JP7634440B2 - App log analyzer - Google Patents

Description

This disclosure relates to an application log analysis device.

In recent years, the number of terminal devices with camera functions has increased, and users are increasingly taking pictures casually. For this reason, for example, a method is being considered in which location information of multiple terminals and the number of times images have been taken at those locations are acquired, and recommended shooting locations are identified based on this information (see, for example, Patent Document 1).

JP 2015-19147 A

However, in recent years, as the environment has become one in which users are more likely to take photos casually, it has become difficult to properly estimate locations that could be recommended photo locations simply by aggregating information on the number of times photos have been taken obtained from multiple devices.
The present disclosure has been made in consideration of the above, and aims to provide a technology capable of estimating areas suitable for photography based on app usage logs obtained from multiple user devices.

To achieve the above object, an application log analysis device according to one embodiment of the present disclosure includes a terminal information acquisition unit that acquires, from a plurality of user terminals, information identifying the user terminal, a camera application usage log including date and time information, and location information identifying the location where the camera application was used for each of the usage logs; a filtering unit that filters the camera application usage logs that satisfy a specific condition based on terminal-related information that is information about each of the plurality of user terminals from each of the camera application usage logs associated with the location information associated with the plurality of user terminals acquired by the terminal information acquisition unit; and a POI estimation unit that estimates a POI suitable for using the camera application by classifying the camera application usage logs associated with location information extracted by the filtering unit based on the location information and the date and time information.

According to the above application log analysis device, the POI is estimated by filtering each of the camera application usage logs associated with location information acquired from multiple user terminals based on device-related information, which is information about each of the multiple user terminals, and then classifying the logs based on the location information and date and time information. As described above, the POI is estimated using the camera application usage logs after filtering based on the device-related information about each of the multiple user terminals, so that areas suitable for photography can be estimated based on the application usage logs acquired from the multiple user terminals. In particular, since the POI is estimated after filtering to remove camera application usage logs that may become noise in estimating the POI, a location more suitable as a POI can be estimated.

The present disclosure provides a technology that can estimate areas suitable for photography based on app usage logs acquired from multiple user devices.

FIG. 1 is a block diagram illustrating an overview of the process performed in an application log analysis device. FIG. 2 is a block diagram illustrating the configuration of the application log analysis device. FIG. 3 is a diagram illustrating an example of an application usage log. FIG. 4 is a diagram illustrating an example of the location information. 5A and 5B are diagrams showing an example of the auxiliary information. FIG. 6 is a diagram showing an example of POI information. FIG. 7 is a diagram showing an example of category-specified POI information. FIG. 8 is a flow diagram showing an example of an application log analysis method. FIG. 9 is a diagram showing an example of a filtering method. FIG. 10 is a diagram showing an example of a filtering method (taking into account a home range). FIG. 11 is a diagram showing an example of a filtering method (taking into account information on applications used before and after). FIG. 12 is a diagram showing an example of a category estimation method (an example in which information on apps used before and after is taken into consideration). FIG. 13 shows an example of a filtering and category estimation method (taking into account user segments). FIG. 14 is a diagram showing an example of a filtering and category estimation method (an example in which the type of camera application is taken into consideration). FIG. 15 is a diagram showing an example of a POI estimation method (an example using population per mesh). FIG. 16 is a diagram showing an example of a change over time in the number of uses of the camera application. FIG. 17 is a diagram illustrating an example of a hardware configuration of an application log analysis device.

The following describes the embodiments in detail with reference to the drawings. In the description, the same elements or elements having the same functions are given the same reference numerals, and duplicate descriptions are omitted.

[Application log analysis device]
FIG. 1 is a diagram conceptually explaining the operation of an application log analysis device 1 according to an embodiment.

The application log analysis device 1 has a function of acquiring usage logs of applications (applications) on the user terminals 9 from multiple user terminals 9, and estimating POIs (Points of Interests) that indicate places of interest to users of the user terminals 9 by analyzing the logs. In particular, the application log analysis device 1 according to this embodiment has a function of estimating POIs by focusing on applications that take images among the applications. An "application that takes images" is not limited to an application that only takes images, but may also include an application that has a function of taking and editing images and then posting the images to a social networking service (SNS). In other words, an application that has a shooting function as part of the application is called an "application that takes images". In the following embodiment, an "application that takes images" may be described as a "camera application".

The application log analysis device 1 collects camera application usage logs from among the application usage logs in the user terminal 9, and estimates the location of the POI. This is based on the assumption that a location where the camera application is used means that the user is interested in that location. In recent years, when it has become standard for user terminals 9 to be equipped with a camera function, the number of applications with a shooting function has increased. In addition, camera applications themselves have been realized to perform operations up to posting to SNS, and there are more opportunities for users to take photos when they come across something that interests them. Therefore, the application log analysis device 1 focuses on where the camera application was used, and estimates that locations where the camera application operates frequently are locations that are likely to interest users. Note that in the following embodiments, the "camera application usage log" may be described as the "camera application log."

The application log analysis device 1 may identify a POI on a regional basis. The region to be estimated for the presence or absence of a POI is divided into a number of smaller regions (for example, mesh units), and it is estimated whether a POI exists in each mesh. FIG. 1 shows an image of identifying a POI on a mesh basis by the application log analysis device 1. FIG. 1 shows a state in which a map M is divided into nine meshes of user terminals. The map M shows a location P where a camera app was used on each user terminal 9 based on a camera app usage log (camera app log) obtained by the application log analysis device 1 collecting app usage logs of a number of user terminals 9. In the example shown in FIG. 1, it is shown that the camera app was used four times in mesh m1, and that the camera app was used once in mesh m2. In this case, since mesh m1 is estimated to be a region where the camera app is used frequently, the application log analysis device 1 may estimate that mesh m1 is a POI.

