Device and method

Abstract

A demand estimation device 10 comprises: a learning unit 12 that performs training on pieces of information related to the positions where a plurality of users are staying in a first area in which a prescribed service has already been introduced, and pieces of information related to degrees of matching with the service in association with each other, thereby constructing a training model 130 that outputs information indicating a degree of matching with the service on the basis of information related to a staying position; an estimation unit 14 that inputs, into the training model 130, pieces of information related to the positions where a plurality of users are staying in a second area to be estimated, thereby estimating pieces of information indicating degrees of matching with the service for the plurality of users in the second area; and an output unit 15 that outputs the pieces of information related to the estimation results by the estimation unit 14.

Description

Apparatus and method

[0001] One aspect of the present invention relates to an apparatus and a method.

[0002] Patent Literature 1 discloses a hobby and preference estimation device including: a history acquisition unit that acquires visit history data for a predetermined period including visited POI (Point Of Interest) candidates and categories of the visited POI candidates; a distribution information acquisition unit that acquires information regarding the distribution of categories of the visited POI candidates based on the acquired visit history data; and a hobby and preference estimation unit that estimates a user's hobby and preference based on the acquired information including at least information regarding the distribution of categories of the visited POI candidates. As such, conventionally, techniques for estimating a user's preferences using the user's location information are known.

[0003] WO2019 / 202791

[0004] For example, some services, such as apps, are first launched domestically and then launched overseas. In such cases, it is necessary to estimate whether the service will be accepted in overseas markets based on macro trends, survey results, the status of similar services, etc. However, this estimation method makes it difficult to accurately estimate demand overseas.

[0005] One aspect of the present invention has been made in view of the above circumstances, and aims to provide an apparatus and method that can accurately identify demand for a service in an area where the service is planned to be developed.

[0006] An apparatus according to one aspect of the present invention includes a learning unit that constructs a learning model that outputs information indicating the degree of match with a service based on the information related to the locations of stay of multiple users in a first region where a specified service has already been introduced, by learning by correlating information related to the locations of stay of multiple users with information related to the degree of match with the service; an estimation unit that estimates information indicating the degree of match with the service for multiple users in a second region, which is the region to be estimated, by inputting information related to the locations of stay of multiple users in the second region into the learning model; and an output unit that outputs information related to the estimation results by the estimation unit.

[0007] In such a device, a learning model is constructed that correlates and learns information related to the locations of multiple users in a first region where the service is available with information related to their match with the service, and outputs information indicating their match with the service from the information related to their locations. The learning model is constructed assuming a correlation between information related to the locations (e.g., places frequently visited by users) and their match with the service (e.g., service usage). For example, the relationship that "users who frequently watch sports games frequently use a baseball watching app" is likely to hold true in other regions (e.g., other countries). That is, if users who frequently watch sports games in Japan frequently use a baseball watching app, if the baseball watching app is deployed overseas, users who similarly frequently watch sports games are likely to use the baseball watching app. Therefore, by inputting information related to the locations of multiple users in a second region, which is the region to be estimated, information indicating their match with the service for the multiple users in the second region can be estimated with high accuracy. By outputting such an estimation result, it is possible to identify with high accuracy the demand for the service in the area where the service will be newly developed (second area).

[0008] According to one aspect of the present invention, it is possible to provide an apparatus and method that can accurately identify demand for a service in an area where the service is planned to be developed.

[0009] FIG. 1 is a diagram illustrating an overview of a demand estimation method according to this embodiment. FIG. 2 is a block diagram illustrating the functional configuration of a demand estimation device according to this embodiment. FIG. 3(a) is a table illustrating an example of GPS information, and FIG. 3(b) is a table illustrating an example of app usage information. FIG. 4(a) is a table illustrating an example of dwell time for each facility category, and FIG. 4(b) is a table illustrating an example of dwell time in areas estimated to be home and work, in addition to the information shown in FIG. 4(a). FIG. 5(a) is a table illustrating an example of facility categories intended for use only domestically, and FIG. 5(b) is a table illustrating an example of facility categories intended for overseas expansion. FIG. 6(a) is a table illustrating an example of user proportions for each facility category when sampling is performed without considering balance, and FIG. 6(b) is a table illustrating an example of user proportions for each facility category when sampling is performed with balance. FIG. 7 is a table illustrating setting of target flags according to service usage status. FIG. 8 is a diagram illustrating construction of a learning model. FIG. 9(a) is a table illustrating an example of data input to the learning model, and FIG. 9(b) is a table illustrating an example of output data from the learning model. Fig. 10 is a diagram illustrating an example of output according to an estimation result. Fig. 11 is a flowchart illustrating an example of a process for constructing a learning model. Fig. 12 is a flowchart illustrating an example of a demand estimation process. Fig. 13 is a diagram illustrating a hardware configuration of a demand estimation device.

