Cinema peripheral product layout method and system based on ticket buying information atlas assistance

Abstract

The invention belongs to the field of cinema systems, and provides a cinema peripheral product layout method and system based on ticket buying information atlas assistance, and the method comprises the steps: building an atlas based on ticket buying information; commodity nodes are constructed to represent products around the cinema, and when a shopping behavior is generated, the weight of a corresponding path is increased by 1; after the user places a movie ticket, ticket buying information is obtained in real time, a corresponding commodity list is obtained according to the path, when the maximum path weight is not smaller than a preset value, the commodities are sorted according to the path weight, and the sorted commodity recommendation list is displayed to the user; if the maximum path weight is smaller than a preset value, obtaining a movie type according to the movie name, obtaining a movie of the same type which is shown recently according to the movie type, sorting commodities according to the path weight of the movie of the same type, and displaying a sorted commodity recommendation list to the user; and after the movie is mapped, deleting the corresponding movie name node and the associated path thereof in the map.

Description

Technical Field

[0001] The present invention belongs to the field of cinema systems, and more particularly relates to a method and system for laying out cinema peripheral products assisted by a ticket purchase information graph. Background Art

[0002] With the rapid development of the cinema industry, the demand for peripheral products (such as food, beverages, movie derivatives, etc.) during the movie-watching process by audiences is increasing day by day. How to accurately recommend cinema peripheral products, improve the consumption experience of audiences, and at the same time increase the sales conversion rate of cinemas has become an important issue of concern in the industry.

[0003] In traditional personalized recommendation systems, it is usually necessary to collect information such as the gender, age, movie-watching preferences, and historical consumption records of users, and build user portraits based on this to generate personalized recommendation results. However, during the cinema ticket purchase process, users do not need to provide real-name information, and it is difficult for the system to directly obtain the basic attribute data of users. At the same time, the same device or account may be shared by different users, making it difficult to ensure the accuracy of user portraits based on accounts. This situation of information shortage has greatly limited the application of existing recommendation methods in the cinema scenario.

[0004] In addition, machine learning or deep learning models are generally used in the prior art for product recommendation, such as methods like collaborative filtering, matrix factorization, deep neural network (DNN), graph neural network (GNN), etc. These methods usually require a large amount of training data and need to go through complex model training and parameter optimization processes to achieve a high recommendation accuracy. However, in the actual application of cinema peripheral product recommendation, the amount of data is relatively limited, especially in the initial stage of a single cinema or a specific movie release, the problem of data sparsity is serious. In addition, the training and deployment of machine learning models often require high computing resources, increasing the difficulty of system implementation and operation costs, making it difficult for some cinemas to promote and apply in actual business.

[0005] In addition, the release cycle of movies is usually short, and their popularity has a great deal of uncertainty. The release time of the vast majority of movies is only 30 to 60 days, and for some films, the main box office period may even be completed within just a few weeks, followed by a rapid decline in the screening schedule. Since the market performance of movies is affected by various factors such as word-of-mouth, cast, promotional strategies, competing films, etc., it is difficult to accurately predict the popularity of movies. Even if predictions are made before release through historical data or box office prediction models, it is very difficult to precisely grasp the market performance of individual movies. Traditional prediction methods, such as time series analysis, regression models, etc., usually require a long period of data accumulation, and for movies with a short release cycle, these methods are difficult to provide effective prediction results. Some movies may experience a reverse box office decline due to the fermentation of word-of-mouth after release, while some movies may see a rapid decline in box office due to market competition or negative reviews. This unpredictability makes recommendation systems based on historical box office or fixed models difficult to adapt to market changes. Summary of the Invention

[0006] To solve the problems in the prior art, the present invention provides a method for arranging products around cinemas assisted by a ticket purchase information graph, including the following steps:

