Vehicle-mounted intelligent networking advertisement publishing and displaying method and system

Abstract

The invention discloses a vehicle-mounted intelligent networking advertisement release and display method and system, and relates to the technical field of advertisement release and display, and the method comprises the steps: enabling an advertiser to carry out the advertisement release through constructing a cloud vehicle-mounted advertisement release platform, and setting an advertisement display scheme according to the vehicle owner information, the vehicle owner and advertisement interaction record and scene data, constructing a vehicle-mounted advertisement feature vector; carrying out group classification on the vehicle owners by adopting a clustering analysis algorithm to obtain a grouping result of a vehicle owner group; constructing an advertisement transaction item set of the vehicle owner group through an association rule learning algorithm, and generating a plurality of frequent transaction items by calculating the support degree of each single transaction in the advertisement transaction item set; and generating a group specific advertisement matching rule set, screening and matching candidate group specific advertisement matching rules to generate a preferred push advertisement, and pushing the advertisement to the vehicle owner. According to the method, intelligent matching and personalized recommendation of the advertisement content are realized through a double-layer optimization mechanism of clustering analysis and association rule learning.

Description

Technical Field

[0001] This invention relates to the field of advertising display technology, and in particular to a method and system for in-vehicle intelligent network advertising publishing and display. Background Technology

[0002] In-vehicle advertising technology has emerged with the rapid development of intelligent connected vehicles and is widely used in commercial scenarios within the connected vehicle ecosystem. In recent years, advertising platforms based on in-vehicle terminals have achieved preliminary personalized recommendations by combining vehicle location, user behavior, and environmental data. Conventional methods typically utilize geolocation technologies (such as GPS) to push location-based ads or perform coarse-grained content matching based on simple user profiles (such as age and gender). Furthermore, some systems have introduced machine learning algorithms, such as collaborative filtering or simple classification models, to analyze users' historical interaction data, thereby optimizing the accuracy of ad delivery. These technologies have achieved certain results in improving ad relevance and user experience.

[0003] However, existing in-vehicle advertising technologies still have significant shortcomings in adaptability and accuracy in dynamic scenarios. First, existing methods largely rely on static or infrequently updated user profiles, making it difficult to capture drivers' dynamic preferences in complex driving environments (such as different road conditions, weather, or driving modes) in real time. This results in insufficient matching between advertising content and actual needs. For example, in high-speed driving or congested traffic, existing systems often fail to adequately consider safety and interaction methods, and the pushed visual ads may distract the driver. Second, existing technologies lack refined segmentation of driver groups and in-depth mining of group characteristics, making it difficult to achieve two-tiered accurate recommendations from group preferences to individual needs. This limits the timeliness and targeting of advertising, especially in the rapidly changing in-vehicle environment, making it difficult to meet the diverse real-time needs of drivers. Summary of the Invention

[0004] In view of the aforementioned existing problems, the present invention is proposed.

[0005] Therefore, this invention provides a method for publishing and displaying intelligent networked advertisements in vehicles to solve the problem of insufficient accuracy of personalized recommendations in in-vehicle advertising systems.

[0006] To solve the above-mentioned technical problems, the present invention provides the following technical solution:

[0007] In a first aspect, the present invention provides a method for publishing and displaying in-vehicle intelligent networked advertisements, comprising,

[0008] By building a cloud-based in-vehicle advertising platform, advertisers can publish ads and set up ad display schemes based on vehicle owner information, vehicle owner-ad interaction records, and scenario data. They can also construct in-vehicle ad feature vectors and normalize them.

[0009] Based on the feature vectors of in-vehicle advertisements, a clustering analysis algorithm is used to classify car owners into groups, and the grouping results of car owner groups are obtained.

[0010] Based on the clustering results and the in-vehicle advertising feature vectors before normalization, an advertising transaction item set for the car owner group is constructed through an association rule learning algorithm. By calculating the support of each individual transaction in the advertising transaction item set, frequent individual transaction items are selected. By iteratively combining and selecting the frequent individual transaction items, multiple frequent transaction items are generated.

[0011] By breaking down multiple frequent transactions, a set of group-specific advertising matching rules is generated. Then, by filtering and matching the candidate group-specific advertising matching rules, preferred push ads are generated and pushed to car owners.