However, simply estimating a POI from the number of times the camera app is used in this way may not provide information with sufficient accuracy to be used for, for example, recommendations. For example, if the user who collected the app usage log lives in mesh m1 or nearby, they may take many photos of scenes from their daily life. In this case, it is expected that the camera app will be launched to take photos of scenes different from those that interest other users (for example, family members), so it cannot be said that there is a strong relationship between the number of times the camera app is launched and the POI. In this way, there is a possibility that a place that should not actually be identified as a POI will be identified as a POI.

In addition, while some POIs have characteristics that interest many users, others are of particular interest to users with specific hobbies. The former POIs are easy to identify from the number of times the camera app is launched, whereas the latter POIs are likely to have fewer launches of the camera app than the former. Also, while a location is estimated to be a POI because the number of times the camera app is used is high, it may not be possible to identify what type of POI it is based on this information alone. In other words, it is expected that there is room for improvement in the accuracy of POI estimation based solely on the number of times the camera app is used.

In the following embodiment, a method is described for identifying a POI while improving the accuracy of the information as a POI by combining the usage log of a camera app with other information in the application log analysis device 1. A method is also described for identifying the type of POI identified in the application log analysis device 1.

Note that the above "mesh unit" is an example, and the basic unit for estimating POI is not limited to "mesh". For example, when a certain area includes multiple POIs (places suitable for photography), it may be preferable to distinguish and handle POIs in finer units for that area. If the mesh is divided more finely, it can be called "location". In addition, the "time series" may be divided into time periods of a predetermined length, but POI estimation may be performed focusing on a specific time. In this way, the application log analysis device 1 can classify the camera application log based on information specifying the place (location) where the camera application was used and information specifying the date and time, and perform statistical processing related to POI based on the classification result. In addition, the application log analysis device 1 has a function of preparing and aggregating data that can be used for the above-mentioned statistical processing.

Returning to FIG. 2, the components of the application log analysis device 1 and the user terminal 9 will be described. First, the user terminal 9 will be described. A "user terminal" is a terminal device carried by a user, and may be a smartphone, a tablet terminal, or the like. The user terminal 9 includes a user application 91, an image information storage unit 92, an application usage log acquisition unit 93, an application usage log transmission unit 94, a location information acquisition unit 95, and a location information transmission unit 96.

The user apps 91 represent a group of apps (applications) used by the user. In FIG. 2, a camera app A, a camera app B, a hobby app A, and a hobby app B are shown as examples. The user operates the apps included in the user apps 91 to perform operations using various user terminals 9.

The image information storage unit 92 has a function of storing images captured by any of the camera apps included in the user app 91. Note that images captured by the camera app may include both still images and videos.

The application usage log acquisition unit 93 has a function of acquiring an application usage log included in the user application 91 in the user terminal 9. The application usage log may include information identifying the application and information identifying the usage time of the application. It may also include information identifying the type of application. The application usage log transmission unit 94 has a function of transmitting the application usage log acquired by the application usage log acquisition unit 93 to the application log analysis device 1. In this embodiment, the application usage log includes information identifying the time period in which the application was used. That is, the application usage log includes information identifying the usage time of the application. However, the application usage log may be a log that operates a specific function of the application. For example, in the case of a camera application, a log of the shooting operation of the camera may be the "application usage log". In this case, the information identifying the usage time of the application is information identifying the time when the function was operated. In the configuration described in the following embodiment, the "application usage log" can be read as the "log of a specific operation in the application" as described above.

The location information acquisition unit 95 has a function of acquiring location information of the user terminal 9. There are no particular limitations on the method of acquiring the location information, and any known method can be used. The information handled here is mainly used to identify the location of the user terminal 9 when the camera app is used, but the location of the user terminal 9 may basically be acquired constantly (e.g., repeatedly at intervals of several tens of seconds to several minutes). The location information transmission unit 96 has a function of transmitting the location information of the user terminal 9 acquired by the location information acquisition unit 95 to the application log analysis device 1.

Next, the application log analysis device 1 will be described. The application log analysis device 1 includes an application usage log receiving unit 11, an application usage log storage unit 12, a location information receiving unit 13, a location information storage unit 14, a filtering unit 15, an auxiliary information storage unit 16, a POI estimation unit 17, a POI information storage unit 18, a POI category estimation unit 19, a category-specified POI information storage unit 20, and an output unit 21. Of the above units, the application usage log receiving unit 11 and the location information receiving unit 13 function as a terminal-related information acquisition unit.

The app usage log receiving unit 11 has a function of receiving and acquiring app usage logs transmitted from the user terminal 9. FIG. 3 shows an example of an app usage log acquired by the app usage log receiving unit 11. As shown in FIG. 3, the app usage log may include the date and time of app use, the app name, and the type of app. Note that, as information for identifying an app, character string information for identifying an app may be used instead of the app name. The app usage log may include a camera app usage log, but may also include usage logs of apps other than the camera app.

The app usage log storage unit 12 has a function of storing app usage logs acquired by the app usage log receiving unit 11. The app usage log storage unit 12 stores app usage logs related to multiple user terminals 9.

The location information receiving unit 13 has a function of acquiring the location information of the user terminal 9 transmitted from the user terminal 9. FIG. 4 shows an example of the location information acquired by the location information receiving unit 13. As shown in FIG. 4, the location information includes a date and time and location information at the date and time. Naturally, this information includes information for identifying the user terminal 9. In FIG. 2, the location information receiving unit 13 is configured to acquire the location information of the user terminal acquired by the user terminal 9 from the user terminal 9, but the location information does not have to be acquired directly from the user terminal 9. In the application log analysis device 1, for example, the location information of the base station in which the user terminal 9 is located may be used as the location information of the user terminal 9. In such a case, the information (such as the location information of the base station) used as the location information of the user terminal 9 may be acquired from a device different from the user terminal 9 (such as a device that manages the location of the user terminal 9).