[0010] DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS Hereinafter, embodiments of the present invention will be described in detail with reference to the accompanying drawings. In the description of the drawings, the same or equivalent elements are designated by the same reference numerals, and redundant description will be omitted.

[0011] First, an outline of the demand estimation method performed by the demand estimation device according to this embodiment will be described with reference to Fig. 1. Fig. 1 is a diagram for explaining the outline of the demand estimation method according to this embodiment. Fig. 1 shows an outline of the processing up to the construction of a learning model in the demand estimation method.

[0012] As shown in FIG. 1( a), the demand estimation device first acquires location information (latitude and longitude information) of multiple users in, for example, Japan (first region), specifically GPS information. The GPS information is continuously acquired, for example, by each of the multiple users' terminals (not shown) and transmitted to the demand estimation device. In the example shown in FIG. 1( a), GPS information indicating locations G1 to G5 is acquired. Such GPS information indicates that the users have transitioned from location G1 to location G2, G3, G4, and G5 in this order.

[0013] 1(b), the demand estimation device identifies, among the acquired positions G1 to G5, a position where the user has stayed for a certain period of time or more, in this case, only position G4, as a stay point to be used for learning. A stay point is not a position that the user has simply passed through, but is assumed to be a position where the user has visited with some purpose and spent time.

[0014] As shown in FIG. 1( c), the demand estimation device searches for surrounding POIs for each stay point. A POI here refers to a location that may be visited by a person. Furthermore, at least location information and a facility category (e.g., a sports facility, a convenience store, a school, etc.) are associated with each POI and set in advance. The demand estimation device identifies POIs surrounding each stay point and further identifies the facility category of the identified POI. In the example shown in FIG. 1( c), three POIs are identified as surrounding the stay point, and each facility category is identified as Category A, Category B, and Category C. Note that in the example shown in FIG. 1( c), only the POI in Category A is close to the stay point, while the other POIs in Categories B and C are far from the stay point. Therefore, it is estimated that the person stayed only at the POI in Category A.

[0015] As shown in Fig. 1(d), the demand estimation device generates a dwell time feature by linking and vectorizing the dwell times (or scores according to the dwell times) for the above-mentioned categories A, B, and C. In the example shown in Fig. 1(d), only the POI in category A is assigned a score of 10, and the other POIs are estimated to have short dwell times (or no dwell times) and are assigned a score of 0.

[0016] 1(e), the demand estimation device further estimates whether the area is near the home or the workplace based on information on the day of the week and the stay time, associates the stay time with the area, and adds it to the stay time feature amount. In the example shown in FIG. 1(e), the score of the stay time at home is set to 2, and the score of the stay time at the workplace is set to 10, and they are added to the stay time feature amount.

[0017] As shown in FIG. 1( f), the demand estimation device associates and learns, for each user, the dwell time feature (vector) and information related to the degree of match with a predetermined service, thereby constructing a machine learning model that outputs information indicating the degree of match with a service based on the dwell time feature. In the example shown in FIG. 1( f), information indicating whether or not a service (here, an app) is liked is exemplified as information indicating the degree of match with a service. In this way, the learning model constructed by the demand estimation device receives as input information related to the user's stay location (dwell time feature) and outputs, for that user, information indicating the degree of match with a predetermined service.

[0018] In the demand estimation method according to this embodiment, the above-described learning model is constructed in an area where a service has already been introduced (e.g., Japan). Then, in an area where the same service is planned to be rolled out in the future (e.g., overseas), the constructed learning model is used to estimate demand for the service. Specifically, information related to the locations of multiple overseas users (dwell time feature values) is input into the learning model, and information indicating the degree of match for the service in the overseas area is estimated, thereby estimating demand for the service. Below, detailed functions of the demand estimation device are described with reference to FIGS. 2 to 10.

[0019] FIG. 2 is a block diagram showing the functional configuration of a demand estimation device 10 according to this embodiment. The demand estimation device 10 includes a reception unit 11, a learning unit 12, a memory unit 13, an estimation unit 14, and an output unit 15. The demand estimation device 10 estimates the demand for a predetermined service in a certain area. The demand estimation device 10 performs two main processes: a model construction process for constructing a learning model 130 in an area (first area) where the service has already been introduced, and a demand estimation process for estimating demand using the learning model in an area (second area) where the service is planned to be deployed in the future. The first area and the second area may be distinct from each other, and their sizes and scales are not limited. In this embodiment, the first area is assumed to be Japan (first country) and the second area is assumed to be a foreign country (second country different from the first country). A service may be provided to a user and may be for-profit or non-profit. In this embodiment, an example is described in which the service is an app used on a smartphone or the like. The following describes the functions of each functional component in the model construction process and the demand estimation process.