[0007] Construct a graph based on ticket purchase information. The nodes of the graph include: movie name nodes, time period nodes, number of people nodes, ticket price nodes, and seat area nodes. Among them, the connection relationships between the nodes are as follows: movie name nodes are connected to time period nodes, time period nodes are connected to number of people nodes, number of people nodes are connected to ticket price nodes, and ticket price nodes are connected to seat area nodes;

[0008] Construct product nodes representing products around cinemas. When a shopping behavior occurs, establish connections between movie name nodes, time period nodes, number of people nodes, ticket price nodes, seat area nodes and product nodes. Each time a shopping behavior occurs, the weight of the corresponding path increases by 1;

[0009] After the user places an order for a movie ticket, real-time obtain the movie name, time period, number of people, ticket price, and seat area information. According to the movie name - time period - number of people - ticket price - seat area path, obtain the corresponding product list. When the maximum path weight is not less than the preset value, sort the products according to the path weight and display the sorted product recommendation list to the user;

[0010] If the maximum path weight is less than the preset value, obtain the movie type according to the movie name, and according to the movie type, obtain the most recently released movies of the same type. Sort the products according to the path weight of the movies of the same type and display the sorted product recommendation list to the user;

[0011] After the movie is taken off the screen, delete the corresponding movie name node and its associated paths in the graph.

[0012] The present invention also provides a layout system for theater peripheral products assisted by a ticket purchase information graph, including the following modules:

[0013] A first construction module for constructing a graph based on ticket purchase information. The nodes of the graph include: a movie name node, a time period node, a number of people node, a ticket price node, and a seat area node. Among them, the connection relationships between the nodes are as follows: the movie name node is connected to the time period node, the time period node is connected to the number of people node, the number of people node is connected to the ticket price node, and the ticket price node is connected to the seat area node;

[0014] A second construction module for constructing a commodity node representing theater peripheral products. When a shopping behavior occurs, a connection is established between the movie name node, the time period node, the number of people node, the ticket price node, the seat area node and the commodity node. Each time a shopping behavior occurs, the corresponding path weight is increased by 1;

[0015] A recommendation module for, when a user places an order for a movie ticket, real-time obtaining the movie name, time period, number of people, ticket price, and seat area information, obtaining the corresponding commodity list according to the movie name - time period - number of people - ticket price - seat area path, sorting the commodities according to the path weight when the maximum path weight is not less than a preset value, and presenting the sorted commodity recommendation list to the user; if the maximum path weight is less than the preset value, obtaining the movie type according to the movie name, obtaining the recently released movies of the same type according to the movie type, sorting the commodities according to the path weight of the same type of movie, and presenting the sorted commodity recommendation list to the user;

[0016] A cleaning module for deleting the corresponding movie name node and its associated paths in the graph after the movie is off the screen.

[0017] Further, when there is no corresponding shopping behavior within a set time period, the path weight is exponentially decayed to prevent outdated data from affecting commodity recommendations:

[0018]

[0019] Where represents the weight at time t + 1, represents the weight at time t.

[0020] Further, the generated ticket purchase information graph is visually processed;

[0021] The ticket purchase information graph is presented in a node - edge structure, where:

[0022] The movie name nodes use different colors or icons to distinguish different types of movies;

[0023] The time period nodes are presented in a time axis manner;

[0024] The number of people node, ticket price node, and seat area node are presented in a hierarchical layout;

[0025] The product nodes are marked in size or color according to the purchase frequency and association strength to highlight high-demand products;

[0026] According to the path weights of ticket purchase behavior and shopping behavior, adjust the thickness, transparency, or color of the edges in the graph to reflect different association strengths.

[0027] Furthermore, if the maximum weight of the current movie is less than the preset weight, the following rules are used to adjust the product recommendation weight of the current movie:

[0028] Set the inheritance weight coefficient α to control the weight inheritance ratio and calculate the inheritance path weight:

[0029] W i, ′ j = αW similar,i,j +(1 - α)W current,i,j

[0030] Among them, W i, ′ j is the weight finally used for recommendation ranking;

[0031] W similar,i,j is the product path weight of the most recent movie of the same type;

[0032] W current,i,j is the original product path weight of the current movie.