Note that FIG. 4 also includes columns for Mesh and Activity Range. Mesh is information that identifies the mesh that corresponds to the location information at each date and time. This information is based on the mesh used when identifying POIs in the application log analysis device 1, and is therefore assigned by the application log analysis device 1.

The area of activity is information that specifies whether the location of the user terminal 9 at each date and time corresponds to the user's daily life area or not (outside the daily life area) based on the location information accumulated about the user who owns the user terminal 9. The criteria for determining whether or not the location is in the daily life area are not particularly limited, but for example, a method of setting it within a predetermined range based on the user's home address (for example, within a radius of 1.5 km assuming a walking distance) may be used. In addition, after specifying the maximum value of the user's travel distance from the user's home for each day based on the user's past travel history, the distance that includes 90% of the maximum value for each day may be specified as the daily life area. In addition, the user's daily life area may be distinguished from the rest by using existing technology. Note that here, the case where the user's stay location (the location of the user terminal 9) is distinguished into two types, the daily life area and the rest, but it may also be divided according to the reason for the move, such as daily life/business trip/travel. The result of distinguishing the user's stay location (information that specifies the area of activity) may be used when specifying the POI described later. Therefore, this information is given in the application log analysis device 1.

The location information storage unit 14 has a function of storing location information of the user terminal 9 acquired by the location information receiving unit 13. The location information storage unit 14 stores location information related to multiple user terminals 9. The above-mentioned association with the mesh and the identification of the area of activity may be performed when the location information is stored in the location information storage unit 14, or may be performed at a different time (for example, when estimating the POI).

The filtering unit 15 has a function of performing filtering using the app usage information stored in the app usage log storage unit 12 and the location information stored in the location information storage unit 14. In this embodiment, filtering is a process of associating the camera app usage log (camera app log) included in the app usage log with location information and then sorting whether or not to use it for estimating a POI. Filtering methods include a method of selecting the camera app log itself and a method of selecting users who use the camera app. Filtering will be described in detail later.

The auxiliary information storage unit 16 has the function of storing, as auxiliary information, information other than the information obtained from the user terminal 9 that can be used for filtering in the filtering unit 15.

Figure 5 shows an example of auxiliary information stored in the auxiliary information storage unit 16. Figure 5 (a) is an example of information related to population changes on a mesh-by-mesh basis. In Figure 5 (a), information indicating population changes per unit time is stored in association with information identifying the mesh (mesh ID). By using such information, it is possible to perform analysis such as how many people are present in an area (mesh) where the camera app is used frequently.

5B shows an example of information indicating the attributes of a user of the user terminal 9. In FIG. 5B, information indicating the attributes (parenting generation, camera hobby, fishing, mountain climbing, etc.) is associated with information identifying the user (user ID) and information on which the attributes are identified (apps, SNS, questionnaires, etc.). For example, for a user identified by user ID AA01, it is shown that the attribute is identified as "parenting generation" based on the usage history of apps for parenting generation and the response history to questionnaires for parenting generation. With such information, for example, by using the camera app log of a user with a specific attribute, it is possible to estimate a POI specialized for a person with a specific attribute. In this way, the auxiliary information storage unit 16 stores auxiliary information that can be used for filtering. Note that auxiliary information is not limited to that shown in FIG. 5, and can be freely collected and used according to the filtering concept, etc.

The POI estimation unit 17 has a function of estimating a POI based on the camera app log after filtering by the auxiliary information storage unit 16. A basic method of estimating a POI is to extract a mesh with a high number of times the camera app has been used (for example, a predetermined value or more) and set it as a POI.

The POI information storage unit 18 has a function of storing information on meshes estimated to be POIs by the POI estimation unit 17. FIG. 6 shows an example of POI information stored in the POI information storage unit 18. As shown in FIG. 6, the POI information includes information identifying the POI (POI ID) and information identifying the mesh estimated to be the POI (mesh ID). Note that, as described below, when the category of the POI is identified, the information storage destination may be changed from the POI information storage unit 18 to the category-identified POI information storage unit 20.

The POI category estimation unit 19 has a function of assigning a category to the POI estimated by the POI estimation unit 17. The category assigned in association with the POI indicates a classification according to the characteristics of the POI. For example, if a specific structure (castle, temple, shrine, etc.) exists in a mesh estimated to be a POI, the type of this structure may be the category. Also, if it is estimated to be a POI for a user with a specific hobby (fishing, mountain climbing, etc.), the hobby may be the category. In addition, for a mesh that becomes a POI only at a specific time of year, such as a famous cherry blossom viewing spot, information specifying the reason for becoming a POI (for example, a famous cherry blossom viewing spot) may be the category. In this way, the category can be set from various information as information indicating the characteristics of the POI. Also, when specifying the category, for example, information used for filtering in the filtering unit 15 can be used. This point will be described later.

The category-specified POI information storage unit 20 has a function of storing information on POIs (meshes) to which categories have been assigned in the POI category estimation unit 19. FIG. 7 shows an example of category-specified POI information stored in the category-specified POI information storage unit 20. As shown in FIG. 7, the category-specified POI information is assigned information that specifies the category, in comparison with the information shown in FIG. 6.

The output unit 21 has a function of outputting information stored in the POI information storage unit 18 and the category-identified POI information storage unit 20. The output unit 21 has a function of preparing and transmitting information on the POI to be extracted, for example, based on a request from an external device. The output unit 21 may also output information to a display device such as a monitor or a printer.

[How to analyze app logs]
Next, the application log analysis method by the application log analysis device 1 will be described with reference to Fig. 8 to Fig. 16. As described above, various filtering methods are possible. Therefore, first, the overall flow will be described, and then, detailed procedures according to the filtering method will be described.