[0020] (Model Building Process) The reception unit 11 acquires GPS information (location information) and usage information of an app (service) for which a model is to be built from multiple users in Japan (first region, first country). The reception unit 11 acquires information from communication devices (not shown), such as smartphones, carried by the multiple users. The GPS information is information related to the user's location and indicates the latitude and longitude of the user's location. The GPS information may be information measured by the communication device (not shown) carried by the user. The reception unit 11 continuously acquires the GPS information at predetermined time intervals. The app usage information is information related to the user's match with the app (service), such as information indicating the number of times the app is used and the amount of usage (amount charged for the app). The reception unit 11 periodically acquires the app usage information (e.g., once a week or once a day). Note that in this embodiment, an example is described in which the information related to the user's match with a service is app usage information; however, the information related to the match with a service may be any information indicating the user's level of interest in the service. For example, information relating to the degree of match with a service may be information indicating whether or not the user likes or needs the service.

[0021] 3(a) is a table showing an example of GPS information, and FIG. 3(b) is a table showing an example of app (service) usage information. As shown in FIG. 3(a), the GPS information is information indicating latitude and longitude, and is continuously acquired for each of multiple users. As shown in FIG. 3(b), the app usage information is information indicating, for example, the total number of times the app is used and the usage amount (amount charged for the app) for each user. The receiving unit 11 outputs the received information to the learning unit 12.

[0022] 2 , the learning unit 12 constructs a learning model 130 by associating and learning GPS information (information related to the location of stay) and app usage information (information related to the degree of match with the service) of multiple users in a country where the app has already been introduced. The learning model 130 is a machine learning model that outputs information indicating the degree of match with the service based on the information related to the location of stay.

[0023] The learning unit 12 performs predetermined preprocessing on the GPS information acquired by the receiving unit 11. Specifically, the learning unit 12 extracts only the location (here, location G4) where the user stayed for a certain period of time or more as a stay point to be used for learning, from among locations G1 to G5, which are the user's stay locations along a timeline as shown in FIG. 1A, as shown in FIG. 1B. Then, as shown in FIG. 1C, the learning unit 12 searches for POIs around the stay point based on distance information from the stay point. Here, each POI is managed in a database within the demand estimation device 10 or a database external to the demand estimation device 10, with at least its location information linked to a facility category. The facility category may be information that can identify an overview of the facility, such as a sports facility, convenience store, or school. The facility category managed in the database may be specified in detail, for example, down to the specific store name. The learning unit 12 identifies the facility category for POIs located within a predetermined distance from the stay point. The learning unit 12 then assigns a score corresponding to the dwell time to the facility category of each identified POI based on the likelihood based on the distance information. FIG. 4A is a table showing an example of scores corresponding to the dwell time for each facility category. In the example shown in FIG. 4A, a score corresponding to the dwell time is assigned only to the POI in category A based on the distance information (10 points), and no score corresponding to the dwell time is assigned to the other POIs in categories B and C (0 points). As shown in FIG. 4A, the learning unit 12 vectorizes the scores corresponding to the dwell time assigned for each category as dwell time features. Note that the learning unit 12 may also estimate facility categories in which the user is likely to stay by further considering information other than distance information, such as information indicating the user's characteristics (age, gender, hobbies, occupation, annual income, etc.), and assign scores for each category as described above.

[0024] The learning unit 12 may estimate the area around the user's home and the area around the workplace based on the user's location information for each day of the week and time period, and may link the time spent in each of these areas and add the link to the above-mentioned dwell time feature. It is likely that users often stay at home, for example, between late night and early morning. It is likely that users often stay at work, for example, during the daytime on weekdays. FIG. 4(b) is a table showing an example of scores according to the time spent in the areas estimated to be the home and workplace in addition to the information shown in FIG. 4(a). As shown in FIG. 4(b), by adding score information according to the time spent at home and workplace to the dwell time feature, parameters indicating the user's lifestyle pattern can be added to the learning data.

[0025] When setting facility categories to be included in the dwell time feature, the learning unit 12 may set facility categories (information indicating POI categories) taking into account differences in category names both domestically and internationally (first and second regions). FIG. 5( a) is a table showing an example of facility categories intended for use only in Japan, and FIG. 5( b) is a table showing an example of facility categories intended for overseas expansion. Among the facility categories shown in FIG. 5( a), shrines are a domestic facility category that does not exist overseas. In this regard, by setting facility categories to include facilities (e.g., churches) visited for similar purposes overseas (see FIG. 5( b)), the learning model 130 can be constructed in a format more suitable for overseas expansion. Similarly, for a detailed facility category such as a baseball stadium, the facility category can be set as a superordinate concept, such as a sports facility. While learning can be performed in more detail based on facility names when considering only domestic use, it is expected that similar facility names will not exist overseas. Therefore, it is preferable to include superordinate facility categories in the dwell time feature.