[0033] Furthermore, after the movie is off the screen, after deleting the corresponding movie name node and its associated paths in the graph, based on the product purchase records of the off-screen movie, extract the high-frequency products associated with the movie and store them in the historical recommendation database.

[0034] The present invention provides a method and system for the layout of products around a cinema assisted by a ticket purchase information graph, which overcomes the following deficiencies of the prior art in the cinema scenario and has the following beneficial effects:

[0035] This method constructs a graph based on ticket purchase information and only uses ticket purchase behavior data such as movie name, time period, number of people, ticket price, and seat area, without the need to obtain sensitive information such as the gender, age, and historical movie viewing records of users, and can achieve accurate recommendation in a non-real-name ticket purchase environment.

[0036] It is applicable to all cinema audiences. Whether it is a first-time ticket purchaser or a long-term movie viewer, effective product recommendations can be generated based on the current ticket purchase behavior.

[0037] Through the path weight mechanism of the ticket - purchase information graph, this method can make recommendations based on the existing ticket - purchase - shopping relationships even when the amount of data is limited. It is applicable to a single cinema or newly released movies. Even with less data, it can infer relatively relevant peripheral products based on the graph structure, avoiding the problem of ineffective recommendations in the case of sparse data by traditional recommendation methods.

[0038] The present invention updates the product recommendation results in real - time based on path weights, enabling the recommendation system to adapt to the impact of different movies, time periods, seat areas, and other factors on consumer behavior. After ticket purchase, the system can immediately calculate the weight of the current ticket - purchase information path and recommend the most relevant products, significantly improving the accuracy of recommendations and user acceptance.

[0039] This method adopts a dynamic weight adjustment strategy based on the graph, without the need to predict movie popularity in advance. Even in the case of newly released movies or niche movies, it can make recommendations based on the historical consumption data of movies of the same type.

[0040] When the path weight of the ticket - purchase - shopping for the current movie is low, it can automatically inherit the consumption pattern of the most recently released movie of the same type, avoiding recommendation failures due to insufficient data and enhancing the generalization ability of the system.

[0041] In summary, through technical means such as ticket - purchase information graph construction, path weight update, dynamic recommendation adjustment, and inheritance of data of movies of the same type, this method solves problems in the prior art such as user information loss, sparse data problems, and difficulty in predicting the movie market caused by non - real - name ticket purchase. It can accurately recommend peripheral products of the cinema under limited data conditions, improving the operating efficiency of the cinema and the consumption experience of the audience. BRIEF DESCRIPTION OF THE DRAWINGS

[0042] In order to more clearly illustrate the technical solutions in the embodiments of the present invention or in the prior art, the following will briefly introduce the drawings required in the embodiments. Obviously, the drawings in the following description are only some embodiments of the present invention. For those of ordinary skill in the art, without creative efforts, other drawings can also be obtained based on these drawings.

[0043] Figure 1 is the flowchart of the method of the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0044] Next, a preferred description of the invention will be made in combination with the drawings and specific embodiments.

[0045] This embodiment solves the above - mentioned problems through the following steps:

[0046] In one embodiment, refer to Figure 1, the present invention provides a method for optimizing the layout of cinema peripheral products assisted by a ticket purchase information graph, which combines the ticket purchase behavior, consumption preferences of moviegoers, and cinema sales data to intelligently optimize the display and layout of cinema peripheral products.

[0047] Specifically, the method includes the following steps:

[0048] Construct a graph based on ticket purchase information. The nodes of the graph include: movie name node, time period node, number of people node, ticket price node, and seat area node. Among them, the connection relationships between the nodes are as follows: the movie name node is connected to the time period node,

[0049] the time period node is connected to the number of people node, the number of people node is connected to the ticket price node, and the ticket price node is connected to the seat area node.

