A method for mobile advertising based on crowd tags

Abstract

The present invention discloses a method for delivering mobile advertisements based on crowd labels, comprising the following steps: first, obtaining personnel trajectory data, transportation route and station data, and outputting a mapping relationship between stations and personnel sets, and a mapping relationship between routes and personnel; second, collecting crowd label data within a preset range of the station, and aggregating and calculating bus route crowd label data based on the mapping relationship between bus routes and personnel, as well as the crowd label data of the personnel; subsequently, setting advertisement target population labels and weights; and selecting one or more bus routes from each bus route according to the distribution of the crowd label data, the advertisement target population labels and weights, to generate an advertisement delivery plan.

Description

Technical Field

[0001] The present application belongs to the field of multimedia, and specifically relates to a method for delivering mobile advertisements based on crowd tags. Background Art

[0002] Public transportation is an indispensable mode of urban travel. Every day, hundreds of millions of people across the country choose public transportation for travel. Therefore, mobile advertising using public transportation as a carrier has significant advantages over other advertising media in terms of coverage area and density, and is an important branch of outdoor advertising.

[0003] At present, when advertisers select public transportation vehicle-mounted advertising media for placement, they generally use routes (or the geographic attributes associated with the routes themselves) as the basic unit for placement. For example, they calculate and screen based on indicators such as the passenger flow of the route, the attributes of nearby POI landmarks of the route-associated stations (such as shopping malls, schools, parks, office buildings, etc.), and administrative regions. Typically, patent application number 201711311519.1 discloses a method and device for selecting bus routes for on-vehicle advertising. The method obtains the bus routes and bus stops covered by each bus route within a specified area, determines the administrative division attributes of each bus stop based on the geographic location information of the bus stop, sets the advertising target administrative region, and selects one or more bus routes that match the advertising target administrative division attributes from the bus routes based on the administrative division attributes of each bus stop as the on-vehicle advertising placement route.

[0004] However, it is obvious that advertisers are most concerned about the characteristics of the advertising audience (such as the elderly, young people, white-collar workers, and high-spending people). The existing solutions do not take the characteristics of the people around the bus routes into consideration, resulting in advertisers being unable to directly place advertisements on routes based on the characteristics of the target audience, and thus unable to place advertisements on routes with the highest coverage of the target audience attributes, and unable to guarantee the effectiveness of advertising.

[0005] Although there are currently methods for advertising based on human factors, such as an advertising delivery evaluation and optimization system based on big data disclosed in patent application number 201910403598.1, its implementation method is as follows: the advertising delivery data acquisition module is used to collect data information on advertising delivery, and the data information includes click area information, pageview information, click information, population information and account information of the clicker, and the age information of the clicker is recorded in the account information. The present invention analyzes and judges the effect of advertising delivery through the setting of the advertising delivery evaluation module, combines the results of the analysis and judgment with the data information involved in the analysis, and judges the reasons that affect the effect of advertising delivery, so that people can understand which age group the advertisement attracts, thereby making improvements to the shortcomings of the advertisement. As we all know, this kind of static data collection is very simple to implement, and this kind of static data collection is completely different from the dynamic data collection logic, and is completely unsuitable for the field of dynamic advertising delivery.

[0006] The difference compared with the existing technology is that the present invention proposes a method for obtaining crowd tags. In contrast, the patent application number 202110101918.5 discloses a method and system for delivering IPTV advertising resources according to crowd tags, and the patent application number 201710880519.7 discloses a method and device for delivering content to advertising screens. These patents all disclose some methods for precise information delivery through crowd tags, but without exception, the above-mentioned related technologies cannot cope with the relevant fields involved in this patent. Summary of the Invention

[0007] The technical problem to be solved by the present invention is that bus advertising only considers the geographical attributes of the bus route, and cannot be combined with the characteristics of the target audience of the bus advertisement, resulting in the defect that advertisers cannot maximize the value of advertising. In response to the above-mentioned defects of the prior art, a method for mobile advertising based on crowd labels is provided, which includes the following steps: first, obtaining personnel trajectory data, transportation route and station data, and outputting the mapping relationship between the station and the personnel set, and the mapping relationship between the route and the personnel; second, collecting the crowd label data within the preset range of the station, and based on the mapping relationship between the bus route and the personnel, and the crowd label data of the personnel, aggregating and calculating the bus route crowd label data; then, setting the advertising target population label and weight; finally, selecting one or more bus routes from each bus route according to the distribution of the crowd label data, the advertising target population label and weight, and generating an advertising delivery plan.

