KR20220149810A - Method for extracting users base location information - Google Patents

Abstract

According to one embodiment of the present disclosure, a method for extracting base location information of a user performed by a processor of a computing device is disclosed. The method comprises the following steps of: identifying a plurality of boarding points and a plurality of drop-off points by using public transportation use data related to a user; identifying at least one initial boarding point and at least one final drop-off point excluding at least one transfer point from the plurality of boarding points and the plurality of drop-off points; and clustering the at least one initial boarding point and the at least one final drop-off point by an unsupervised learning model to obtain at least two base location groups. According to the present invention, base locations of a user are extracted based on public transportation information.

Description

How to extract user base information {METHOD FOR EXTRACTING USERS BASE LOCATION INFORMATION}

The present disclosure relates to a method of extracting base information of a user based on public transportation information, and more particularly, to a technology for extracting base information of a user by using the user's individual rough transit use information.

With the development of mobile devices, mobile devices provide various services such as calling, listening to music or movies, the Internet, playing games, and taking pictures. In particular, with the spread of smart phones, various services using location information of users are being provided. For example, a service for providing information related to the current location by using the user's current location information is being provided.

The service using the user's current location depends on the GPS signal of the user terminal. However, the GPS signal of the user terminal has a lot of noise, and it may be difficult to determine the exact location of the user in a shaded area such as an underground space. In addition, in the case of a service using the user's current location, when the user moves to another location, the value of information provided to the user based on the current location may decrease.

Accordingly, there may be a demand in the art for a technology for predicting a user's base in order to provide a service using the user's moving route or final destination location.

Unexamined Patent Publication No. 10-2013-0064548

The present disclosure has been devised in response to the above-described background technology, and is intended to provide a method of extracting a user's base based on public transportation information.

The technical problems of the present disclosure are not limited to the technical problems mentioned above, and other technical problems not mentioned will be clearly understood by those skilled in the art from the following description.

According to some embodiments of the present disclosure for solving the problems as described above, a method of extracting user base information performed by a processor of a computing device is disclosed. The method includes: recognizing a plurality of boarding points and a plurality of alighting points using public transportation usage data associated with the user; recognizing at least one initial boarding point and at least one final getting off point excluding at least one transfer point from the plurality of boarding points and the plurality of alighting points; and clustering the at least one initial boarding point and the at least one final getting off point through an unsupervised learning model to obtain at least two base groups.

In addition, the unsupervised learning model calculates a distance for each of the combinable point pairs among the at least one initial boarding point and the at least one final get off point, and uses the calculated distances to perform a density-based spatial clustering technique (DBSCAN). ), the at least one initial boarding point and the at least one final getting off point may be clustered.

In addition, in order to cluster the at least one initial boarding point and the at least one final disembarkation point through the unsupervised learning model, it is used when determining the minimum number of points that can belong to the same base group or points belonging to the same base. At least one of the critical radii may be determined.

In addition, each of the at least two base groups includes: at least one core point designated when the number of other points existing within a distance shorter than the critical radius with respect to itself is equal to or greater than the minimum number; and at least one designated when the number of other points existing within a distance shorter than the critical radius with respect to itself is less than the minimum number, and within a distance shorter than the critical radius from any one of the at least one core point Border points; may include.

Also, each of the core points included in a different base group among the at least two base groups may be at a distance greater than the critical radius.

In addition, among the at least one initial boarding point and the at least one final disembarkation point, a noise point is not included in the at least two base groups, and the noise point exists within a distance shorter than the critical radius with respect to itself. The number of other points may be less than the minimum number, and may be points that exist at a distance greater than the critical radius from each of the at least one core point.

In addition, the method may include: recognizing the characteristics of each of the at least two base groups by using the public transportation use data corresponding to each of the at least two base groups and a preset rule; and tagging each of the at least two base groups with the personality, and storing the tag in a memory.

In addition, the step of recognizing the characteristics of each of the at least two base groups by using the public transport use data corresponding to each of the at least two base groups and a preset rule includes: calculating a first score associated with the first personality and a second score associated with a second personality different from the first personality; and recognizing the characteristics of each of the two base groups based on the size of the first score and the size of the second score of each of the at least two base groups.

In addition, for each of the at least two base groups, calculating a first score related to a first character and a second score related to a second character different from the first character may include: recognizing the first boarding point, the first boarding time, the last getting off point, and the last getting off time of the group-by-group public transportation usage information by analyzing the group-by-group public transportation use information; The first score of each of the at least two base groups based on whether the first boarding point, the first boarding time, the last getting off point, and the last getting off time of the group-specific public transportation use information satisfy preset conditions adding a preset value to ; analyzing the group-by-group public transportation use information for each of the at least two base groups to recognize an intermediate boarding point, an intermediate boarding time, an intermediate getting-off point, and an intermediate getting-off time of the group-by-group public transportation use information; and adding a preset value to the second score of each of the at least two base groups based on whether the intermediate boarding point, the intermediate boarding time, the intermediate getting off point, and the intermediate getting off time satisfy preset conditions. step; may include.

In addition, the step of recognizing the characteristics of each of the two base groups based on the size of the first score and the size of the second score of each of the at least two base groups may include: A group that exceeds the value and the magnitude of the second score does not exceed a preset value is recognized as the first personality, the magnitude of the first score does not exceed the preset value and the magnitude of the second score does not exceed the preset value Recognizing a group exceeding a preset value as a second personality, and recognizing a group in which the magnitude of the first score and the magnitude of the second score do not exceed a preset value as a third personality; have.

In addition, the step of recognizing the characteristics of each of the two base groups based on the size of the first score and the size of the second score of each of the at least two base groups includes the size of the first score and the size of the second score. and determining the nature of the unprocessed group based on a preset personality when there is an unprocessed group having the same size of 2 scores.

In addition, when the user's boarding point information is input, recognizing which group the boarding point information belongs to among the at least two base groups; acquiring base information by recognizing the characteristics of the group to which the boarding point information belongs; and inputting the user's boarding point and the base information to a destination prediction model related to the user from among a plurality of destination prediction models stored in the memory to obtain the user's alighting data.

In addition, the destination prediction model, in the learning process of the destination prediction model, is to be learned using a training data set including training data generated by labeling the training ride data related to the user to the training getting off data related to the user. can

In addition, the step may include: recognizing a disembarkation point included in the user's disembarkation data; recognizing a first group associated with the disembarkation point among the at least two base groups; and transmitting customized service information to the user terminal based on the characteristics of the first group.

According to some embodiments of the present disclosure for solving the problems as described above, a computer program stored in a computer-readable storage medium is disclosed. The computer program includes instructions for causing a processor of a server to perform the following steps, the steps of: recognizing a plurality of boarding points and a plurality of alighting points using public transportation usage data associated with the user; recognizing at least one initial boarding point and at least one final getting off point excluding at least one transfer point from the plurality of boarding points and the plurality of alighting points; and clustering the at least one initial boarding point and the at least one final getting off point through an unsupervised learning model to obtain at least two base groups.

The technical solutions obtainable in the present disclosure are not limited to the above-mentioned solutions, and other solutions that are not mentioned are clearly to those of ordinary skill in the art to which the present disclosure belongs from the description below. can be understood

The present disclosure may extract user's base information by using the user's public transportation usage information, and provide a user-customized service based on the predicted destination using this.

The effects obtainable in the present disclosure are not limited to the above-mentioned effects, and other effects not mentioned will be clearly understood by those of ordinary skill in the art to which the present disclosure belongs from the description below. .

Various aspects are now described with reference to the drawings, wherein like reference numerals are used to refer to like elements collectively. In the following examples, for purposes of explanation, numerous specific details are set forth in order to provide a thorough understanding of one or more aspects. It will be evident, however, that such aspect(s) may be practiced without these specific details.
1 is a diagram for explaining an example of a destination prediction system including a server and a user terminal in which various aspects of the present disclosure may be implemented.
2 is a flowchart illustrating an example of a destination prediction technique according to some embodiments of the present disclosure.
3 and 4 are tables for explaining a destination prediction model according to some embodiments of the present disclosure.
5 is a flowchart for explaining an example of a method of acquiring at least two node groups according to some embodiments of the present disclosure.
6 is a view for explaining at least two base groups according to some embodiments of the present disclosure.
7 is a diagram for explaining an example of clustering the final points of the unsupervised learning model according to some embodiments of the present disclosure.
8 and 9 are diagrams for explaining a method of recognizing the characteristics of at least two base groups according to some embodiments of the present disclosure.
10 is a flowchart illustrating an example of a method of transmitting customized service information to a user terminal according to some embodiments of the present disclosure.
11 is a flowchart illustrating a method of determining whether to use a destination prediction model according to some embodiments of the present disclosure.
12 depicts a simplified, general schematic diagram of an example computing environment in which embodiments of the present disclosure may be implemented.

Various embodiments and/or aspects are now disclosed with reference to the drawings. In the following description, for purposes of explanation, numerous specific details are set forth to provide a thorough understanding of one or more aspects. However, it will also be appreciated by one of ordinary skill in the art that such aspect(s) may be practiced without these specific details. The following description and accompanying drawings set forth in detail certain illustrative aspects of one or more aspects. These aspects are illustrative, however, and some of the various methods in principles of various aspects may be employed, and the descriptions set forth are intended to include all such aspects and their equivalents. Specifically, as used herein, “embodiment”, “example”, “aspect”, “exemplary”, etc. is not to be construed as an advantage or advantage of any aspect or design described over other aspects or designs. It may not be.