[0050] Construct a graph based on ticket purchase information. The nodes of the graph include:

[0051] Movie name node: Represents the name of the movie involved in the ticket purchase. Each movie is used as an independent node to reflect the ticket purchase and consumption behaviors of audiences for different movies.

[0052] Time period node: Represents the time interval of ticket purchase or movie viewing, such as "weekday daytime", "weekday evening", "weekend afternoon", "weekend evening", etc., to distinguish the influence of different time periods on users' ticket purchase and consumption habits.

[0053] The time period of audience ticket purchase directly affects their consumption patterns. For example:

[0054] Morning shows may attract students and the elderly, and their consumption preferences tend to be low-price food.

[0055] Evening shows may mainly target office workers and couples, and they are more inclined to purchase high-end packages or peripheral products.

[0056] The time period node can analyze the impact of different time periods on the sales of cinema peripheral products and optimize the product recommendation strategies for different shows.

[0057] Number of people node: Represents the number of people watching the movie corresponding to each ticket purchase order, such as single-person viewing, two-person viewing, or multiple-person viewing (such as family viewing).

[0058] The number of people purchasing tickets determines the consumption preferences of users. For example:

[0059] Single-person viewing may be more inclined to purchase simple meals, such as a cup of coffee or a small bucket of popcorn.

[0060] Two-person viewing may purchase couple packages or double servings of drinks.

[0061] Multiple-person or family viewing tends to purchase family packages and even movie derivatives (such as children's toys).

[0062] The number of people node can analyze the characteristics of the audience group in different time periods to formulate targeted product recommendation strategies.

[0063] Ticket price node: Represents the ticket price category purchased by the user, such as "low ticket price (discounted ticket)", "medium ticket price (ordinary 2D ticket)", "high ticket price (IMAX / 4DX ticket)", to measure the relationship between the ticket price and the user's consumption ability.

[0064] The ticket price reflects the user's consumption ability, and the number of tickets purchased also affects the ticket price choice:

[0065] Students or price-sensitive users are more likely to choose low-priced tickets (such as group discount tickets).

[0066] Couples and movie fans may choose ordinary ticket prices or high-end seats (such as IMAX tickets).

[0067] Family viewing may choose medium-priced package tickets or family screening rooms.

[0068] The ticket price node helps to analyze the differences in the consumption of peripheral products between high-ticket-price users and low-ticket-price users, and optimize the product layout.

[0069] Seat area node: Represents the seat area selected by the user when purchasing tickets, such as "front row", "middle row", "rear row" or "VIP seat", etc., to reflect the user's viewing habits and consumption behaviors.

[0070] Seat selection reflects the user's viewing preferences and may also affect their consumption ability:

[0071] Users who choose VIP seats are usually willing to pay extra for high-end food or movie peripheral products.

[0072] Users who choose the front row may be more inclined to low ticket prices and have a higher price sensitivity to products.

[0073] Audience who choose the middle row may pay more attention to the viewing experience and are willing to buy package products.

[0074] The seat area node can optimize the internal product layout of the cinema. For example, high-end food and limited movie peripheral products are preferentially laid out near the IMAX screening hall or VIP area.

[0075] Construct a product node, representing the peripheral products of the cinema. When a shopping behavior occurs, establish a connection between the movie name node, time period node, number of people node, ticket price node, seat area node and product node. Each time a shopping behavior occurs, the corresponding path weight is increased by 1.

[0076] When a ticket purchase behavior occurs, obtain information such as the movie name, time period, number of people, ticket price, and seat area, and establish corresponding movie name nodes, time period nodes, number of people nodes, ticket price nodes, and seat area nodes in the ticket purchase information graph;

[0077] When a shopping behavior occurs, establish a connection relationship between the movie name node, time period node, number of people node, ticket price node, seat area node and the commodity node in the ticket purchase information graph, so that the shopping behavior can be reflected in the graph as the association between the commodity node and the ticket purchase related nodes.

