TW202401329A - System and method for speculating advertisement site location based on telecommunication data which allows data of different accuracies to be quickly integrated together - Google Patents

Abstract

A system and a method for speculating an advertisement site location based on telecommunication data are provided, which acquires, from a plurality of telecommunication data, a plurality of crowd description data and a plurality of point-of-interest (POI) data; a processor using a honeycomb hexagonal space indexing algorithm to redefine the accuracy of crowd description data and POI data, so as to have the crowd description data and the POI data corresponding to a first accuracy and a second accuracy of the honeycomb hexagonal space indexing algorithm, respectively; then, integrating the POI data of the second accuracy and the crowd description data of the first accuracy together; the processor searching the integrated data of the crowd description data and the POI data to speculate at least one site location that matches both the crowd description and the POI. Based on the system and the method of the present invention, the data of different accuracies can be quickly integrated together.

Description

System and method for inferring advertising placement locations based on telecommunications data

The present invention relates to an advertising placement technology, and in particular to a system and method for inferring the location of advertising slots based on telecommunications data.

Advertising agencies or media agencies need to accurately deliver advertisements to target customers, thereby increasing the probability of products being sold. One of the keys to achieving precise marketing is to deliver content that accurately touches the appropriate attributes of the population. For example, near the park Set up a public broadcasting system. If you can know the description of the people near the park, for example, there are mostly young people who like sports, or there are mostly middle-aged and elderly people who like pets. For different groups, the attributes of the advertisements will be different. , in order to resonate with people who watch the advertisement and further generate commercial behavior. This is currently a major solution to increase advertising effectiveness.

Generally speaking, crowd description data is obtained through telecommunications data, or by cooperating with telecommunications providers to obtain the description of the number of people with gender, age, and interest preferences within a 35*35 meter radius of a designated location every day. This part except In addition to the acquisition of telecommunications data, there are also technical components of data analysis and statistics, so that correct crowd description information can be screened out from the massive amount of data. It is also difficult to obtain the point of interest (POI) data described in the location. How to know how many banks, schools, parks, etc. are nearby? In addition to manual collection, the most effective way is to The identification method is obtained using the Google Points of Interest Application Programming Interface (POI API). Although the current technology is organic It will be possible to obtain the above two types of data, but integrating telecommunications data that "describes groups of people" and point-of-interest data that "describes locations" together is a problem that cannot yet be solved technically. Because the accuracy of telecommunications data is an area of 35*35 square meters centered on a point, if the same accuracy is used to collect information on points of interest, the range is too small. In addition to being unable to find out a few points of interest, In addition, a large amount of money is also needed to build the API. For example, the size of Taiwan is 36197000000 square meters. If the telecommunications data is 35*35=1225 square meters, 36197000000/1225=29533877, which represents the use of To type the Google Points of Interest API with an accuracy of 35*35(m), it will take nearly 30 million times to type one category; to type all 90 categories, you need to type the Google Points of Interest API. It takes 30 million times * 90 categories = 2.7 billion times to complete the treatment, which is quite an astronomical figure and expense. And even so, it may not be possible to collect complete points of interest, because the accuracy of 35*35 (m) is too small, and the amount of information returned by the Google Points of Interest application interface each time is incomplete.

In view of this, the present invention proposes a system and method for inferring advertising placement locations based on telecommunications data to address the deficiencies of the above-mentioned conventional technologies and future needs, in order to solve the above-mentioned deficiencies. The specific architecture and implementation methods will be described in detail in Down:

The main purpose of the present invention is to provide a system and method for inferring the location of advertising slots based on telecommunications data. It uses the Uber H3 cellular hexagonal spatial index algorithm to integrate two types of data with different precisions to solve the problem of inability to describe crowds in the past. The problem of integrating telecommunications data and data describing people's points of interest allows the location of advertising placements to be more in line with needs.

Another object of the present invention is to provide a system and method for inferring the location of advertising slots based on telecommunications data, which uses the Uber H3 cellular hexagonal spatial indexing algorithm to expand the search range from one cell to six adjacent cells. .

Another object of the present invention is to provide a system and method for inferring the location of advertising slots based on telecommunications data. The Uber H3 cellular hexagonal spatial index algorithm can be used to obtain the surrounding index value of a cell, and the surrounding area can be quickly searched and guaranteed to be Proximity points to increase search speed.