[0026] Furthermore, when sampling target users whose dwell time features are used as learning data, the learning unit 12 may perform adjustments to prevent bias in the characteristics of the sample. Specifically, when sampling target users, adjustments may be performed to prevent bias in the expected gender ratio and age distribution. Such adjustments are typically difficult to determine from dwell time features, which only contain location information. In this regard, the bias in the characteristics described above can be estimated by calculating the percentage of users confirmed to have visited each facility category. FIG. 6( a) is a table showing an example of the user percentage for each facility category when sampling is performed without considering balance, and FIG. 6( b) is a table showing an example of the user percentage for each facility category when sampling is performed with balance in mind. In the example shown in FIG. 6( a), when the percentage of users who visited each facility category for the sampled users is organized, it is estimated that, for example, the percentage of users who visited beauty salons and clinics is low, and that there are few women among the sampled users. It is also estimated that, for example, the percentage of users who visited schools is low, and that there are few students among the sampled users. Taking this into consideration, as shown in Figure 6(b), by resampling to increase the proportion of users who visited beauty salons / clinics or schools, for example, it is possible to build a more balanced learning model.

[0027] The learning unit 12 may perform predetermined preprocessing on the app usage information of the user whose dwell time feature is used as learning data (sampled) from the app usage information acquired by the receiving unit 11. For example, as shown in FIG. 7 , the learning unit 12 may determine whether or not the app is used daily based on the app usage information, and set the information on whether or not the app is used daily as a target flag.

[0028] After the preprocessing described above is completed, the learning unit 12 constructs (trains) the learning model 130. Specifically, as shown in FIG. 8 , the learning unit 12 constructs the learning model 130, which outputs information indicating whether or not a service is regularly used, based on information related to the location of stay (information indicating the facility category of the POI and the length of stay at that POI). The output result may be, for example, a score indicated by a number between 0 and 1. That is, the learning model 130 may output a score closer to 1 as the likelihood of daily use of the service increases, and closer to 0 as the likelihood of no daily use increases. The learning unit 12 stores the thus constructed learning model 130 in the storage unit 13.

[0029] (Demand Estimation Process) The reception unit 11 acquires GPS information (location information) from multiple users in one overseas country (second region, second country), which is the region to be estimated. The reception unit 11 acquires information from communication terminals (not shown), such as smartphones, carried by the multiple users. The GPS information is information related to the user's location and indicates the latitude and longitude where the user is staying. The GPS information may be information measured by the communication terminal (not shown) carried by the user. The reception unit 11 continuously acquires the GPS information at predetermined time intervals. The reception unit 11 outputs the received information to the estimation unit 14.

[0030] The estimation unit 14 inputs the information received by the reception unit 11 (information relating to the location of stay in one of the overseas countries that is the region to be estimated) into the learning model 130, and estimates information indicating the degree of match with the above-mentioned app for multiple users in the overseas country to be estimated.

[0031] The estimation unit 14 performs a predetermined preprocessing on the GPS information acquired by the receiving unit 11. This preprocessing is similar to the preprocessing performed by the learning unit 12 on the GPS information described above. Specifically, the estimation unit 14 extracts stay points from the GPS information of the overseas user, searches for POIs around the stay points, identifies facility categories of the surrounding POIs, and vectorizes scores corresponding to the stay times for each facility category as stay time features. The estimation unit 14 also adds score information corresponding to the stay time at home and at work to the stay time features (see FIG. 9A). Note that, similar to the preprocessing performed by the learning unit 12, the estimation unit 14 may adjust the sample characteristics to prevent bias. The estimation unit 14 may then input each user's stay time feature into the learning model 130 and output the score output from the learning model 130 as an estimation result (see FIG. 9B). Note that the score indicates the degree of match with the app and the degree of app usage. As described above, the score approaches 1 as the likelihood of daily use of the app (service) increases, and approaches 0 as the likelihood of no daily use increases. In this way, the estimation unit 14 may estimate a score indicating the degree of use of the service as information indicating the degree of match with the service.

[0032] The estimation unit 14 may estimate an average score, which is an average value of scores among multiple users. Furthermore, the estimation unit 14 may estimate a predicted number of users of the service in the overseas country based on the average score. Such a predicted number of users may be estimated based on, for example, the average score and the app user population in the country.