[0078] In the ticket purchase information graph, define the path weight parameter W i,j To represent the purchase association strength between the ticket purchase information node and the commodity node.

[0079] Whenever a viewer makes a commodity purchase behavior after buying a ticket, adjust the weight of the corresponding path in the ticket purchase information graph so that the path weight increases by 1;

[0080] W i,j =W i,j +1

[0081] Further optionally, when no corresponding shopping behavior occurs within a set time period, perform exponential decay on the path weight to prevent outdated data from affecting commodity recommendations:

[0082]

[0083] Where represents the weight at time t + 1, represents the weight at time t.

[0084] The accumulation of the path weight can be used to calculate the popularity of different commodities, optimize the commodity layout and recommendation strategies.

[0085] Through the above method, the cinema can realize intelligent commodity recommendation and layout optimization based on ticket purchase behavior, improve the viewing experience of the audience, and at the same time improve the sales conversion rate of the cinema's peripheral products.

[0086] Furthermore, perform visualization processing on the generated ticket purchase information graph to intuitively display the association relationship between the movie name, time period, number of people, ticket price, seat area and commodities, and provide dynamic interaction functions to support data analysis and optimization decisions.

[0087] Use a node-edge structure to intuitively display the ticket purchase information graph, where:

[0088] The movie name nodes use different colors or icons to distinguish different types of movies;

[0089] The time period nodes are presented in a timeline manner to observe the ticket purchase and consumption trends in different time periods;

[0090] The number of people nodes, ticket price nodes, and seat area nodes are presented in a hierarchical layout to reveal the impact of audience characteristics on shopping behavior;

[0091] The product nodes are marked by size or color according to the purchase frequency and association strength to highlight high-demand products.

[0092] The graph can adopt a force-directed layout, a hierarchical layout, or a circular layout to optimize the visualization effect.

[0093] According to the path weights of ticket purchase behavior and shopping behavior, adjust the thickness, transparency, or color of the edges in the graph to reflect different association strengths:

[0094] Paths with higher weights can be shown with thicker and darker-colored edges to highlight strong correlation relationships;

[0095] Paths with lower weights can be shown with thinner, semi-transparent, or dashed lines to reduce visual interference;

[0096] Dynamically adjust the path color or animation effect to show the real-time impact of ticket purchase behavior on product recommendations.

[0097] Through the above visualization method, theater managers can intuitively understand the relationship between ticket purchase behavior and the consumption of peripheral products, improve the efficiency of data analysis, and thus optimize the recommendation strategy and layout plan of theater peripheral products.

[0098] When the user places an order for a movie ticket, real-time obtain the movie name, time period, number of people, ticket price, and seat area information. According to the movie name - time period - number of people - ticket price - seat area path, obtain the corresponding product list. When the maximum path weight is not less than the preset value, sort the products according to the path weight and display the sorted product recommendation list to the user.

[0099] When the user places an order for a movie ticket, real-time obtain the ticket purchase information and collect the key information when the user purchases the ticket, including:

[0100] Movie name, time period, number of people, ticket price, seat area. Obtain the above ticket purchase information through the theater ticket purchase system, online ticket purchase platform, or theater self-service terminal, and store it in the ticket purchase information graph database in real time.

[0101] In the ticket purchase information graph, according to the movie name - time period - number of people - ticket price - seat area path, retrieve all product lists associated with the ticket purchase information. The product lists include food products, movie derivatives, and souvenirs;

[0102] In the ticket purchase information graph, each commodity node is connected to a ticket purchase information node, and its relevance is characterized by a path weight.

[0103] Calculate the path weights of all commodity nodes in the ticket purchase information graph. The larger the path weight, the higher the frequency of purchase of the commodity under the same ticket purchase information conditions;

[0104] Determine whether the maximum path weight is greater than a preset threshold T:

[0105] If the maximum path weight is greater than or equal to T, sort the commodities in descending order according to the path weight, and display the sorted commodity recommendation list to the user;

[0106] If the maximum path weight is less than T, execute a recommendation strategy based on movies of the same type to improve the rationality and accuracy of the recommendation.

