CN108171534B

CN108171534B - Vehicle-mounted advertisement bus route recommendation method and device

Info

Publication number: CN108171534B
Application number: CN201711310513.2A
Authority: CN
Inventors: 王浩; 杨康; 庞旭林; 张晨
Original assignee: Beijing Qihoo Technology Co Ltd
Current assignee: Beijing Qihoo Technology Co Ltd
Priority date: 2017-12-11
Filing date: 2017-12-11
Publication date: 2021-06-15
Anticipated expiration: 2037-12-11
Also published as: CN108171534A

Abstract

The invention provides a vehicle-mounted advertisement bus route recommendation method and device, wherein the method comprises the steps of obtaining bus routes in a designated area and bus stations covered by the bus routes; collecting data related to buses and mining multidimensional attributes of all bus stations from the data related to the buses; determining an advertisement putting scene according to the advertisement putting intention, matching the advertisement putting scene with the multidimensional attributes of each bus station, and acquiring the matched multidimensional attributes of the bus stations; and selecting at least one bus line from the bus lines according to the matched multidimensional attribute of the bus station, and recommending the bus line as an advertisement delivery bus line of the advertisement delivery scene. The embodiment of the invention deeply excavates the space-time mobility of the bus and passengers by means of an advanced data excavation technology, provides accurate information support for the selection of the advertisement delivery bus route, and pertinently selects a proper advertisement delivery route according to an advertisement delivery scene, so that the advertisement of the bus body can exert the maximum effect.

Description

Vehicle-mounted advertisement bus route recommendation method and device

Technical Field

The invention relates to the technical field of data mining, in particular to a vehicle-mounted advertisement bus route recommendation method and device.

Background

In each large city, a large number of residents can choose a public transport system to go out, for example, more than ten million people in Beijing choose buses to commute every day. Public transport advertisement covers the crowd extensively, the region is big, therefore has more obvious advantage compared with static advertisement, has become outdoor advertisement mainstream form.

One of the most critical contents in the bus body advertisement scheme is to select an advertisement delivery route. However, the public transportation network in the city is complicated, the geographical position distribution of different public transportation lines is different, the functional areas of the surrounding cities are different, and the population of the covered passengers is different, so that the audiences of the advertisements in different public transportation lines and different time periods are different, and the responses to the advertisements in different types are different. The accurate delivery of the bus body advertisements becomes the target of all advertisers, namely, the accurate delivery of the advertisements aiming at target population, target areas and target subjects spreads the most suitable advertisements to the most suitable audiences in the most suitable time and place, thereby exerting the maximum value of the advertisements.

The public transport line selection method has the common fault in the prior art, namely the public transport line is usually selected by taking the large passenger capacity as the only standard, however, the public transport line and the passengers have respective characteristics which have space-time dynamic characteristics, and the characteristics are not captured and analyzed in the prior art, so that the advertisement cannot be spread to the most appropriate audience, and the advertisement putting effect cannot be guaranteed. For example, some public transportation routes cover more business offices, some public transportation routes cover more tourist sites, and the audience of passengers on the two routes and their psychological state while taking the public transportation are different, so that the sensitivity and the reverberation for different types of advertisements are different.

In addition, in cities, particularly in large and medium-sized cities in which bus body advertisements are widely used for delivery, the bus routes are complicated, the manual delivery route selection method is low in efficiency, poor in expandability, scalability and operability, and the advertisement delivery effect cannot be estimated.

In conclusion, the prior art generally does not analyze the characteristics of the bus lines and passengers and the space-time dynamic characteristics of the characteristics, so that the bus body advertisement is difficult to exert the maximum effect.

Disclosure of Invention

In view of the above, the present invention has been made to provide a method and apparatus for recommending an on-board advertising bus route that overcomes or at least partially solves the above-mentioned problems.

According to one aspect of the invention, a vehicle-mounted advertisement bus route recommendation method is provided, which comprises the following steps:

acquiring bus lines in a designated area and bus stations covered by the bus lines;

collecting data related to buses, and mining multidimensional attributes of all bus stations from the data related to the buses;

determining an advertisement putting scene according to the advertisement putting intention, matching the advertisement putting scene with the multidimensional attributes of each bus station, and acquiring the matched multidimensional attributes of the bus stations;

and selecting at least one bus line from the bus lines according to the matched multidimensional attribute of the bus station, and recommending the bus line as an advertisement delivery bus line of the advertisement delivery scene.

Optionally, the bus-related data comprises at least one of: map data, bus station data, bus route data, intelligent bus card data SCT and user interest point POI data.

Optionally, mining the multidimensional attribute of each bus stop from the bus-related data includes:

processing the intelligent bus card data, calculating to obtain the track data of passengers, and calculating the junction of each bus station according to the passing station sequence in the track data of the passengers, wherein the junction is used for measuring the number of the bus stations as transit stations;

acquiring the geographical position information of the bus station according to the bus station data, and calculating the function theme distribution of each bus station based on the geographical position information of the bus station and POI data within a specified distance around the geographical position information of the bus station;

and determining administrative division attribution of each bus station according to the geographical position information and the map data of the bus station.

Optionally, including bus route number, passenger time of getting on or off the bus, passenger number of getting on or off the bus station in the intelligent bus card data, right intelligent bus card data are handled, calculate the trajectory data who obtains the passenger, include:

recording the number of the passenger getting on or off the bus station in the intelligent bus card data as a station charging number;

based on the charging number of the station for passengers to get on and off the bus, the logic number of the station for passengers to get on and off the bus in the intelligent bus card data is calculated according to a preset strategy;

and determining the passenger passing station number sequence according to the logic number of the passenger getting-on and getting-off station, thereby obtaining passenger track data comprising the bus line number, the passenger getting-on and getting-off time and the passenger passing station number sequence.

Optionally, based on the passenger getting-on and getting-off station charging number, the logic number of the passenger getting-on station is calculated according to a preset strategy, including:

sequencing the intelligent bus card data on the same bus according to the passenger boarding time to obtain the sequenced intelligent bus card data;

grouping the sorted intelligent bus card data, and grouping the passengers in the sorted intelligent bus card data into a group with the same charging number of the stations where the passengers get on the bus;

clustering each group of intelligent bus card data by using a clustering algorithm, wherein the obtained clustering number is used as the actual station number corresponding to the station charging number;

calculating the average value of the getting-on time aiming at the intelligent bus card data in each cluster;

and sequencing the clusters according to the average value of the boarding time, and determining the logical number of the boarding stations of the passengers of the intelligent bus card data in each cluster according to the sequencing.

Optionally, calculating a hub degree of each bus stop according to a passing stop sequence in the track data of the passenger includes:

mining a sequence with the length larger than n based on the passenger track data, and recording the sequence as a target sequence, wherein n is a positive integer;

counting the number of passenger tracks in the target sequence, which are the same as the passenger tracks in each sequence, and taking the number as the weight of the corresponding sequence;

merging sequences containing the same station based on the weight of each sequence in the target sequence to obtain a connection mode network;

and calculating the hub degree of each bus station on the bus line according to the connection mode network.

Optionally, calculating the function theme distribution of each bus station based on the geographical location information of the bus station and the POI data within the peripheral designated distance thereof, including:

acquiring all POI data within a specified distance around each bus station based on the geographical position information of the bus station and the geographical position information in the POI data;

extracting description information in all POI data and combining the description information into a document; analyzing the functional theme distribution of each bus station by using a theme distribution model LDA based on the merged documents;

and determining the functional theme to be reserved, and counting the number of POIs which accord with the functional theme to be reserved within the peripheral appointed distance of each bus station, thereby obtaining the final functional theme distribution of each bus station.

Optionally, determining an advertisement placement scenario according to the advertisement placement intention includes:

if the advertisement putting intention does not distinguish advertisement audiences, determining that an advertisement putting scene is a widely spread scene;

if the advertisement putting intention has the functional theme attribute, determining that an advertisement putting scene is a directional functional theme scene;

and if the advertisement putting intention has the administrative division attribute, determining that the advertisement putting scene is a directed administrative region scene.

Optionally, matching the advertisement delivery scene with the multidimensional attributes of each bus station, and obtaining the matched multidimensional attributes of the bus station, including:

if the advertisement putting scene is a broad propagation scene, the multidimensional attribute matched from the multidimensional attributes of the bus stations according to the broad propagation scene is the hub degree of each bus station;

if the advertisement putting scene is a directional function theme scene, the multi-dimensional attributes matched from the multi-dimensional attributes of the bus stations according to the directional function theme scene are the function theme distribution of the bus stations;

and if the advertisement putting scene is a directed administrative region scene, attributing administrative regions of the bus stations according to the multidimensional attributes matched from the multidimensional attributes of the bus stations in the directed administrative region scene.

Optionally, if the advertisement delivery scene is a widely spread scene, selecting at least one bus route from the bus routes according to the matched multidimensional attribute of the bus station, and recommending the bus route as the advertisement delivery bus route of the advertisement delivery scene, including:

forming a candidate line set by bus lines in a designated area, extracting bus stations covered by the bus lines in the candidate line set, and forming a candidate station set;

traversing the passenger track data, extracting passenger track data containing any bus station in the candidate station set, and forming a candidate track set;

calculating the weight of each passenger track in the candidate track set to the bus line in the candidate line set by combining the junction of each bus station;

and calculating the coverage of any bus line in the candidate line set according to the weight of each passenger track in the bus line of the candidate line set, and selecting the bus line with the maximum coverage as an advertisement delivery bus line for recommendation.