[0008] As an improvement to the method for mobile advertising based on crowd labels described in the present invention, the method includes the following steps: first, obtaining personnel trajectory data, transportation route and station data, and outputting the mapping relationship between stations and personnel sets, and the mapping relationship between routes and personnel; second, collecting crowd label data within a preset range of the station, and based on the mapping relationship between bus routes and personnel, as well as the crowd label data of the personnel, aggregating and calculating the bus route crowd label data; then, setting the advertising target population label and weight; finally, selecting one or more bus routes from each bus route based on the distribution of the crowd label data, the advertising target population label and weight, and generating an advertising delivery plan.

[0009] As a further improvement to the method of mobile advertising based on crowd tags described in the present invention, the process of obtaining personnel trajectory data, transportation route and station data is as follows: obtain the total number of all personnel p, output the trajectory data set of all personnel p ; Get all bus stop data of candidate bus routes, remove duplicates according to bus stops, get the total number of bus stops m, and output the latitude and longitude data set of m bus stops ; Get the The bus routes are ,in From 1 to , each line A collection of sites ,in From 1 to , It is The number of stations on the line.

[0010] As a further improvement to the method of delivering mobile advertisements based on crowd tags described in the present invention, a method for obtaining the mapping relationship between sites and personnel sets is as follows: Get the first A set of trajectory points of a person ;pass Get the set of people M that pass through the preset range of the station: Get the mapping of the set of people corresponding to all bus stations .

[0011] As a further improvement to the method of delivering mobile advertisements based on crowd tags described in the present invention, a method for obtaining the mapping relationship between routes and personnel is as follows: Calculate the mapping relationship between bus routes and stops; Calculate the set of people passing a certain station on a certain bus line; Aggregate and calculate the number of stops for each person on each bus line; Filter the number of stations on a certain line that is greater than the preset threshold Get the mapping relationship between bus routes and personnel .

[0012] As a further improvement to the method of mobile advertising based on crowd labels described in the present invention, based on the mapping relationship between bus routes and personnel, and the crowd label data of personnel, the steps of aggregating and calculating the crowd label data of bus routes are as follows: Calculate the crowd label set: by Calculate the number of people with crowd label values; by Calculate the proportion of people with a crowd label value.

[0013] As a further improvement to the method of delivering mobile advertisements based on crowd labels described in the present invention, the steps of setting advertisement target crowd labels and weights are as follows: defining an advertisement target crowd label set and its corresponding weight ,in ;pass Calculate the comprehensive score of the crowd labels of bus routes.

[0014] As a further improvement to the method for mobile advertising based on crowd tags described in the present invention, the index set π is sorted in descending order of the comprehensive scores of the bus routes and meets the following requirements: .

[0015] As a further improvement to the method of mobile advertising based on crowd tags described in the present invention, the line allocation steps are as follows: , so that the sum of the number of screens on the lines in this set does not exceed the advertiser's expected number of screens. , and the routes in this set are the first several routes sorted in descending order of comprehensive scores; bus routes are added in the sorted order until the budget is exhausted: the final route set It is a collection of bus routes that are suitable for the target population.