(Series of steps)
8, the application log analysis device 1 first executes step S01. In step S01, the application usage log receiving unit 11 of the application log analysis device 1 acquires an application usage log from the user terminal 9, and the location information receiving unit 13 acquires location information from the user terminal 9. The acquired data is stored in the application usage log storage unit 12 (application usage log) and the location information storage unit 14 (location information). The information may be acquired from the user terminal 9 periodically (for example, every few hours to every few days), or may be acquired by transmitting an information request from the application log analysis device 1 to the user terminal 9.

Next, the application log analysis device 1 executes step S02. In step S02, the filtering unit 15 of the application log analysis device 1 extracts a camera application log based on the information stored in the application usage log storage unit 12 and the location information storage unit 14, and associates the usage log with the location information. The camera application log includes the time when the application was used. Therefore, location information corresponding to that time is extracted from the information stored in the location information storage unit 14, and association is performed. Note that the processing in step S02 may be performed at the timing when information is obtained from the user terminal 9 in step S01. In addition, the application usage log after the above association is performed may be stored in the application usage log storage unit 12.

Next, the application log analysis device 1 executes step S03. In step S03, the filtering unit 15 extracts (filters) data to be used for estimating the POI based on the above information and the information stored in the application usage log storage unit 12, the location information storage unit 14, and the auxiliary information storage unit 16.

Next, the application log analysis device 1 executes step S04. In step S04, the POI estimation unit 17 estimates the POI from the camera application log extracted in step S03. As a method of estimating the POI, for example, a period to be estimated (for example, several hours to several tens of days) is set, and a mesh unit with a large number of camera application logs (i.e., a large number of times the camera application is started) during the period is determined to be a POI. As an example, a method is given in which a threshold value for estimating the POI is determined in advance, and an area (mesh) in which the number of camera application logs exceeds the threshold is determined to be a POI. However, the criteria for determining the POI are not limited to this. Information related to the area (mesh) estimated to be a POI is stored in the POI information storage unit 18, for example, with information (ID) that identifies the POI attached. Note that, in the estimation of the POI by the POI estimation unit 17, information stored in the application usage log storage unit 12, the position information storage unit 14, and the auxiliary information storage unit 16 may also be used as necessary.

Next, the application log analysis device 1 executes step S05. In step S05, the POI category estimation unit 19 estimates the POI category for the area (mesh) estimated to be a POI in step S04. Information related to the POI whose category has been estimated is stored in the category-identified POI information storage unit 20, for example, in a state in which the information is associated with the category that is the estimation result.

Note that various specific procedures for steps S03 to S05 are possible. Therefore, the specific procedures will be described later.

Next, the application log analysis device 1 executes step S06. In step S06, the output unit 21 extracts and outputs POI information that satisfies a predetermined condition from the POI information stored in the POI information storage unit 18 and the category-identified POI information storage unit 20. Examples of the predetermined condition include conditions included in a request from an external device, etc. Through the above series of processes, POI estimation and category estimation are completed.

Next, a specific example of the procedure relating to steps S03 to S05 above will be described with reference to Figures 9 to 16. As described above, various types of information can be assumed to be used for filtering and POI category estimation. Below, we will explain how to use representative information used for filtering and/or category estimation.

(Example of filtering based on user activity area)
9 is a diagram for explaining a procedure of filtering using information on a user's area of activity. In this case, in step S11, the filtering unit 15 of the application log analysis device 1 calculates the user's daily living area based on the information stored in the location information storage unit 14. Note that the calculation of the daily living area (step S11) may be performed in advance.

Next, in step S12, the filtering unit 15 checks whether the location where the camera app was used is outside the daily living area. This is determined based on whether the location information associated with the camera app log is outside the daily living area. In the example shown in this embodiment, when FIG. 3 and FIG. 4 are associated with each other, it can be determined that the user terminal 9 was outside the daily living area when camera app A was used between "13:00 and 13:05". If the camera app was used outside the daily living area (step S12-YES), it is determined that the log is to be used to estimate the POI (step S13). On the other hand, if the camera app was used in the daily living area (i.e., it was not used outside the daily living area) (step S12-NO), it is determined that the log is not to be used to estimate the POI (step S14).

In this way, the camera app log to be used for estimating the POI may be extracted by focusing on the user's area of activity and the timing of use of the camera app. In this case, for example, a configuration can be realized in which logs from when the camera app is used in the user's daily life are not used for estimating the POI. Note that in the above example, a distinction is made between the daily life area and other areas, but the filtering conditions may be switched depending on the purpose of estimating the POI.

(An example of filtering based on user behavior)
Fig. 10 is a diagram for explaining a procedure of filtering according to the type of apps used before and after the camera app, as filtering focusing on user behavior. In this case, the filtering unit 15 of the app log analysis device 1 identifies apps used by the user (operated on the user terminal 9) during a predetermined time period (e.g., 10 minutes before and after) before and after the camera app based on information stored in the app usage log storage unit 12 in step S21. In the example shown in Fig. 3, for example, a sightseeing guide app, a SNS app, and a plant encyclopedia app shown 10 minutes before and after (12:50-13:00 and 13:05-13:15) the usage time period (13:00-13:05) of the camera app A may be identified as apps used before and after the camera app A.

Next, in step S22, the filtering unit 15 checks whether the apps used before and after the camera app include a social networking app. This is determined based on whether the apps extracted in step S21 above include a social networking app.

In the example shown in this embodiment, an "SNS app" is used, but this is just one example, and is an example that takes into consideration the high affinity between camera apps and SNS apps. In this embodiment, when an SNS app is used before or after a camera app, it is assumed that the user intends to link (or has attempted to link) images captured with the camera app to the SNS app, and filtering is performed based on the idea that this camera app log is used to estimate the POI. If an SNS app is used before or after camera app A (step S22-YES), it is determined that the log will be used to estimate the POI (step S23). On the other hand, if an SNS app is not used before or after camera app A (step S22-NO), it is determined that the log will not be used to estimate the POI (step S24).

