.

[0014] The present invention also provides a computer storage medium, wherein the computer readable storage medium stores a computer program, and the computer program is executed by a processor to implement any of the above methods.

[0015] The present invention also provides a computer program product, wherein the computer program product includes computer code, and when the computer code is executed by a processor of an electronic device, any of the above methods is implemented.

[0016] The solution of the present invention combines the ticket purchase rules and preferences of multiple types of users and the personalized conditions of individual users to determine more appropriate ticket recommendation information, so that individual users can obtain more diverse ticket recommendation information that meets their ticket purchase rules and preferences. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings required for use in the embodiments are briefly introduced below. It should be understood that the following drawings only show certain embodiments of the present invention and therefore should not be regarded as limiting the scope. For ordinary technicians in this field, other related drawings can be obtained based on these drawings without creative work.

[0018] Figure 1 It is a flowchart of a ticket recommendation method based on artificial intelligence disclosed in an embodiment of the present invention.

[0019] Figure 2 It is a schematic diagram of the relationship between the semantic analysis component and the second ticket recommendation model disclosed in an embodiment of the present invention.

[0020] Figure 3 It is a structural diagram of an artificial intelligence-based ticket recommendation system disclosed in an embodiment of the present invention. DETAILED DESCRIPTION

[0021] The following is a description of the implementation of the present application by specific specific embodiments. People familiar with the technology can easily understand other advantages and effects of the present application from the contents disclosed in this specification. 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 the field without creative work are within the scope of protection of the present application.

[0022] In addition, the technical features involved in the different embodiments of the present application described below can be combined with each other as long as they do not conflict with each other.

[0023] The existing ticket recommendation system is based on the analysis of the big data of ticket purchase and viewing of multiple types of users to obtain the viewing habits of different types of users, and then recommend tickets for the corresponding types of users. However, this method does not take into account the personalized needs of individual users, as well as the real and immediate needs of individual users in certain special scenes and locations, resulting in the recommended ticket information not meeting the real needs of individual users, and poor user experience.

[0024] Therefore, how to further optimize the ticket information recommendation method is a technical problem that urgently needs to be solved.

[0025] In response to the above technical issues, such as Figure 1 As shown, an embodiment of the present invention discloses an artificial intelligence-based ticket recommendation method, which includes the following steps: obtaining ticket purchase big data of multiple types of users, constructing a first training data set based on the ticket purchase big data, and using the first training data set to train a first ticket recommendation model; obtaining first ticket purchase history data of a target user, using the first ticket recommendation model to analyze the first ticket purchase history data, and predicting multiple first ticket recommendation information; obtaining second ticket purchase history data of the target user in a recent first time period, using the second ticket recommendation model to analyze the second ticket purchase history data, and predicting multiple second ticket recommendation information; wherein the second ticket recommendation model is trained based on the second training data set, and the second training data set is constructed based on the first ticket purchase history data; integrating the first ticket recommendation information and the second ticket recommendation information to obtain third ticket recommendation information, and pushing the third ticket recommendation information to the target user.

[0026] In the above scheme of the present invention, the present invention sets up two models, namely the first ticket recommendation model and the second ticket recommendation model, and obtains the best ticket recommendation information based on the two models. Specifically: First, obtain the ticket purchase big data of various types of users, and the ticket purchase big data can be obtained by, for example, web crawler technology. A first training data set is constructed based on the ticket purchase big data, and the first ticket recommendation model is trained using the first training data set. The trained first ticket recommendation model has mastered the ticket purchase habits of different types of users, and can then filter out ticket recommendation information that meets the ticket purchase habits (i.e., drama viewing preferences) of users of the corresponding types from the currently running dramas.

[0027] Then, the first ticket purchase history data of a specific individual user, i.e., the target user, is obtained, and the first ticket purchase history data is analyzed using the first ticket recommendation model to predict a plurality of first ticket recommendation information. The first ticket recommendation information is predicted based on the purchase and viewing habits and preferences of the group.

[0028] Next, the target user's second ticket purchase history data in the recent first period (the length of the preset period will be further explained later) is obtained, and the second ticket purchase history data is analyzed using the second ticket recommendation model to predict multiple second ticket recommendation information. The second ticket recommendation model is pre-constructed based on the target user's first ticket purchase history data, that is, a second training data set is constructed based on the first ticket purchase history data, and then the second training data set is used to train the second ticket recommendation model.

[0029] Finally, the first ticket recommendation information predicted by the first ticket recommendation model and the second ticket recommendation information predicted by the second ticket recommendation model are integrated to obtain the final third ticket recommendation information, which is pushed to the target user.