Optionally, if the advertisement delivery scene is a theme scene with a directional function, selecting at least one bus route from the bus routes according to the matched multidimensional attribute of the bus station, and recommending the bus route as the advertisement delivery bus route of the advertisement delivery scene, including:

setting an advertisement target function theme;

calculating the target function theme weight of each bus station in the candidate station set according to the set advertisement target function theme and the function theme distribution of the bus stations;

calculating the weight of each passenger track in the candidate track set to the bus line in the candidate line set based on the target function theme weight of each bus station;

Optionally, calculating a target function theme weight of each bus station in the candidate station set according to the set advertisement target function theme and the function theme distribution of the bus stations, including:

determining the function theme distribution of each bus station in the candidate station set;

counting the number of POIs related to the advertisement target function theme from the POI data contained in the function theme distribution of each bus station;

and calculating the target function theme weight of each bus station in the candidate station set according to the counted number of the POI.

Optionally, if the advertisement delivery scene is a targeted administrative region scene, selecting at least one bus route from the bus routes according to the matched multidimensional attribute of the bus station, and recommending the bus route as the advertisement delivery bus route of the advertisement delivery scene, including:

setting an advertisement target administrative region;

calculating the weight of the target administrative region of each bus station in the candidate station set according to the set advertising target administrative region and the administrative division affiliation of the bus station;

calculating the weight of each passenger track in the candidate track set to the bus line in the candidate line set based on the target administrative region weight of each bus station;

Optionally, calculating the coverage of any bus line in the candidate line set according to the weight of each passenger track in the bus line of the candidate line set, including:

traversing all bus routes in the candidate route set, for any bus route, extracting a bus station set covered by any bus route from the candidate station set, and extracting all passenger tracks covered by the bus station set from the candidate track set;

and according to the weight of each passenger track in the bus route of the candidate route set, summing the weights of all the passenger tracks covered by the bus station set in any bus route to obtain the coverage of any bus route.

Optionally, the selecting the bus route with the largest coverage as the advertisement delivery bus route for recommendation includes:

selecting a bus line with the largest coverage from the candidate line set, adding the selected bus line into a preset selected bus line set, and deleting the selected bus line from the candidate line set;

determining a passenger track covered by the selected bus route, adding the passenger track into a preset covered passenger track set, and deleting the passenger track from the candidate track set;

judging whether the number of the bus routes in the selected bus route set is equal to a preset threshold value or not; if not, continuing to select the bus line with the largest coverage from the deleted candidate lines and adding the bus line to the preset selected bus line set until the number of the bus lines in the selected bus line set is equal to a preset threshold value;

and taking the bus lines in the preset selected bus line set as advertisement delivery bus lines for recommendation.

According to another aspect of the present invention, there is also provided a vehicle-mounted advertisement bus route recommendation device, including:

the acquisition module is suitable for acquiring bus lines in a designated area and bus stations covered by the bus lines;

the mining module is suitable for collecting data related to buses and mining the multi-dimensional attributes of all bus stations from the data related to the buses;

the determining module is suitable for determining an advertisement putting scene according to the advertisement putting intention, matching the advertisement putting scene with the multidimensional attributes of each bus station and acquiring the matched multidimensional attributes of the bus stations;

and the recommendation module is suitable for selecting at least one bus line from the bus lines according to the matched multidimensional attribute of the bus station and recommending the bus line as an advertisement putting bus line of the advertisement putting scene.

Optionally, the excavation module is further adapted to:

processing the intelligent bus card data, calculating to obtain the track data of passengers, and calculating the junction of each bus station according to the passing station sequence in the track data of the passengers; the hub degree is used for measuring the number of bus stations as transit stations;

Optionally, the intelligent bus card data comprises bus line numbers, passenger getting-on and getting-off time, and passenger getting-on and getting-off station numbers, the mining module is further adapted to:

Optionally, the excavation module is further adapted to:

Optionally, the excavation module is further adapted to: acquiring all POI data within a specified distance around each bus station based on the geographical position information of the bus station and the geographical position information in the POI data;

Optionally, the determining module is further adapted to:

Optionally, if the advertisement delivery scenario is a widely spread scenario, the recommendation module is further adapted to:

Optionally, if the advertisement delivery scenario is a targeting function topic scenario, the recommendation module is further adapted to:

setting an advertisement target function theme;

Optionally, the recommendation module is further adapted to:

Optionally, if the advertisement delivery scenario is a targeted administrative area scenario, the recommendation module is further adapted to:

setting an advertisement target administrative region;

Optionally, the recommendation module is further adapted to:

According to still another aspect of the present invention, there is also provided an electronic apparatus including:

a processor; and

a memory arranged to store computer executable instructions that, when executed, cause the processor to perform the in-vehicle advertising bus route recommendation method according to the above.

According to still another aspect of the present invention, there is also provided a computer storage medium, wherein the computer-readable storage medium stores one or more programs that, when executed by an electronic device including a plurality of application programs, cause the electronic device to execute the in-vehicle advertising bus route recommendation method according to the above.

In the embodiment of the invention, firstly, bus routes in a designated area and bus stops covered by the bus routes are obtained. Then, data related to the bus is collected and multidimensional attributes of each bus stop are mined from the data related to the bus. And then, determining an advertisement putting scene according to the advertisement putting intention, matching the advertisement putting scene with the multidimensional attributes of each bus station, and acquiring the matched multidimensional attributes of the bus stations. And finally, selecting at least one bus line from the bus lines according to the matched multidimensional attribute of the bus station, and recommending the bus line as an advertisement putting bus line of the advertisement putting scene. Therefore, the embodiment of the invention deeply excavates the space-time mobility of the bus and the passengers and the multi-dimensional attributes of the bus station by collecting the big data about the bus and the bus passengers and by means of the advanced data mining technology, thereby providing accurate information support for the selection of the advertisement delivery bus route. And moreover, the advertisement putting scene is determined by knowing the advertisement putting intention of the advertisement putting person, and the appropriate bus putting line is selected according to the advertisement putting scene and the multidimensional attribute of the bus station to recommend the optimal bus line for the advertisement putting person, so that the appropriate advertisement putting line can be selected in a targeted manner according to the advertisement putting scene, and the advertisement of the bus body can exert the maximum effect.

Furthermore, the embodiment of the invention also effectively solves the problems that in the prior art, the selection standard of the advertisement delivery route is relatively fixed (the bus route is selected by taking the large passenger capacity or the longest route length as the inherent standard), the most suitable advertisement cannot be transmitted to the most suitable audience at the most suitable time and place, and the maximum value of the advertisement cannot be exerted.

The foregoing description is only an overview of the technical solutions of the present invention, and the embodiments of the present invention are described below in order to make the technical means of the present invention more clearly understood and to make the above and other objects, features, and advantages of the present invention more clearly understandable.

The above and other objects, advantages and features of the present invention will become more apparent to those skilled in the art from the following detailed description of specific embodiments thereof, taken in conjunction with the accompanying drawings.

Drawings

Various other advantages and benefits will become apparent to those of ordinary skill in the art upon reading the following detailed description of the preferred embodiments. The drawings are only for purposes of illustrating the preferred embodiments and are not to be construed as limiting the invention. Also, like reference numerals are used to refer to like parts throughout the drawings. In the drawings:

FIG. 1 illustrates a scene diagram of a prior art public transportation advertisement delivery;

FIG. 2 is a flow chart diagram illustrating a method for vehicle-mounted advertising bus route recommendation in accordance with one embodiment of the present invention;

FIG. 3 shows a flow diagram of a method of calculating trajectory data of a passenger according to another embodiment of the invention;

FIG. 4 is a schematic diagram of station numbers of a Beijing 510-way bus according to one embodiment of the invention;

FIG. 5 shows a schematic flow diagram of the temporal-IdeaGraph algorithm, according to one embodiment of the present invention;

FIG. 6 is a schematic structural diagram of a vehicle-mounted advertising bus route recommendation device according to one embodiment of the invention;

FIG. 7 illustrates a block diagram of a computing device for executing a method for on-board advertising bus route recommendation in accordance with the present invention; and

fig. 8 shows a storage unit for holding or carrying program codes for implementing the on-vehicle advertising bus route recommendation method according to the present invention.

Detailed Description

Exemplary embodiments of the present disclosure will be described in more detail below with reference to the accompanying drawings. While exemplary embodiments of the present disclosure are shown in the drawings, it should be understood that the present disclosure may be embodied in various forms and should not be limited to the embodiments set forth herein. Rather, these embodiments are provided so that this disclosure will be thorough and complete, and will fully convey the scope of the disclosure to those skilled in the art.