[0012] As a preferred embodiment of the in-vehicle intelligent network advertising publishing and display method of the present invention, the specific steps for constructing the advertising transaction item set of the car owner group are as follows:

[0013] Based on the in-vehicle advertising feature vector before normalization and the ID of the car owner group, the advertising transaction items of the car owners are constructed, and the advertising transaction items of all car owners in the same car owner group are aggregated to generate an advertising transaction item set.

[0014] Based on the advertising transaction itemset, the Apriori algorithm is initialized, and the support of individual transactions in the advertising transaction itemset is calculated. Through filtering, frequent individual transaction items are generated.

[0015] The system joins individual frequent transaction items and iteratively filters them to obtain multiple frequent transaction items, and then calculates the support of multiple frequent transaction items.

[0016] In a preferred embodiment of the in-vehicle intelligent network advertising publishing and display method of the present invention, the specific steps for classifying car owners into groups are as follows:

[0017] Summarize all the feature vectors of in-vehicle advertisements to obtain a set of feature vectors;

[0018] Based on the number of advertising types, a range for the number of clusters is set, and the feature vectors of in-vehicle advertisements are randomly selected as the initial cluster centers according to the number of clusters.

[0019] Initialize the configuration parameters of the association rule learning algorithm, including the maximum number of iterations and the convergence threshold;

[0020] Calculate the sum of squared distances from each in-vehicle advertisement feature vector in the feature vector set to the initial cluster center to obtain the sum of squared distances within the group;

[0021] Based on the sum of squares within a group with different numbers of clusters, an elbow curve is plotted to find the point with the largest decrease rate of the sum of squares within a group, and the number of clusters is determined.

[0022] For each in-vehicle advertisement feature vector, calculate the Euclidean distance with each initial cluster center, and assign the in-vehicle advertisement feature vector to the initial cluster center with the smallest Euclidean distance;

[0023] The cluster centers are updated by averaging the feature vectors of the in-vehicle advertisements assigned to the initial cluster centers until the iteration stops, thus obtaining the grouping results of the car owner group.

[0024] As a preferred embodiment of the in-vehicle intelligent network advertising publishing and display method of the present invention, the specific steps for generating a group-specific advertising matching rule set are as follows:

[0025] The frequent transactions of car owners are arbitrarily broken down into pre-events and post-events to generate a set of advertising matching rules specific to the candidate group.

[0026] Calculate the confidence level for each candidate group based on specific ad matching rules;

[0027] Based on the standard distribution of confidence scores of all candidate group-specific ad matching rules, a confidence score threshold is set. Candidate group-specific ad matching rules with confidence scores greater than the confidence score threshold are retained, while those with confidence scores less than the threshold are removed, thus obtaining the group-specific ad matching rule set for the car owner group.

[0028] In a preferred embodiment of the in-vehicle intelligent network advertising publishing and display method of the present invention, the specific steps for generating preferred push advertisements and pushing the advertisements to vehicle owners are as follows:

[0029] Based on the advertising transaction itemset of the car owner group, calculate the matching score of the specific advertising matching rule for each group;

[0030] For each group, the matching scores of specific ad matching rules are aggregated by ad type to obtain the matching scores of each type of ad. The ad type with the highest matching score is then randomly pushed to all car owners in the car owner group.

[0031] As a preferred embodiment of the in-vehicle intelligent network advertising publishing and display method of the present invention, the construction of the in-vehicle advertising feature vector refers to encoding and splicing the vehicle owner information, vehicle owner and advertising interaction records and scene data to obtain the in-vehicle advertising feature vector.

[0032] As a preferred embodiment of the in-vehicle intelligent network advertising publishing and display method of the present invention, wherein: the vehicle owner information refers to the initial preferences set by the vehicle owner on the cloud-based in-vehicle advertising publishing platform, and the initial preferences refer to the types of advertisements selected by the vehicle owner in the preference settings page, and are represented by priority order;

[0033] The car owner's interaction records with the advertisement include the number of clicks, the number of voice interactions, and the type of advertisement.