In this way, the camera app log to be used for estimating the POI may be extracted by focusing on the relationship between the camera app and the apps used before and after the camera app. In this case, for example, a configuration can be realized that focuses on specific user actions before and after using the camera app, and enables the extraction of the camera app log to be used for estimating the POI. Note that instead of the above example, for example, a configuration that focuses only on the time after the camera app is used may be used. Also, extraction may be performed based on the presence or absence of the use of a specific hobby app instead of the SNS app.

(Example of category identification using user behavior)
When the camera app log used for estimating the POI is extracted using the type of app used before and after the camera app shown in Fig. 10, the type of app used as a criterion for extracting the camera app log may be used as the type of category to be assigned to the POI. Fig. 11 is a diagram explaining a procedure for performing category estimation using the same information when filtering according to the type of app used before and after the camera app is performed as filtering focusing on user behavior.

In this case, the filtering unit 15 of the application log analysis device 1 identifies applications used by the user (operated on the user terminal 9) during a predetermined time period (e.g., 10 minutes before and after) before and after the camera application based on the information stored in the application usage log storage unit 12 in step S31. This procedure is the same as step S21 shown in FIG. 10. Furthermore, the filtering unit 15 extracts the camera application log used to estimate the POI by the procedure shown in FIG. 10. The procedure at this time is the same as, for example, steps S22 to S24 in FIG. 10.

Next, the POI estimation unit 17 identifies the camera app log used for estimating the POI according to the procedure shown in FIG. 10, and then performs a process of estimating the POI in step S32. The procedure for estimating the POI is not particularly limited. Next, for the POI estimated in step S32, the POI category estimation unit 19 identifies the category of the POI in step S33. At this time, the POI category estimation unit 19 focuses on the app used as the criterion for extracting the camera app log in the filtering unit 15, and assigns a category to the POI according to the type of the app. In the example shown in FIG. 10, the camera app log used for POI estimation is extracted based on the SNS app. In this case, the POI category may be set to "SNS related (POI with high affinity with SNS app)" for the area (mesh) estimated to be a POI.

If the app used as the criterion when extracting the camera app log to be used for POI estimation is one that has a strong relationship to a more specific hobby (e.g., fishing, mountain climbing, etc.), it is possible that the reason for using the camera app is related to that hobby. Therefore, if a POI is estimated from a camera app log extracted using information related to such a hobby, it is possible that this POI is related to the hobby. Therefore, as described above, by assigning information that specifies the type of app used as the criterion for extraction as a category related to the POI that is the estimated result, it is possible to estimate and assign a category that is thought to be highly related to the POI.

In this way, in addition to extracting the camera app log used to estimate the POI by focusing on the relationship between the camera app and the apps used before and after the camera app, information about the apps used before and after the camera app (here, the type) for the POI can also be used as the category estimation result. In this case, for example, a configuration can be realized that focuses on specific actions of the user before and after using the camera app, making it possible to identify the POI category.

The logic used in FIG. 11 may be used or not used depending on the type of app that extracts the camera app log used to estimate the POI. For example, when extracting a camera app log for POI estimation based on an app related to a hobby, a category may be assigned using the method shown in FIG. 11, but when extracting a camera app log based on another app (e.g., an app that is thought to have little relevance to a hobby), a category may not be assigned using the method shown in FIG. 11.

(Example of filtering and category specification using user segments)
Instead of identifying a filtering category using information related to the type of app used before and after the camera app described with reference to Figures 10 and 11, a similar procedure may be performed using other information. The other information is not limited to information acquired from the user terminal 9, such as an app usage log and location information of the user terminal 9, but may be information stored in the auxiliary information storage unit 16. Below, an example focusing on attributes as a segment for classifying users will be described.

Figure 12 is a diagram explaining the procedure for filtering and category identification based on information on user attributes stored in the auxiliary information storage unit 16 as filtering focusing on user attributes. In this case, first, in step S41, the filtering unit 15 of the application log analysis device 1 identifies a segment for each user who owns the user terminal 9 based on the information stored in the auxiliary information storage unit 16. For example, if the auxiliary information storage unit 16 stores the information shown in Figure 5 (b), the attributes assigned to each user can be used as a segment. Other possible segments include, for example, segments based on the user's gender, place of residence, occupation, etc.

Next, in step S42, the filtering unit 15 classifies (extracts) the camera application log to be used for estimating the POI for each user segment. This procedure corresponds to extracting the camera application log to be used for estimating the POI (e.g., steps S12 to S14 in FIG. 9). Since the camera application log is classified for each segment, there will be as many groups for estimating the POI as there are segments. Note that the groups that actually perform the POI estimation may be only a part of these groups.

Next, in step S43, the POI estimation unit 17 performs a process of estimating a POI. The method of POI estimation is not particularly limited, but a process related to POI estimation is performed for each classification based on a segment. Next, in step S44, the POI category estimation unit 19 identifies the category of the POI. At this time, the POI category estimation unit 19 uses the segment used to classify the camera app log in the filtering unit 15 as the category related to the POI.

As described above, the camera app logs used for POI estimation are classified based on segments, so the POI estimated from the classified camera app logs is likely to be related to that segment. Therefore, when a POI is estimated from a camera app log classified using a segment such as a user-related attribute, it is possible that this POI is related to that segment. Therefore, as described above, by assigning information related to the segment used to classify the camera app log as a category related to the POI that is the estimated result, a category that is thought to be more relevant can be assigned to the POI.

In this way, the camera app log used to estimate the POI may be extracted (classified) by focusing on the segment of the user who owns the user terminal 9, and information related to the segment may be assigned as a category to the estimated POI. In this case, a POI category according to the user's segment may be assigned, so that a configuration can be realized that can assign a category that is estimated to have a certain degree of accuracy. Note that how the information used for the segment (information such as that shown in FIG. 5(b)) is obtained, and which information is used to classify the camera app log, may be changed as appropriate.