In order to more clearly describe the embodiment of the invention, firstly, a delivery scene of the public transportation advertisement is described. Fig. 1 shows a scenario in which public transport advertisement delivery is required. The advertiser needs to select k lines from the public transportation network and put the advertisements on the buses with the k lines. In order to solve this problem, an advertiser needs to consider a plurality of factors, such as advertisement subject, a delivery area, a number of covered people, attributes of a bus route (e.g., distribution of points of interest (POI)) around the bus route, whether the bus stop is a hub or a general bus stop, and the like, in different advertisement delivery scenes. For example, the number of transfer persons in the terminal station is far greater than that of ordinary stations, so that the effect of advertisement delivery to the terminal station is good. However, a difficulty with ad placement is how to measure the multidimensional attributes of a site. To complete the task of accurately delivering advertisements, an important problem to be solved urgently is how to search the best advertisement delivery route in the urban public transportation network.

And more particularly to three challenges. The first mass public transport network makes the solution space of the problem of selecting k lines from mass public transport lines to advertise and promote very large. Secondly, the space-time mobility of the passengers prevents the bus route extraction from being performed in isolation, and the comprehensive solution needs to be performed in consideration of the time and space context (for example, on weekdays, a large number of passengers commute between the residence and the workplace, and on weekends, the starting point of the commute becomes a shopping area and a tourist attraction). Thirdly, the attributes of the lines are diversified, including the positions of the bus stops in the whole bus network, the administrative regions where the bus stops are located, the surrounding POI environments and the like, which are all factors influencing the advertising promotion effect. The user interest point data POI is information of all public places in a city, such as schools, parks, restaurants, shopping malls, cinemas, 4S shops, office buildings, public buildings, communities and the like. One record contains information of a place, specifically including the name, address, latitude and longitude coordinates, description of the place, and the like.

In order to solve the technical problem, the embodiment of the invention provides a vehicle-mounted advertisement bus route recommendation method. Fig. 2 is a flow chart illustrating a method for recommending bus routes for vehicle-mounted advertisements according to an embodiment of the present invention. Referring to fig. 2, the method includes at least steps S202 to S208.

Step S202, bus routes in the designated area and bus stops covered by the bus routes are obtained.

In this step, the designated area may be a designated city (e.g., beijing city, shijiazhuang city, etc.), a designated province (e.g., hebei province, shandong province, etc.), a certain administrative area in the designated city (e.g., beijing hai lake area, beijing chaoyang district, etc.), etc., and the designated area is not specifically limited in the present invention.

And step S204, collecting data related to the bus, and mining the multidimensional attribute of each bus station from the data related to the bus.

In this step, the data related to the bus may include multi-source heterogeneous big data such as map data, bus station data, bus route data, SCT (Smart Card Transaction, Smart bus Card data), and POI data of a user interest point. The intelligent bus card data SCT is the record of the card holder taking the bus. One record comprises a record of one-time taking of the card holder, and specifically comprises a bus card ID, a bus card type, record insertion time, a bus line ID, a bus ID, passenger boarding time, passenger disembarking time, a boarding station ID, a disembarking station ID and the like.

Step S206, determining an advertisement putting scene according to the advertisement putting intention, matching the advertisement putting scene with the multidimensional attributes of each bus station, and obtaining the matched multidimensional attributes of the bus stations.

In this step, the advertisement delivery intentions include delivery intentions that do not distinguish the advertisement audience, delivery intentions that tend to functional topics, delivery intentions that tend to administrative divisions, and the like. The advertisement putting scenes comprise broad spreading scenes, directional function theme scenes, directional administrative region scenes and the like.

And S208, selecting at least one bus line from the bus lines according to the matched multidimensional attribute of the bus station, and recommending the bus line as an advertisement putting bus line of the advertisement putting scene.

According to the embodiment of the invention, by collecting big data about buses and bus passengers and by means of an advanced data mining technology, the space-time mobility of the buses and the passengers and the multi-dimensional attributes of the bus stations are deeply mined, so that accurate information support is provided for the selection of the advertisement delivery bus routes. And moreover, the advertisement putting scene is determined by knowing the advertisement putting intention of the advertisement putting person, and the appropriate bus putting line is selected according to the advertisement putting scene and the multidimensional attribute of the bus station to recommend the optimal bus line for the advertisement putting person, so that the appropriate advertisement putting line can be selected in a targeted manner according to the advertisement putting scene, and the advertisement of the bus body can exert the maximum effect.

Furthermore, the embodiment of the invention also effectively solves the problems that in the prior art, the selection standard of the advertisement delivery route is single (only the passenger capacity is used as the only standard to select the bus route), the most suitable advertisement cannot be transmitted to the most suitable audience at the most suitable time and place, and the maximum value of the advertisement cannot be exerted.

Referring to the above-mentioned steps S202 and S204, when the bus lines in the designated area and the bus stops covered by the bus lines are obtained, the bus lines and the bus stops in the designated area may be obtained from the preset map API by using the JavaScript technology.

The multidimensional attributes of the bus station comprise the hub degree of the bus station, the function theme distribution of the bus station, the administrative division attribution of the bus station and the like. The junction degree is used for measuring the number of the bus stations as the transit stations, the more the number of the buses at the stations is, the larger the transfer number is, the larger the junction degree is, and meanwhile, the passenger flow of the bus stations can be measured. The distribution of the function theme refers to an area with certain function characteristics, such as a school district, a scenic spot, a residential district and the like. Administrative division attribution refers to administrative regions, such as the east city region, the west city region, the hai lake region and the like of Beijing city.

When the hub degrees of all bus stations are mined from the related data of the buses, the intelligent bus card data can be processed, the track data of passengers are obtained through calculation, and then the hub degrees of all bus stations are calculated according to the passing station sequence in the track data of the passengers.

When the function theme distribution of each bus station is mined from the related data of the bus, the geographical position information of the bus station is obtained according to the bus station data, and the function theme distribution of each bus station is calculated based on the geographical position information of the bus station and POI data within a peripheral designated distance.

And when the administrative division attribution of each bus station is mined from the related data of the bus, the administrative division attribution of each bus station is determined according to the geographical position information and the map data of the bus station.

The specific processes of mining the hub degree of each bus station, the function theme distribution of the bus station and the administrative division attribution of the bus station are described below.

The excavation process of the hub degrees of each bus stop is now introduced.

As can be seen from the above, in order to calculate the junction degree of each bus stop, in the first step, the trajectory data of the passenger needs to be calculated first, and the calculation of the trajectory data of the passenger is actually a preprocessing process of the data of the intelligent bus card. Specifically, the calculation process of the trajectory data of the passenger refers to fig. 3 and the following steps:

and step S302, acquiring a bus line number, passenger getting-on/off time and a passenger getting-on/off station number from the intelligent bus card data.

And step S304, recording the number of the passenger getting on or off the bus station in the intelligent bus card data as a station charging number.

And S306, based on the charging number of the station for passengers to get on or off the bus, calculating the logic number of the station for passengers to get on or off the bus in the intelligent bus card data according to a preset strategy.

In the prior art, the station number (getting on and getting off) in the intelligent bus card data is the charging number of the station. Therefore, there is a problem that a plurality of stations share the same number. As shown in fig. 4, the numbers of stations of a 510-way bus in beijing are the same, and the numbers of the "subway forest bridge station" and the "beijing teacher garden north gate" in the smart bus card data are both 4. This numbering method causes a problem that the passenger flow and the hub degree of each station cannot be accurately analyzed, and thus the charging number of the station needs to be mapped to an actual logical number. How to map the charging number of the station to the actual logical number will be described in detail later.

And S308, determining a passenger passing station number sequence according to the logic number of the passenger getting-on and getting-off station, thereby obtaining passenger track data comprising a bus line number, passenger getting-on and getting-off time and the passenger passing station number sequence.

For example, taking the Beijing 510 buses as an example, the station number sequences of the passing stations of the passenger A are "double quanberg east", "subway forest bridge station", "Beijing garden Beimen" and "forest bridge junction north" through analysis, the boarding time of the passenger A is 8:00, and the disembarking time is 8: 40. Therefore, the track data of the passenger A can be obtained by taking 510 buses from 'double quanbutong' at 8:00 and getting off at 'forest intersection north' at 8:40 after taking four stations. The bus line number here is the number of the bus, and of course, other unique codes may be used as the corresponding bus line number, which is not specifically limited in the embodiment of the present invention.

In an embodiment of the present invention, the process of mapping the charging number of the station to the actual logic number may be: firstly, sequencing the intelligent bus card data on the same bus according to the passenger boarding time to obtain the sequenced intelligent bus card data.

And secondly, grouping the sorted intelligent bus card data, and grouping the passengers in the sorted intelligent bus card data into a group with the same charging number of the stations where the passengers get on. As in the beijing 510-route bus in fig. 4, "double quanbutong" may be divided into a first group, "subway forest bridge station" and "beijing opera north gate" may be divided into a second group, "south ditch mud river" may be divided into four groups, and the other stations shown may be divided into three groups.

And then clustering each group of intelligent bus card data by using a clustering algorithm, wherein the obtained clustering number is used as the actual station number corresponding to the station charging number. Here, the K-Means algorithm may be used to cluster each set of smart bus card data. Of course, other clustering algorithms may also be adopted, and this is not specifically limited in the embodiment of the present invention. For example, after the intelligent bus card data corresponding to the first group of stations "double quanburg dong" are clustered, the obtained clustering number is 1, and after the intelligent bus card data corresponding to the second group of stations "subway forest bridge station" and "Beijing teacher garden Beimen" are clustered, the obtained clustering number is 2.