Further, various aspects and features will be presented by a system that may include one or more apparatuses, terminals, servers, devices, components and/or modules, and the like. The various systems may include additional apparatuses, terminals, servers, devices, components and/or modules, etc. and/or the apparatuses, terminals, servers, etc. discussed in connection with the drawings. It should also be understood and appreciated that it may not include all of devices, components, modules, etc.

As used herein, the terms "computer program," "component," "module," "system," and the like, may be used interchangeably with each other, and include computer-related entities, hardware, firmware, software, combinations of software and hardware; or to the execution of software. For example, a component can be, but is not limited to being, a process running on a processor, a processor, an object, a thread of execution, a program, and/or a computer. For example, both an application running on a computing device and the computing device may be a component. One or more components may reside within a processor and/or thread of execution. A component may be localized within one computer. A component may be distributed between two or more computers.

In addition, these components can execute from various computer readable media having various data structures stored therein. Components may communicate via a network such as the Internet with another system, for example via a signal having one or more data packets (eg, data and/or signals from one component interacting with another component in a local system, distributed system, etc.) may communicate via local and/or remote processes depending on the data being transmitted).

Hereinafter, the same or similar components are assigned the same reference numerals regardless of reference numerals, and redundant description thereof will be omitted. In addition, in describing the embodiments disclosed in the present specification, if it is determined that detailed descriptions of related known technologies may obscure the gist of the embodiments disclosed in the present specification, the detailed description thereof will be omitted. In addition, the accompanying drawings are only for easy understanding of the embodiments disclosed in the present specification, and the technical ideas disclosed in the present specification are not limited by the accompanying drawings.

The terminology used herein is for the purpose of describing the embodiments and is not intended to limit the present disclosure. In this specification, the singular also includes the plural, unless specifically stated otherwise in the phrase. As used herein, “comprises” and/or “comprising” does not exclude the presence or addition of one or more other components in addition to the stated components.

Although the first, second, etc. are used to describe various elements or elements, these elements or elements are not limited by these terms, of course. These terms are only used to distinguish one element or component from another. Accordingly, it goes without saying that the first device or component mentioned below may be the second device or component within the spirit of the present disclosure.

Unless otherwise defined, all terms (including technical and scientific terms) used herein may be used with the meaning commonly understood by those of ordinary skill in the art to which this disclosure belongs. In addition, terms defined in a commonly used dictionary are not to be interpreted ideally or excessively unless clearly defined in particular.

In addition, the term “or” is intended to mean an inclusive “or” rather than an exclusive “or.” That is, unless otherwise specified or clear from context, "X employs A or B" is intended to mean one of the natural implicit substitutions. That is, X employs A; X employs B; or when X employs both A and B, "X employs A or B" may apply to either of these cases. It should also be understood that the term “and/or” as used herein refers to and includes all possible combinations of one or more of the listed related items.

In addition, as used herein, the terms “information” and “data” may often be used interchangeably.

The suffixes “module” and “part” for components used in the following description are given or mixed in consideration of only the ease of writing the specification, and do not have distinct meanings or roles by themselves.

Objects and effects of the present disclosure, and technical configurations for achieving them will become clear with reference to the embodiments described below in detail in conjunction with the accompanying drawings. In describing the present disclosure, if it is determined that a detailed description of a well-known function or configuration may unnecessarily obscure the subject matter of the present disclosure, the detailed description thereof will be omitted. In addition, the terms described below are terms defined in consideration of functions in the present disclosure, which may vary according to intentions or customs of users and operators.

However, the present disclosure is not limited to the embodiments disclosed below and may be implemented in various different forms. Only the present embodiments are provided so that the present disclosure is complete, and to fully inform those of ordinary skill in the art to which the present disclosure belongs, the scope of the disclosure, and the present disclosure is only defined by the scope of the claims . Therefore, the definition should be made based on the content throughout this specification.

The scope of the steps in the claims of the present disclosure arises from the functions and features described in each step, and unless the precedence of the order in each step is specified, in the claims The order of description of each step is not affected. For example, in a claim described as a step comprising step A and step B, even if step A is stated before step B, the scope of the rights is not limited that step A must precede step B.

1 is a diagram for explaining an example of a destination prediction system including a server and a user terminal in which various aspects of the present disclosure may be implemented.

Referring to FIG. 1 , a destination prediction system according to some embodiments of the present disclosure may include a server 100 and a user terminal 200 . Additionally, the destination prediction system may further include an external server that provides the user's public transportation usage data and other users' public transportation usage data related to the user terminal 200 . However, since the above-described components are not essential in implementing the destination prediction system, the destination prediction system may have more or fewer components than those listed above.

According to some embodiments of the present disclosure, the server 100 predicts the user's destination by using the user's public transportation usage data related to the user terminal 200, and a server for providing a customized service based on the predicted destination can be However, the present invention is not limited thereto.

Server 100 may include any type of computer system or computer device, such as, for example, microprocessors, mainframe computers, digital processors, portable devices and device controllers, and the like. However, the present invention is not limited thereto.

The server 100 may include a processor 110 , a communication unit 120 , and a memory 130 . However, since the above-described components are not essential in implementing the server 100 , the server 100 may have more or fewer components than those listed above. Here, each component may be configured as a separate chip, module, or device, or may be included in one device.

The processor 110 of the server 100 typically controls the overall operation of the server 100 . The processor 110 may provide or process appropriate information or functions to the user by processing signals, data, information, etc. input or output through the above-described components or driving an application program stored in the memory 130 .

In addition, the processor 110 may control at least some of the components of the server 100 in order to drive an application program stored in the memory 130 . Furthermore, the processor 110 may operate by combining at least two or more of the components included in the server 100 to drive the application program.

The communication unit 120 of the server 100 may include one or more modules that enable communication between the server 100 and the user terminal 200 and between the server 100 and the user terminals of other users. In addition, the communication unit 120 may include one or more modules for connecting the server 100 to one or more networks.

The communication unit 120 of the server 100 according to some embodiments of the present disclosure may communicate with the user terminal 200 by wire or wirelessly. Also, the communication unit 120 may mutually transmit and receive control data for controlling the user terminal 200 with the user terminal 200 .

According to some embodiments of the present disclosure, the communication unit 120 of the server 100 may receive public transportation usage data related to the user from the user terminal 200 or an external server. In this case, the processor 110 of the server 100 may recognize a plurality of boarding points and a plurality of alighting points by using public transportation usage data related to the user. In addition, the processor 110 may recognize at least one initial boarding point and at least one final getting off point from the plurality of boarding points and the plurality of alighting points except for at least one transfer point. In addition, the processor 110 may obtain at least two base groups by clustering at least one initial boarding point and at least one final getting off point through an unsupervised learning model.

Here, the unsupervised learning model of the present disclosure may calculate a distance for each of the combinable point pairs among at least one initial boarding point and at least one final getting off point. In addition, the unsupervised learning model may use the calculated distances to cluster at least one initial boarding point and at least one final getting off point according to a density-based spatial clustering technique.

That is, the unsupervised learning model of the present disclosure may be a density-based spatial clustering of applications with noise (DBSCAN). However, the present invention is not limited thereto.

DBSCAN is a method of clustering high-density data because the data is densely packed. For example, DBSCAN may recognize as one group (cluster) when more than a preset number of points existing within a preset radius based on a specific point exist. In addition, DBSCAN does not need to set the number of clusters, and since clusters are connected to each other according to the density of clusters, clusters having a geometric shape can also be found.

Meanwhile, the processor 110 is configured to cluster at least one initial boarding point and at least one final disembarkation point through an unsupervised learning model. At least one of a critical radius (eg, 5 km) used when determining points belonging to a stronghold may be determined. However, the present invention is not limited thereto, and the processor 110 may receive at least one of information on the minimum number and information on the critical radius from the manager terminal through the communication unit 120 .

The unsupervised learning model of the present disclosure may select any of the final points to cluster the final points (at least one initial boarding point and at least one final getting off point). In addition, the unsupervised learning model may cluster the final points into at least two base groups using a selected random point, a distance between an arbitrary point and other final points, a critical radius, and a minimum number.

Specifically, the unsupervised learning model determines whether the number of other final points existing at a distance shorter than the critical radius is equal to or greater than the minimum number (that is, the minimum number of final points within the critical radius centered on any point is greater than or equal to the minimum number). or not) to cluster the final points into at least two node groups. That is, the processor 110 of the server 100 may acquire at least two base groups (base information) in which the final points are clustered using the unsupervised learning model.

Each of the at least two stronghold groups of the present disclosure may include a core point and a border point.

Here, the core point may be designated when the number of other points existing within a distance shorter than the critical radius with respect to itself among the final points is equal to or greater than the minimum number.

In addition, the border point is designated when the number of other points that exist within a distance shorter than the critical radius from among the final points is less than the minimum number, and exists within a distance shorter than the critical radius from any one of the at least one core point. can be

Meanwhile, among the final points, points not included in at least two base groups may be designated as noise points. The noise point is not included in at least two base groups among the final points, the number of other points existing within a distance shorter than the critical radius with respect to itself is less than the minimum number, and is farther than the critical radius from each of the at least one core point It may be a point existing on the street.

Hereinafter, a method for the processor 110 of the server 100 to cluster the final points using an unsupervised learning model to obtain at least two base groups will be described later with reference to FIGS. 5 to 7 .