[0033] The output unit 15 outputs information related to the estimation result by the estimation unit 14. The output unit 15 may transmit the information related to the estimation result to, for example, a device external to the demand estimation device 10. The output unit 15 may output the score estimated by the estimation unit 14, may output the average score of each user described above, or may output the predicted number of users of the service.

[0034] 10 is a diagram illustrating an example of output according to the estimation result. In the example shown in FIG. 10, for example, for an app related to a "video distribution service," information is output indicating that the average score is 0.4 and the predicted number of users is 9,000. Also, for an app related to a "bicycle sharing service," information is output indicating that the average score is 0.7 and the predicted number of users is 90,000.

[0035] Next, the model construction process (see FIG. 11) and the demand estimation process (see FIG. 12) performed by the demand estimation device 10 will be described.

[0036] As shown in FIG. 11, in the model construction process, first, the demand estimation device 10 acquires GPS information and service usage information of each user in the country (step S1).

[0037] Next, the demand estimation device 10 performs preprocessing on the GPS information and preprocessing on the service usage information (step S2), and prepares residence time features for each user to be sampled and usage information for apps for which target flags are set.

[0038] Next, the demand estimation device 10 constructs a learning model 130 using learning data in which the residence time feature values ​​for each sampled user are associated with usage information of apps for which the target flag is set (step S3). Finally, the learning model 130 is saved (stored), and the model construction process is completed (step S4).

[0039] As shown in FIG. 12, in the demand estimation process, first, the demand estimation device 10 acquires GPS information of each user in one overseas country (step S11).

[0040] Next, the demand estimation device 10 performs preprocessing on the GPS information (step S12), and prepares a residence time feature amount for each user to be sampled.

[0041] Next, the residence time feature for each sampled user is input to the learning model 130, and information (e.g., a score) indicating the match degree of the overseas user to the service is estimated (step S13). Finally, information according to the estimation result is output, and the demand estimation process is completed (step S14).

[0042] Next, the effects of the demand estimation device 10 according to this embodiment will be described.

[0043] The demand estimation device 10 of this embodiment includes a learning unit 12 that constructs a learning model 130 that outputs information indicating the degree of match with the service based on the information related to the stay locations by learning information related to the stay locations of multiple users in a first region where a specified service has already been introduced and information related to the degree of match with the service, an estimation unit 14 that estimates information indicating the degree of match with the service for multiple users in a second region, which is the region to be estimated, by inputting information related to the stay locations of multiple users in the second region into the learning model 130, and an output unit 15 that outputs information related to the estimation result by the estimation unit 14.

[0044] In the demand estimation device 10, a learning model 130 is constructed that correlates and learns information related to the locations of multiple users in a first region where a service is introduced with information related to their match with the service, and outputs information indicating their match with the service from the information related to their locations. The learning model 130 is constructed assuming that there is a correlation between information related to the locations (e.g., places frequently visited by users) and their match with the service (e.g., service usage). For example, the relationship that "users who frequently watch sports games frequently use a baseball watching app" is likely to hold true in other regions (e.g., other countries). In other words, if users who frequently watch sports games in Japan frequently use a baseball watching app, if the baseball watching app is deployed overseas, users who similarly frequently watch sports games are likely to use the baseball watching app. Therefore, by inputting information related to the locations of multiple users in a second region, which is the region to be estimated, into the learning model 130, information indicating the match with the service for multiple users in the second region can be estimated with high accuracy. By outputting such an estimation result, it is possible to identify with high accuracy the demand for the service in the area where the service will be newly developed (second area).

[0045] The information related to the stay location may include information indicating the facility category of the POI (Point of Interest) associated with the stay location and information indicating the stay time at the POI (e.g., a score). By taking the stay time for each facility category into consideration in this way, it is possible to appropriately reflect the characteristics of POIs that are likely to be visited by users, thereby improving the accuracy of the learning model 130 described above.

[0046] The learning unit 12 may use only information related to stay locations where the user stayed for a certain period of time or more for learning. This allows information about facilities that the user simply passed through to be excluded from the learning data, allowing the learning model 130 to be constructed only from information about facilities that the user is likely to have visited out of interest, thereby improving the accuracy of the learning model 130.

[0047] The learning unit 12 may use information related to the use of a service as information related to the degree of match with the service. The information related to the use of a service accurately reflects the user's degree of interest in the service. Therefore, by using the information related to the use of a service as information related to the degree of match, the accuracy of the learning model 130 can be improved.

[0048] The estimation unit 14 may estimate a score indicating the degree of use of the service as information indicating the degree of match with the service, and the output unit 15 may output the score estimated by the estimation unit 14. By outputting the information indicating the degree of match as a score, it is possible to quantitatively evaluate the demand for the service.