And finally, calculating the average value of the getting-on time aiming at the intelligent bus card data in each cluster, sequencing the clusters according to the average value of the getting-on time, and determining the logical serial number of the passenger getting-on station of the intelligent bus card data in each cluster according to the sequencing. For example, the row with the smaller time-to-vehicle average value in each cluster is ranked in front of the row with the larger time-to-vehicle average value in the row with the smaller time-to-vehicle average value in the cluster.

Similarly, the logic number of the passenger getting-off station can be calculated by adopting the above mode, and the details are not repeated here.

And a second step of calculating the junction of each bus station, wherein after the track data of the passengers is obtained through calculation, the junction of each bus station is calculated according to the passing station sequence in the track data of the passengers.

According to the embodiment of the invention, the Temporal-IdeaGraph algorithm can be adopted to mine the hub degree attribute of the bus station, and the algorithm can mine the sequence in the data and the relationship network among all element bodies. Fig. 5 is a schematic flow chart of the temporal-IdeaGraph algorithm, which, referring to fig. 5, may include the following steps:

step one, mining a sequence mode, mining a sequence with the length larger than n (n is a positive integer) based on passenger track data, and marking the sequence as a target sequence. And counting the number of passenger tracks in the target sequence, which are the same as the passenger tracks in each sequence, and taking the number as the weight of the corresponding sequence. In this embodiment, all sequence sets are denoted as P, a certain sequence is denoted as P, and the weight of the sequence P is denoted as w (P);

and step two, combining the sequence modes, combining the sequences containing the same station based on the weight of each sequence in the target sequence, and obtaining a connection mode network.

And step three, hub stations find out, and the hub degree of each bus station on the bus line is calculated according to the connection mode network. The calculation formula of the pivot degree is shown as formula 1-1:

the calculation of each parameter in the above formula 1 is referred to the following formulas 1-2 to 1-5:

wherein in the above formula, s_iIs station i, H(s)_i) As station s_iPivot, Lstart_iIs the starting station s_iSet of sequences of (1), left_iFor stopping station as s_iSet of sequences of l_i→jFor its real station is s_iAnd the stop station is s_jThe sequence of (a).

The mining process of the functional theme distribution of each bus stop is now introduced.

Firstly, all POI data within a specified distance around each bus stop are acquired aiming at each bus stop based on the geographical position information of the bus stop and the geographical position information in the POI data. For example, for each bus stop, all POI data within a distance of 1km around it are acquired. The designated distance may also be other values, which are not specifically limited in this embodiment of the present invention.

Then, extracting description information in all POI data and combining the description information into a document; analyzing the functional theme distribution of each bus stop by using a theme distribution model LDA (latent Dirichlet allocation) based on the merged documents;

and finally, determining the functional theme to be reserved, and counting the number of POIs which accord with the functional theme to be reserved within the peripheral appointed distance of each bus station, thereby obtaining the final functional theme distribution of each bus station. For example, it is possible to remove irrelevant topics, merge similar topics, and assume that only 10 topics are ultimately reserved for residential, travel, home, office, company, automobile, shopping, education, market, and conspirate, based on the general knowledge in the field of advertising promotion. Then, for each reserved topic, counting the number of POIs which are close to the bus stop and accord with the topic, so as to obtain the topic distribution of the bus stop:

wherein, s is a bus station,

the number of POIs for the ith topic of bus stop s, i ∈ 1,2, …, 10.

Now, the excavation process of administrative division affiliation at each bus stop will be described.

Compared with the mining process of the hub degree and the function subject distribution of the bus station, the mining of the administrative division attribution of the bus station is simple, namely the administrative division attribute of the bus station can be determined directly according to the longitude and latitude coordinates of the geographic position of the bus station and by means of map data. The map data can be acquired from a preset map API using JavaScript technology.

Referring to the above document step S206, as described above, the advertisement placement scenes include a broad propagation scene, a targeted function topic scene, a targeted administrative area scene, and the like. How to determine an advertisement putting scene according to the advertisement putting intention of an advertisement putting person is described below.

The method has the advantages that the scene is widely spread, and the bus route scheme can cover the widest passenger track, namely, the advertisement audience is the most. For example, the delivered advertisements are advertisements in vacation areas for traveling, the advertisements do not distinguish audiences, the wider the transmission range, the better the transmission range, and therefore, a wide transmission scene is selected. Therefore, when the advertisement putting intention does not distinguish advertisement audiences, the advertisement putting scene can be determined to be a widely spread scene.

The directional function theme scene is suitable for the delivered advertisements, has obvious theme tendency, and can be propagated aiming at corresponding audiences, thereby greatly improving the effect and the cost performance of the advertisements. For example, a recruitment website advertisement should be placed on a large number of public transportation lines passing through businesses and offices. Therefore, when the advertisement putting intention has the function theme attribute, the advertisement putting scene is determined to be the directional function theme scene.

And the targeted administrative region scene is suitable for the delivered advertisements, has obvious administrative region tendency and needs to be spread only aiming at audiences of relevant regions. For example, the advertisement in the mall in the core urban area should be placed heavily on the public traffic route in the core urban area (not in the suburban area). Thus, when the advertisement placement intention has the administrative division attribute, the advertisement placement scene is determined to be a targeted administrative area scene.

Based on the introduction of the advertisement putting scene, the advertisement putting scene is matched with the multidimensional attribute of each bus station, and the process of the multidimensional attribute of the bus station obtained by matching is introduced.

Specifically, if the advertisement delivery scene is a broad propagation scene, the multidimensional attribute matched from the multidimensional attributes of the bus stations according to the broad propagation scene is the hub degree of each bus station. And if the advertisement putting scene is a directional function theme scene, the matched multidimensional attribute is the function theme distribution of each bus station. And if the advertisement putting scene is a directed administrative region scene, the matched multidimensional attribute is administrative division attribution of each bus station.

Referring to the step S208, the matched bus stops have different multidimensional attributes, and the basis for selecting the bus routes in each bus route is different, so that the recommended bus routes are also different. At least one bus route is selected from the bus routes according to different multidimensional attributes of the bus station, and the process of recommending the bus routes serving as the advertisement putting scenes is introduced respectively.

In an embodiment of the present invention, when the advertisement delivery scene is a widely spread scene and the matched multidimensional attribute is the hub degree of each bus stop, the process of recommending the advertisement delivery bus route is as follows.

Firstly, forming a candidate line set by the bus lines in the designated area, extracting bus stops covered by the bus lines in the candidate line set, and forming the candidate station set.

Secondly, traversing passenger track data, extracting passenger track data containing any bus station in the candidate station set, and forming the candidate track set.

And then, calculating the weight of each passenger track in the candidate track set to the bus line in the candidate line set by combining the junction degree of each bus stop.

And finally, calculating the coverage of any bus line in the candidate line set according to the weight of each passenger track in the bus line of the candidate line set, and selecting the bus line with the maximum coverage as an advertisement delivery bus line for recommendation. In this embodiment, the coverage may be calculated as follows. Firstly, traversing all bus routes in the candidate route set, for any bus route, extracting a bus station set covered by any bus route from the candidate station set, and extracting all passenger tracks covered by the bus station set from the candidate track set. Secondly, according to the weight of each passenger track in the bus line of the candidate line set, summing the weights of all the passenger tracks covered by the bus station set in any bus line to obtain the coverage of any bus line.

In an embodiment of the present invention, in the process of selecting a bus route with the largest coverage as an advertisement delivery bus route for recommendation, first, a bus route with the largest coverage is selected from a candidate route set, the selected bus route is added to a preset selected bus route set (the initial state of the bus route set is empty), and the selected bus route is deleted from the candidate route set. Secondly, determining the passenger track covered by the selected bus route, adding the passenger track into a preset covered passenger track set (the initial state of the passenger track set is empty), and deleting the passenger track from the candidate track set. Then, judging whether the number of the bus routes in the selected bus route set is equal to a preset threshold (for example, the preset threshold is 3 or 5); and if not, continuing to select the bus line with the largest coverage from the deleted candidate lines and adding the bus line to the preset selected bus line set until the number of the bus lines in the selected bus line set is equal to the preset threshold value. And finally, recommending the bus routes in the preset selected bus route set as advertisement delivery bus routes.