[0016] Compared with related technologies, the method for mobile advertising delivery based on crowd tags provided by the present invention has the following beneficial effects: the present invention provides a method for mobile advertising delivery based on crowd tags, which can combine the crowd tag data of bus stops and surrounding crowds to mine the crowd tag distribution of each bus route, determine the target population for advertising delivery by understanding the advertiser's delivery intention, and then generate a plan for on-board advertising delivery on bus routes that meets the advertiser's delivery intention, so as to improve the advertising delivery effect. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] Figure 1A schematic diagram of public transportation advertising placement according to target audience labels;

[0018] Figure 2 To generate a schematic diagram of the process of placing on-board advertising on bus routes;

[0019] Figure 3 A schematic diagram of the relationship between bus routes, stops, and personnel;

[0020] Figure 4 Detailed flowchart of the technical solution of the present invention;

[0021] Figure 5 Schematic diagram of the modules of the present invention. DETAILED DESCRIPTION

[0022] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. All other embodiments obtained by ordinary technicians in this field based on the embodiments of the present invention without making any creative efforts shall fall within the scope of protection of the present invention.

[0023] When placing bus advertisements, advertisers need to select n routes from the bus network that have the highest coverage of target audience characteristics (i.e., demographic tags), and then place advertisements on buses on these n routes. Taking bus advertising as an example, the present invention includes the following steps:

[0024] Example 1 Figures 1 to 5 A method for mobile advertising based on crowd labeling is presented. The system consists of three components: a data acquisition module, a route crowd label distribution determination module, and a plan generation module. The data acquisition module includes a route data acquisition unit, a station data acquisition unit, and a personnel trajectory data acquisition unit. The route crowd label distribution determination module includes a station-person relationship calculation unit, a route-person relationship calculation unit, and a route crowd label calculation unit. The plan generation module includes a user input processing unit, a route scoring calculation unit, a route ranking unit, and a route recommendation unit.

[0025] First, obtain the bus routes and bus stops covered by each bus route within the specified area. A bus route consists of multiple stops, each with latitude and longitude information. For example, route A consists of stops 1, 2, and 3.

[0026] Secondly, the crowd label data within the preset range of each bus stop (for example, a certain area is defined as the preset range with the bus stop as the center) is collected, and the crowd label data distribution of the bus line is calculated based on the relationship between the bus line and the bus stop and the crowd label data.

[0027] Among them, the crowd is composed of several people, and people are independent natural individuals, such as Figure 3 In the example, p1, p2, p3, p4, and p5 are represented by the following: A person has multiple population label values, such as male, middle-class consumption, and white-collar occupation.

[0028] Finally, the target audience labels and weights are set, and one or more bus routes are selected from each bus route according to the distribution of the audience labels of each bus route to generate a bus route plan for on-board advertising that meets the target audience labels of the advertisement.

[0029] The specific implementation process is as follows;

[0030] 1. The data acquisition module obtains relevant data:

[0031] 1.1. The station data acquisition unit obtains all bus station data of the candidate bus route and removes duplicates according to the bus station (for example, by removing duplicates by station name) to obtain the total number of bus stations m; the latitude and longitude data (λ, Φ) of each bus station, as shown in the first The latitude and longitude information of each site is , thus, the station data acquisition unit can output the longitude and latitude data set of m bus stations .

[0032] 1.2. The personnel trajectory data acquisition unit obtains the total number of personnel p (all personnel corresponding to the trajectory data collected by the system), the latitude and longitude data of each person (λ, Φ), and the latitude and longitude data of each person's stay constitute the trajectory points, as shown in the first The number of trajectory points of a person is , No. Personnel in the The longitude and latitude information of each trajectory point is , thus, the personnel trajectory data acquisition unit can output the trajectory data set of all personnel .

[0033] 1.3. Line data acquisition unit obtains the The bus routes are ,in From 1 to , each line A collection of sites ,in From 1 to , It is The number of stations on the line.

[0034] 2. The station-personnel relationship calculation unit calculates the trajectories of all people passing through the preset range of each bus station, and aggregates the trajectories into people, thereby obtaining the mapping relationship between bus stations and people.

[0035] 2.1. Calculate the personnel trajectory set T:

[0036] Randomly select The trajectory points of each person are calculated. A set of trajectory points of a person , .in, For the Personnel in the The latitude and longitude information of each trajectory point.

[0037] 2.2. Calculate the set of people M that pass through the preset range of the site:

[0038] ,in For two longitude and latitude points (i.e., Personnel in the The latitude and longitude of each track point , and The latitude and longitude of each site ), For the bus stops, For site The preset range ( The value can be set to a global unified value or configured according to the actual situation of the site).