[0049] The estimation unit 14 may estimate an average score, which is an average value of the scores among a plurality of users, and the output unit 15 may output the average score estimated by the estimation unit 14. By outputting the average score, it is possible to more accurately evaluate the demand for the service in the second region.

[0050] The estimation unit 14 may estimate the predicted number of users of the service in the second region based on the average score, and the output unit 15 may output the predicted number of users of the service estimated by the estimation unit 14. By outputting the predicted number of users, it is possible to more accurately evaluate the demand for the service in the second region.

[0051] The learning unit 12 may set information indicating the category of the POI, taking into consideration the difference in category names between the first region and the second region, thereby appropriately constructing a learning model 130 that can be commonly used across different regions.

[0052] The first region may be a first country, and the second region may be a second country different from the first country. Typically, it is difficult to predict demand for a service because the information that can be obtained from different countries is limited. In this regard, by using the demand estimation method according to this embodiment, even when the countries are different, it is possible to appropriately predict demand using a learning model 130 constructed in a country where the service has already been introduced.

[0053] Next, the hardware configuration of the above-mentioned demand estimation device 10 will be described with reference to Fig. 13. The demand estimation device 10 may be physically configured as a computer device including a processor 1001, a memory 1002, a storage 1003, a communication device 1004, an input device 1005, an output device 1006, a bus 1007, etc.

[0054] In the following description, the term "apparatus" may be interpreted as a circuit, a device, a unit, etc. The hardware configuration of the demand estimation apparatus 10 may be configured to include one or more of the apparatuses shown in the figure, or may be configured to exclude some of the apparatuses.

[0055] Each function of the demand estimation device 10 is realized by loading specified software (programs) onto hardware such as the processor 1001 and memory 1002, causing the processor 1001 to perform calculations and control communication via the communication device 1004 and the reading and / or writing of data in the memory 1002 and storage 1003.

[0056] The processor 1001 controls the entire computer by running, for example, an operating system. The processor 1001 may be configured as a central processing unit (CPU) including an interface with peripheral devices, a control unit, an arithmetic unit, a register, etc. For example, the control functions of the reception unit 11 and the like may be realized by the processor 1001.

[0057] The processor 1001 also reads programs (program codes), software modules, and data from the storage 1003 and / or the communication device 1004 into the memory 1002, and executes various processes in accordance with these. The programs used are those that cause a computer to execute at least some of the operations described in the above embodiments.

[0058] For example, the control functions of the reception unit 11 and the like may be realized by a control program stored in the memory 1002 and running on the processor 1001, and similar functions may be realized for other functional blocks. Although the above-described various processes have been described as being executed by one processor 1001, they may be executed simultaneously or sequentially by two or more processors 1001. The processor 1001 may be implemented on one or more chips. The program may be transmitted from a network via a telecommunications line.

[0059] The memory 1002 is a computer-readable recording medium and may be composed of at least one of, for example, a read-only memory (ROM), an erasable programmable ROM (EPROM), an electrically erasable programmable ROM (EEPROM), a random access memory (RAM), etc. The memory 1002 may also be called a register, a cache, a main memory (primary storage device), etc. The memory 1002 can store executable programs (program codes), software modules, etc. for implementing a wireless communication method according to one embodiment of the present invention.

[0060] Storage 1003 is a computer-readable recording medium, and may be, for example, at least one of an optical disk such as a CD-ROM (Compact Disc ROM), a hard disk drive, a flexible disk, a magneto-optical disk (e.g., a compact disk, a digital versatile disk, a Blu-ray disc), a smart card, a flash memory (e.g., a card, a stick, a key drive), a floppy disk, a magnetic strip, etc. Storage 1003 may also be referred to as an auxiliary storage device. The above-mentioned storage medium may be, for example, a database, a server, or other appropriate medium including memory 1002 and / or storage 1003.

[0061] The communication device 1004 is hardware (transmission / reception device) for communicating between computers via a wired and / or wireless network, and is also called, for example, a network device, a network controller, a network card, or a communication module.

[0062] The input device 1005 is an input device (e.g., a keyboard, a mouse, a microphone, a switch, a button, a sensor, etc.) that receives input from the outside. The output device 1006 is an output device (e.g., a display, a speaker, an LED lamp, etc.) that outputs to the outside. The input device 1005 and the output device 1006 may be integrated into one device (e.g., a touch panel).

[0063] Furthermore, each device such as the processor 1001 and the memory 1002 is connected to a bus 1007 for communicating information. The bus 1007 may be configured as a single bus, or may be configured as different buses between the devices.