In order to more clearly embody the scheme of the present invention, a specific embodiment specifically introduces a process of recommending an advertisement delivery bus route in a widely spread scenario. The designated area of this example is Beijing. The embodiment measures the hub degree of each bus station, the larger the hub degree is, the larger the station passenger flow is, and the more advertisement audiences passing through the station are. The method is named Hub-KRQ and comprises the following specific steps:

step 1-1, setting the selected bus route set as null, and recording as BR_result(ii) a The set of covered passenger trajectories is set to null, denoted Traj_covered；

Step 1-2, taking all bus lines in Beijing as a candidate line set and recording the candidate line set as BR_can；

Step 1-3, extracting BR according to bus route data_canAll covered bus stations form a candidate station set which is marked as S_can；

Step 1-4, traversing passenger track data, if the passenger track sequence contains S_canAnd (4) extracting passenger track data from any station, and forming a candidate track set which is recorded as Traj_can；

Step 1-5, calculating Traj according to equations 2-1 and 2-2_canAll trajectories for BR_canWeight w (traj; br) of all lines in (A):

w(traj；br)＝∑_s∈itsh(s) formula 2-1; is ═ I_ts[traj]∩I_ts[br]Equation 2-2

Wherein, I_rs[br]For bus lines br coverBus station set of covers, I_ts[traj]Is a bus station set covered by the track traj, its is the intersection of the bus line br and the bus line covered by the track traj, and H(s) is the junction degree of the bus station;

step 1-6, traversing BR_canFor any bus route br, at S_canExtracting all bus station sets covered by the bus station set and recording as S_brAt Traj_canMiddle extraction of S_brCovered passenger trajectory, denoted Traj_brTo Traj_brSumming the line weights of all the passenger tracks to obtain the coverage C of the bus line br_brThe following equations 2-3:

wherein, I_rs[br]Bus station set S for br coverage extraction_br，I_st[I_rs[br]]To extract S_brCovered set of passenger trajectories Traj_br；

Step 1-7, selecting BR_canBus route br with maximum coverage_maxAdding it into the selected bus line set BR_resultAnd from the candidate bus route set BR_canRemoving;

step 1-8, extracting BR according to bus route data_canAll covered bus stations form a candidate station set which is marked as S_can；

Step 1-9, mix br_maxCovered passenger trajectory Traj_brmaxFrom the set of candidate trajectories Traj_canRemoving;

step 1-10, repeating steps 1-6 to 1-9 until a bus route set BR is selected_resultThe number of the medium bus lines meets the requirements of advertisement placers.

The sequence of the steps in the above process is not specifically limited in the embodiments of the present invention.

In an embodiment of the present invention, when the advertisement delivery scene is a targeted function topic scene and the matched multidimensional attribute is the function topic distribution of each bus station, the process of recommending the advertisement delivery bus route is as follows.

Secondly, setting an advertisement target function theme, and calculating the target function theme weight of each bus station in the candidate station set according to the set advertisement target function theme and the function theme distribution of the bus stations.

In an optional embodiment of the present invention, when calculating the weight of the target function topic of each bus stop in the candidate station set, it is necessary to determine the function topic distribution of each bus stop in the candidate station set, and count the number of POIs related to the advertisement target function topic from the POI data included in the function topic distribution of each bus stop. And then calculating the target function theme weight of each bus station in the candidate station set according to the counted number of POI.

Then, traversing the passenger track data, extracting the passenger track data containing any bus station in the candidate station set, and forming the candidate track set.

Then, based on the target function theme weight of each bus stop, the weight of each passenger track in the candidate track set to the bus line in the candidate line set is calculated.

And finally, calculating the coverage of any bus line in the candidate line set according to the weight of each passenger track in the bus line of the candidate line set, and selecting the bus line with the maximum coverage as an advertisement delivery bus line for recommendation. The calculation of the coverage can be referred to the above embodiments, and is not described herein.

In addition, the sequence of the steps in the above process is not specifically limited in the embodiments of the present invention.

In order to more clearly embody the scheme of the present invention, a specific embodiment specifically introduces a process of recommending an advertisement delivery bus route in a subject scene with a targeting function. The designated area of this example is Beijing. If the advertisement putting scene is a subject scene with a directional function, the bus route recommendation method considers the subject characteristics of the passing landmarks of the passenger track, is named as Topic-KRQ, and comprises the following specific steps:

step 2-1, setting the target function theme of the advertisement, and recording the theme as a set Traj_targrt；

Step 2-2, setting the selected bus route set as null and recording as BR_result(ii) a The set of covered passenger trajectories is set to null, denoted Traj_covered；

Step 2-3, taking all bus lines in Beijing as a candidate line set and recording the candidate line set as BR_can；

Step 2-4, extracting BR according to bus route data_canAll covered bus stations form a candidate station set which is marked as S_can；

Step 2-5, calculating S according to the following formula 3-1_canTarget function theme weight of each bus station:

wherein, TD_s[t]The number of POIs related to the subject t is the bus station s;

step 2-6, traversing the passenger track data, if the passenger track sequence contains S_canExtracting passenger track candidate data from any bus station, and forming a candidate track set which is recorded as Traj_can；

Step 2-7, calculating Traj according to the following equations 3-2 and 3-3_canAll passenger trajectories to BR_canWeight w (traj; br) of all lines in (A):

w(traj；br)＝∑_s∈itsw_target(s) formula 3-2;

its＝I_ts[traj]∩I_rs[br]formula 3-3;

wherein, I_rs[br]Set of bus stations covered by bus route br, I_ts[traj]Is a bus station set covered by the track traj, and its is the intersection of the bus line br and the bus line covered by the track traj;

step 2-8, traversing BR_canFor all bus routes, at S, for any bus route br_canExtracting all bus station sets covered by the bus station set and recording as S_brAt Traj_canMiddle extraction of S_brCovered passenger trajectory, denoted Traj_brTo Traj_brSumming the line weights of all the tracks to obtain the coverage C of the bus line br_brAs in equations 3-4 below:

wherein I_rs[br]Bus station set S for br coverage extraction_br，I_st[I_rs[br]]To extract S_brCovered set of passenger trajectories Traj_br；

Step 2-9, selecting BR_canBus route br with maximum coverage_maxAdding it into the selected bus line set BR_resultAnd from candidate bus routes and the BR_canRemoving;

step 2-10, extracting BR according to bus route data_canAll covered bus stations are combined into a candidate station set, and S is recorded_can；

Step 2-11, mix br_maxCovered passenger trajectory Traj_brmaxFrom the set of candidate trajectories Traj_canRemoving;

step 2-12, repeating step 2-6 to step 2-11 until a bus route set BR is selected_resultThe number of the medium bus lines meets the requirements of advertisement placers.

In an embodiment of the present invention, when the advertisement delivery scene is a targeted administrative area scene and the matched multidimensional attribute is administrative division affiliation of each bus stop, a process of recommending an advertisement delivery bus route is as follows, which is similar to the process of recommending a bus route when the advertisement delivery scene is a targeted function subject scene.

And secondly, setting an advertisement target administrative region, and calculating the weight of the target administrative region of each bus station in the candidate station set according to the set advertisement target administrative region and the administrative division affiliation of the bus station.

In an optional embodiment of the present invention, when calculating the target administrative region weight of each bus stop in the candidate stop set, the administrative division affiliation of each bus stop in the candidate stop set is determined first. And then judging whether the administrative division attribution of the bus station is the same as the advertising target administrative area or not, if so, the weight of the target administrative area of the bus station is 1, and if not, the weight of the target administrative area of the bus station is 0.

And further, calculating the weight of each passenger track in the candidate track set to the bus line in the candidate line set based on the target administrative region weight of each bus stop.

In order to more clearly embody the scheme of the present invention, a specific embodiment specifically describes a process of recommending an advertisement delivery bus line in a targeted administrative area scene. The designated area of this example is Beijing. If the advertisement putting scene is a targeted administrative region scene, the bus route recommendation method considers the administrative region characteristics of the passenger track passing landmarks, is named as District-KRQ, and comprises the following specific steps:

step 3-1, setting an advertisement target administrative region and recording the region as a set D_Target；

Step 3-2, the selected bus route set is set to be empty and is recorded as RR_result(ii) a The set of covered passenger trajectories is set to null, denoted Traj_covered；

Step 3-3, taking all bus lines in Beijing as a candidate line set and recording as BR_can；

Step 3-4, extracting BR according to bus route data_canAll covered bus stations form a candidate station set which is marked as S_can；

Step 3-5, calculating S_canAnd (3) the weight of the target administrative region of each bus station, if the administrative region to which the bus station belongs is the target administrative region, the weight is 1, otherwise, the weight is 0: is marked as w_Target(s)；

Step 3-6, traversing the passenger track data, if the passenger track sequence contains S_canExtracting passenger track candidate data from any bus station, and forming a candidate track set which is recorded as Traj_can；

Step 3-7, calculating Traj according to the following equations 4-1 and 4-2_canAll passenger trajectories to BR_canWeight w (traj; br) of all lines in (A):

w(traj；br)＝∑_s∈itsW_target(s) formula 4-1;

its＝I_ts[traj]∩I_rs[br]formula 4-2;

step 3-8, traversing BR_canFor all bus routes, at S, for any bus route br_canExtracting all bus station sets covered by the bus station set and recording the bus station sets asS_brAt Traj_canMiddle extraction of S_brCovered passenger trajectory, denoted Traj_brTo Traj_brSumming the line weights of all the tracks to obtain the coverage C of the bus line br_brThe following equation 4-3:

Step 3-9, selecting BR_canBus route br with maximum coverage_maxAdding it into the selected bus line set BR_resultAnd from candidate bus routes and the BR_canRemoving;

step 3-10, extracting BR according to bus route data_canAll covered bus stations form a candidate station set which is marked as S_can；

Step 3-11, mix br_maxCovered passenger trajectory Traj_brmaxFrom the set of candidate trajectories Traj_canRemoving;

step 3-12, repeating 3-6 to 3-11 until a bus route set BR is selected_resultThe number of the medium bus lines meets the requirements of advertisement placers.