[0030] Therefore, the solution of the present invention combines the ticket purchase rules and preferences of multiple types of users and the personalized conditions of individual users to determine more appropriate ticket recommendation information, so that individual users can obtain more diverse ticket recommendation information that meets their ticket purchase rules and preferences.

[0031] In some embodiments, the first training data set is constructed based on the ticket purchase big data, and the first ticket recommendation model is trained using the first training data set, including: classifying the acquired ticket purchase big data according to user type, and constructing multiple first training data sets based on part of the classified ticket purchase big data; screening multiple collaborative nodes corresponding to the number of classifications of the ticket purchase big data in the blockchain network, sending the first ticket recommendation model and the corresponding first training data set to the corresponding collaborative nodes respectively, and each collaborative node uses the received first training data set to train the first ticket recommendation model, and when the convergence conditions corresponding to each collaborative node are met, the training of the first ticket recommendation model is stopped, and the key model parameters of the first ticket recommendation model are fed back to the local node; the local node integrates the fedback groups of the key model parameters to obtain target key model parameters, and inputs the target key model parameters into the first ticket recommendation model of the local node; randomly selects part of the training data from each of the first training data sets to form a third training data set, and uses the third training data set to train the first ticket recommendation model with the target key model parameters input until the convergence conditions of the local node are met.

[0032] In this embodiment, the present invention uses a distributed training mode to train the first ticket recommendation model. Specifically, the acquired ticket purchase big data is first classified according to user type, and the user type is a preset classification, for example, it can be divided into movie lovers, drama lovers, stage play lovers, etc. A plurality of first training data sets are constructed based on the classified part of the ticket purchase big data.

[0033] Then, multiple collaborative nodes corresponding to the number of classifications of the ticket purchase big data (i.e., the number of preset user types) are screened in the blockchain network. The first ticket recommendation model and a set of first training data sets are sent to a collaborative node. These collaborative nodes each determine their own convergence conditions and use the received first training data sets to train the first ticket recommendation model. When the convergence conditions are met, the training of the first ticket recommendation model is stopped, and the key model parameters in the trained first ticket recommendation model are fed back to the local node.

[0034] Finally, the local node integrates the feedback of each group of key model parameters, such as taking the average, to obtain the target key model parameters, and inputs the target key model parameters into the first ticket recommendation model of the local node. Also, randomly select some training data from each first training data set, use these training data to form a third training data set, and the third training data set contains training data of multiple types of users. Use the third training data set to perform secondary training on the first ticket recommendation model with the target key model parameters input until the convergence condition of the local node is reached. At this point, the final first ticket recommendation model is obtained.

[0035] Due to the use of a distributed training method, the convergence conditions adopted by each collaborative node are different, and the local node uses training data based on multiple user types to perform secondary training on the model. The first ticket recommendation model obtained in this way has undergone a richer training process, which is conducive to subsequent predictions to obtain more accurate first ticket recommendation information.

[0036] In some embodiments, the first ticket recommendation model is constructed based on a convolutional neural network, and the second ticket recommendation model is constructed based on a large language model.

[0037] In this embodiment, the present invention preferably uses a convolutional neural network to construct the first ticket recommendation model, such as LeNet, AlexNet, VGGNet, GoogLeNet, ResNet, etc.

[0038] At the same time, since the second ticket recommendation model needs to conduct a more in-depth analysis of the second ticket purchase history data of the target user in the recent first time period, its analysis complexity is higher than that of the first ticket recommendation model. Therefore, the present invention preferably sets the second ticket recommendation model to be constructed based on a large language model. Among them, the large language model based on is, for example, a large model of the GPT series, a large model of the BERT series, a large model of T5, an ERNIE large model, etc. At the same time, the significant difference from conventional models such as convolutional neural networks, LSTM, etc. is that the large model does not need to be trained with a large amount of training samples, but only needs to be trained with a small number of small sample training data. The specific training process will not be repeated here.

[0039] It should be noted that the key model parameters of the first ticket recommendation model at this time refer to the size of the convolution kernel, the size and step size of the pooling window, the weight and bias term of the fully connected layer, the learning rate, etc. Of course, due to the different specific algorithms used by the convolutional neural network model, the corresponding key model parameters also have some differences, which will not be described in detail.

[0040] In some embodiments, the obtaining of the second ticket purchase history data of the target user within a recent first time period includes: obtaining the activity venue information of the target user within a recent preset time period, performing semantic analysis and semantic clustering on the activity venue information to obtain several activity venue types, and calculating the total number of each activity venue type; determining a duration adjustment coefficient based on the total number of the activity venue types, and using the duration adjustment coefficient to adjust the preset time period to the first time period; obtaining the second ticket purchase history data of the target user within the recent first time period.