[0064] The demand estimation device 10 may also be configured to include hardware such as a microprocessor, a digital signal processor (DSP), an application specific integrated circuit (ASIC), a programmable logic device (PLD), or a field programmable gate array (FPGA), and some or all of the functional blocks may be realized by the hardware. For example, the processor 1001 may be implemented by at least one of these pieces of hardware.

[0065] Although the present embodiment has been described in detail above, it is clear to those skilled in the art that the present embodiment is not limited to the embodiment described in this specification. The present embodiment can be implemented in modified and altered forms without departing from the spirit and scope of the present invention as defined by the claims. Therefore, the description in this specification is intended to be illustrative and does not have any limiting meaning on the present embodiment.

[0066] Each aspect / embodiment described herein may be applied to systems utilizing LTE (Long Term Evolution), LTE-Advanced (LTE-A), SUPER 3G, IMT-Advanced, 4G, 5G, FRA (Future Radio Access), W-CDMA, GSM, CDMA2000, UMB (Ultra Mobile Broadband), IEEE 802.11 (Wi-Fi), IEEE 802.16 (WiMAX), IEEE 802.20, UWB (Ultra-Wide Band), Bluetooth, or other suitable systems and / or next generation systems enhanced thereon.

[0067] The order of the procedures, sequences, flowcharts, etc. of each aspect / embodiment described herein may be rearranged unless it is consistent. For example, the methods described herein present elements of various steps in an example order and are not limited to the particular order presented.

[0068] Input and output information may be stored in a specific location (for example, memory) or managed in a management table. Input and output information may be overwritten, updated, or added to. Output information may be deleted. Input information may be sent to another device.

[0069] The determination may be made based on a value represented by one bit (0 or 1), a Boolean value (true or false), or a numerical comparison (e.g., comparison with a predetermined value).

[0070] The aspects / embodiments described herein may be used alone or in combination, or may be switched depending on the implementation. Notification of predetermined information (e.g., notification that "X is true") is not limited to explicit notification, but may be implicit (e.g., not notifying the predetermined information).

[0071] Software shall be construed broadly to mean instructions, instruction sets, code, code segments, program code, programs, subprograms, software modules, applications, software applications, software packages, routines, subroutines, objects, executable files, threads of execution, procedures, functions, etc., whether referred to as software, firmware, middleware, microcode, hardware description language, or otherwise.

[0072] Software, instructions, etc. may also be transmitted or received over a transmission medium. For example, if the software is transmitted from a website, server, or other remote source using wired technologies such as coaxial cable, fiber optic cable, twisted pair, and Digital Subscriber Line (DSL), and / or wireless technologies such as infrared, radio, and microwave, these wired and / or wireless technologies are included within the definition of transmission media.

[0073] The information, signals, etc. described herein may be represented using any one of a variety of different technologies. For example, data, instructions, commands, information, signals, bits, symbols, chips, etc. that may be referred to throughout the above description may be represented by voltages, currents, electromagnetic waves, magnetic fields or magnetic particles, optical fields or photons, or any combination thereof.

[0074] It should be noted that terms explained in this specification and / or terms necessary for understanding this specification may be replaced with terms having the same or similar meanings.

[0075] Furthermore, the information, parameters, etc. described in this specification may be expressed as absolute values, as relative values ​​from a predetermined value, or as corresponding other information.

[0076] A communications terminal may also be referred to by those skilled in the art as a mobile communications terminal, subscriber station, mobile unit, subscriber unit, wireless unit, remote unit, mobile device, wireless device, wireless communications device, remote device, mobile subscriber station, access terminal, mobile terminal, wireless terminal, remote terminal, handset, user agent, mobile client, client, or some other suitable terminology.

[0077] As used herein, the phrase "based on" does not mean "based only on," unless expressly specified otherwise. In other words, the phrase "based on" means both "based only on" and "based at least on."

[0078] When designations such as "first," "second," etc. are used herein, any reference to such elements does not generally limit the quantity or order of those elements. These designations may be used herein as a convenient method of distinguishing between two or more elements. Thus, a reference to a first and a second element does not imply that only two elements may be employed therein or that the first element must precede the second element in some way.

[0079] To the extent that the terms "include," "including," and variations thereof are used herein or in the claims, these terms are intended to be inclusive, similar to the term "comprising." Furthermore, the term "or," as used herein or in the claims, is not intended to be an exclusive or.

[0080] In this specification, a plurality of devices is also included unless the context or the technology clearly indicates that only one device exists.

[0081] Throughout this disclosure, the plural is intended to be included unless the singular is clearly indicated by the context.