In order to achieve the above purpose, the present invention provides a system for inferring the location of advertising slots based on telecommunications data. It is installed in a server and is suitable for collecting crowd description data and point-of-interest data, and infers the location of advertising slots based on this. , a system for inferring advertising placement locations based on telecommunications data includes: a first collection module, which obtains plural crowd description information from plural telecommunications data; a second collection module, which obtains plural point-of-interest information; and a first database, Connected to the first collection module for storing crowd description data; a second database connected to the second collection module for storing interest point data; and a processor connected to the first database and the second database, A honeycomb hexagonal spatial index algorithm is used to redefine the accuracy of crowd description data and interest point data, the crowd description data is mapped to a first accuracy of the honeycomb hexagonal spatial index algorithm, and the interest point data is mapped to the honeycomb hexagonal A second precision of the shape space indexing algorithm is used, and then the second precision point of interest data is integrated with the first precision crowd description data to infer at least one location that meets both the crowd description and the point of interest.

According to an embodiment of the present invention, the crowd description information is obtained from at least one telecommunications provider, and the crowd description information includes gender, age, and interest preferences.

According to an embodiment of the present invention, the point of interest data is obtained using a point of interest collection application interface, and the point of interest data includes public facilities in a specific area.

According to an embodiment of the present invention, a first comparison table corresponding to the first accuracy of the group description data is stored in the first database.

According to an embodiment of the present invention, a second comparison table corresponding to the point of interest data with a second accuracy is stored in the second database.

According to an embodiment of the present invention, both the first precision and the second precision divide a map into a honeycomb grid composed of regular hexagonal plural cells, and the processor calculates for any coordinate point the coordinate point at the first precision Or the second precision index value.

According to an embodiment of the present invention, when the processor searches for crowd description data in one of the cells of the first precision, it can further extrapolate to the upper level of precision of the first precision, and merge the six adjacent cells around the cell into cells to expand the search scope.

According to an embodiment of the present invention, when the processor searches for crowd description data or point of interest data in one of the cells of the first precision or the second precision, if the required data cannot be found, the cell can be used as a reference point to the surrounding related information. Search adjacent cells.

The present invention further provides a method for inferring the location of advertising slots based on telecommunications data, which is set up in a server and includes the following steps: obtaining plural crowd description data and obtaining plural point-of-interest data from plural telecommunications data; a processor using a The honeycomb hexagon spatial index algorithm redefines the accuracy of crowd description data and interest point data, and maps the crowd description data to the first accuracy of the honeycomb hexagon spatial index algorithm; the processor maps the interest point data to the honeycomb hexagon A second precision of the spatial index algorithm, and then integrates the second precision point of interest data with the first precision crowd description data; and the processor searches the integrated data of the crowd description data and the point of interest data to infer that the data matches the Crowd description and at least one slot location of the point of interest.

10: System for inferring advertising placement locations based on telecommunications data

12:Server

13:The first collection module

14:Second collection module

15:First database

152: First comparison table

16: Second database

162: Second comparison table

18: Processor

20:Telecommunications provider

22:Internet service provider

30:Cell

32: First precision

34: Second precision

36:Center cell

38: Upper cell

Figure 1 is a block diagram of a system for inferring advertising space locations based on telecommunications data according to the present invention.

Figure 2 is a flow chart of the method of estimating the location of advertising slots based on telecommunications data according to the present invention.

Figure 3 is a schematic diagram of the Uber H3 cellular hexagonal spatial indexing algorithm.

Figure 4 is a schematic diagram of the present invention using the Uber H3 cellular hexagonal spatial indexing algorithm to integrate crowd description data and point-of-interest data.

Figure 5 is a schematic diagram of the present invention using the Uber H3 cellular hexagonal spatial indexing algorithm to find adjacent cells.

The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of them. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without making any progressive efforts shall fall within the scope of protection of the present invention.

It should be understood that, when used in this specification and the appended claims, the terms "comprise" and "include" indicate the presence of described features, integers, steps, operations, elements and/or elements but do not exclude the presence of one or The presence or addition of various other features, integers, steps, operations, elements, components and/or collections thereof.