[0041] In this embodiment, the present invention sets a preset time period, which is a fixed value, such as one week or two weeks. First, the activity venue information (preferably the drama viewing venue) of the target user in the recent preset time period is obtained, and the obtained activity venue information is semantically analyzed and semantically clustered to obtain several activity venue types, and the total number of each activity venue type is calculated. The total number of activity venue types is related to the change in the recent drama viewing habits of the target user. Specifically, when the total number of activity venue types is larger, it indicates that the change range of the recent drama viewing habits of the target user is larger. Correspondingly, the duration adjustment coefficient can be set to a smaller value, that is, based on more recent data to analyze its new viewing habits; on the contrary, when the total number of activity venue types is smaller, it indicates that the change range of the recent drama viewing habits of the target user is smaller, that is, it is more consistent with historical habits. Correspondingly, the duration adjustment coefficient can be set to a larger value, that is, based on more historical data to analyze its viewing habits. Obviously, the duration adjustment coefficient is negatively correlated with the total number of activity venue types.

[0042] Therefore, the time span of the preset period can be adjusted using the time adjustment coefficient to adjust the amount of historical data or recent data covered. Here, the adjustment of the time span of the preset period refers to adjusting from the current time as the starting point to the historical time direction.

[0043] In some embodiments, the using of the second ticket recommendation model to analyze the second ticket purchase history data to predict multiple pieces of second ticket recommendation information includes: using a semantic analysis component to perform semantic transformation and clustering on the second ticket purchase history data to obtain simplified third ticket purchase history data, inputting the third ticket purchase history data into the second ticket recommendation model to obtain predicted multiple pieces of second ticket recommendation information.

[0044] In this embodiment, since the target user's recent second ticket purchase history data may contain a lot of irrelevant and repeated data, the present invention first uses a semantic analysis component to perform semantic transformation and clustering on the second ticket purchase history data to achieve simplified processing of the second ticket purchase history data, and then obtain the third ticket purchase history data. Then, the simplified third ticket purchase history data is analyzed in depth using the second ticket recommendation model, so as to predict multiple second ticket recommendation information. These second ticket recommendation information are predicted based on the target user's recent drama viewing habits.

[0045] In some embodiments, the first ticket recommendation information and the second ticket recommendation information are integrated to obtain the third ticket recommendation information, including: determining the integration ratio according to the duration adjustment coefficient, determining the first integrated quantity of the second ticket recommendation information according to the preset total number of ticket recommendation information and the integration ratio, and calculating the second integrated quantity of the first ticket recommendation information according to the total number and the first integrated quantity; sorting the second ticket recommendation information from large to small according to the recommendation degree value of each piece of the second ticket recommendation information, and selecting the first integrated quantity of the second ticket recommendation information at the head of the sorting queue; similarly, screening the first ticket recommendation information from each piece of the first ticket recommendation information to obtain the second integrated quantity; sorting the selected first ticket recommendation information and second ticket recommendation information from large to small according to the recommendation degree value, so as to obtain the third ticket recommendation information.

[0046] In this embodiment, when integrating, part of the first ticket recommendation information predicted by the first ticket recommendation model and part of the second ticket recommendation information predicted by the second ticket recommendation model are integrated, and the recommendation values ​​predicted by the first ticket recommendation model and the second ticket recommendation model are sorted from large to small, and then output to the target user in sequence. Specifically: First, the integration ratio is determined according to the aforementioned duration adjustment coefficient, and the integration ratio is used to determine the number of second ticket recommendation information in the third ticket recommendation information. The total number of the third ticket recommendation information is preset, and the total number is multiplied by the integration ratio to obtain the first integrated number of the second ticket recommendation information, and the remainder is the second integrated number of the first ticket recommendation information. Among them, there is a negative correlation between the integration ratio and the aforementioned duration adjustment coefficient, but the present invention does not specifically limit the corresponding form of the negative correlation.

[0047] Then, sort the recommendation values ​​of each first ticket recommendation information from large to small, and sort the recommendation values ​​of each second ticket recommendation information from large to small, and select the first ticket recommendation information of the second integrated quantity at the head of the queue and the second ticket recommendation information of the first integrated quantity at the head of the queue respectively. Then sort the selected first ticket recommendation information and second ticket recommendation information again from large to small according to the recommendation values, and obtain the third ticket recommendation information. The third ticket recommendation information includes the total number of ticket recommendation information mentioned above.