[0039] 2.3. Mapping of the corresponding set of people for all bus stops:

[0040] ,in, Represents the set of all people passing through the station. This formula represents each bus station and the mapping relationship between the set of people passing through its preset range.

[0041] 3. Based on the mapping relationship between bus stops and personnel, and the mapping relationship between bus routes and stops, the route-personnel relationship calculation unit aggregates and calculates the number of stops for each person on each bus route. According to the preset stop number threshold, the personnel with a stop number greater than the preset stop number threshold on each bus route are screened out, thereby obtaining the mapping relationship between bus routes and personnel.

[0042] 3.1. Calculate the mapping relationship between bus routes and stops:

[0043] ,in Indicates the bus routes, including From 1 to , each line A collection of sites ,in From 1 to , It is The number of stations on the line.

[0044] 3.2. Calculate the set of people passing a certain station on a certain bus line:

[0045] , for bus routes and its site , Passing through the site A collection of people.

[0046] 3.3. Aggregate and calculate the number of stops for each person on each bus route:

[0047] ,in, Indicates the definition Personnel The first The number of stations on the line, is the indicator function (when exist The value is 1 when it is in, otherwise it is 0).

[0048] 3.4. Filtering the number of stations on a certain route that is greater than the preset threshold A collection of people:

[0049] ,in The preset site number threshold.

[0050] 3.5. Get the final mapping relationship between bus routes and personnel:

[0051] .

[0052] 4. Based on the mapping relationship between bus routes and personnel, as well as the personnel's crowd label data, the route crowd label data, the number of people and the proportion of people with each crowd label value are aggregated and calculated through the route crowd label calculation unit.

[0053] 4.1. Calculate the crowd label set:

[0054] ,in is the total number of crowd labels, is the set of all possible crowd labels. Associated crowd tags yes A value in .

[0055] 4.2. Calculate the number of people in the crowd label value:

[0056] , For the Label of each group on the route The number of personnel, including is the indicator function (when When , the value is 1, otherwise the value is 0).

[0057] 4.3. Calculate the proportion of people with crowd label values:

[0058] ,in, Indicates the Label of each group on the route The proportion of personnel, including Indicates the The total number of people on the route.

[0059] 5. The target population labels of the advertisement and the weights of each population label are set through the route scoring calculation unit. The comprehensive population label score of each bus route is calculated based on the proportion of people corresponding to each population label value and the label weight.

[0060] 5.1. Define the advertising target audience label set and its corresponding weight ,in .

[0061] 5.2. Calculate the comprehensive score of the crowd labels of bus routes:

[0062] ,in, Indicates the The comprehensive score of the crowd labels of the bus routes is the sum of the product of the proportion of each label and its weight.

[0063] 6. The route sorting unit sorts the bus routes in reverse order according to the comprehensive scores of the population tags. Based on the number of screens on each bus route and the advertiser's budget for screens, the top-ranked route set is obtained, which is the bus route set that meets the target population.

[0064] 6.1 Calculate the reverse-order index set π of the bus route comprehensive scores, which meets the following requirements:

[0065] ,in Indicates the An index of bus routes.

[0066] 6.2. Define the number of screens on bus routes and the advertiser's budget for screens:

[0067] Indicates the Number of vehicle screens on bus routes, Indicates the advertiser's budget for the number of screens to be served.

[0068] 7. The route recommendation unit calculates a set of bus routes that are suitable for the target population.

[0069] We need to find a set of lines , so that the sum of the number of screens on the lines and vehicles in this set does not exceed the advertiser's budget for the number of screens to be delivered , and the routes in this set are the first several routes sorted in descending order of comprehensive scores.

[0070] definition: Represents a set of bus routes that meet the conditions. Indicates the cumulative number of vehicle screens.

[0071] initialization: , .

[0072] Traverse in sorted order and try to Accumulated to ,if , then Add to bus route collection and will Accumulated to , where i ranges from 1 to q, Indicates the The index of the bus route, For the The number of vehicle screens for the bus route.