[0107] The sorted commodity recommendation list can be displayed through the following channels:

[0108] Online ticket purchase platform: After the user completes the ticket purchase, a commodity recommendation page pops up to display the sorted commodities;

[0109] Cinema self-service ticket vending machine: During the ticket collection process, display personalized commodity recommendations according to the ticket purchase information;

[0110] Cinema APP / mini-program: Push personalized commodity recommendations to the user after ticket purchase;

[0111] Screen at the entrance of the screening hall: Dynamically display popular commodities in the area at the entrance of the screening hall according to the session information.

[0112] If the user selects to purchase a recommended commodity, in the ticket purchase information graph, the path weight corresponding to the commodity node increases by 1;

[0113] If the user does not purchase the recommended commodity, the path weight remains unchanged, but the recommendation strategy can be adjusted in subsequent analysis to optimize the commodity recommendation effect.

[0114] If the maximum path weight is less than the preset value, obtain the movie type according to the movie name, obtain the recently released movies of the same type according to the movie type, sort the commodities according to the path weights of the movies of the same type, and display the sorted commodity recommendation list to the user.

[0115] If the maximum path weight is less than the preset value, it means that the release time of this movie is too short and there is not enough data support, so further optimize the commodity recommendation based on the movie type.

[0116] Based on the movie name of the current user's ticket purchase, retrieve the movie database or the ticket purchase information graph to obtain the movie type of this movie (such as science fiction, action, comedy, animation, etc.);

[0117] In the ticket purchase information graph, retrieve the recently released movies of this type. The recently released movies can be determined according to the following rules:

[0118] Filter by release time and select the movie closest to the current time;

[0119] Filter by box office or audience rating and select the most popular movie of the same type;

[0120] Filter by ticket purchase data and select the movie with a high purchase record in the ticket purchase information graph.

[0121] Based on the recently released movies of the same type obtained, retrieve the product recommendation path in the ticket purchase information graph and obtain the corresponding path weight.

[0122] Further optionally, adjust the product recommendation weight of the current movie using the following rules:

[0123] Set an inheritance weight coefficient α to control the weight inheritance ratio and calculate the inheritance path weight:

[0124] W i, ′ j = αW similar,i,j +(1 - α)W current,i,j

[0125] where W i, ′ j is the weight finally used for recommendation ranking;

[0126] W similar,i,j is the product path weight of the recently released movies of the same type;

[0127] W current,i,j is the original product path weight of the current movie;

[0128] The inheritance coefficient α can be dynamically adjusted according to historical data.

[0129] Sort the product list in descending order according to the adjusted path weight and display the recommended product list to the user.

[0130] In this step, in the case of less shopping behavior for the current movie, based on the historical purchase data of popular movies of the same type, supplement the recommendation information to improve the reliability and accuracy of the recommendation.

[0131] Inherit the product recommendation weight of movies of the same type to make the recommended products more in line with user preferences and optimize the shopping experience.

[0132] Adopt the path weight of the recently released movies so that the recommendation system can adapt to market trends in real time and avoid the drawbacks of fixed recommendation strategies.

[0133] After the movie is taken off the screen, delete the corresponding movie name node and its associated paths in the graph.

[0134] Monitor the release status of the movie, determine whether the movie has reached the off-screen time or has been removed from the theater schedule; mark the off-screen movie based on the information in the theater management system or movie database.

[0135] In the ticket purchase information graph, locate the movie name node of the movie that has been taken off the screen;

[0136] Remove all paths associated with this movie name node in sequence, including:

[0137] The connection between the movie name node and the time period node;

[0138] The connection between the movie name node and the number of people node;

[0139] The connection between the movie name node and the ticket price node;

[0140] The connection between the movie name node and the seat area node;

[0141] The connection between the movie name node and the commodity node;

[0142] Completely delete the movie name node to release storage resources and improve the graph query efficiency.