Of course, no matter which advertisement putting scene is aimed at, when at least one bus route is selected in each bus route according to the multidimensional attribute of the bus station, the putting requirements of the advertisement putting person need to be considered, such as the putting time range (specific to hours), the number of the bus routes (such as 2-3 routes, 4 routes and the like), the bus route length range (such as 5000 meters, 3000 meters and the like), the bus route passenger flow range and the like, so that at least one bus route is selected in each bus route according to the multidimensional attribute of the bus station by combining the putting requirements, and the advertisement putting bus route serving as the advertisement putting scene is recommended.

In an embodiment of the invention, before selecting at least one bus route according to the scheme and recommending the bus route as the advertisement putting bus route of the advertisement putting scene, the putting effect of the bus body advertisement putting route scheme based on the statistical characteristic attributes according to a preset strategy can be measured, and when the measurement result reaches a preset advertisement putting reference, the recommendation is carried out.

The modes for measuring the delivery effect of the bus body advertisement delivery route scheme comprise the following three modes.

The method comprises the steps of measuring the delivery effect by utilizing a PTS (PTS To See) index, wherein the PTS index is used for averaging the number of preset theme areas passed by each audience along the way, namely, the number of the preset theme areas passed by each advertisement audience along the way when the advertisement audience takes a bus is measured based on the statistical characteristic attribute, and the delivery effect is better when the PTS value is larger.

And secondly, measuring the influence of the functional theme and/or the administrative region generated by the advertisements to be launched in the tracks of the passengers in the preset number based on the statistical characteristic attributes, wherein the larger the influence value is, the better the launching effect is.

In this manner, T _ GRP (Topic ranking Points) is used to represent the influence of the functional Topic, and the T _ GRP is similar to GRP (Topic ranking Points) and is used to evaluate the influence of the functional Topic generated by the advertisement in N passenger tracks, and the calculation formula is T _ GRP is PTS × Coverage × N. Where PTS indicates audience sensitivity to an advertisement, Coverage is advertisement Coverage, and T _ GRP can evaluate the impact of an advertisement on N audiences. The advertisement Coverage ratio Coverage refers to the proportion of advertisement audiences in a target population, namely the proportion of passenger tracks covered by the delivered lines in all passenger tracks, and the calculation formula is as follows:

in the formula, traj is a set of all passenger tracks, traj _ covered is a set of tracks of advertisement audiences covered by the bus line, and num_trajFor all that isNumber of tracks in passenger track set, num_{traj_covered}The number of tracks in the advertisement audience track set covered by the bus line.

In this method, D _ GRP (three Gross Rating Points) is used to represent the influence of the administrative area. D _ GRP is similar to GRP and used to evaluate the impact of the administrative area caused by the advertisement in the N passenger tracks, and the formula is D _ GRP TDC × OTS × N. The TDC index is the proportion of the audience track covering the target administrative region, and the OTS is the average number of times of exposure of the audience to the advertisement.

Measuring the delivery effect by using a TDC (targeted delivery coverage) index, wherein the TDC is the proportion of the delivered bus route covering the trajectory passing through the target administrative area, namely the proportion of the delivered bus route covering the trajectory passing through the target administrative area is measured based on the statistical characteristic attribute, and the delivery effect is better when the TDC value is larger. Of course, the three modes can be combined in any form to measure the delivery effect of the bus body advertisement delivery route scheme.

In an embodiment of the invention, before measuring the delivery effect of the advertisement delivery route scheme of the bus body, the advertisement delivery route can be visually displayed, so that the advertisement delivery person can conveniently measure and adjust the advertisement delivery route. For example, if the advertisement delivery routes are all public transportation routes of a certain city, the advertisement delivery routes can be displayed on a map of the city. The specific display adjustment steps are as follows:

1) and visually displaying the route putting scheme, calling a map through a preset interface, newly building a layer on the map, and performing formatting division on the newly built layer, so that a grid result of the urban map is displayed in a grid mode. Meanwhile, the passenger flow volume of each grid is displayed in a heat map mode. In the newly-built layer, bus routes in the scheme are displayed in a linear graph mode, and grids along the way are filled with obvious marks (such as colors and symbols); by adopting the method, the area covered along the bus route, namely the area for advertisement propagation, can be visually displayed very intuitively;

in this step, when the newly-built map layer on the urban map is gridded, longitude and latitude information of the urban bus station and the bus route can be obtained first, the urban center coordinates and the advertisement delivery area are set, and the newly-built map layer on the map data is subjected to grid segmentation by taking 1km (or other distances) as a grid interval.

The statistical process of the passenger flow of each grid is as follows, in each time interval (each hour) of the whole day, the total number of the passengers getting on and off the bus at each bus station is calculated, and the number of the passengers with repeated bus card IDs is subtracted, so that the passenger flow of each bus station in the time interval is obtained. And calculating to obtain the passenger flow of each map grid all day according to the longitude and latitude information of the bus station and the passenger flow information of each bus station.

Certainly, the passenger flow of each bus line can also be calculated according to the method, for example, in each time period (each hour) of the whole day, the sum of the passenger flows of all bus stops where each bus line passes is calculated, and the passenger flow of each bus line in the time period is obtained by subtracting the repeated passenger number of the bus card ID.

2) The method comprises the following steps of carrying out statistics on bus line characteristics of an advertisement delivery line scheme, and carrying out statistics on all bus line characteristics in the current delivery line scheme, such as the number of bus lines, the length of the bus lines, the passenger flow, administrative divisions or functional subject areas, administrative divisions or functional subject area coverage areas and total area, so as to measure the advertisement delivery effect of the bus line scheme, wherein the measurement process refers to the above embodiment;

3) personalized adjustment is carried out on the advertisement delivery route scheme, if adjustment is needed, a new bus route can be freely added into the delivery scheme or the existing bus route can be removed through the newly-built layer according to the current bus route delivery scheme, a new delivery scheme is formed, and all bus route characteristics of the new delivery scheme are counted according to the step 2);

for example, the number of the advertisement delivery bus lines is 2, the length is 56km, the passenger flow is 27722 people, and the coverage area of the subject area of the travel function is 25km²The shopping function subject coverage area is 17km²Adding bus lines in the advertisement putting linesAnd 3, the circuit covers areas meeting the functional theme requirements, such as a western bill shopping center, a Tiananmen square and the like. Therefore, the adjusted number of the bus lines is 3, the length is 77km, the passenger flow volume is 77300, the coverage area of the subject area of the travel function is 34km2, and the coverage area of the subject area of the shopping function is 25km2, wherein the length, the passenger flow volume and the coverage area of the bus lines refer to the sum of characteristic attribute values of the bus line 1, the bus line 2 and the bus line 3.

4) The method comprises the following steps that manual comparison, analysis and evaluation are carried out on multiple release schemes, the generated new release schemes and the release schemes obtained by selecting at least one bus route are displayed on different newly-built layers respectively, and switching is carried out in different release scheme views, so that different schemes can be compared and analyzed from two aspects of view effect and attribute characteristics, and manual full understanding and comparison and measurement of the release schemes are assisted;

5) and repeating the step 3) and the step 4) until an optimal advertisement putting bus route scheme is formed, and recommending the optimal advertisement putting bus route scheme.

Based on the same inventive concept, the embodiment of the invention also provides a vehicle-mounted advertisement bus route recommendation device. Fig. 6 shows a schematic structural diagram of a vehicle-mounted advertising bus route recommendation device according to an embodiment of the invention. Referring to fig. 6, the vehicle-mounted advertising bus route recommending apparatus 600 at least comprises an obtaining module 610, a mining module 620, a determining module 630 and a recommending module 640.

Now, the functions of the components or devices of the vehicle-mounted advertisement bus route recommendation device 600 and the connection relationship between the components are introduced:

the acquisition module 610 is suitable for acquiring bus routes in a designated area and bus stops covered by the bus routes;

the mining module 620 is coupled with the obtaining module 610 and is suitable for collecting data related to the buses and mining the multi-dimensional attributes of all bus stations from the data related to the buses;

the determining module 630 is coupled with the mining module 620 and is adapted to determine an advertisement putting scene according to the advertisement putting intention, match the advertisement putting scene with the multidimensional attributes of each bus station and obtain the matched multidimensional attributes of the bus stations;

and the recommending module 640 is coupled with the determining module 630 and is adapted to select at least one bus route from the bus routes according to the matched multidimensional attribute of the bus station and recommend the bus route as an advertisement delivery bus route of the advertisement delivery scene.

The bus related data comprises at least one of: map data, bus station data, bus route data, intelligent bus card data SCT and user interest point POI data.

In an embodiment of the present invention, the mining module 620 is further adapted to process the data of the intelligent bus card, calculate trajectory data of the passenger, and calculate a hub degree of each bus station according to a sequence of passing stations in the trajectory data of the passenger; the hub degree is used for measuring the number of bus stations as transit stations. And acquiring the geographical position information of the bus station according to the bus station data, and calculating the function theme distribution of each bus station based on the geographical position information of the bus station and the POI data within the peripheral designated distance. And furthermore, according to the geographic position information and the map data of the bus stations, the administrative division attribution of each bus station is determined.