[0048] like Figure 3 As shown, an embodiment of the present invention further discloses an artificial intelligence-based ticket recommendation system, the system comprising a data processor and a data storage device; the data processor calls and runs the computer program code in the data storage device to implement the following steps: obtaining ticket purchase big data of multiple types of users, constructing a first training data set based on the ticket purchase big data, and using the first training data set to train a first ticket recommendation model; obtaining first ticket purchase history data of a target user, using the first ticket recommendation model to analyze the first ticket purchase history data, and predicting multiple first ticket recommendation information; obtaining second ticket purchase history data of the target user in a recent first time period, using the second ticket recommendation model to analyze the second ticket purchase history data, and predicting multiple second ticket recommendation information; wherein the second ticket recommendation model is trained based on the second training data set, and the second training data set is constructed based on the first ticket purchase history data; integrating the first ticket recommendation information and the second ticket recommendation information to obtain third ticket recommendation information, and pushing the third ticket recommendation information to the target user.

[0049] An embodiment of the present invention further discloses an electronic device, comprising: at least one processor, a memory, and a computer program stored in the memory and executable on the at least one processor, wherein the processor executes the computer program to implement the method described in the above embodiment.

[0050] An embodiment of the present invention further discloses a computer storage medium, wherein the computer storage medium stores a computer program, and the computer program is executed by a processor to implement the method described in the above embodiment.

[0051] An embodiment of the present invention further discloses a computer program product, which includes computer code. When the computer code is executed by a processor of an electronic device, the method described in the above embodiment is implemented.

[0052] The various component embodiments of the present invention can be implemented in hardware, or in software modules running on one or more processors, or in a combination thereof. It should be understood by those skilled in the art that a microprocessor or digital signal processor (DSP) can be used in practice to implement some or all functions of some or all components in the device according to an embodiment of the present invention. The present invention can also be implemented as a device or apparatus program (e.g., computer program and computer program product) for executing part or all of the methods described herein. Such a program implementing the present invention can be stored on a computer-readable medium, or can have the form of one or more signals. Such a signal can be downloaded from an Internet website, or provided on a carrier signal, or provided in any other form.

[0053] References herein to "one embodiment," "embodiment," or "one or more embodiments" mean that a particular feature, structure, or characteristic described in conjunction with the embodiment is included in at least one embodiment of the present invention. In addition, please note that examples of the term "in one embodiment" do not necessarily all refer to the same embodiment.

[0054] In the description provided herein, a large number of specific details are described. However, it is understood that embodiments of the present invention can be practiced without these specific details. In some instances, well-known methods, structures and techniques are not shown in detail so as not to obscure the understanding of this description.

[0055] It should be noted that the above embodiments illustrate the present invention rather than limit it, and that those skilled in the art may devise alternative embodiments without departing from the scope of the appended claims. In the claims, any reference symbol between brackets shall not be construed as a limitation on the claims. The word "comprising" does not exclude the presence of elements or steps not listed in the claims. The word "one" or "an" preceding an element does not exclude the presence of a plurality of such elements. The present invention may be implemented by means of hardware comprising a number of different elements and by means of a suitably programmed computer. In a unit claim enumerating a number of devices, several of these devices may be embodied by the same hardware item. The use of the words first, second, and third, etc., does not indicate any order. These words may be interpreted as names.

[0056] In addition, it should be noted that the language used in this specification is selected primarily for readability and teaching purposes, rather than for explaining or limiting the subject matter of the present invention. Therefore, many modifications and variations will be apparent to those of ordinary skill in the art without departing from the scope and spirit of the appended claims. The disclosure of the present invention is illustrative rather than restrictive with respect to the scope of the present invention, and the scope of the present invention is defined by the appended claims.

Claims

1. A ticket recommendation method based on artificial intelligence, characterized in that: The method comprises the following steps: obtaining ticket purchase big data of multiple types of users, constructing a first training data set based on the ticket purchase big data, and using the first training data set to train a first ticket recommendation model; obtaining first ticket purchase history data of a target user, analyzing the first ticket purchase history data using the first ticket recommendation model, and predicting a plurality of first ticket recommendation information; Acquire the second ticket purchase history data of the target user in a recent first time period, use the second ticket recommendation model to analyze the second ticket purchase history data, and predict a plurality of second ticket recommendation information; wherein the second ticket recommendation model is trained based on a second training data set, and the second training data set is constructed based on the first ticket purchase history data; integrate the first ticket recommendation information and the second ticket recommendation information to obtain third ticket recommendation information, and push the third ticket recommendation information to the target user.