Meanwhile, the communication unit 120 of the server 100 may receive ride data from the user terminal 200 . In this case, the processor 110 of the server 100 may obtain the user's alighting data by inputting the boarding data into the destination prediction model. In addition, the processor 110 may control the communication unit 120 to transmit user-customized service information to the user terminal 200 by using the get-off data.

Here, the destination prediction model may be learned using a training data set including training data generated by labeling the user-related training ride data to the user-related training alight data in the training process of the destination prediction model. In addition, the destination prediction model may be composed of an artificial neural network or a neural network.

Specifically, the processor 110 may acquire riding data for learning by using the user's public transportation usage history. Then, the processor 110 may obtain the labeling data by labeling the learning riding data associated with the learning riding data for learning. In addition, the processor 110 may derive an error by comparing the labeling data with the alighting data output as the train ride data for training is input to the destination prediction model. Also, the processor 110 may update the weight of the destination prediction model by backpropagating the error to the destination prediction model.

In addition, the processor 110 may input the boarding data received from the user terminal 200 into the destination prediction model learned as described above, and obtain alighting data related to an alighting point at which the user is expected to get off. In addition, the processor 110 may provide a user-customized service by using the alighting data.

Accordingly, when the user uses public transportation, the server 100 according to some embodiments of the present disclosure may quickly identify a disembarkation point and provide customized service information related to the identified disembarkation point to the user.

The destination prediction model according to some embodiments of the present disclosure may be formed as a decision tree. However, the present invention is not limited thereto.

The decision tree may mean making a prediction/classification model in the form of a tree that represents patterns inherent in data as a combination of variables. Specifically, the decision tree forms a tree structure from an upper node to a lower node, and may be branched according to a classification variable (question) at every step of forming the tree structure from an upper node to a lower node. And, the final lower node may be a prediction value of the decision tree. Here, the order of the classification variables (ie, the order of the lower node from the upper node) may be determined by a preset algorithm. For example, the classification variable may be the day and time the user got on or off the public transportation. However, the present invention is not limited thereto.

On the other hand, models related to machine learning can be classified into black box models and white box models. Here, the black box model refers to a model in which it is difficult to visualize or understand the processes occurring inside the model, and to understand the relationship between explanatory variables and outcome variables. In addition, the white box model is the opposite of the black box model, and refers to a model that is easy to visualize or understand the process occurring inside the model.

The decision tree is a representative white box model and has the advantage of easy visualization and interpretation.

That is, the destination prediction model formed of the decision tree may facilitate interpretation of the destination prediction result.

Accordingly, the server 100 according to some embodiments of the present disclosure may improve service quality by analyzing the correct answer or incorrect answer for the destination predicted by the destination prediction model. Also, the server 100 may facilitate setting a direction for model improvement.

Additionally, decision trees can be modeled using a small amount of data. Accordingly, the server 100 according to some embodiments of the present disclosure may generate a destination prediction model using a small amount of data. In other words, the server 100 of the present disclosure may generate an individual destination prediction model by using a small amount of data that an individual uses public transportation. In this case, the server 100 of the present disclosure may provide a personalized service specialized for an individual through individual destination prediction.

Hereinafter, a description of a method in which the processor 110 of the server 100 acquires the user's alighting data using the destination prediction model and transmits the customized service information to the user terminal 200 by using the user's alighting data is It will be described later with reference to FIGS. 2 to 11 .

Meanwhile, the memory 130 of the server 100 may store a program for the operation of the processor 110 , and may temporarily or permanently store input/output data. The memory 130 may include a flash memory type, a hard disk type, a multimedia card micro type, a card type memory (eg, SD or XD memory), and a RAM. (Random Access Memory, RAM), SRAM (Static Random Access Memory), ROM (Read-Only Memory, ROM), EEPROM (Electrically Erasable Programmable Read-Only Memory), PROM (Programmable Read-Only Memory), magnetic memory, magnetic It may include at least one type of storage medium among a disk and an optical disk. The memory 130 may be operated under the control of the processor 110 .

Various embodiments described herein may be implemented in, for example, a computer or similar device-readable recording medium and storage medium using software, hardware, or a combination thereof.

According to the hardware implementation, the embodiments described herein are ASICs (application specific integrated circuits), DSPs (digital signal processors), DSPDs (digital signal processing devices), PLDs (programmable logic devices), FPGAs (field programmable gate arrays, It may be implemented using at least one of processors, controllers, micro-controllers, microprocessors, and other electrical units for performing functions. The described embodiments may be implemented by the processor 110 itself of the server 100 .

According to the software implementation, embodiments such as the procedures and functions described in this specification may be implemented as separate software modules. remind

According to some embodiments of the present disclosure, the user terminal 200 may be related to a user who uses a customized service provided by the server 100 . Here, the user may be a user who regularly uses public transportation. However, the present invention is not limited thereto.

The user terminal 200 is, for example, a mobile phone, a smart phone, a laptop computer, a personal digital assistant (PDA), a slate PC, a tablet PC (tablet PC), an ultrabook ( ultrabook), and a wearable device (eg, a watch-type terminal (smartwatch), a glass-type terminal (smart glass), and a head mounted display (HMD)). However, the present invention is not limited thereto.

The user terminal 200 may include a processor 210 , a communication unit 220 , and a memory 230 . However, since the above-described components are not essential for implementing the user terminal, the user terminal 200 may have more or fewer components than those listed above. Here, each component may be configured as a separate chip, module, or device, or may be included in one device.

The processor 210 of the user terminal 200 typically controls the overall operation of the user terminal 200 . The processor 210 may provide or process appropriate information or functions to the user by processing signals, data, information, etc. input or output through the above-described components or by driving an application program stored in the memory 230 .

In addition, the processor 210 may control at least some of the components of the user terminal 200 in order to drive an application program stored in the memory 230 . Furthermore, the processor 210 may operate by combining at least two or more of the components included in the user terminal 200 to drive the application program.

The communication unit 220 of the user terminal 200 may include one or more modules that enable communication between the user terminal 200 and the user terminal 200, and between the user terminal 200 and the user terminals of other users. can In addition, the communication unit 220 may include one or more modules for connecting the user terminal 200 to one or more networks.

The communication unit 220 of the user terminal 200 according to some embodiments of the present disclosure may communicate with the user terminal 200 by wire or wirelessly. In addition, the communication unit 220 may mutually transmit and receive control data for controlling the user terminal 200 with the user terminal 200 .

The memory 230 of the user terminal 200 may store a program for the operation of the processor 210 , and may temporarily or permanently store input/output data. The memory 230 may include a flash memory type, a hard disk type, a multimedia card micro type, a card type memory (eg, SD or XD memory), and a RAM. (Random Access Memory, RAM), SRAM (Static Random Access Memory), ROM (Read-Only Memory, ROM), EEPROM (Electrically Erasable Programmable Read-Only Memory), PROM (Programmable Read-Only Memory), magnetic memory, magnetic It may include at least one type of storage medium among a disk and an optical disk. The memory 230 may be operated under the control of the processor 210 .

According to some embodiments of the present disclosure, the user terminal 200 may transmit ride data to the server 100 in order to receive a user-customized service provided by the server 100 . Here, the riding data may include information about the card number, information on the riding time, information on the riding day, and information on the fare. However, the present invention is not limited thereto.

For example, the user terminal 200 may be a smart phone with a built-in traffic card function. In this case, when the user tags the user terminal 200 to the transportation card terminal provided to pay the public transportation fee, the user terminal 200 may transmit the riding data to the server 100 . However, the present invention is not limited thereto.

As another example, the user terminal 200 may be a transportation card. In this case, when the user tags the transportation card to the transportation card terminal provided to pay the public transportation fee, the transportation card terminal may transmit the ride data to the server 100 . Meanwhile, the server 100 receiving the ride data may recognize the card number of the tagged transportation card, recognize the user, and transmit customized service information to the user terminal 200 of the user. However, the present invention is not limited thereto.

The network 300 according to some embodiments of the present disclosure is a Public Switched Telephone Network (PSTN), x Digital Subscriber Line (xDSL), Rate Adaptive DSL (RADSL), Multi Rate DSL (MDSL), VDSL ( A variety of wired communication systems such as Very High Speed DSL), Universal Asymmetric DSL (UADSL), High Bit Rate DSL (HDSL), and Local Area Network (LAN) can be used.

In addition, the network 300 presented here is CDMA (Code Division Multi Access), TDMA (Time Division Multi Access), FDMA (Frequency Division Multi Access), OFDMA (Orthogonal Frequency Division Multi Access), SC-FDMA (Single Carrier-) A variety of wireless communication systems may be used, such as FDMA) and other systems.

The network 300 according to embodiments of the present disclosure may be configured regardless of its communication aspects, such as wired and wireless, and may be configured in various ways, such as a local area network (LAN), a wide area network (WAN) It may consist of a communication network. In addition, the network may be a well-known World Wide Web (WWW), and may use a wireless transmission technology used for short-range communication such as Infrared Data Association (IrDA) or Bluetooth.

The techniques described herein may be used in the networks mentioned above, as well as in other networks.

2 is a flowchart illustrating an example of a destination prediction technique according to some embodiments of the present disclosure.

Referring to FIG. 2 , the processor 110 of the server 100 may receive ride data from the user terminal 200 through the communication unit 120 ( S110 ). Here, the riding data may include information about the card number, information on the riding time, information on the riding day, and information on the fare. However, the present invention is not limited thereto.