In an embodiment of the present invention, the intelligent bus card data includes a bus route number, a time for a passenger to get on or off the bus, and a number for a passenger to get on or off the bus station, and the mining module 620 is further adapted to, first, record the number for the passenger to get on or off the bus station in the intelligent bus card data as a station charging number. And then, based on the charging number of the station for passengers to get on or off the bus, calculating the logic number of the station for passengers to get on or off the bus in the intelligent bus card data according to a preset strategy. And finally, determining a passenger passing station number sequence according to the logic number of the passenger getting-on and getting-off station, thereby obtaining passenger track data comprising a bus line number, passenger getting-on and getting-off time and the passenger passing station number sequence.

In an embodiment of the present invention, the mining module 620 is further adapted to, first, sort the smart bus card data on the same bus according to the passenger boarding time to obtain the sorted smart bus card data. And then, grouping the sorted intelligent bus card data, and grouping the passengers in the sorted intelligent bus card data into a group with the same charging number of the stations where the passengers get on. And clustering each group of intelligent bus card data by using a clustering algorithm, taking the obtained cluster number as the actual station number corresponding to the station charging number, and calculating the average getting-on time of the intelligent bus card data in each cluster. And finally, sequencing the clusters according to the average value of the boarding time, and determining the logical number of the boarding stations of the passengers of the intelligent bus card data in each cluster according to the sequencing.

In an embodiment of the present invention, the mining module 620 is further adapted to mine a sequence with a length greater than n based on the passenger trajectory data, and record the sequence as a target sequence, where n is a positive integer, and count the number of passenger trajectories in the target sequence that is the same as each sequence, and use the number as a weight of the corresponding sequence. And combining the sequences containing the same station based on the weight of each sequence in the target sequence to obtain a connection mode network, and calculating the hub degree of each bus station on the bus line according to the connection mode network.

In an embodiment of the present invention, the mining module 620 is further adapted to, based on the geographic location information of the bus stops and the geographic location information in the POI data, acquire, for each bus stop, all POI data within a specified distance around the bus stop, extract description information in all POI data, and merge the description information into one document; based on the merged documents, the functional topic distribution of each bus stop is analyzed using a topic distribution model LDA. Therefore, the function theme needing to be reserved is determined, the number of POIs which accord with the function theme needing to be reserved within the peripheral appointed distance of each bus station is counted, and the final function theme distribution of each bus station is obtained.

In an embodiment of the present invention, the determining module 630 is further adapted to determine that the advertisement delivery scene is a widely spread scene if the advertisement delivery intention does not distinguish advertisement audiences; if the advertisement putting intention has the functional theme attribute, determining that the advertisement putting scene is a directional functional theme scene; and if the advertisement putting intention has the administrative division attribute, determining that the advertisement putting scene is a directed administrative region scene.

In an embodiment of the present invention, the determining module 630 is further adapted to:

In an embodiment of the present invention, if the advertisement delivery scenario is a widely spread scenario, the recommendation module 640 is further adapted to, first, form the bus routes in the designated area into a candidate route set, extract the bus stops covered by the bus routes in the candidate route set, and form the candidate route set. Then, traversing the passenger track data, extracting the passenger track data containing any bus station in the candidate station set, and forming the candidate track set. And then, calculating the weight of each passenger track in the candidate track set to the bus line in the candidate line set by combining the junction degree of each bus stop. And finally, calculating the coverage of any bus line in the candidate line set according to the weight of each passenger track in the bus line of the candidate line set, and selecting the bus line with the maximum coverage as an advertisement delivery bus line for recommendation.

In an embodiment of the present invention, if the advertisement delivery scene is a targeted function topic scene, the recommendation module 640 is further adapted to, first, set an advertisement target function topic, form a candidate route set from the bus routes in the specified area, extract the bus stops covered by the bus routes in the candidate route set, and form a candidate stop set. And calculating the target function theme weight of each bus station in the candidate station set according to the set advertisement target function theme and the function theme distribution of the bus stations. And traversing the passenger track data, extracting the passenger track data comprising any bus station in the candidate station set, and forming the candidate track set. And calculating the weight of each passenger track in the candidate track set to the bus line in the candidate line set based on the target function theme weight of each bus stop. And calculating the coverage of any bus line in the candidate line set according to the weight of each passenger track in the bus line of the candidate line set, and selecting the bus line with the maximum coverage as an advertisement delivery bus line for recommendation.

In an embodiment of the present invention, the recommending module 640 is further adapted to determine the distribution of the function topics of each bus stop in the candidate stop set. And counting the number of POIs related to the function theme of the advertisement target from the POI data contained in the function theme distribution of each bus stop. And calculating the target function theme weight of each bus station in the candidate station set according to the counted number of the POI.

In an embodiment of the present invention, if the advertisement delivery scene is a targeted administrative area scene, the recommendation module 640 is further adapted to set an advertisement target administrative area, form a candidate route set from the bus routes in the specified area, extract the bus stops covered by the bus routes in the candidate route set, and form a candidate stop set. And then, calculating the weight of the target administrative region of each bus station in the candidate station set according to the set advertising target administrative region and the administrative division affiliation of the bus station. And traversing the passenger track data, extracting the passenger track data comprising any bus station in the candidate station set, and forming the candidate track set. And then calculating the weight of each passenger track in the candidate track set to the bus line in the candidate line set based on the target administrative region weight of each bus stop. And finally, calculating the coverage of any bus line in the candidate line set according to the weight of each passenger track in the bus line of the candidate line set, and selecting the bus line with the maximum coverage as an advertisement delivery bus line for recommendation.

In an embodiment of the present invention, the recommending module 640 is further adapted to, first, traverse all bus routes in the candidate route set, for any bus route, extract a bus stop set covered by any bus route from the candidate stop set, and extract all passenger tracks covered by the bus stop set from the candidate track set. And then, according to the weight of each passenger track in the bus line of the candidate line set, summing the weights of all the passenger tracks covered by the bus station set in any bus line to obtain the coverage of any bus line.

In an embodiment of the present invention, the recommending module 640 is further adapted to, first, select a bus route with the largest coverage from the candidate route set, add the selected bus route to the preset selected bus route set, and delete the selected bus route from the candidate route set. And then determining the passenger track covered by the selected bus route, adding the passenger track to a preset covered passenger track set, and deleting the passenger track from the candidate track set. Then, judging whether the number of the bus routes in the selected bus route set is equal to a preset threshold value or not; and if not, continuing to select the bus line with the largest coverage from the deleted candidate lines and adding the bus line to the preset selected bus line set until the number of the bus lines in the selected bus line set is equal to the preset threshold value. And finally, recommending the bus routes in the preset selected bus route set as advertisement delivery bus routes.

According to any one or a combination of the above preferred embodiments, the following advantages can be achieved by the embodiments of the present invention:

firstly, bus routes in a designated area and bus stops covered by the bus routes are obtained. Then, data related to the bus is collected and multidimensional attributes of each bus stop are mined from the data related to the bus. And then, determining an advertisement putting scene according to the advertisement putting intention, matching the advertisement putting scene with the multidimensional attributes of each bus station, and acquiring the matched multidimensional attributes of the bus stations. And finally, selecting at least one bus line from the bus lines according to the matched multidimensional attribute of the bus station, and recommending the bus line as an advertisement putting bus line of the advertisement putting scene. Therefore, the embodiment of the invention deeply excavates the space-time mobility of the bus and the passengers and the multi-dimensional attributes of the bus station by collecting the big data about the bus and the bus passengers and by means of the advanced data mining technology, thereby providing accurate information support for the selection of the advertisement delivery bus route. And moreover, the advertisement putting scene is determined by knowing the advertisement putting intention of the advertisement putting person, and the appropriate bus putting line is selected according to the advertisement putting scene and the multidimensional attribute of the bus station to recommend the optimal bus line for the advertisement putting person, so that the appropriate advertisement putting line can be selected in a targeted manner according to the advertisement putting scene, and the advertisement of the bus body can exert the maximum effect.

In the description provided herein, numerous specific details are set forth. It is understood, however, that embodiments of the invention may be practiced without these specific details. In some instances, well-known methods, structures and techniques have not been shown in detail in order not to obscure an understanding of this description.

Similarly, it should be appreciated that in the foregoing description of exemplary embodiments of the invention, various features of the invention are sometimes grouped together in a single embodiment, figure, or description thereof for the purpose of streamlining the disclosure and aiding in the understanding of one or more of the various inventive aspects. However, the disclosed method should not be interpreted as reflecting an intention that: that the invention as claimed requires more features than are expressly recited in each claim. Rather, as the following claims reflect, inventive aspects lie in less than all features of a single foregoing disclosed embodiment. Thus, the claims following the detailed description are hereby expressly incorporated into this detailed description, with each claim standing on its own as a separate embodiment of this invention.

Those skilled in the art will appreciate that the modules in the device in an embodiment may be adaptively changed and disposed in one or more devices different from the embodiment. The modules or units or components of the embodiments may be combined into one module or unit or component, and furthermore they may be divided into a plurality of sub-modules or sub-units or sub-components. All of the features disclosed in this specification (including any accompanying claims, abstract and drawings), and all of the processes or elements of any method or apparatus so disclosed, may be combined in any combination, except combinations where at least some of such features and/or processes or elements are mutually exclusive. Each feature disclosed in this specification (including any accompanying claims, abstract and drawings) may be replaced by alternative features serving the same, equivalent or similar purpose, unless expressly stated otherwise.