(An example of filtering and categorizing based on the type of camera app)
Next, the filtering category specification focusing on the type of camera app will be described with reference to FIG. 13. This method can be effective, for example, when a camera app specialized for a specific purpose (for example, a camera app having a function suitable for recording related to a specific hobby) is used in the user terminal 9. As shown in FIG. 13, first, the POI estimation unit 17 of the application log analysis device 1 classifies the camera application log acquired from the user terminal 9 by the type of camera app in step S51. Since the user terminal 9 is often equipped with multiple types of camera apps, the camera application log transmitted from one user terminal 9 may also include those originating from multiple types of camera apps. Therefore, classification is performed by the type.

Next, in step S52, the POI estimation unit 17 performs a process of estimating a POI for each of the classified types of camera apps. The method of POI estimation is not particularly limited, but the process of POI estimation is performed for each group of camera app logs classified based on the type of camera app.

Next, in steps S53 to S55, the POI category estimation unit 19 determines whether or not to estimate the category of the POI based on the type of camera app. If the camera app is a common type (e.g., one that is installed as standard on the user terminal 9), it is unlikely that the classification results will be biased towards a specific one based on the type of camera app. On the other hand, the startup log of a camera app that is expected to be used for a specific purpose is highly related to the specific purpose, so it is considered effective to use that purpose as the app category. As described above, the camera app logs used for POI estimation are classified based on segments, so it is likely that the POI estimated from the camera app logs after classification is related to that segment.

In step S53, the POI category estimation unit 19 determines whether the camera app used as the classification standard is for a specific category. If the camera app is for a specific category (step S53-YES), the category of the camera app (use, features, etc.) is assigned to the POI category (step S54). On the other hand, if the camera app is not for a specific category (step S53-NO), the category of the camera app is not assigned to the POI category (step S55), and no category is assigned to the POI. In this way, the camera app log used to estimate the POI may be extracted (classified) with attention paid to the type of camera app, and the information may be assigned as a category to the estimated POI according to the type of camera app. In this case, it is possible to identify the POI category by using the relationship between the camera app and the category, and it is possible to realize a configuration that can assign a category with high accuracy, especially when a distinctive camera app is used. In addition, whether or not the type of camera app is used for the POI category can be determined in advance, and the information can be stored in the auxiliary information storage unit 16 or the like in advance as information that serves as a criterion for whether or not to assign a category.

(POI estimation taking into account population in each mesh)
Next, an example of a method for estimating a POI in which other requirements are taken into consideration, rather than simply determining a mesh with a large number of camera application logs as a POI, will be described. In the above embodiment, the application log analysis device 1 was described assuming that a mesh with a large number of camera application logs within a predetermined period is basically estimated as a POI, as shown in FIG. 1. However, for example, if there are a large number of user terminals 9 present in a mesh, it is expected that the number of camera application logs will increase regardless of the presence or absence of a POI in the mesh. Therefore, a method of estimating a POI after performing calculations taking into account the population in the mesh may be adopted. FIG. 14 shows an example of this.

First, as shown in step S61, when estimating a POI, the POI estimation unit 17 calculates the total number of times the camera app log is logged in each mesh. Next, as shown in step S62, the POI estimation unit 17 calculates the ratio of the number of times the camera app log is logged to the number of population in each mesh. To perform this calculation, for example, population information as shown in FIG. 5(a) is acquired in advance. When the auxiliary information holding unit 16 holds the information shown in FIG. 5(a) in advance, for example, the number of times the camera app log is logged in each time period in each mesh (for example, the time period corresponding to the demographic statistics shown in FIG. 5(a)) is first calculated, and this is divided by the population of the corresponding time period in the corresponding mesh shown in FIG. 5(a). As a result, the number of times the camera app log is logged relative to the population is calculated. After performing the above calculation for each mesh, the POI estimation unit 17 estimates the mesh in which the "camera app log count/population" is equal to or greater than a predetermined value as a POI in step S63. With this configuration, it is possible to prevent a location where the number of camera app logs is increasing simply due to an increase in population from being estimated as a POI.

When estimating POI using camera app logs with a time span wider than the unit time slot (1 hour in FIG. 5(a)) that is the unit of population calculation shown in FIG. 5(a), various calculations may be added or modified. For example, after calculating "number of camera app logs/population" for each unit time slot, the calculation results for each unit time slot may be averaged. Alternatively, the average population fluctuation for each unit time slot may be calculated in advance, and the ratio of the sum of the number of camera app logs in all time slots to be used for POI estimation may be calculated for each mesh. In this way, the calculation method may be modified as appropriate within the concept of using the ratio of the number of camera app logs to population.

(Identifying a POI category using data of previously categorized POIs)
Next, another method for identifying a POI category will be described with reference to Fig. 15 and Fig. 16. The above examples have basically been described with reference to cases where a POI is estimated from the number of camera application logs in mesh units, and a category is assigned to the POI from the camera application logs or other peripheral information. In contrast to this, a method will be described in which a category is assigned to a POI whose category has not been identified yet by using data of a POI whose category has been identified in the past (whose category is known).

First, in step S71, the POI category estimation unit 19 acquires information on the change over time of the number of times the camera app is used for the information of the category-specified POI stored in the category-specified POI information storage unit 20. As described above, the category-specified POI information storage unit 20 stores information for identifying the POI to which a category is assigned. In addition, information related to the location information of the mesh can be extracted from the location information storage unit 14 using the mesh ID. Using these pieces of information, the POI category estimation unit 19 creates information related to the change over time of the number of times the camera app is used for each POI. FIG. 16 is a diagram showing an example of the change over time. FIG. 16 shows, for example, the change over time of the number of times the camera app is used at a POI that is a famous cherry blossom viewing spot. In a famous cherry blossom viewing spot, it is assumed that the number of times the camera app is used increases extremely during the cherry blossom blooming period (period T1 shown in FIG. 16), and that the distribution has a peak on the day when the cherry blossoms are in full bloom (date Td shown in FIG. 16). In this way, there may be spots in the POI where the number of times the camera app is used increases only in a particular season. Also, for example, in spots that are characterized by sunset scenery, the number of times the camera app is used may increase only in the evening. In this way, for POIs where the number of times the camera app is used may vary during a particular period or time period, it may be possible to identify the category of the POI by utilizing that variation. Specifically, this is a method of identifying the category of a POI whose category is unknown, by referring to the variation (seasonal variation or time variation) in the camera app log of an existing POI whose category is known.