Specifically, the processor 110 may receive the ride data from the user terminal 200 when the user tags the transportation card to the transportation card terminal provided to pay the public transportation fee. That is, the user terminal 200 for transmitting the ride data to the server 100 may be a smart phone with a built-in traffic card function. However, the present invention is not limited thereto.

Meanwhile, the processor 110 of the server 100 may recognize the user of the user terminal 200 based on the ride data (S120). Specifically, the processor 110 may recognize the user of the user terminal 200 by using the card number included in the ride data. Additionally, the processor 110 may recognize the user of the user terminal 200 by recognizing the user terminal 200 that has transmitted the riding data. However, the present invention is not limited thereto.

The processor 110 of the server 100 may recognize a base recognition model related to a user from among a plurality of base recognition models stored in the memory 130 . Then, the processor 110 may input the boarding point information included in the boarding data to the recognized base recognition model to obtain base information related to the boarding data ( S130 ). Here, the base information may include information about the base group to which the boarding point belongs, and information about the characteristics of the base group to which the boarding point belongs. However, the present invention is not limited thereto.

On the other hand, the base recognition model may be built by the base recognition model building steps executed through the processor 110 . Specifically, the base recognition model may obtain at least two groups by extracting the first boarding point and the last getting off point from the user's public transportation use data, and clustering the plurality of extracted points. In addition, the characteristics of each of the at least two groups may be recognized using a preset rule.

Hereinafter, a method for the processor 110 of the server 100 to obtain user's base information (or a method for constructing a base recognition model) will be described later with reference to FIG. 5 .

The processor 110 of the server 100 may recognize a destination prediction model related to a user from among a plurality of destination prediction models stored in the memory 130 . Then, the processor 110 may obtain the user's alighting data by inputting the base information and the boarding data to the destination prediction model (S140). Here, the boarding data input to the destination prediction model may include boarding point information, time information, and day information. However, the present invention is not limited thereto.

According to some additional exemplary embodiments of the present disclosure, the processor 110 of the server 100 may recognize a disembarkation point included in the user's disembarkation data. In addition, the processor 110 may transmit customized service information to the user terminal 200 based on the disembarkation point. Here, the customized service information may include store information including at least one of advertisement information or discount information received from at least one store existing within a preset distance from the drop off point, or received from an administrative agency related to an area including the drop off point. It may include at least one of local information and preset marketing information.

For example, the processor 110 of the server 100 may transmit, to the user terminal 200 , store information including at least one of advertisement information or discount information received from a cafe existing within a preset distance from the disembarkation point. . Here, when the character of the first group is a work place, the store information may include a discount coupon available at a restaurant near the drop off point and a coffee coupon available at a cafe near the Hachi point. However, the present invention is not limited thereto.

As another example, the processor 110 of the server 100 may transmit the area information received from the city hall, which is an administrative agency related to the area including the alighting point, to the user terminal 200 . Here, the regional information may include information on trends in the occurrence of infectious diseases in the region, information on traffic control zones in the region, and information on requesting questionnaires for residents of the region. However, the present invention is not limited thereto.

As another example, the processor 110 of the server 100 may transmit the marketing information received from the advertiser company terminal or the server to the user terminal 200 in response to the drop off point. Here, the marketing information may include various advertisements that the advertiser wants to advertise. For example, the marketing information may be information for advertising real estate around a disembarkation point. However, the present invention is not limited thereto.

As described above in the description of FIG. 1, the destination prediction model uses a training data set including training data generated by labeling user-related training alight data to user-related training ride data in the learning process of the destination prediction model. can be learned by Here, the train data for learning may include first location information and first station information, and the alight data for learning may include second location information and second station information.

According to some embodiments of the present disclosure, the destination prediction model may be pre-stored in the memory 130 of the server 100 for each user as described above. That is, the destination prediction model of the present disclosure may be an individual user model.

As described above, the learning time of the destination prediction model generated individually for each user may be shortened compared to the destination prediction model generated for each user group. Also, the size (capacity) of the destination prediction model generated for each user may be smaller than that of the destination prediction model generated for each user group. Also, the accuracy of the destination prediction model generated for each user may be higher than that of the destination prediction model generated for each user group.

Hereinafter, the accuracy, size, learning time, and high-accuracy section share of each destination prediction model created for each user and destination prediction model created for each user group will be described below with reference to FIGS. 3 and 4 .

In general, depending on the type of public transportation, there may be stations having the same location information and different station information.

For example, location information of 'Near Gangnam Station' is the same depending on the type of public transportation such as subway, intercity bus, city bus, and village bus. Station information may be different, such as 'Gangnam Station'.

Meanwhile, when the server 100 recognizes a specific station frequently used by a user among a plurality of stations having the same location information, it may be easy to provide a user-customized service.

Therefore, before the destination prediction model is trained, the processor 110 of the server 100 determines that the first location information among the training ride data is the same and the first station information has the largest number of appearances in the different unprocessed boarding data. 1 Can recognize the most frequent station information. Then, the processor 110 may change the first station information of the unprocessed boarding data to the first most frequent station information.

In addition, the processor 110 may recognize the information of the second most appearing station having the largest number of appearances in the unprocessed alighting data having the same second location information and different second station information among the alighting data for learning. Then, the processor 110 may change the second station information of the unprocessed getting off data to the second most frequent station information.

Here, each of the first location information and the second location information may be information about a radius on a map including public transportation stations set to have a specific name (eg, Gangnam Station). In addition, the first station information may include information on the specific name and information on the type of public transportation having the specific name. However, the present invention is not limited thereto.

That is, the processor 110 of the server 100 may recognize a station frequently used by the user through the first most frequent station information and the second most frequent station information.

Accordingly, the server 100 according to some embodiments of the present disclosure may provide a customized service based on a station frequently used by a user.

3 and 4 are tables for explaining a destination prediction model according to some embodiments of the present disclosure.

According to some embodiments of the present disclosure, the processor 110 of the server 100 may train a destination prediction model using public transportation data.

Specifically, the processor 110 may train a destination prediction model by using individual public transportation use data. In addition, the processor 110 may train a destination prediction model by using public transportation use data for each of a plurality of user groups.

Referring to FIG. 3 , it is a table showing the accuracy, size, and learning time of each destination prediction model learned for each individual user or each user group by time period.

The processor 110 of the server 100 may train the destination prediction model for each 6-hour division group, 3-hour division group, and individual. Here, each of the 6-hour division group and the 3-hour division group may be a target that groups users with similar moving patterns and is considered for generating a destination prediction model. For example, a 6-hour division group divides a 24-hour day into 04-10 o'clock, 10-16 o'clock, 16-22 o'clock, and 22-04 o'clock, can In addition, the 3-hour division group has a 24-hour period of 04~07, 07~10, 10~13, 13~16, 16~19, 19~22:00, 22~01, and 01~04. It may be a group of users having a similar movement pattern for each divided time period. However, the present invention is not limited thereto.

Referring to the table shown, for the destination prediction model trained as a 6-hour division group, the maximum value of accuracy was 85.71%, the minimum value was 19.01%, the average value was 37.92%, and the median value was 37.59%. In addition, for the destination prediction model trained in the 3-hour division group, the maximum value of accuracy was 100%, the minimum value was 0%, the average value was 40.22%, and the median value was 38.69%. And, for the destination prediction model trained for each individual, the maximum value of accuracy was 100%, the minimum value was 0%, the average value was 44.2%, and the median value was 44.44%.

Specifically, referring to FIG. 4 , it is a table showing the number of distributions of users for each accuracy section of each destination prediction model learned for each individual user or for each user group by time period. Here, the total number of users to be compared for prediction accuracy is about 930,000.

Looking at the distribution ratio of users having an accuracy of 70 to 100 for which the accuracy of prediction is judged to be effective in the table shown, for the destination prediction model trained as a 6-hour division group, 70 to 100 of about 930,000 users There were about 97,000 users with accuracy, and a rate of 10.38% was derived. In addition, with respect to the destination prediction model trained in the 3-hour division group, about 116,000 users with an accuracy of 70 to 100 among about 930,000 users, a ratio of 12.3% was derived. In addition, with respect to the destination prediction model trained for each individual, the number of users having an accuracy of 70 to 100 among about 930,000 users was about 172,000, and a ratio of 18.21% was derived.

That is, the destination prediction model trained for each individual used by the processor 110 of the server 100 according to some embodiments of the present disclosure has relatively higher accuracy than other models, and the accuracy distribution ratio determined to be valid is also a different model. relatively larger than

Referring back to Figure 3, for the destination prediction model trained as a 6-hour division group, the size of the model has 0.15G to 20G for each group (that is, in the case of the entire group (1448), the size of 217G to 2896G ), and the learning time is 13 hours. On the other hand, with respect to the destination prediction model trained for each individual, the total sum of the sizes of each individual model is 86G, and the total learning time of each individual model is 56 minutes.

That is, a destination prediction model trained in a 6-hour division group or a 3-hour division group requires a long learning time and a large capacity, and thus may not be suitable for a real-time service. On the other hand, the destination prediction model trained for each individual used by the server 100 according to some embodiments of the present disclosure has a shorter learning time than other models and occupies a relatively small capacity in the memory 130, so it is more suitable for a real-time service. may be suitable.