[0034] The scenario data includes vehicle location, speed, driving mode, weather conditions and road conditions at the vehicle's location.

[0035] Secondly, the present invention provides an in-vehicle intelligent network advertising publishing and display system, comprising,

[0036] The feature vector module is used to build a cloud-based in-vehicle advertising platform, enabling advertisers to publish advertisements and set advertising display schemes based on vehicle owner information, vehicle owner-ad interaction records, and scene data. It also constructs in-vehicle advertising feature vectors and performs normalization processing.

[0037] The car owner segmentation module is used to classify car owners into groups based on the feature vectors of in-vehicle advertisements using clustering analysis algorithms, and obtain the segmentation results of the car owner groups.

[0038] The association rule module is used to construct an advertising transaction item set for car owners based on the clustering results and the in-vehicle advertising feature vector before normalization, and to filter out frequent single transaction items by calculating the support of each single transaction in the advertising transaction item set. By iteratively combining and filtering the frequent single transaction items, multiple frequent transaction items are generated.

[0039] The ad push module is used to generate a set of group-specific ad matching rules by breaking down multiple frequent transactions, and to generate preferred push ads by filtering and matching candidate group-specific ad matching rules, and then push the ads to car owners.

[0040] Thirdly, the present invention provides a computer device, including a memory and a processor, wherein the memory stores a computer program, wherein: when the computer program is executed by the processor, it implements any step of the in-vehicle intelligent network advertising publishing and display method as described in the first aspect of the present invention.

[0041] Fourthly, the present invention provides a computer-readable storage medium having a computer program stored thereon, wherein: when the computer program is executed by a processor, it implements any step of the in-vehicle intelligent network advertising publishing and display method as described in the first aspect of the present invention.

[0042] The beneficial effects of this invention are as follows: By combining K-means clustering optimized by the elbow rule with an improved Apriori algorithm, this invention first determines the optimal number of clusters by analyzing the rate of change of the sum of squares within a group, establishing a scene-adaptive car owner clustering model. Then, using association rule mining technology, it extracts frequent itemsets containing group IDs, environmental parameters, and advertising types from the clustered advertising transaction item sets, generating matching rules with confidence assessments. This two-layer optimization mechanism enables the system to capture common group characteristics (such as a general preference for restaurant advertisements while driving in commercial areas) and achieve intelligent matching and personalized recommendations of advertising content through antecedent matching. Attached Figure Description

[0043] To more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings used in the following description of the embodiments will be briefly introduced. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0044] Figure 1 This is a flowchart of a method for publishing and displaying intelligent connected vehicle advertisements.

[0045] Figure 2 This is a module diagram of an in-vehicle intelligent network advertising publishing and display system.

[0046] Figure 3 A flowchart for classifying car owners.

[0047] Figure 4 A flowchart for generating association rules and pushing ads. Detailed Implementation

[0048] To make the above-mentioned objects, features and advantages of the present invention more apparent and understandable, the specific embodiments of the present invention will be described in detail below with reference to the accompanying drawings.

[0049] Many specific details are set forth in the following description in order to provide a full understanding of the invention. However, the invention may also be practiced in other ways different from those described herein, and those skilled in the art can make similar extensions without departing from the spirit of the invention. Therefore, the invention is not limited to the specific embodiments disclosed below.

[0050] Secondly, the term "one embodiment" or "embodiment" as used herein refers to a specific feature, structure, or characteristic that may be included in at least one implementation of the present invention. The phrase "in one embodiment" appearing in different places in this specification does not necessarily refer to the same embodiment, nor is it a single or selective embodiment that is mutually exclusive with other embodiments.

[0051] Reference Figures 1-4 As one embodiment of the present invention, this embodiment provides a method for publishing and displaying in-vehicle intelligent network advertisements, including the following steps:

[0052] S1. By building a cloud-based in-vehicle advertising platform, advertisers can publish advertisements and set up advertising display schemes based on vehicle owner information, vehicle owner and advertising interaction records and scene data. They can also construct in-vehicle advertising feature vectors and normalize them.

[0053] Deploy a cloud-based vehicle advertising platform on cloud infrastructure, such as Alibaba Cloud ECS. Specifically, use container orchestration tools, such as Kubernetes, to configure a microservice architecture and set up a distributed server cluster, including front-end services, back-end services, and database services, such as user interaction, data processing logic, and data storage.