To realize the above method, first, in step S71, the POI category estimation unit 19 acquires the change over time of the camera app log for POIs whose categories are known. Then, in step S72, the POI category estimation unit 19 acquires the change over time of the camera app log for POIs whose categories are not specified. Then, the change over time may be quantified, for example, using time series clustering. In addition, a value that evaluates the magnitude of the peak may be obtained using, for example, the ratio of the number of camera app logs at the peak (for example, the day when the cherry blossoms are in full bloom) to the average number of camera app logs on other days.

As described above, in order to identify the category using the change over time in the camera app log, for example, for POIs where the change over time in the camera app log is small, the subsequent processing may be omitted. In addition, whether to calculate daily fluctuations (those that change on a daily to monthly basis, such as famous cherry blossom spots) or hourly fluctuations (those with distinctive changes over time, such as famous sunset spots) may be selected, for example, from the variation or tendency of data for POIs with unspecified categories.

Next, in step S73, the POI category estimation unit 19 determines whether any of the changes over time in the camera app log counts for each POI calculated in step S71 (changes over time in the number of times the camera app is used) have a similar tendency to the changes over time in the camera app log counts calculated in step S72 (changes over time in the number of times the camera app is used). At this time, for example, if the changes over time have been quantified in the previous process, similar ones may be searched for by comparing the numerical values. Furthermore, if quantification has not been performed, POIs showing similar fluctuations may be identified using a classification method using known statistical processing, or classification may be performed focusing on, for example, the peak date and time and the distribution of the number of times the camera app is used centered on the peak.

If the data of the POI with a specified category calculated in step S71 contains something that has a similar trend to the change over time of the number of camera app uses calculated in step S72 (change over time in the number of camera app logs) (step S73-YES), it is determined that the POI with a specified category and the POI with an unspecified category are included in the same category, and the POI with an unspecified category is assigned the same category as the POI with a similar trend. On the other hand, if the data of the POI with a specified category calculated in step S71 does not contain something that has a similar trend to the change over time of the number of camera app uses calculated in step S72 (change over time in the number of camera app logs), in this procedure it is determined that no category is assigned to the POI with an unspecified category (step S75).

In this way, by using data on changes over time in the number of camera app logs (number of times the camera app is used) at POIs for which a category has already been identified, and using a method to assign the same category as POIs with similar changes over time, it becomes possible to accurately assign categories to POIs that have characteristics that may cause changes over time in the number of times the camera app is used.

POIs where seasonal (daily) variations are expected include areas where there are plants whose best viewing times change. Seasonal variations in the number of times camera apps are used are also expected in places affected by climate change (e.g., ski resorts, mountain areas, etc.).

On the other hand, POIs where time fluctuations are expected include, for example, locations that are affected by the position of the sun. Locations where some kind of event is held at a specific time of day (for example, animals are fed at a fixed time every day) are also likely to have time fluctuations in the number of times the camera app is used. There may also be areas where the number of times the camera app is used increases only at specific times of the day during specific seasons.

As described above, when identifying a POI from the number of times the camera app is used, attention is paid to the time or daily fluctuations, and POIs that show similar fluctuations are determined to belong to the same category. This makes it possible to use information about known POIs to identify the category of POIs whose category has not yet been identified.

Instead of "determining that POIs showing similar fluctuations belong to the same category," POIs showing similar fluctuations may be determined to belong to the same type of category (not the same category, but a category with similar characteristics). For example, camera app logs are expected to be concentrated in specific periods for famous cherry blossom viewing spots and famous autumn foliage viewing spots. However, the period when camera app usage logs are concentrated differs between spring and autumn. In this way, a configuration may be used to supplement information for estimating the category of an unspecified POI by using information on POIs showing similar fluctuations but in different periods.

The filtering procedures and categorization procedures shown in Figures 9 to 15 can be combined as appropriate. For example, a method of filtering that combines the above-mentioned methods (for example, a combination of filtering using the area of movement and filtering using the apps used before and after) can also be used.

[Action]
According to the application log analysis device 1, the POI is estimated by filtering each of the camera application usage logs associated with the location information acquired from the multiple user terminals 9 based on the terminal-related information, which is information about each of the multiple user terminals, and then classifying the logs based on the location information and the date and time information. In this way, the POI is estimated using the camera application usage logs after filtering based on the terminal-related information about each of the multiple user terminals 9, so that an area suitable for photography can be estimated based on the application usage logs. In particular, since the POI is estimated after filtering to remove the camera application usage logs that may become noise in the estimation of the POI, a place more suitable as a POI can be estimated.

In addition, an area with a large number of camera app usage logs can be assumed to be a POI suitable for camera app use for more users. Therefore, by estimating an area with a large number of camera app usage logs as a POI, it is possible to estimate an area suitable for photography.

The number of camera app usage logs can vary depending on the attractiveness of the POI as well as the influence of the population of the area. Therefore, as in the above example, by configuring an area where the ratio of the number of camera app usage logs to the population in the area during a specific period is high as the POI, it is possible to estimate the POI without being influenced by differences in population.

In addition, the camera app usage log in the user's daily life area may be related to personal photography, and therefore may not be suitable for presenting to other users as a POI. Therefore, as described in the above embodiment, the user's area of activity of the user terminal is identified based on the location information of the user terminal during times when the camera app is not being used, and the camera app usage log in a location that matches the user's daily life area as the area of activity is determined not to be used for estimating the POI, thereby preventing the POI from being estimated based on the camera app usage log in the daily life area. Note that the location information used to identify the area of activity may be combined with the location information of the user terminal during times when the camera app is being used.