[0143] At the same time, based on the commodity purchase records of the off-screen movie, extract the high-frequency commodities associated with the movie and store them in the historical recommendation database; when a movie of the same type is released in the future, give priority to inheriting the commodity recommendation weight of the off-screen movie to optimize the commodity recommendation list of the newly released movie.

[0144] On the other hand, the present invention also provides a theater peripheral product layout system assisted by a ticket purchase information graph, including:

[0145] The first construction module is used to construct a graph based on ticket purchase information. The nodes of the graph include: movie name node, time period node, number of people node, ticket price node, seat area node. Among them, the connection relationships between the nodes are as follows: the movie name node is connected to the time period node, the time period node is connected to the number of people node, the number of people node is connected to the ticket price node, and the ticket price node is connected to the seat area node;

[0146] The second construction module is used to construct a commodity node, representing theater peripheral products. When a shopping behavior occurs, establish the connection between the movie name node, time period node, number of people node, ticket price node, seat area node and the commodity node. Each time a shopping behavior occurs, the corresponding path weight is increased by 1;

[0147] A recommendation module, which is used to obtain the movie name, time period, number of people, ticket price, and seat area information in real time after the user places an order for movie tickets. According to the movie name - time period - number of people - ticket price - seat area path, obtain the corresponding product list. When the maximum path weight is not less than the preset value, sort the products according to the path weight, and display the sorted product recommendation list to the user; if the maximum path weight is less than the preset value, obtain the movie type according to the movie name, and according to the movie type, obtain the recently released movies of the same type, sort the products according to the path weight of the movies of the same type, and display the sorted product recommendation list to the user;

[0148] A cleaning module, which is used to delete the corresponding movie name node and its associated paths in the graph after the movie is taken off the screen.

[0149] For the part of the module structure not specifically defined in the present invention, the content recorded in the prior art shall prevail. The prior art mentioned in the foregoing background art part and the specific embodiment part of the present invention can be used as a part of the present invention to understand the meaning of some technical features or parameters.

Claims

1. A method for arranging cinema peripheral products based on ticket purchase information graph, characterized in that: The method comprises the following steps: A graph is constructed based on the ticket purchase information, wherein the nodes of the graph include: a movie name node, a time period node, a number of people node, a ticket price node, and a seat area node, wherein the connection relationship between the nodes is as follows: the movie name node is connected to the time period node, the time period node is connected to the number of people node, the number of people node is connected to the ticket price node, and the ticket price node is connected to the seat area node; Construct product nodes to represent cinema peripheral products. When a shopping behavior occurs, establish connections between the movie name node, time period node, number of people node, ticket price node, seat area node and product node. Each time a shopping behavior occurs, the corresponding path weight increases by 1. When a user places an order for movie tickets, the movie name, time period, number of people, ticket price, and seat area information are obtained in real time. According to the path of movie name-time period-number of people-ticket price-seat area, the corresponding product list is obtained. When the maximum path weight is not less than the preset value, the products are sorted according to the path weight, and the sorted product recommendation list is displayed to the user; If the maximum path weight is less than the preset value, the movie type is obtained according to the movie name, and the recently released movies of the same type are obtained according to the movie type. The products are sorted according to the path weights of the movies of the same type, and the sorted product recommendation list is displayed to the user; After a movie is taken off the screen, the corresponding movie name node and its associated path are deleted from the graph.

2. The method for arranging cinema peripheral products based on ticket purchase information graph assistance according to claim 1 is characterized in that: When no corresponding shopping behavior occurs within a set time period, the path weight is exponentially decayed to prevent outdated data from affecting product recommendations.