It should also be understood that the terminology used in the description of the invention is for the purpose of describing particular embodiments only and is not intended to be limiting of the invention. As used in the specification of the invention and the appended claims, the singular forms "a", "an" and "the" are intended to include the plural forms unless the context clearly dictates otherwise.

It should be further understood that the term "and/or" as used in the description of the present invention and the appended claims refers to any and all possible combinations of one or more of the associated listed items and includes such combinations. .

The present invention provides a system and method for inferring advertising placement locations based on telecommunications data. Please refer to Figure 1 , which is a block diagram of a system for inferring advertising placement locations based on telecommunications data according to the present invention. The system 10 for inferring advertising slot locations based on telecommunications data of the present invention includes a server 12. The system 10 for inferring advertising slot locations based on telecommunications data includes a first collection module 13, a second collection module 14, A first database 15, a second database 16 and a processor 18. The first collection module 13 is connected to a host signal of a telecommunications provider 20 to obtain telecommunications data of the telecommunications provider 20, including users of the telecommunications provider 20. Gender, age, interest preferences (such as which types of web pages to use mobile phones to watch), movement trajectories and other crowd description information. The first collection module 13 is connected to the first database 15 and stores the collected crowd description data in the first database 15 . The second collection module 14 is connected to a host signal of an Internet service provider 22. The Internet service provider 22, such as Google POI application programming interface, obtains location description information, including banks, schools, parks, etc. The location of facilities (i.e. points of interest). The second collection module 14 is connected to the second database 16 and stores the collected point-of-interest data in the second database 16 . The processor 18 connects the first database 15 and the second database 16 . The processor 18 uses a cellular hexagonal spatial indexing algorithm to redefine the accuracy of the crowd description data and the interest point data, and combine the crowd description data and the interest point data together.

Please also refer to Figure 2, which is a flowchart of the method of using the present invention to estimate the location of advertising slots based on telecommunications data. In step S10, a plurality of crowd description data and a plurality of interest point data are obtained from the plurality of telecommunications data. In step S12, the processor 18 uses a cellular hexagonal spatial index algorithm to redefine the accuracy of the crowd description data and the point of interest data, and corresponds the crowd description data to A first accuracy to the cellular hexagon spatial indexing algorithm. Then, as described in step S14, the processor 18 maps the interest point data to a second precision of the cellular hexagonal spatial index algorithm, and then integrates the second precision interest point data with the first precision crowd description data. Finally, as described in step S16, after the crowd description data and the point of interest data are integrated, the processor 18 performs a keyword search on the integrated data to deduce at least one location that matches both the crowd description and the point of interest.

The cellular hexagonal spatial index algorithm used in the present invention is the Uber H3 algorithm. This algorithm divides the earth's space into identifiable cells, and encodes the longitude and latitude H3 into a hexagonal grid index, as shown in Figure 3 , the plurality of cells 30 form a honeycomb grid, and each cell 30 is surrounded by 6 adjacent cells 30. These seven cells 30 can form a larger hexagon, as shown in Figure 4 As shown in the figure. The H3 index value of every location in the world is unified and is divided into 16 stages (Level). Each stage represents a different accuracy. Stage 0 is the maximum diameter (side length 1107712 meters), and stage 15 is the minimum. diameter (side length 0.5 meters), the smaller the diameter, the higher the accuracy, so stage 15 has the highest accuracy. The index value can be derived upwards and downwards, for example the hexagon of stage 10 covers 7 times the area of stage 11 (1884*7=13188). The index value (cell id) of cell 30 can be deduced from each other. Suppose the index value in stage 11 is 8b4ba010e0a1fff, then through the H3 algorithm, it can be deduced that the index value in stage 10 is 8a4ba010e0a7fff. Therefore, the Uber H3 algorithm can effectively deduct two types of data with different precisions from each other into data of the same precision and mix them for use.