[0054] Develop a React-based web portal on a cloud-based in-vehicle advertising platform, set up an advertiser registration page, and allow registered advertisers to access the ad upload interface, supporting the upload of ad content including voice, head-up display dynamic images, and touch interaction.

[0055] Develop a Flutter-based mobile application (supporting iOS and Android) on a cloud-based in-vehicle advertising platform. Set up a car owner registration page. Registered car owners can access the preference settings page and select initial preferences through a drop-down menu. The initial preferences refer to the types of advertisements selected by the car owner on the preference settings page, and are indicated by priority, such as restaurant advertisements first, retail advertisements second, etc., indicating the types of advertisements that the car owner needs or prefers.

[0056] The system communicates with the vehicle's electronic control unit via the in-vehicle bus to obtain scenario data for different car owners, including vehicle location, speed, driving mode, weather conditions, and road conditions at the vehicle's location. Vehicle location is represented by its environment and encoded, e.g., commercial area 1, residential area 2. Speed ​​is categorized as high speed and low speed and encoded, e.g., high speed 1, low speed 2. Driving mode is categorized as parked, cruising, and low speed based on the vehicle's status and encoded, e.g., parked 1, cruising 2, and low speed 3. Weather conditions are categorized as sunny, rainy, and snowy and encoded, e.g., sunny 1, rainy 2, snowy 3. Road conditions are encoded using real-time traffic data, e.g., smooth traffic 1, congested traffic 2. By accessing log files stored in the in-vehicle terminal, the system extracts interaction records between car owners and advertisements, including click counts, voice interaction records, and advertisement types. Based on the advertiser's business scope, the system determines the type of advertisement published and encodes it, e.g., food retail advertisement 1 published by a food company, restaurant advertisement 2, etc.

[0057] Furthermore, based on the priority ranking of ad types in the initial preferences set by the car owner, the linear normalization method is used to calculate the preference score for each type of ad, and the preference scores of all types of ads are summarized to generate the car owner's preference vector.

[0058] It should be noted that vehicle location is key to achieving precise geographic targeting for advertising. For example, when a vehicle approaches a commercial area, it may push advertisements for discounts at nearby shopping malls, while when it enters a residential area, it may recommend advertisements for local services. Vehicle speed determines the safety and format of the advertisement display. For example, when driving at high speeds (>80 km / h), only voice advertisements are allowed to avoid visual interference. Driving mode (parking, cruising, low speed) reflects the vehicle's operating status. For example, users are more likely to interact with the device when parked, while voice interaction is more suitable when cruising. The external environment affects user needs and advertising effectiveness. For example, rain gear advertisements are recommended on rainy days, while voice advertisements are prioritized in congested traffic to reduce interference.

[0059] The vehicle location, speed, driving mode, number of clicks, number of voice interaction records, preference vector, ad type, weather conditions and road conditions are integrated into the in-vehicle ad feature vector;

[0060] The set of in-vehicle advertising feature vectors is normalized. Specifically, the Min-Max normalization method is used to map the numerical range of each component of the in-vehicle advertising feature vector to between 0 and 1, so as to eliminate the influence of different component dimensions.

[0061] S2. Based on the feature vectors of in-vehicle advertisements, a clustering analysis algorithm is used to classify car owners into groups, obtaining the grouping results of car owner groups. The specific steps are as follows:

[0062] By analyzing the changing trend of clustering costs under different numbers of clusters, the number of clusters is determined. Specifically, the elbow rule is adopted. By calculating the sum of squares within a cluster, the sum of squared distances from the feature vectors of in-vehicle advertisements to their respective cluster centers is measured. The "elbow" point of the sum of squares within a cluster is found to determine the number of clusters. For example, all feature vectors of in-vehicle advertisements are aggregated to obtain a feature vector set. Based on the number of advertisement types, the range of the number of clusters is set. , This indicates the number of clusters, with basic parameters configured for each cluster size, such as a maximum number of iterations of 150, and random selection from the feature vector set. Using a set of vehicle advertising feature vectors as initial cluster centers, the sum of squared distances from each vehicle advertising feature vector in the feature vector set to the initial cluster centers is calculated to obtain the in-group sum of squares. Based on the in-group sum of squares with different numbers of clusters, an elbow curve is plotted, where the horizontal axis represents the number of clusters and the vertical axis represents the in-group sum of squares. Observing the shape of the elbow curve, the point with the largest decrease rate of the in-group sum of squares is identified as the elbow point. For example... At that time, the rate of decrease in the sum of squares within the group was 9.41%. At that time, the rate of decrease in the sum of squares within the group was 11.1%, while At that time, the rate of decline in the sum of squares within the group decreased to 6.25%, indicating After that, the benefit of the number of clusters decreases, that is... The corresponding sum of squares within the group is the "elbow" point, and the number of clusters is determined to be 6;

[0063] The K-means clustering analysis method is initialized by, for example, setting the number of cluster centers based on the determined number of clusters, i.e., the number of clusters and the number of cluster centers are numerically equal, setting the maximum number of iterations for the K-means clustering analysis method, and setting an early stopping rule (such as stopping iteration when the change in cluster centers is less than 0.005). For each in-vehicle advertising feature vector, the Euclidean distance with each cluster center is calculated, and the in-vehicle advertising feature vector is assigned to the cluster center with the smallest Euclidean distance. The cluster centers are updated by averaging the in-vehicle advertising feature vectors assigned to the cluster centers until iteration stops, thus obtaining the grouping results of the car owner group.

[0064] S3. Based on the clustering results and the in-vehicle advertising feature vector before normalization, an advertising transaction item set for the car owner group is constructed through an association rule learning algorithm. By calculating the support of each individual transaction in the advertising transaction item set, frequent individual transaction items are selected. By iteratively combining and selecting frequent individual transaction items, multiple frequent transaction items are generated.

[0065] Based on the car owner segmentation results and the in-vehicle advertising feature vector before normalization, an advertising transaction itemset is constructed. The specific steps are as follows:

[0066] Extract each car owner's vehicle location, speed, preference vector, ad type, and interaction record from the in-vehicle ad feature vector before normalization. Combine these with the car owner group ID to form the car owner's ad transaction item. The car owner group ID can be presented in an encoded form, such as a shopping hobby group, where the car owner group is encoded as 1. The car owner group's ad transaction item is presented in the form of a six-tuple, where each element in the six-tuple represents a transaction. Summarize the ad transaction items of all car owners within the car owner group to obtain the car owner group's ad transaction item set.

[0067] Based on the advertising transaction itemset, the Apriori algorithm is initialized, and the support of individual transactions in the advertising transaction itemset is calculated. Through filtering, frequent individual transaction items are generated. The specific steps are as follows:

[0068] Based on the statistical distribution of the frequency of each transaction in the advertising transaction item set, for example, based on the mean and standard deviation of the frequency of each transaction in the advertising transaction item set, a support threshold is set using the three-standard-deviation principle as a screening criterion for frequent items of a single transaction, so as to reflect the lower limit of the frequency of a single transaction in the advertising transaction item set.

[0069] The system counts the occurrence frequency of each transaction in the advertising transaction set and calculates the support of each transaction, which is the ratio of the occurrence frequency of each transaction to the total number of transactions. The support of each transaction is compared with the support threshold. If the support of a transaction is greater than the support threshold, it is retained as a single frequent transaction; otherwise, it is removed.

[0070] Based on the Apriori algorithm, we join single frequent transaction items and iteratively filter them to obtain multiple frequent transaction items. The specific steps are as follows:

[0071] The Apriori algorithm is used for join operations to pairwise combine the single frequent transactions of the car owner group to generate two candidate transactions. It should be noted that at least one candidate transaction must contain the ID of the car owner group to ensure the preservation of the group label. At the same time, the join rules require that the two combined single frequent transactions cannot be the same. The support calculation method of each transaction in the same advertising transaction set is used to count and calculate the support of the two candidate transactions and compare it with the support threshold to generate two frequent transactions. This process is repeated to connect and filter until no higher-order frequent transactions can be generated, resulting in multiple frequent transactions, and the support is calculated.