[0073] The final line set It is a collection of bus routes that are suitable for the target population.

[0074] Although the embodiments of the present application have been shown and described, those skilled in the art will appreciate that various changes, modifications, substitutions, and variations may be made to the embodiments without departing from the principles and intent of the present application, and that the scope of the present application is defined by the claims and their equivalents.

Claims

1. A method for delivering mobile advertisements based on crowd tags, characterized in that: The method includes the following steps: first, obtaining personnel trajectory data, transportation route and station data, and outputting the mapping relationship between stations and personnel sets, and the mapping relationship between routes and personnel; Secondly, the crowd label data within the preset range of the station is collected, and based on the mapping relationship between bus routes and people, as well as the crowd label data of people, the bus route crowd label data is aggregated and calculated; Then, set the advertising target audience labels and weights; Finally, one or more bus routes are selected from each bus route based on the distribution of crowd label data, advertising target crowd labels, and weights to generate an advertising delivery plan. The process of obtaining personnel trajectory data, transportation route and station data is as follows: Get the total number of all personnel p, The number of trajectory points of a person is , No. Personnel in the The longitude and latitude information of each trajectory point is , output the trajectory data set of all personnel p ； Get all bus stop data of candidate bus routes, The latitude and longitude information of each site is , and after deduplication according to bus stops, the total number of bus stops m is obtained, and the latitude and longitude data set of m bus stops is output ; The method for obtaining the mapping relationship between sites and personnel sets is as follows: Randomly select The trajectory points of each person are calculated. A set of trajectory points of a person , pass Get the first A set of trajectory points of a person ,in, For the Personnel in the The latitude and longitude information of each trajectory point; pass Get the set of people who pass through the preset range of the site, where For the Personnel in the The latitude and longitude of each trajectory point and The latitude and longitude of each site The distance between For the bus stops, For site The preset range; Get the mapping of the corresponding personnel sets for all bus stops ,in, Indicates passing site The collection of all persons; The method for obtaining the mapping relationship between routes and personnel is as follows: pass Calculate the mapping relationship between bus routes and stops, where Indicates the bus routes, including From 1 to , each line A collection of sites ,in From 1 to , It is Number of stations on the line; pass Calculate the set of people passing a certain station on a certain bus line. and its site , Passing through the site A collection of people; pass Aggregate and calculate the number of stops for each person on each bus line, where: Indicates the definition Personnel The first The number of stations on the line, is the indicator function, when exist The value is 1 when it is in the middle, otherwise it is 0; pass Filter the number of stations on a certain line that is greater than the preset threshold A collection of people; Get the mapping relationship between bus routes and personnel .

2. A method for delivering mobile advertisements based on crowd tags according to claim 1, characterized in that: Based on the mapping relationship between bus routes and people, as well as the people's crowd label data, the steps for aggregating and calculating the bus route crowd label data are as follows: pass Calculate the crowd label set, where is the total number of crowd labels, is the set of all possible crowd labels; each person Associated crowd tags yes A value in: pass Calculate the number of people with crowd label values, For the Label of each group on the route The number of personnel, including is the indicator function, when When , the value is 1, otherwise the value is 0; pass Calculate the proportion of people with crowd label values, where: Indicates the Label of each group on the route The proportion of personnel, including Indicates the The total number of people on the route.

3. A method for delivering mobile advertisements based on crowd tags according to claim 2, characterized in that: The steps to set the target audience label and weight are as follows: Define the target audience label set for advertising and its corresponding weight ,in ; pass Calculate the comprehensive score of the crowd labels of the bus routes, where Indicates the The comprehensive score of the crowd labels of the bus routes is the sum of the product of the proportion of each label and its weight; Sort the index set π in descending order of the comprehensive scores of bus routes, and meet the following requirements: ,in Indicates the An index of bus routes with names; The line allocation steps are as follows: By line collection , so that the sum of the number of screens on the lines and vehicles in this set does not exceed the advertiser's budget for the number of screens to be delivered , and the routes in this set are the first several routes sorted in descending order of comprehensive scores. The final route set It is a collection of bus routes that are suitable for the target population.

Images