2. The ticket recommendation method based on artificial intelligence according to claim 1, characterized in that: A first training data set is constructed based on the ticket purchase big data, and a first ticket recommendation model is trained using the first training data set, including: classifying the acquired ticket purchase big data according to user types, and constructing multiple first training data sets based on part of the classified ticket purchase big data; screening multiple collaborative nodes corresponding to the number of classifications of the ticket purchase big data in the blockchain network, sending the first ticket recommendation model and the corresponding first training data set to the corresponding collaborative nodes respectively, and each collaborative node uses the received first training data set to train the first ticket recommendation model, and when the convergence conditions corresponding to each collaborative node are met, the training of the first ticket recommendation model is stopped, and the key model parameters of the first ticket recommendation model are fed back to the local node; the local node integrates the fedback groups of the key model parameters to obtain target key model parameters, and inputs the target key model parameters into the first ticket recommendation model of the local node; randomly selects part of the training data from each of the first training data sets to form a third training data set, and uses the third training data set to train the first ticket recommendation model with the target key model parameters input until the convergence conditions of the local node are met.

3. The ticket recommendation method based on artificial intelligence according to claim 2, characterized in that: The first ticket recommendation model is built based on a convolutional neural network, and the second ticket recommendation model is built based on a large language model.

4. The ticket recommendation method based on artificial intelligence according to claim 3, characterized in that: The second ticket purchase history data of the target user within a recent first time period is obtained, including: obtaining the activity venue information of the target user within a recent preset time period, performing semantic analysis and semantic clustering on the activity venue information, obtaining a number of activity venue types, and calculating the total number of each activity venue type; determining a duration adjustment coefficient based on the total number of the activity venue types, and using the duration adjustment coefficient to adjust the preset time period to the first time period; obtaining the second ticket purchase history data of the target user within the recent first time period.

5. The ticket recommendation method based on artificial intelligence according to claim 4, characterized in that: The second ticket purchase history data is analyzed using a second ticket recommendation model to predict multiple second ticket recommendation information, including: using a semantic analysis component to perform semantic transformation and clustering on the second ticket purchase history data to obtain simplified third ticket purchase history data, and inputting the third ticket purchase history data into the second ticket recommendation model to obtain predicted multiple second ticket recommendation information.

6. The ticket recommendation method based on artificial intelligence according to claim 5, characterized in that: The first ticket recommendation information and the second ticket recommendation information are integrated to obtain the third ticket recommendation information, including: determining the integration ratio according to the duration adjustment coefficient, determining the first integrated quantity of the second ticket recommendation information according to the total number of preset ticket recommendation information and the integration ratio, and calculating the second integrated quantity of the first ticket recommendation information according to the total number and the first integrated quantity; sorting the second ticket recommendation information from large to small according to the recommendation degree value of each piece of the second ticket recommendation information, and selecting the first integrated quantity of the second ticket recommendation information at the head of the sorting queue; similarly, selecting the second integrated quantity of the first ticket recommendation information from each piece of the first ticket recommendation information; sorting the selected first ticket recommendation information and second ticket recommendation information from large to small according to the recommendation degree value, so as to obtain the third ticket recommendation information.

7. A ticket recommendation system based on artificial intelligence, the system comprising a data processor and a data storage device; characterized in that: The data processor calls and runs the computer program code in the data storage to implement the following steps: obtaining ticket purchase big data of multiple types of users, constructing a first training data set based on the ticket purchase big data, and using the first training data set to train a first ticket recommendation model; obtaining first ticket purchase history data of a target user, analyzing the first ticket purchase history data using the first ticket recommendation model, and predicting a plurality of first ticket recommendation information; Acquire the second ticket purchase history data of the target user in a recent first time period, use the second ticket recommendation model to analyze the second ticket purchase history data, and predict a plurality of second ticket recommendation information; wherein the second ticket recommendation model is trained based on a second training data set, and the second training data set is constructed based on the first ticket purchase history data; integrate the first ticket recommendation information and the second ticket recommendation information to obtain third ticket recommendation information, and push the third ticket recommendation information to the target user.

8. An electronic device, comprising: At least one processor, a memory, and a computer program stored in the memory and executable on the at least one processor, wherein the processor executes the computer program to implement the method according to any one of claims 1 to 6.

9. A computer storage medium storing a computer program, characterized in that: The computer program is executed by a processor to implement the method according to any one of claims 1 to 6.

10. A computer program product, characterized in that: The computer program product includes computer codes, and when the computer codes are executed by a processor of an electronic device, the method according to any one of claims 1 to 6 is implemented.

Images