[0072] S4. By splitting multiple frequent transaction items, a set of group-specific advertising matching rules is generated. Then, by filtering and matching the candidate group-specific advertising matching rules, preferred push ads are generated and pushed to car owners.

[0073] The frequent transactions of car owners are arbitrarily broken down into antecedents and consequents to generate a set of ad matching rules specific to the candidate group. Each ad matching rule can be represented as follows: ,in, Indicates the antecedent, To indicate the consequent, it should be noted that the antecedent must include the IDs of the car owner group and a separate transaction. For each candidate group-specific ad matching rule, a confidence score is calculated to assess the rule's reliability, as shown in the following formula:

[0074] ;

[0075] In the formula, Indicates the specific ad matching rules for the candidate group. confidence level For including the antecedent and successor Support for multiple frequent transaction sets, For the preceding Support levels, specifically, for the advertising transaction set of all car owner groups, statistically analyze whether the preceding conditions are met. The number of transactions under all conditions, and the percentage;

[0076] Based on the standard distribution of confidence scores of all candidate group-specific ad matching rules, a confidence score threshold is set. Candidate group-specific ad matching rules with confidence scores greater than the confidence score threshold are retained, while those with confidence scores less than the threshold are removed, thus obtaining the group-specific ad matching rule set for the car owner group.

[0077] Based on the advertising transaction item set of the car owner group, the matching score of the group-specific advertising matching rule is calculated. Specifically, when all transactions in the antecedent of the group-specific advertising matching rule are completely consistent with the transactions in the advertising transaction item set, the matching score is 1. For example, all transactions in the antecedent are "commercial area 1, preference vector 2, shopping enthusiast 1", and the transactions in the advertising transaction item set are "commercial area 1, low speed 2, preference vector 2, food retail advertisement 1, number of clicks, number of voice interaction records, and shopping enthusiast group 1". When any transaction in the antecedent of the group-specific advertising matching rule is inconsistent with any transaction in the advertising transaction item set, the matching score is 0.

[0078] For each group, the matching score of the specific ad matching rule is summarized by ad type to obtain the matching score of each type of ad. The ad type with the highest matching score is selected as the preferred push ad and randomly pushed to all car owners in the car owner group.

[0079] It should be noted that, based on the segmentation results of car owners, the Apriori algorithm is used to generate multiple frequent transaction items exclusive to each car owner group, enabling the advertising content to adapt to the rapid changes in the in-vehicle environment, improving the timeliness and targeting of advertising push. By filtering and matching group-specific advertising matching rules (association rules), a two-layer personalized recommendation from group characteristics to individual preferences is achieved.

[0080] This embodiment also provides an in-vehicle intelligent network advertising publishing and display system, including: a feature vector module, used to build a cloud-based in-vehicle advertising publishing platform, where advertisers publish advertisements, and set advertising display schemes based on vehicle owner information, vehicle owner-ad interaction records, and scene data, and construct in-vehicle advertising feature vectors and normalize them; a vehicle owner grouping module, used to classify vehicle owners into groups based on the in-vehicle advertising feature vectors using a clustering analysis algorithm to obtain the grouping results of the vehicle owner groups; an association rule module, used to construct an advertising transaction item set for the vehicle owner groups based on the grouping results and the in-vehicle advertising feature vectors before normalization using an association rule learning algorithm, and to filter out single frequent transaction items by calculating the support of each single transaction in the advertising transaction item set, and to generate multiple frequent transaction items by iteratively combining and filtering the single frequent transaction items; and an advertising push module, used to generate a set of group-specific advertising matching rules by splitting the multiple frequent transaction items, and to generate preferred push advertisements by filtering and matching the candidate group-specific advertising matching rules, and to push the advertisements to the vehicle owners.

[0081] This embodiment also provides a computer device applicable to the in-vehicle intelligent network advertising publishing and display method, including: a memory and a processor; the memory is used to store computer-executable instructions, and the processor is used to execute the computer-executable instructions to implement the in-vehicle intelligent network advertising publishing and display method proposed in the above embodiment.