[0082] Finally, various exemplary embodiments included in the present disclosure are described below in [E1] to [E10].

[0083] [E1] A device comprising: a learning unit that constructs a learning model that outputs information indicating the degree of match with a service based on information related to the stay locations of multiple users in a first region where a specified service has already been introduced, by learning by correlating information related to the stay locations with information related to the degree of match with the service; an estimation unit that estimates information indicating the degree of match with the service for multiple users in a second region that is an estimation target region, by inputting information related to the stay locations of multiple users in the second region into the learning model; and an output unit that outputs information related to the estimation result by the estimation unit.

[0084] [E2] The device according to [E1], wherein the information relating to the stay position includes information indicating a category of a POI (Point of Interest) associated with the stay position, and information indicating a stay time at the POI.

[0085] [E3] The device according to [E1] or [E2], wherein the learning unit uses only information relating to the stay position where the user has stayed for a certain period of time or more for learning.

[0086] [E4] The device according to any one of [E1] to [E3], wherein the learning unit uses information related to use of the service as information related to the degree of match with the service.

[0087] [E5] The device according to [E4], wherein the estimation unit estimates a score indicating a degree of use of the service as information indicating a degree of match with the service, and the output unit outputs the score estimated by the estimation unit.

[0088] [E6] The device according to [E5], wherein the estimation unit estimates an average score that is an average value of the scores among the plurality of users, and the output unit outputs the average score estimated by the estimation unit.

[0089] [E7] The device described in [E6], wherein the estimation unit estimates a predicted number of users of the service in the second region based on the average score, and the output unit outputs the predicted number of users of the service estimated by the estimation unit.

[0090] [E8] The device according to [E2], wherein the learning unit sets information indicating the category of the POI, taking into consideration differences in category names between the first region and the second region.

[0091] [E9] The device according to any one of [E1] to [E8], wherein the first region is a first country, and the second region is a second country different from the first country.

[0092] [E10] A method executed by a device, the method including: a step of constructing a learning model that outputs information indicating the degree of match with a service based on information related to the stay locations by correlating and learning information related to the stay locations of multiple users in a first region where a specified service has already been introduced with information related to the degree of match with the service; a step of estimating information indicating the degree of match with the service for multiple users in a second region that is an estimation target region by inputting information related to the stay locations of multiple users in the second region into the learning model; and a step of outputting information related to the estimation result in the estimating step.

[0093] 10...demand estimation device, 12...learning unit, 14...estimation unit, 15...output unit, 130...learning model

Claims

1. A device comprising: a learning unit that constructs a learning model that outputs information indicating the degree of match with a service based on information related to the locations of stay of multiple users in a first region where a specified service has already been introduced, by learning by correlating information related to the locations of stay with information related to the degree of match with the service; an estimation unit that estimates information indicating the degree of match with the service for multiple users in a second region, which is the region to be estimated, by inputting information related to the locations of stay of multiple users in the second region into the learning model; and an output unit that outputs information related to the estimation results by the estimation unit.

2. The device of claim 1, wherein the information relating to the stay location includes information indicating the category of a POI (Point of Interest) associated with the stay location and information indicating the duration of stay at the POI.

3. The device according to claim 2, wherein the learning unit uses only information relating to the stay position where the user has stayed for a certain period of time or more for learning.

4. The device according to claim 1, wherein the learning unit uses information related to the use of the service as information related to the degree of match with the service.

5. The device according to claim 4, wherein the estimation unit estimates a score indicating the degree of use of the service as information indicating the degree of match with the service, and the output unit outputs the score estimated by the estimation unit.

6. The device according to claim 5, wherein the estimation unit estimates an average score that is an average value of the scores among the plurality of users, and the output unit outputs the average score estimated by the estimation unit.

7. The device according to claim 6, wherein the estimation unit estimates a predicted number of users of the service in the second region based on the average score, and the output unit outputs the predicted number of users of the service estimated by the estimation unit.

8. The device according to claim 2, wherein the learning unit sets information indicating the category of the POI, taking into consideration differences in category names between the first region and the second region.

9. The device according to any one of claims 1 to 8, wherein the first region is a first country, and the second region is a second country different from the first country.

10. A method executed by a device, comprising: a step of constructing a learning model that outputs information indicating the degree of match with a service based on information related to the locations of stay of multiple users in a first region where a specified service has already been introduced, by correlating and learning information related to the locations of stay of multiple users in the first region where a specified service has already been introduced with information related to the degree of match with the service; a step of estimating information indicating the degree of match with the service for multiple users in the second region, which is the region to be estimated, by inputting information related to the locations of stay of multiple users in the second region into the learning model; and a step of outputting information related to the estimation results in the estimating step.

Images