5 is a flowchart for explaining an example of a method of acquiring at least two node groups according to some embodiments of the present disclosure. 6 is a view for explaining at least two base groups according to some embodiments of the present disclosure. 7 is a diagram for explaining an example of clustering the final points of the unsupervised learning model according to some embodiments of the present disclosure.

Referring to FIG. 5 , the processor 110 of the server 100 may recognize a plurality of boarding points and a plurality of alighting points by using public transportation usage data related to the user ( S210 ). Here, the public transportation usage data related to the user includes at least one of boarding point information, boarding time information, boarding day information, transfer point information, transfer time information, transfer day information, alighting point information, alighting time information, and alighting day information can do. However, the present invention is not limited thereto.

The processor 110 of the server 100 may recognize at least one initial boarding point and at least one final getting off point from the plurality of boarding points and the plurality of alighting points except at least one transfer point (S220). Then, the processor 110 may obtain at least two groups by clustering at least one initial boarding point and at least one final getting off point through an unsupervised learning model (S230).

Referring to FIG. 6 , as shown, the processor 110 of the server 100 uses an unsupervised learning model to cluster the final points on the map 10 to form a plurality of base groups 11 to 14 . can recognize

The unsupervised learning model of the present disclosure may calculate a distance for each of the combinable point pairs among at least one initial boarding point and at least one final getting off point. In addition, the unsupervised learning model may use the calculated distances to cluster at least one initial boarding point and at least one final getting off point according to a density-based spatial clustering technique.

Specifically, the unsupervised learning model of the present disclosure may select any of the final points. And, the unsupervised learning model may cluster the final points into at least two base groups using a selected arbitrary point, a critical radius, and the minimum number of final points that may belong to the same base group. Here, the threshold radius and the minimum number may be determined by the processor 110 , preset and stored in the memory 130 , or may be received from the manager terminal through the communication unit 120 .

Hereinafter, it will be described in more detail with reference to each drawing of FIG. 7 . Here, FIG. 7 shows an example in which one base group is clustered for convenience of description.

First, referring to FIG. 7A , the unsupervised learning model may select an arbitrary first point 21 among final points. Also, the unsupervised learning model may recognize whether the number of other final points existing at a distance shorter than a critical radius (eps: epsilon) from the first point 21 is equal to or greater than the minimum number.

In the unsupervised learning model, when the number of other final points 22 to 25 existing at a distance shorter than the critical radius (eps) with respect to the first point 21 is at least the minimum number (eg, three) , the first point 21 may be designated as the first core point of the first sub-base group. In addition, the unsupervised learning model may designate other final points 22 to 25 existing at a distance shorter than a critical radius from the first core point as at least one border point. In addition, the unsupervised learning model may cluster the first core point and at least one border point into the first sub base group.

Meanwhile, in the unsupervised learning model, among the points included in the first sub-base group, the number of other final points existing at a distance shorter than the critical radius (eps) based on each of the remaining points other than the first core point It can be recognized whether is greater than or equal to the minimum number.

For example, referring to FIG. 7B , the unsupervised learning model is shorter than the critical radius (eps) with respect to the second point 24 or the third point 25 included in the first sub-base group. It can be recognized that the number of other final points existing at a distance by a distance is two. That is, in the unsupervised learning model, the number of other final points existing at a distance shorter than the critical radius (eps) with respect to the second point 24 or the third point 25 is the minimum number (e.g., Assuming 3), less than 3 can be recognized. In this case, the unsupervised learning model may designate the second point 24 or the third point 25 as a border point.

That is, in the border point of the present disclosure, the number of other points existing within a distance shorter than the critical radius from among the final points is less than the minimum number, and exists within a distance shorter than the critical radius from any one of the at least one core point. If so, it can be specified.

As another example, referring to (c) of FIG. 7 , the unsupervised learning model is less than the critical radius (eps) with respect to the fourth point 22 or the fifth point 23 included in the first sub-base group. It can be recognized that the number of other final points existing at a location separated by a short distance is three. In the unsupervised learning model, the number of other final points existing at a distance shorter than the critical radius (eps) with respect to the fourth point 22 or the fifth point 23 is the minimum number (for example, three Assuming that ), an abnormality can be recognized. In this case, the unsupervised learning model may designate the fourth point 22 or the fifth point 23 as the core point.

Specifically, the unsupervised learning model may designate the fourth point 22 as the second core point of the second sub-base group. In addition, the unsupervised learning model may designate other final points existing at a distance shorter than a critical radius from the second core point as at least one border point. In addition, the unsupervised learning model may cluster the second core point and at least one border point into a second sub base group.

In addition, the unsupervised learning model may designate the fifth point 23 as the third core point of the third sub-base group. In addition, the unsupervised learning model may designate other final points existing at a distance shorter than a critical radius from the third core point as at least one border point. In addition, the unsupervised learning model may cluster the third core point and at least one border point into a third sub base group.

That is, the core point of the present disclosure may be designated when the number of other points existing within a distance shorter than the critical radius with respect to itself among the final points is equal to or greater than the minimum number.

According to some embodiments of the present disclosure, each of the first sub-node group, the second sub-base group, and the third sub-base group having a point that is shared with each other (a core point or a border point) may be clustered into one base group. . That is, since the unsupervised learning model of the present disclosure connects clusters to each other according to the density of the clusters, clusters having a geometric shape can also be found. However, the present invention is not limited thereto.

Meanwhile, referring to FIG. 7D , the unsupervised learning model may designate a noise point.

Specifically, the unsupervised learning model may select the sixth point 26 not included in the base group among the final points. And, the unsupervised learning model recognizes that the number of other final points existing at a distance shorter than the critical radius (eps) from the sixth point 26 is less than the minimum number, and the sixth point 26 is It may be recognized that the at least one core point is at a distance greater than a critical radius (eps) from each of the points. In this case, the unsupervised learning model may designate the sixth point 26 as a noise point.

That is, the noise point of the present disclosure is not included in at least two base groups among the final points, and the number of other points existing within a distance shorter than the critical radius with respect to itself is less than the minimum number, and each of at least one core point It may be a point that exists at a distance greater than a critical radius from .

Referring to FIG. 7E , an example of the base group 30 clustered by the unsupervised learning model of the present disclosure is shown.

As described above, the base group 30 may be composed of a first sub base group, a second sub base group, and a third sub base group, each having a common point (a core point or a border point).

That is, the base group 30 may include at least one core point 31 of each of the sub base groups and at least one border point 32 of each of the sub base groups. Meanwhile, the noise point 33 designated by the unsupervised learning model may not be included in the base group 30 .

The unsupervised learning model of the present disclosure may cluster at least two base groups based on the user's getting on and off points through the above-described method. Here, at least two bases may be used when the server 100 predicts the user's destination or provides user-customized information to the user.

On the other hand, the characteristics of each of the first base group 11, the second base group 12, the third base group 13 and the fourth base group 14 can be determined through a rule set by the server 100. have.

Referring back to FIG. 5 , the processor 110 of the server 100 may recognize the characteristics of each of the at least two base groups using a preset rule ( S240 ).

Specifically, the processor 110 may calculate a score related to a preset personality for each of the at least two base groups. In addition, the processor 110 may recognize the characteristics of each of the at least two base groups based on the calculated score.

Additionally, the processor 110 may store the recognized personality in the memory 130 by tagging each of the at least two base groups.

According to some additional embodiments of the present disclosure, the processor 110 of the server 100 may transmit customized service information to the user terminal 200 based on the characteristics of each of the at least two base groups. Here, the customized service information includes store information including at least one of advertisement information or discount information received from at least one store existing within a preset distance from each of at least two specific points included in each of the at least two base groups, or at least It may include at least one of regional information or preset marketing information received from an administrative agency related to a region including each of the two specific points.

For example, when the character of the base group is a residence, the processor 110 may transmit customized service information related to the residence to the user terminal 200 . Specifically, the processor 110 may transmit customized service information including information that can be usefully used in a residence to the user terminal 200 . More specifically, the processor 110 may transmit regional information received from the city hall, which is an administrative agency in which a residence is located, to the user terminal 200 . Here, the regional information may include information on trends in the occurrence of infectious diseases in the region, information on traffic control zones in the region, and information on requesting questionnaires for residents of the region. However, the present invention is not limited thereto.

For another example, when the character of the base group is a work place, the processor 110 may transmit customized service information related to the work place to the user terminal 200 . Specifically, the processor 110 may transmit customized service information including information that can be usefully used in the workplace to the user terminal 200 . More specifically, the processor 110 may transmit, to the user terminal 200 , store information including at least one of advertisement information or discount information received from a cafe existing within a preset distance from the work place. Here, when the character of the first group is a work place, the store information may include a discount coupon available at a restaurant near the drop off point and a coffee coupon available at a cafe near the Hachi point. That is, the processor 110 may provide the user with customized service information including shop information related to a cafe that the user frequently uses at work. However, the present invention is not limited thereto.

That is, the processor 110 of the server 100 according to some embodiments of the present disclosure may provide customized service information including information usefully available to the user in the base group based on the nature of the base group of the user. have.

Hereinafter, a description of a method for the processor 110 of the server 100 to recognize the characteristics of each of at least two groups according to some embodiments of the present disclosure will be described later with reference to FIGS. 8 and 9 .

According to some embodiments of the present disclosure, as described above in step S130 of FIG. 2 , the processor 110 may obtain the base information related to the boarding data by inputting the boarding point information into the recognized base recognition model. have. Here, the base recognition model may be built or executed by the above-described steps ( S210 to S240 ) executed through the processor 110 .