[0082] The computer device can be a terminal, comprising a processor, memory, communication interface, display screen, and input devices connected via a system bus. The processor provides computing and control capabilities. The memory includes non-volatile storage media and internal memory. The non-volatile storage media stores the operating system and computer programs. The internal memory provides an environment for the operation of the operating system and computer programs stored in the non-volatile storage media. The communication interface is used for wired or wireless communication with external terminals; wireless communication can be achieved through Wi-Fi, carrier networks, NFC (Near Field Communication), or other technologies. The display screen can be an LCD screen or an e-ink screen. The input devices can be a touch layer covering the display screen, buttons, a trackball, or a touchpad on the computer device's casing, or an external keyboard, touchpad, or mouse.

[0083] This embodiment also provides a storage medium storing a computer program, which, when executed by a processor, implements the method for publishing and displaying in-vehicle intelligent network advertisements as proposed in the above embodiments. The storage medium can be implemented by any type of volatile or non-volatile storage device or a combination thereof, such as Static Random Access Memory (SRAM), Electrically Erasable Programmable Read-Only Memory (EEPROM), Erasable Programmable Read Only Memory (EPROM), Programmable Red-Only Memory (PROM), Read-Only Memory (ROM), magnetic storage, flash memory, magnetic disk, or optical disk.

[0084] In summary, this invention combines elbow-law-optimized K-means clustering with an improved Apriori algorithm. First, it determines the optimal number of clusters by analyzing the rate of change of the sum of squares within a group, establishing a scene-adaptive car owner clustering model. Then, it utilizes association rule mining technology to extract frequent itemsets containing group ID, environmental parameters, and ad type from the clustered ad transaction item sets, generating matching rules with confidence assessments. This two-layer optimization mechanism enables the system to capture common group characteristics (such as a general preference for restaurant ads while driving in commercial areas) and achieve precise individual-level recommendations through antecedent matching.

[0085] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit it. Although the present invention has been described in detail with reference to preferred embodiments, those skilled in the art should understand that modifications or equivalent substitutions can be made to the technical solutions of the present invention without departing from the spirit and scope of the technical solutions of the present invention, and all such modifications or substitutions should be covered within the scope of the claims of the present invention.

Claims

1. A method for publishing and displaying in-vehicle intelligent networked advertisements, characterized in that: include, By building a cloud-based in-vehicle advertising platform, advertisers can publish their ads and set up ad display schemes based on vehicle owner information, vehicle owner-ad interaction records, and scenario data. They can also construct in-vehicle ad feature vectors and perform normalization processing. Based on the feature vectors of in-vehicle advertisements, a clustering analysis algorithm is used to classify car owners into groups, and the grouping results of car owner groups are obtained. Based on the clustering results and the in-vehicle advertising feature vectors before normalization, an advertising transaction item set for the car owner group is constructed through an association rule learning algorithm. By calculating the support of each individual transaction in the advertising transaction item set, frequent individual transaction items are selected, and iterative combination and selection are performed to generate multiple frequent transaction items. By breaking down multiple frequent transactions, a set of group-specific advertising matching rules is generated. Then, by filtering and matching the candidate group-specific advertising matching rules, preferred push ads are generated and pushed to car owners.

2. The in-vehicle intelligent network advertising publishing and display method as described in claim 1, characterized in that: The specific steps for constructing the advertising transaction set for the car owner group are as follows: Based on the in-vehicle advertising feature vector before normalization and the ID of the car owner group, the advertising transaction items of the car owners are constructed, and the advertising transaction items of all car owners in the same car owner group are aggregated to generate an advertising transaction item set. Based on the advertising transaction itemset, the Apriori algorithm is initialized, and the support of individual transactions in the advertising transaction itemset is calculated. Through filtering, frequent individual transaction items are generated. The system joins individual frequent transaction items and iteratively filters them to obtain multiple frequent transaction items, and then calculates the support of multiple frequent transaction items.