Based on the above-mentioned principles of Uber H3, as shown in Figure 4, the processor 18 maps the crowd description data to a first precision 32 of the cellular hexagonal spatial index algorithm, and maps the interest point data to the cellular hexagonal spatial index. A second precision 34 of the algorithm then integrates the interest point data of the second precision 34 and the crowd description data of the first precision 32 . Since the accuracy of telecommunications data is 35 square meters, the most suitable H3 index value is stage 11 (radius of about 25 meters). Therefore the first A precision of 32 is stage 11. In addition, the accuracy of the most important point of interest data is based on stage 8 (radius of about 460 meters) as the second accuracy 34. This will greatly reduce the number of calls to the Google Point of Interest API. The total area of Taiwan is 36197000000 square meters. The hexagonal area of each H3 stage 8 is 646367 square meters, 36197000000/646367=56000 times. In other words, it only takes 56000*90=5040000 times to collect all 90 categories. The number has been reduced from the 2.7 billion times mentioned in the previous technology to 5 million times after using the H3 algorithm, but the collected interests The quality of the point data will not be degraded as a result. In this way, the problem of mixed use of two different precision data can be solved. The processor 18 calculates the index value of the coordinate point at the first precision 32 or the second precision 34 for any coordinate point to infer that it meets the crowd description at the same time. and at least one location of the point of interest. In addition, the present invention can also save the number of calls to the POI application program interface by applying the H3 algorithm.

In the present invention, in addition to storing crowd description information, the first database 15 also stores the first comparison table 152 corresponding to the first accuracy 32 of telecommunications data in the first database 15 . Similarly, in addition to storing interest point data, the second database 16 also stores the second comparison table 162 corresponding to the second precision 34 in the second database 16 .

In addition, since the index value of the H3 algorithm can be extrapolated from the small-stage index value to the large-stage index value, the index superposition function can be achieved. When the processor 18 searches for crowd description information in one of the cells 30 of the first precision 32 , it can further extrapolate to the upper level of precision of the first precision 32 and merge the six adjacent cells around the cell 30 into the cell 30 , to expand the search scope. For example, assume that stage 8 is currently used to define this piece of telecommunications data. Of course, going up one level to stage 7 has a larger diameter. When obtaining a larger range, the surrounding six stage 8 cell data can be integrated. , as a larger range of data index. Take Figure 4 as an example. Assume that a central cell 36 only finds one convenience store. If the number is insufficient, go up a level and frame the surrounding six adjacent cells. It becomes By forming a larger upper cell 38 to the large basket, you can find more convenience stores close to the target point (lat, lon).

In the same way, based on the H3 algorithm, the characteristics of the surrounding indexes can be known. When the index value of the center cell 36 is known, the index values of the six surrounding cells can be known. Therefore, when the processor 18 searches for crowd description data or point of interest data in one of the cells 30 of the first precision 32 or the second precision 34, if the required data cannot be found, the cell 30 can be used as the reference point to the surrounding areas. Search adjacent cells. This method can ensure that the data found are adjacent points, without the need to calculate the straight-line distance between the interest points by yourself, and can also avoid the shortcomings of scanning each interest point in the previous technology. This method can speed up data indexing and can greatly improve search speed.

In the present invention, the placement device can be an Android box, an electronic billboard, a display screen at a bus stop or convenience store, a screen in an elevator, etc., and can play designated advertisements at designated times. This invention uses the Uber H3 algorithm to accurately integrate crowd description data and point-of-interest data, and can quickly search for the desired advertising location. For example, if you want a location where there are mostly young men who like to watch movies, and there is a business district nearby and there are department stores, this condition is applicable to the prediction based on telecommunications data provided by the present invention. Systems and methods for advertising placement locations.

In summary, the present invention provides a system and method for inferring the location of advertising slots based on telecommunications data, which can integrate crowd description data and point-of-interest data with different accuracy through the index results of the Uber H3 algorithm, effectively saving data purchases. and the number of times data is collected. In addition, because as long as you confirm a geographical coordinate (lat, lon) at a point on the earth, you can know all the index values of this coordinate falling into stages 1 to 15, so you can also perform integrated queries among various data. Take out materials from various stages and mix them.

The above are only preferred embodiments of the present invention and are not intended to limit the scope of the present invention. Therefore, all equivalent changes or modifications made in accordance with the characteristics and spirit described in the scope of the present invention shall be included in the patent scope of the present invention.