Specifically, when boarding point information is input to the base recognition model, the processor 110 may recognize which group the boarding point information belongs to among at least two base groups through the base recognition model. Then, the processor 110 may acquire the base information by recognizing the characteristics of the group to which the boarding point information belongs.

8 and 9 are diagrams for explaining a method of recognizing the characteristics of at least two base groups according to some embodiments of the present disclosure.

First, referring to FIG. 8 , the processor 110 of the server 100 generates, for each of at least two base groups, a first score related to a first personality and a second score related to a second personality different from the first personality. can be calculated (S310).

Specifically, the processor 110 analyzes the group-by-group public transportation use information for each of the at least two base groups to recognize the first boarding point, the first boarding time, the last getting off point, and the last getting off time of the group-by-group public transportation usage information. can And, the processor 110, based on whether the first boarding point, the first boarding time, the last getting off point, and the last getting off time of the group-by-group public transportation use information satisfies a preset condition, the first of each of the at least two base groups A preset value may be added to the score.

In addition, the processor 110 analyzes the group-by-group public transportation use information for each of the at least two base groups to recognize the intermediate boarding point, the intermediate boarding time, the intermediate getting off point, and the intermediate getting off time of the group-by-group public transportation use information. have. Then, the processor 110 adds a preset value to the second score of each of the at least two base groups based on whether the intermediate boarding point, the intermediate boarding time, the intermediate getting off point, and the intermediate getting off time satisfy preset conditions. can do.

For example, a method in which the processor 110 of the server 100 adds a preset value to the scores of each of at least two base groups will be described with reference to FIG. 9 as follows.

First, referring to FIG. 9A , the processor 110 of the server 100 may calculate a first score related to the first personality by using information related to the first boarding and the final getting off at a specific point. . Here, the first personality may be, for example, a residence. However, the present invention is not limited thereto.

For example, when the first boarding point included in group A and the last getting off point included in group D are the same, the processor 110 of the server 100 may add a first score related to the first characteristics of group A and group D. A preset value of 1 may be added. In this case, the processor 110 may ignore the B and C groups including the intermediate getting off point and the intermediate boarding point. Here, the processor 110 adds 3, which is three times the preset value of 1, to the first score related to the first characteristics of group A and group D when the same condition is satisfied for group A and group D for 3 days. can Also, group A and group D may be the same group. However, the present invention is not limited thereto.

For another example, the processor 110 of the server 100 boards at the first boarding point included in group A, stays at another point included in group B for a preset time (eg, 4 hours), and then in group A When disembarking at the first boarding point included in , a preset value of 1 may be added to the first score related to the first characteristic of group A. Here, when the same condition is satisfied for the group A for 2 days, the processor 110 may add 2, which is twice the preset value of 1, to the first score related to the first characteristic of the group A. However, the present invention is not limited thereto.

For another example, the processor 110 of the server 100 boards at the first boarding point included in group A before a preset time (eg, 10:00 am), and then a preset time (eg, 2 hours) If you get off at another point included in group B within the period, a preset value of 1 may be added to the first score related to the first characteristic of group A. Here, when the same condition is satisfied for the group A for 5 days, the processor 110 may add 5, which is 5 times the preset value of 1, to the first score related to the first characteristic of the group A. However, the present invention is not limited thereto.

Referring to FIG. 9B , the processor 110 of the server 100 may calculate a second score related to the second personality by using information related to intermediate boarding and intermediate getting off at a specific point. Here, the second personality may be, for example, a place of work. However, the present invention is not limited thereto.

For example, the processor 110 of the server 100 alights at the intermediate getting off point included in group B, and the time interval at which boarding at the intermediate boarding point included in group C is a preset time (eg, 4 hours) or more In this case, a preset value of 1 may be added to the second score related to the second characteristics of group B and group C. Here, the processor 110 may add 3, which is three times the preset value 1, of the group B and the group C when the same condition is satisfied for the group B and the group C for 3 days. Also, group B and group C may be the same group. However, the present invention is not limited thereto.

For another example, in the processor 110 of the server 100, the travel time from the intermediate boarding point included in group C to the final disembarkation point included in group D is within a preset time (eg, 2 hours), and the final When the get-off time at the Hachi point is after a preset time (eg, 5:00 pm), a preset value of 1 is added to the second score related to the second character of group C, and the second score related to the first character of group D is added. A preset value of 1 may be added to the 1 score. However, the present invention is not limited thereto.

As another example, in the processor 110 of the server 100, the movement time from the first boarding point included in group A to the intermediate getting off point included in group B is within a preset time (eg, 2 hours), When the boarding time at the first boarding point is before a preset time (eg, 10:00 am), a preset value of 1 is added to the second score related to the second character of group B, and the value related to the first character of group A is added. A preset value of 1 may be added to the first score. However, the present invention is not limited thereto.

Referring back to FIG. 8 , the processor 110 of the server 100 recognizes the characteristics of each of the two base groups based on the size of the first score and the size of the second score of each of the at least two base groups. can be (S320).

Specifically, the processor 110 may recognize a group in which the magnitude of the first score exceeds a preset value and the magnitude of the second score does not exceed the preset value as the first personality (eg, residence). In addition, the processor 110 may recognize a group in which the size of the first score does not exceed the preset value and the size of the second score exceeds the preset value as the second personality (eg, work place). In addition, the processor 110 may recognize a group in which the size of the first score and the size of the second score do not exceed a preset value as a third characteristic (eg, a base other than a residence or work place).

Meanwhile, when there is an unprocessed group having the same size of the first score and the size of the second score, the processor 110 may determine the nature of the unprocessed group as a preset characteristic.

For example, when the first score and the second score of the specific group are the same, the processor 110 may determine the character of the specific group as a residence. As another example, when the first score and the second score of a specific group are the same, the characteristic of the specific group may be determined as a work place.

That is, with respect to a specific group, the processor 110 may give priority to a score related to a specific characteristic. However, the present invention is not limited thereto.

On the other hand, when there is an unprocessed group in which the size of the first score and the size of the second score each exceed a preset value, the processor 110 may recognize a score having a larger size among the first score and the second score. have. In addition, the characteristics of the untreated group may be determined by the characteristics related to the large score. However, the present invention is not limited thereto.

As described above, the size of each of the first score and the second score is compared with a value preset by the processor 110 of the server 100 and may be used as a value for determining the nature of the base group. Also, there may be a third score for judging a third personality different from the first personality and the second personality related to the base group. However, the present invention is not limited thereto.

Accordingly, the processor 110 of the server 100 according to some embodiments of the present disclosure recognizes the personality corresponding to each of the at least two base groups through the scores for each different personality, and more accurately identifies the personality of each group. can recognize

In addition, the processor 110 may provide a personalized service by using the recognized character of the base group. For example, the processor 110 may provide different services depending on whether the point where the user is expected to get off is a residence or a work place.

According to some additional embodiments of the present disclosure, the processor 110 of the server 100 may recognize a specific base group as a third characteristic (eg, a base other than a place of residence or work).

In this case, the processor 110 may subdivide and recognize the characteristics of the base group. For example, a base group having a third characteristic may be recognized as a weekday base, a weekend base, or a base for each day of the week.

For a detailed example, with respect to the base group having the third characteristic, the processor 110 may recognize a base frequently visited by the user on weekdays as a weekday base. Also, with respect to the base group having the third characteristic, a base frequently visited by the user on a weekend may be recognized as a weekend base. However, the present invention is not limited thereto.

Accordingly, the processor 110 of the server 100 according to some additional exemplary embodiments of the present disclosure may provide a user with a customized service tailored to the date (ie, weekday, weekend, or day of the week) visited by the user.

10 is a flowchart illustrating an example of a method of transmitting customized service information to a user terminal according to some embodiments of the present disclosure.

Referring to FIG. 10 , the processor 110 of the server 100 may recognize a disembarkation point included in the user's disembarkation data ( S410 ).

Here, the alighting data may be data obtained by the processor 110 inputting base information and boarding data into the destination prediction model as described above in step S140 of FIG. 2 . In addition, the disembarkation data may include a disembarkation point, a base group including the disembarkation point, and characteristics of the base group. However, the present invention is not limited thereto.

The processor 110 of the server 100 may recognize a first group related to a drop off point among at least two base groups (S420).

Here, at least two base groups are obtained by the processor 110 clustering at least one initial boarding point and at least one final getting off point through an unsupervised learning model, as described above in step S230 of FIG. 3 . It can be one group.

The processor 110 of the server 100 may control the communication unit 120 to transmit customized service information to the user terminal 200 based on the characteristics of the first group. Here, the customized service information includes store information including at least one of advertisement information or discount information received from at least one store existing within a preset distance from the drop off point, or an area received from an administrative agency related to an area including the drop off point. It may include at least one of information or preset marketing information. However, the present invention is not limited thereto.

For example, the processor 110 of the server 100 may transmit, to the user terminal 200 , store information including at least one of advertisement information or discount information received from a cafe existing within a preset distance from the disembarkation point. . Here, when the character of the first group is a work place, the store information may include a discount coupon available at a restaurant near the drop off point and a coffee coupon available at a cafe near the Hachi point. However, the present invention is not limited thereto.