3. The in-vehicle intelligent network advertising publishing and display method as described in claim 2, characterized in that: The specific steps for classifying car owners into groups are as follows. Summarize all the feature vectors of in-vehicle advertisements to obtain a set of feature vectors; Based on the number of advertising types, a range for the number of clusters is set, and the feature vectors of in-vehicle advertisements are randomly selected as the initial cluster centers according to the number of clusters. Initialize the configuration parameters for the association rule learning algorithm; Calculate the sum of squared distances from each in-vehicle advertisement feature vector in the feature vector set to the initial cluster center to obtain the sum of squared distances within the group; Based on the sum of squares within a group with different numbers of clusters, an elbow curve is plotted to find the point with the largest decrease rate of the sum of squares within a group, and the number of clusters is determined. For each in-vehicle advertisement feature vector, calculate the Euclidean distance with each initial cluster center, and assign the in-vehicle advertisement feature vector to the initial cluster center with the smallest Euclidean distance; The cluster centers are updated by averaging the feature vectors of the in-vehicle advertisements assigned to the initial cluster centers until the iteration stops, thus obtaining the grouping results of the car owner group.

4. The in-vehicle intelligent network advertising publishing and display method as described in claim 3, characterized in that: The specific steps for generating a set of group-specific ad matching rules are as follows. The frequent transactions of car owners are arbitrarily broken down into pre-events and post-events to generate a set of advertising matching rules specific to the candidate group. Calculate the confidence level for each candidate group based on specific ad matching rules; Based on the standard distribution of confidence scores of all candidate group-specific ad matching rules, a confidence score threshold is set. Candidate group-specific ad matching rules with confidence scores greater than the confidence score threshold are retained, while those with confidence scores less than the threshold are removed, thus obtaining the group-specific ad matching rule set for the car owner group.

5. The in-vehicle intelligent network advertising publishing and display method as described in claim 4, characterized in that: The process of generating preferred push ads and sending them to car owners involves the following steps. Based on the advertising transaction itemset of the car owner group, calculate the matching score of the specific advertising matching rule for each group; For each group, the matching scores of specific ad matching rules are aggregated by ad type to obtain the matching scores of each type of ad. The ad type with the highest matching score is then randomly pushed to all car owners in the car owner group.

6. The in-vehicle intelligent network advertising publishing and display method as described in claim 5, characterized in that: The construction of the in-vehicle advertising feature vector refers to encoding and concatenating vehicle owner information, vehicle owner-ad interaction records, and scene data to obtain the in-vehicle advertising feature vector.

7. The in-vehicle intelligent network advertising publishing and display method as described in claim 6, characterized in that: The vehicle owner information refers to the initial preferences set by the vehicle owner on the cloud-based in-vehicle advertising platform; The initial preference refers to the type of advertisement selected by the car owner in the preference settings page, and is indicated by priority order. The car owner's interaction records with the advertisement include the number of clicks, the number of voice interactions, and the type of advertisement. The scenario data includes vehicle location, speed, driving mode, weather conditions and road conditions at the vehicle's location.

8. A vehicle-mounted intelligent networked advertising publishing and display system, based on the vehicle-mounted intelligent networked advertising publishing and display method according to any one of claims 1 to 7, characterized in that: include, The feature vector module is used to build a cloud-based in-vehicle advertising platform, enabling advertisers to publish advertisements and set advertising display schemes based on vehicle owner information, vehicle owner-ad interaction records, and scene data. It also constructs in-vehicle advertising feature vectors and performs normalization processing. The car owner segmentation module is used to classify car owners into groups based on the feature vectors of in-vehicle advertisements using clustering analysis algorithms, and obtain the segmentation results of the car owner groups. The association rule module is used to construct an advertising transaction item set for car owners based on the clustering results and the in-vehicle advertising feature vector before normalization, and to filter out frequent single transaction items by calculating the support of each single transaction in the advertising transaction item set. By iteratively combining and filtering the frequent single transaction items, multiple frequent transaction items are generated. The ad push module is used to generate a set of group-specific ad matching rules by breaking down multiple frequent transactions, and to generate preferred push ads by filtering and matching candidate group-specific ad matching rules, and then push the ads to car owners.

9. A computer device comprising a memory and a processor, wherein the memory stores a computer program, characterized in that: When the processor executes the computer program, it implements the steps of the in-vehicle intelligent network advertising publishing and display method according to any one of claims 1 to 7.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by the processor, it implements the steps of the in-vehicle intelligent network advertising publishing and display method according to any one of claims 1 to 7.

Images