10: System for inferring advertising placement locations based on telecommunications data

12:Server

13:The first collection module

14:Second collection module

15:First database

152: First comparison table

16: Second database

162: Second comparison table

18: Processor

20:Telecommunications provider

22:Internet service provider

Claims (16)

A system for inferring the location of advertising slots based on telecommunications data. It is set up in a server and is suitable for collecting crowd description information and point-of-interest data, and infers the location of advertising slots based on this. The system is used to infer the location of advertising slots based on telecommunications data. Location systems include: A first collection module that obtains description data of plural groups of people from plural telecommunications data; A second collection module to obtain multiple points of interest data; A first database connected to the first collection module for storing the crowd description data; a second database connected to the second collection module for storing the point of interest data; and A processor connects the first database and the second database, uses a cellular hexagonal spatial index algorithm to redefine the accuracy of the crowd description data and the point of interest data, and maps the crowd description data to a first precision of the cellular hexagonal spatial index algorithm, and map the interest point data to a second precision of the cellular hexagonal spatial index algorithm, and then combine the second precision of the interest point data with The first-precision crowd description data are integrated together to infer at least one location that matches both the crowd description and the point of interest. A system for inferring advertising placement locations based on telecommunications data as described in request item 1, wherein the crowd description information is obtained from at least one telecommunications provider. A system for inferring advertising placement locations based on telecommunications data as described in request 1, wherein the demographic information includes gender, age and interest preferences. A system for inferring advertising placement locations based on telecommunications data as described in request 1, wherein the point-of-interest data is obtained using a point-of-interest collection application interface. A system for inferring the location of advertising slots based on telecommunications data as described in request 1, wherein the point-of-interest data includes public facilities in a specific area. As claimed in claim 1, the system for inferring the location of advertising slots based on telecommunications data, wherein a first comparison table corresponding to the first accuracy of the crowd description data is stored in the first database. As claimed in claim 1, the system for inferring the location of advertising slots based on telecommunications data, wherein the point-of-interest data corresponds to a second comparison table of the second accuracy and is stored in the second database. The system for estimating the location of advertising slots based on telecommunications data as described in request 1, wherein both the first precision and the second precision divide a map into a honeycomb grid composed of regular hexagonal plural cells, and The processor calculates, for any coordinate point, the index value of the coordinate point in the first precision or the second precision. The system for estimating the location of advertising slots based on telecommunications data as described in request item 8, wherein the processor further infers to the upper level of the first precision when searching for the crowd description information in one of the cells of the first precision. Precision, merge the six adjacent cells around the cell into this cell to expand the search range. The system for inferring the location of advertising slots based on telecommunications data as described in request item 9, wherein when the processor searches for the crowd description information or the point of interest information in one of the cells of the first precision or the second precision, If the required data cannot be found, the cell is used as the reference point to search the surrounding adjacent cells. A method of inferring the location of advertising slots based on telecommunications data, which is set up in a server. The method of inferring the location of advertising slots based on telecommunications data includes the following steps: Obtain plural group description information from plural telecommunications data, and obtain plural point-of-interest information; A processor uses a cellular hexagonal spatial index algorithm to redefine the accuracy of the crowd description data and the interest point data, and maps the crowd description data to a first accuracy of the cellular hexagonal spatial index algorithm; The processor maps the interest point data to a second precision of the cellular hexagonal spatial index algorithm, and then integrates the interest point data of the second precision with the crowd description data of the first precision. together; and The processor searches the integrated data of the crowd description data and the point-of-interest data to infer at least one location that matches both the crowd description and the point-of-interest data. As described in request 11, the method of inferring the location of advertising slots based on telecommunications data, wherein the crowd description information is obtained from at least one telecommunications provider, and the point-of-interest information is obtained using a point-of-interest collection application interface. As described in request 11, the method of inferring the location of advertising slots based on telecommunications data, wherein the population description information includes gender, age and interest preferences, and the point of interest information includes public facilities in a specific area. The method of estimating the location of advertising slots based on telecommunications data as described in request 11, wherein the first precision and the second precision both divide a map into a honeycomb grid composed of regular hexagonal plural cells, and The processor calculates, for any coordinate point, the index value of the coordinate point in the first precision or the second precision. The method of inferring the location of advertising slots based on telecommunications data as described in request 14, wherein the processor searches for the crowd description information in one of the cells of the first precision At that time, the upper level precision of the first precision is further deduced, and the six adjacent cells around the cell are merged into the cell to expand the search range. The method of inferring the location of advertising slots based on telecommunications data as described in request item 15, wherein when the processor searches for the crowd description information or the point of interest information in one of the cells of the first precision or the second precision, If the required data cannot be found, the cell is used as the reference point to search the surrounding adjacent cells.

Images