Furthermore, those skilled in the art will appreciate that while some embodiments herein include some features included in other embodiments, rather than other features, combinations of features of different embodiments are meant to be within the scope of the invention and form different embodiments. For example, in the claims, any of the claimed embodiments may be used in any combination.

The various component embodiments of the invention may be implemented in hardware, or in software modules running on one or more processors, or in a combination thereof. Those skilled in the art will appreciate that a microprocessor or Digital Signal Processor (DSP) may be used in practice to implement some or all of the functions of some or all of the components in an on-board advertising bus route recommendation device according to embodiments of the present invention. The present invention may also be embodied as apparatus or device programs (e.g., computer programs and computer program products) for performing a portion or all of the methods described herein. Such programs implementing the present invention may be stored on computer-readable media or may be in the form of one or more signals. Such a signal may be downloaded from an internet website or provided on a carrier signal or in any other form.

An embodiment of the present invention also provides an electronic device, comprising a processor and a memory arranged to store computer executable instructions that, when executed, cause the processor to perform a method for vehicle-mounted advertising bus route recommendation according to any of the above embodiments.

An embodiment of the present invention further provides a computer storage medium, where the computer storage medium stores one or more programs, and when the one or more programs are executed by an electronic device including a plurality of application programs, the electronic device is caused to execute the vehicle-mounted advertisement bus route recommendation method according to any one of the above embodiments.

For example, FIG. 7 illustrates a computing device that may implement a method for in-vehicle advertising bus route recommendation. The computing device conventionally includes a computer program product or computer-readable medium in the form of a processor 710 and memory 720. The memory 720 may be an electronic memory such as a flash memory, an EEPROM (electrically erasable programmable read only memory), an EPROM, a hard disk, or a ROM. The memory 720 has a storage space 730 for storing program code 731' for performing any of the method steps of the above-described method. For example, the storage space 730 storing the program codes may include respective program codes 731' for respectively implementing various steps in the above method. The program code can be read from or written to one or more computer program products. These computer program products comprise a program code carrier such as a hard disk, a Compact Disc (CD), a memory card or a floppy disk. Such a computer program product is typically a portable or fixed storage unit as shown for example in fig. 8. The memory unit may have memory segments, memory spaces, etc. arranged similarly to memory 720 in the computing device of fig. 7. The program code may be compressed, for example, in a suitable form. Typically, the memory unit comprises computer readable code 731', i.e. code that can be read by a processor such as 710, for performing the steps of the method of the invention, which when run by a computing device causes the computing device to perform the steps of the method described above.

It should be noted that the above-mentioned embodiments illustrate rather than limit the invention, and that those skilled in the art will be able to design alternative embodiments without departing from the scope of the appended claims. In the claims, any reference signs placed between parentheses shall not be construed as limiting the claim. The word "comprising" does not exclude the presence of elements or steps not listed in a claim. The word "a" or "an" preceding an element does not exclude the presence of a plurality of such elements. The invention may be implemented by means of hardware comprising several distinct elements, and by means of a suitably programmed computer. In the unit claims enumerating several means, several of these means may be embodied by one and the same item of hardware. The usage of the words first, second and third, etcetera do not indicate any ordering. These words may be interpreted as names.

Thus, it should be appreciated by those skilled in the art that while a number of exemplary embodiments of the invention have been illustrated and described in detail herein, many other variations or modifications consistent with the principles of the invention may be directly determined or derived from the disclosure of the present invention without departing from the spirit and scope of the invention. Accordingly, the scope of the invention should be understood and interpreted to cover all such other variations or modifications.

Claims

1. A vehicle-mounted advertisement bus route recommendation method comprises the following steps:

collecting data related to buses, and mining multidimensional attributes of all bus stations from the data related to the buses; the method comprises the following steps of mining multidimensional attributes of each bus station from bus related data, wherein the multidimensional attributes comprise:

acquiring geographical position information of the bus stations according to the bus station data, and calculating function theme distribution of each bus station based on the geographical position information of the bus stations and POI data within a specified distance around the geographical position information of the bus stations;

determining administrative division attribution of each bus station according to the geographical position information and the map data of the bus station;

selecting at least one bus line from the bus lines according to the matched multidimensional attribute of the bus station, and recommending the bus line as an advertisement delivery bus line of the advertisement delivery scene;

wherein, confirm the advertisement putting scene according to the advertisement putting intention, include:

2. The method of claim 1, wherein the bus-related data comprises at least one of: map data, bus station data, bus route data, intelligent bus card data SCT and user interest point POI data.

3. The method as claimed in claim 2, wherein the smart bus card data includes a bus route number, a passenger getting-on/off time, and a passenger getting-on/off station number, and the processing the smart bus card data to calculate the passenger trajectory data includes:

4. The method as claimed in claim 3, wherein the calculating of the logical number of the boarding station of the passenger according to a preset strategy based on the charging number of the boarding and disembarking station of the passenger comprises:

5. The method of claim 4, wherein calculating the hub of each bus stop from the sequence of past stops in the passenger's trajectory data comprises:

6. The method of claim 5, wherein calculating the functional topic distribution of each bus stop based on the geographic location information of the bus stop and the POI data within the specified distance around the bus stop comprises:

7. The method of claim 6, wherein matching the advertisement delivery scenario with multidimensional attributes of each bus stop to obtain matched multidimensional attributes of the bus stop comprises:

8. The method of claim 7, wherein if the advertisement placement scene is a widely spread scene, selecting at least one bus route from the bus routes according to the matched multidimensional attribute of the bus station, and recommending the bus route as the advertisement placement bus route of the advertisement placement scene, comprising:

9. The method according to claim 7 or 8, wherein if the advertisement delivery scene is a theme scene with a directional function, selecting at least one bus route from the bus routes according to the matched multidimensional attribute of the bus station, and recommending the bus route as the advertisement delivery bus route of the advertisement delivery scene, comprising:

setting an advertisement target function theme;

10. The method of claim 9, wherein calculating the weight of the target function theme of each bus stop in the candidate stop set according to the set advertisement target function theme and the function theme distribution of the bus stops comprises:

11. The method of claim 10, wherein if the advertisement placement scene is a targeted administrative area scene, selecting at least one bus route from the bus routes according to the matched multidimensional attribute of the bus station, and recommending the bus route as the advertisement placement bus route of the advertisement placement scene, comprising:

setting an advertisement target administrative region;

12. The method of claim 11, wherein calculating a coverage of any bus line in the set of candidate lines as a function of a weight of each passenger trajectory in the bus lines of the set of candidate lines comprises:

13. The method of claim 12, wherein the selecting the bus line with the largest coverage as the advertisement delivery bus line for recommendation comprises:

14. An on-vehicle advertisement bus route recommendation device, comprising:

the mining module is suitable for collecting data related to buses and mining the multi-dimensional attributes of all bus stations from the data related to the buses; the method comprises the following steps of mining multidimensional attributes of each bus station from bus related data, wherein the multidimensional attributes comprise:

the recommendation module is suitable for selecting at least one bus line from the bus lines according to the matched multidimensional attribute of the bus station and recommending the bus line as an advertisement putting bus line of the advertisement putting scene;

wherein the determination module is further adapted to:

15. The apparatus of claim 14, wherein the bus-related data comprises at least one of: map data, bus station data, bus route data, intelligent bus card data SCT and user interest point POI data.

16. The device of claim 15, wherein the smart bus card data includes a bus route number, a passenger getting-on/off time, a passenger getting-on/off station number, and the mining module is further adapted to:

17. The apparatus of claim 16, wherein the excavation module is further adapted to:

18. The apparatus of claim 17, wherein the excavation module is further adapted to:

19. The apparatus of claim 18, wherein the excavation module is further adapted to: acquiring all POI data within a specified distance around each bus station based on the geographical position information of the bus station and the geographical position information in the POI data;

20. The apparatus of claim 19, wherein the determining module is further adapted to:

21. The apparatus of claim 20, wherein if the advertising placement scenario is a widely disseminated scenario, the recommendation module is further adapted to:

22. The apparatus of claim 20 or 21, wherein if the advertising placement scenario is a targeted function topic scenario, the recommendation module is further adapted to:

setting an advertisement target function theme;

23. The apparatus of claim 22, wherein the recommendation module is further adapted to:

24. The apparatus of claim 23, wherein if the advertising scene is a targeted administrative area scene, the recommendation module is further adapted to:

setting an advertisement target administrative region;

25. The apparatus of claim 24, wherein the recommendation module is further adapted to:

26. The apparatus of claim 25, wherein the recommendation module is further adapted to:

27. An electronic device, comprising:

a processor; and

a memory arranged to store computer executable instructions that when executed cause the processor to perform the method of on-board advertising bus route recommendation according to any of claims 1-13.

28. A computer storage medium, wherein the computer-readable storage medium stores one or more programs that, when executed by an electronic device including a plurality of application programs, cause the electronic device to perform the in-vehicle advertising bus route recommendation method according to any one of claims 1-13.