3. The method for arranging cinema peripheral products based on ticket purchase information graph assistance according to claim 1 is characterized in that: Visualize the generated ticket purchase information graph; The node-edge structure is used to display the ticket purchase information map, where: Movie name nodes use different colors or icons to distinguish different types of movies; Time period nodes are presented in a timeline format; The number of people nodes, ticket price nodes, and seat area nodes are displayed in a hierarchical layout; Product nodes are marked with size or color based on purchase frequency and association strength to highlight high-demand products; According to the path weights of ticket purchasing behavior and shopping behavior, the thickness, transparency or color of the edges in the graph are adjusted to reflect different association strengths.

4. The method for arranging cinema peripheral products based on ticket purchase information graph assistance according to claim 1, characterized in that: If the maximum weight of the current movie is less than the preset weight, the following rules are used to adjust the product recommendation weight of the current movie: Set the inheritance weight coefficient α to control the weight inheritance ratio and calculate the inheritance path weight.

5. The method for arranging cinema peripheral products based on ticket purchase information graph assistance according to claim 1 is characterized in that: After a movie is taken off the screen, the corresponding movie name node and its associated path are deleted from the graph, and then the high-frequency products associated with the movie are extracted based on the product purchase records of the movie, and stored in the historical recommendation database.

6. A cinema peripheral product layout system based on ticket purchase information graph, characterized in that: The system includes the following modules: The first construction module is used to construct a graph based on the ticket purchase information, wherein the nodes of the graph include: a movie name node, a time period node, a number of people node, a ticket price node, and a seat area node, wherein the connection relationship between the nodes is as follows: the movie name node is connected to the time period node, the time period node is connected to the number of people node, the number of people node is connected to the ticket price node, and the ticket price node is connected to the seat area node; The second building module is used to build commodity nodes, which represent the peripheral products of the cinema. When a shopping behavior occurs, the connection between the movie name node, time period node, number of people node, ticket price node, seat area node and commodity node is established. Each time a shopping behavior occurs, the corresponding path weight increases by 1; The recommendation module is used to obtain the movie name, time period, number of people, ticket price, and seat area information in real time after the user places an order for movie tickets. According to the path of movie name-time period-number of people-ticket price-seat area, the corresponding product list is obtained. When the maximum path weight is not less than the preset value, the products are sorted according to the path weight, and the sorted product recommendation list is displayed to the user; if the maximum path weight is less than the preset value, the movie type is obtained according to the movie name, and according to the movie type, the recently released movies of the same type are obtained, and the products are sorted according to the path weight of the movies of the same type, and the sorted product recommendation list is displayed to the user; The cleaning module is used to delete the corresponding movie name node and its associated path in the graph after the movie is taken off the screen.

7. The cinema peripheral product layout system based on ticket purchase information graph assistance according to claim 6 is characterized in that: When no corresponding shopping behavior occurs within a set time period, the path weight is exponentially decayed to prevent outdated data from affecting product recommendations.

8. The cinema peripheral product layout system based on ticket purchase information graph assistance according to claim 6 is characterized in that: Visualize the generated ticket purchase information graph; The node-edge structure is used to display the ticket purchase information map, where: Movie name nodes use different colors or icons to distinguish different types of movies; Time period nodes are presented in a timeline format; The number of people nodes, ticket price nodes, and seat area nodes are displayed in a hierarchical layout; Product nodes are marked with size or color based on purchase frequency and association strength to highlight high-demand products; According to the path weights of ticket purchasing behavior and shopping behavior, the thickness, transparency or color of the edges in the graph are adjusted to reflect different association strengths.

9. The cinema peripheral product layout system based on ticket purchase information graph assistance according to claim 6 is characterized in that: If the maximum weight of the current movie is less than the preset weight, the following rules are used to adjust the product recommendation weight of the current movie: Set the inheritance weight coefficient α to control the weight inheritance ratio and calculate the inheritance path weight.

10. The cinema peripheral product layout system based on ticket purchase information graph assistance according to claim 6 is characterized in that: After a movie is taken off the screen, the corresponding movie name node and its associated path are deleted from the graph, and then the high-frequency products associated with the movie are extracted based on the product purchase records of the movie, and stored in the historical recommendation database.

Images