As another example, the processor 110 of the server 100 may transmit the area information received from the city hall, which is an administrative agency related to the area including the alighting point, to the user terminal 200 . Here, the regional information may include information on trends in the occurrence of infectious diseases in the region, information on traffic control zones in the region, and information on requesting questionnaires for residents of the region. However, the present invention is not limited thereto.

As another example, the processor 110 of the server 100 may transmit the marketing information received from the advertiser company terminal or the server to the user terminal 200 in response to the drop off point. Here, the marketing information may include various advertisements that the advertiser wants to advertise. For example, the marketing information may be information for advertising real estate around a disembarkation point. However, the present invention is not limited thereto.

11 is a flowchart illustrating a method of determining whether to use a destination prediction model according to some embodiments of the present disclosure.

Referring to FIG. 11 , the processor 110 of the server 100 may acquire alighting data for a preset period (eg, 7 days). In addition, the processor 110 may acquire actual alighting data that the user actually got off for during a preset period (eg, 7 days) ( S510 ).

The processor 110 of the server 100 may calculate the accuracy by comparing the disembarkation data obtained from the destination prediction model with the actual disembarkation data (S520).

For example, the processor 110 may compare the alighting data obtained through the destination prediction model for a preset period of 7 days with the actual alighting data received from the user terminal 200 . Specifically, the processor 110 may calculate the accuracy of the destination prediction model by dividing the number of times the first getting off point included in the getting off data and the second getting off point included in the actual getting off data match by the total number of getting off. For example, when the number of times that the first and second get-off points coincide with each other is 7 and the total number of get-offs is 10, 0.7 divided by 7 may be the accuracy of the destination prediction model.

Meanwhile, the processor 110 of the server 100 may determine whether to use the destination prediction model based on whether the accuracy is less than a preset value.

For example, when the accuracy is less than a preset value (S530, Yes), the processor 110 of the server 100 may determine not to use the destination prediction model associated with the user. Then, the processor 110 may collect the user's public transportation use data during a preset additional learning period in order to re-train the destination prediction model (S540).

Then, the processor 110 may re-train the destination prediction model by using the user's public transportation use data collected during the additional learning period. In addition, the processor 110 may not provide a service related to the destination prediction during the additional learning period in order to prevent an error in the destination prediction. In this case, the processor 110 may transmit a request signal to use public transportation to the user terminal 200 in order to collect public transportation use data. However, the present invention is not limited thereto.

As another example, when the accuracy is greater than or equal to a preset value (S530, No), the processor 110 of the server 100 may determine to continue using the destination prediction model related to the user (S550).

Accordingly, the server 100 according to some embodiments of the present disclosure may periodically manage the accuracy related to the destination prediction of the destination prediction model. In addition, the server 100 of the present disclosure can preserve the quality of service by periodically managing the accuracy related to the destination prediction of the destination prediction model.

12 depicts a simplified, general schematic diagram of an example computing environment in which embodiments of the present disclosure may be implemented.

Although the present disclosure has been described above generally in the context of computer-executable instructions that may be executed on one or more computers, those skilled in the art will appreciate that the present disclosure may be implemented in combination with other program modules and/or in a combination of hardware and software. will be.

Generally, modules herein include routines, procedures, programs, components, data structures, etc. that perform particular tasks or implement particular abstract data types. In addition, those skilled in the art will appreciate that the methods of the present disclosure are suitable for single-processor or multiprocessor computer systems, minicomputers, mainframe computers as well as personal computers, handheld computing devices, microprocessor-based or programmable consumer electronics, etc. (each of which is It will be appreciated that other computer system configurations may be implemented, including those that may operate in connection with one or more associated devices.

The described embodiments of the present disclosure may also be practiced in distributed computing environments where certain tasks are performed by remote processing devices that are linked through a communications network. In a distributed computing environment, program modules may be located in both local and remote memory storage devices.

Computers typically include a variety of computer-readable media. Any medium accessible by a computer can be a computer readable medium, and such computer readable media includes volatile and nonvolatile media, transitory and non-transitory media, removable and non-transitory media. including removable media. By way of example, and not limitation, computer-readable media may include computer-readable storage media and computer-readable transmission media.

Computer-readable storage media includes volatile and non-volatile media, temporary and non-transitory media, removable and non-removable media implemented in any method or technology for storage of information such as computer readable instructions, data structures, program modules or other data. includes media. A computer-readable storage medium may be RAM, ROM, EEPROM, flash memory or other memory technology, CD-ROM, digital video disk (DVD) or other optical disk storage device, magnetic cassette, magnetic tape, magnetic disk storage device, or other magnetic storage device. device, or any other medium that can be accessed by a computer and used to store the desired information.

Computer readable transmission media typically embodies computer readable instructions, data structures, program modules or other data in a modulated data signal such as a carrier wave or other transport mechanism, and Includes any information delivery medium. The term modulated data signal means a signal in which one or more of the characteristics of the signal is set or changed so as to encode information in the signal. By way of example, and not limitation, computer-readable transmission media includes wired media such as a wired network or direct-wired connection, and wireless media such as acoustic, RF, infrared, and other wireless media. Combinations of any of the above are also intended to be included within the scope of computer-readable transmission media.

An exemplary environment 1100 implementing various aspects of the disclosure is shown including a computer 1102 , the computer 1102 including a processing unit 1104 , a system memory 1106 , and a system bus 1108 . do. A system bus 1108 couples system components, including but not limited to system memory 1106 , to the processing device 1104 . The processing device 1104 may be any of a variety of commercially available processors. Dual processor and other multiprocessor architectures may also be used as processing unit 1104 .

The system bus 1108 may be any of several types of bus structures that may further interconnect a memory bus, a peripheral bus, and a local bus using any of a variety of commercial bus architectures. System memory 1106 includes read only memory (ROM) 1110 and random access memory (RAM) 1112 . A basic input/output system (BIOS) is stored in non-volatile memory 1110, such as ROM, EPROM, EEPROM, etc., which BIOS is the basic input/output system (BIOS) that helps transfer information between components within computer 1102, such as during startup. contains routines. RAM 1112 may also include high-speed RAM, such as static RAM, for caching data.

Computer 1102 may also include an internal hard disk drive (HDD) 1114 (eg, EIDE, SATA) - this internal hard disk drive 1114 may also be configured for external use within a suitable chassis (not shown). Yes—a magnetic floppy disk drive (FDD) 1116 (eg, for reading from or writing to removable diskette 1118), and an optical disk drive 1120 (eg, a CD-ROM) for reading from, or writing to, disk 1122, or other high capacity optical media such as DVD. The hard disk drive 1114 , the magnetic disk drive 1116 , and the optical disk drive 1120 are connected to the system bus 1108 by the hard disk drive interface 1124 , the magnetic disk drive interface 1126 , and the optical drive interface 1128 , respectively. ) can be connected to Interface 1124 for external drive implementation includes, for example, at least one or both of Universal Serial Bus (USB) and IEEE 1394 interface technologies.

These drives and their associated computer readable media provide non-volatile storage of data, data structures, computer executable instructions, and the like. For computer 1102, drives and media correspond to storing any data in a suitable digital format. Although the description of computer-readable storage media above refers to HDDs, removable magnetic disks, and removable optical media such as CDs or DVDs, those skilled in the art will use zip drives, magnetic cassettes, flash memory cards, cartridges, It will be appreciated that other tangible computer-readable storage media and the like may also be used in the exemplary operating environment and any such media may include computer-executable instructions for performing the methods of the present disclosure. .

A number of program modules may be stored in the drive and RAM 1112 , including an operating system 1130 , one or more application programs 1132 , other program modules 1134 , and program data 1136 . All or portions of the operating system, applications, modules, and/or data may also be cached in RAM 1112 . It will be appreciated that the present disclosure may be implemented in various commercially available operating systems or combinations of operating systems.

A user may enter commands and information into the computer 1102 via one or more wired/wireless input devices, for example, a pointing device such as a keyboard 1138 and a mouse 1140 . Other input devices (not shown) may include a microphone, IR remote control, joystick, game pad, stylus pen, touch screen, and the like. Although these and other input devices are often connected to the processing unit 1104 through an input device interface 1142 that is connected to the system bus 1108, parallel ports, IEEE 1394 serial ports, game ports, USB ports, IR interfaces, It may be connected by other interfaces, etc.

A monitor 1144 or other type of display device is also coupled to the system bus 1108 via an interface, such as a video adapter 1146 . In addition to the monitor 1144, the computer typically includes other peripheral output devices (not shown), such as speakers, printers, and the like.

Computer 1102 may operate in a networked environment using logical connections to one or more remote computers, such as remote computer(s) 1148 via wired and/or wireless communications. Remote computer(s) 1148 may be workstations, server computers, routers, personal computers, portable computers, microprocessor-based entertainment devices, peer devices, or other common network nodes, and are generally Although including many or all of the components described, only memory storage device 1150 is shown for simplicity. The logical connections shown include wired/wireless connections to a local area network (LAN) 1152 and/or a larger network, eg, a wide area network (WAN) 1154 . Such LAN and WAN networking environments are common in offices and companies, and facilitate enterprise-wide computer networks, such as intranets, all of which can be connected to a worldwide computer network, for example, the Internet.

When used in a LAN networking environment, the computer 1102 is connected to the local network 1152 through a wired and/or wireless communication network interface or adapter 1156 . Adapter 1156 may facilitate wired or wireless communication to LAN 1152 , which also includes a wireless access point installed therein for communicating with wireless adapter 1156 . When used in a WAN networking environment, the computer 1102 may include a modem 1158 , connected to a communication server on the WAN 1154 , or otherwise establishing communications over the WAN 1154 , such as over the Internet. have the means A modem 1158 , which may be internal or external and a wired or wireless device, is coupled to the system bus 1108 via a serial port interface 1142 . In a networked environment, program modules described for computer 1102 or portions thereof may be stored in remote memory/storage device 1150 . It will be appreciated that the network connections shown are exemplary and other means of establishing a communication link between the computers may be used.

Computer 1102 may be associated with any wireless device or object that is deployed and operates in wireless communication, for example, printers, scanners, desktop and/or portable computers, portable data assistants (PDAs), communication satellites, wireless detectable tags. It operates to communicate with any device or place, and phone. This includes at least Wi-Fi and Bluetooth wireless technologies. Accordingly, the communication may be a predefined structure as in a conventional network or may simply be an ad hoc communication between at least two devices.

Wi-Fi (Wireless Fidelity) makes it possible to connect to the Internet, etc. without a wire. Wi-Fi is a wireless technology such as cell phones that allows these devices, eg, computers, to transmit and receive data indoors and outdoors, ie anywhere within range of a base station. Wi-Fi networks use a radio technology called IEEE 802.11 (a, b, g, etc.) to provide secure, reliable, and high-speed wireless connections. Wi-Fi can be used to connect computers to each other, to the Internet, and to wired networks (using IEEE 802.3 or Ethernet). Wi-Fi networks may operate in unlicensed 2.4 and 5 GHz radio bands, for example, at 11 Mbps (802.11a) or 54 Mbps (802.11b) data rates, or in products that include both bands (dual band). have.

A person of ordinary skill in the art of the present disclosure will recognize that the various illustrative logical blocks, modules, processors, means, circuits and algorithm steps described in connection with the embodiments disclosed herein include electronic hardware, (convenience For this purpose, it will be understood that it may be implemented by various forms of program or design code (referred to herein as "software") or a combination of both. To clearly illustrate this interchangeability of hardware and software, various illustrative components, blocks, modules, circuits, and steps have been described above generally in terms of their functionality. Whether such functionality is implemented as hardware or software depends upon the particular application and design constraints imposed on the overall system. A person skilled in the art of the present disclosure may implement the described functionality in various ways for each specific application, but such implementation decisions should not be interpreted as a departure from the scope of the present disclosure.

The various embodiments presented herein may be implemented as methods, apparatus, or articles of manufacture using standard programming and/or engineering techniques. The term “article of manufacture” includes a computer program, carrier, or media accessible from any computer-readable device. For example, computer-readable storage media include magnetic storage devices (eg, hard disks, floppy disks, magnetic strips, etc.), optical disks (eg, CDs, DVDs, etc.), smart cards, and flash drives. memory devices (eg, EEPROMs, cards, sticks, key drives, etc.). The term “machine-readable medium” includes, but is not limited to, wireless channels and various other media capable of storing, retaining, and/or carrying instruction(s) and/or data.

It is understood that the specific order or hierarchy of steps in the presented processes is an example of exemplary approaches. Based on design priorities, it is to be understood that the specific order or hierarchy of steps in the processes may be rearranged within the scope of the present disclosure. The appended method claims present elements of the various steps in a sample order, but are not meant to be limited to the specific order or hierarchy presented.

The description of the presented embodiments is provided to enable any person skilled in the art to make or use the present disclosure. Various modifications to these embodiments will be readily apparent to those skilled in the art, and the generic principles defined herein may be applied to other embodiments without departing from the scope of the present disclosure. Thus, the present disclosure is not intended to be limited to the embodiments presented herein, but is to be construed in the widest scope consistent with the principles and novel features presented herein.

Claims (15)

In the method of extracting the user's base information performed by the processor of the computing device,
Recognizing a plurality of boarding points and a plurality of alighting points by using public transportation usage data related to the user;
recognizing at least one initial boarding point and at least one final getting off point excluding at least one transfer point from the plurality of boarding points and the plurality of alighting points; and
obtaining at least two base groups by clustering the at least one first boarding point and the at least one final getting off point through an unsupervised learning model;
containing,
How to extract user base information.
The method of claim 1,
The unsupervised learning model is
calculating a distance for each of the combinable point pairs among the at least one initial boarding point and the at least one final getting off point;
Clustering the at least one first boarding point and the at least one last getting off point according to a density-based spatial clustering technique (DBSCAN) using the calculated distances,
How to extract user base information.
The method of claim 1,
In order to cluster the at least one initial boarding point and the at least one final disembarkation point through the unsupervised learning model, the minimum number of points that can belong to the same base group or a threshold used when determining points belonging to the same base station determining at least one of the radii,
How to extract user base information.
4. The method of claim 3,
Each of the at least two base groups,
at least one core point designated when the number of other points existing within a distance shorter than the critical radius with respect to itself is equal to or greater than the minimum number; and
At least one border designated when the number of other points existing within a distance shorter than the critical radius with respect to itself is less than the minimum number, and within a distance shorter than the critical radius from any one of the at least one core point Point;
containing,
How to extract user base information.
5. The method of claim 4,
Each of the core points included in a different base group among the at least two base groups,
exist at a distance from each other greater than the critical radius,
How to extract user base information.
5. The method of claim 4,
A noise point among the at least one initial boarding point and the at least one final getting off point is not included in the at least two base groups,
The noise point is
The number of other points existing within a distance shorter than the critical radius with respect to itself is less than the minimum number, and a point existing at a distance greater than the critical radius from each of the at least one core point,
How to extract user base information.
The method of claim 1,
recognizing the characteristics of each of the at least two base groups by using the public transport use data corresponding to each of the at least two base groups and a preset rule; and
tagging each of the at least two base groups and storing the personality in a memory;
further comprising,
How to extract user base information.
8. The method of claim 7,
The step of recognizing the characteristics of each of the at least two base groups by using the public transport use data corresponding to each of the at least two base groups and a preset rule,
calculating, for each of the at least two base groups, a first score related to a first personality and a second score related to a second personality different from the first personality; and
recognizing the characteristics of each of the two base groups based on the magnitude of the first score and the magnitude of the second score of each of the at least two base groups;
containing,
How to extract user base information.
9. The method of claim 8,
calculating, for each of the at least two base groups, a first score related to a first personality and a second score related to a second personality different from the first personality,
analyzing the group-by-group public transportation use information for each of the at least two base groups to recognize the first boarding point, the first boarding time, the last getting off point, and the last getting off time of the group-by-group public transportation usage information;
The first score of each of the at least two base groups based on whether the first boarding point, the first boarding time, the last getting off point, and the last getting off time of the group-specific public transportation use information satisfy preset conditions adding a preset value to ;
analyzing the group-by-group public transportation use information for each of the at least two base groups to recognize an intermediate boarding point, an intermediate boarding time, an intermediate getting-off point, and an intermediate getting-off time of the group-by-group public transportation use information; and
adding a preset value to the second score of each of the at least two base groups based on whether the intermediate boarding point, the intermediate boarding time, the intermediate getting off point, and the intermediate getting off time satisfy preset conditions ;
containing,
How to extract user base information.
9. The method of claim 8,
Recognizing the characteristics of each of the two base groups based on the size of the first score and the size of the second score of each of the at least two base groups comprises:
A group in which the magnitude of the first score exceeds a preset value and the magnitude of the second score does not exceed a preset value is recognized as the first personality, and the magnitude of the first score does not exceed the preset value and a group in which the magnitude of the second score exceeds the preset value is recognized as the second personality, and the group in which the magnitude of the first score and the magnitude of the second score do not exceed the preset value is recognized as the third personality Recognizing as;
containing,
How to extract user base information.
9. The method of claim 8,
Recognizing the characteristics of each of the two base groups based on the size of the first score and the size of the second score of each of the at least two base groups comprises:
determining the nature of the untreated group as a preset characteristic when there is an untreated group having the same size of the first score and the size of the second score;
containing,
How to extract user base information.
The method of claim 1,
recognizing which group the boarding point information belongs to among the at least two base groups when the user's boarding point information is received;
acquiring base information by recognizing the characteristics of the group to which the boarding point information belongs; and
acquiring the user's alighting data by inputting the user's boarding point and the base information into a destination prediction model related to the user among a plurality of destination prediction models stored in a memory;
further comprising,
How to extract user base information.
13. The method of claim 12,
The destination prediction model is
In the learning process of the destination prediction model, it is learned using a training data set including training data generated by labeling the training getting off data related to the user to the training boarding data related to the user,
How to extract user base information.
13. The method of claim 12,
recognizing a disembarkation point included in the user's disembarkation data;
recognizing a first group associated with the disembarkation point among the at least two base groups; and
transmitting customized service information to a user terminal based on the characteristics of the first group;
further comprising,
How to extract user base information.
A computer program stored in a computer readable storage medium, comprising:
The computer program includes instructions for causing a processor of a server to perform the following steps, the steps comprising:
Recognizing a plurality of boarding points and a plurality of alighting points by using public transportation usage data related to the user;
recognizing at least one initial boarding point and at least one final getting off point excluding at least one transfer point from the plurality of boarding points and the plurality of alighting points; and
obtaining at least two base groups by clustering the at least one first boarding point and the at least one final getting off point through an unsupervised learning model;
containing,
A computer program stored on a computer-readable storage medium.

Images