When a camera app is used on a user terminal, it is assumed that it will be linked to other applications as described above. Furthermore, if the use of the camera app is based on a particular hobby or the like, it is also assumed that a related app will be used before and after that. Therefore, as in the above embodiment, by extracting the camera app usage log used to estimate the POI in the POI estimation unit 17 based on the type of application used during a predetermined period before and after the use of the camera app on the user terminal 9, it is possible to estimate the POI from the camera app usage log where a related application was used before and after. Therefore, it is possible to estimate a POI that has a strong relationship with a related application.

The terminal-related information may be information that identifies a segment related to the user of the user terminal 9. In this case, the information that identifies the segment related to the user may be used to extract a usage log of a camera app to be used for estimating the POI in the POI estimation unit 17. In this case, it becomes possible to estimate a POI that is related to the user's segment.

Depending on the type of camera app, some may be specialized for a particular purpose. Therefore, as described above, by configuring the POI estimation unit 17 to extract the usage log of the camera app used to estimate the POI based on the type of camera app, it becomes possible to estimate the POI associated with a specific camera app.

The system may further include a POI category estimation unit 19 that assigns a category related to the type of POI to the POI estimated by the POI estimation unit 17. In addition, when assigning a category related to the type of POI to the estimated POI, a configuration is adopted in which a category based on the device-related information used for filtering in the filtering unit is assigned, thereby making it possible to assign a category based on the device-related information.

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 as an illustrative example and does not have any restrictive meaning with respect to the present embodiment.

For example, as described above, the application log analysis device 1 described in the above embodiment can be modified in various ways. Therefore, the functions of each part may be changed in response to the changes in each part.

For example, in the above embodiment, various modifications can be made to the application log analysis device 1. However, it is sufficient that the application log analysis device 1 includes a configuration for filtering the usage log of the camera application in the filtering unit 15. To this extent, various modifications can be made to the application log analysis device 1. For example, in the application log analysis device 1, the procedure for estimating the category by the POI category estimation unit 19 may be omitted.

[others]
The block diagrams used in the description of the above embodiments show functional blocks. These functional blocks (components) are realized by any combination of at least one of hardware and software. In addition, the method of realizing each functional block is not particularly limited. That is, each functional block may be realized using one device that is physically or logically coupled, or may be realized using two or more devices that are physically or logically separated and directly or indirectly (for example, using wires, wirelessly, etc.) connected to each other and using these multiple devices. The functional block may be realized by combining the one device or the multiple devices with software. In addition, the wording "apparatus" described in this embodiment can be replaced with circuit, device, unit, etc.

Functions include, but are not limited to, judgment, determination, judgment, calculation, computation, processing, derivation, investigation, search, confirmation, reception, transmission, output, access, resolution, selection, selection, establishment, comparison, assumption, expectation, consideration, broadcasting, notifying, communicating, forwarding, configuring, reconfiguring, allocating, mapping, and assignment. For example, a functional block (component) that performs the transmission function is called a transmitting unit or a transmitter. As mentioned above, there are no particular limitations on the method of realization for each of these.

Figure 17 is a diagram showing an example of the hardware configuration of the application log analysis device 1 according to this embodiment. The application log analysis device 1 may be physically configured as a computer device including a processor C1, a memory C2, a storage C3, a communication device C4, an input device C5, an output device C6, a bus C7, etc.

In the following description, the term "apparatus" may be interpreted as a circuit, device, unit, etc. The hardware configuration of the application log analysis apparatus 1 may be configured to include one or more of the devices shown in FIG. 17. It may also be configured to exclude some of the devices.

Each function of the application log analysis device 1 is realized by loading specific software onto hardware such as the processor C1 and memory C2, which causes the processor C1 to perform calculations and control communication via the communication device C4, and the reading and/or writing of data in the memory C2 and storage C3.

The processor C1, for example, operates an operating system to control the entire computer. The processor C1 may be configured as a central processing unit (CPU) including an interface with peripheral devices, a control unit, an arithmetic unit, a register, etc. The processor C1 may also be configured to include a GPU (Graphics Processing Unit). For example, each functional unit of the application log analysis device 1 may be realized by the processor C1.

The processor C1 also reads out programs (program codes), software modules, and data from the storage C3 and/or the communication device C4 into the memory C2, and executes various processes according to these. The programs used are those that cause a computer to execute at least some of the operations described in the above-mentioned embodiments. For example, each functional unit of the application log analysis device 1 may be realized by a control program stored in the memory C2 and operated by the processor C1. Although the above-mentioned various processes have been described as being executed by one processor C1, they may be executed simultaneously or sequentially by two or more processors C1. The processor C1 may be implemented in one or more chips. The programs may be transmitted from a network via a telecommunications line.

Memory C2 is a computer-readable recording medium and may be composed of at least one of, for example, ROM (Read Only Memory), EPROM (Erasable Programmable ROM), EEPROM (Electrically Erasable Programmable ROM), RAM (Random Access Memory), etc. Memory C2 may also be called a register, cache, main memory (primary storage device), etc. Memory C2 can store executable programs (program codes), software modules, etc. for implementing a payment information management method according to one embodiment of the present invention.

Storage C3 is a computer-readable recording medium, and may be composed of at least one of, for example, 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 (registered trademark) disk), a smart card, a flash memory (e.g., a card, a stick, a key drive), a floppy (registered trademark) disk, a magnetic strip, etc. Storage C3 may be referred to as an auxiliary storage device. The above-mentioned storage medium may be, for example, a database, a server, or other suitable medium including memory C2 and/or storage C3.

The communication device C4 is hardware (transmitting/receiving 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, a communication module, etc.

The input device C5 is an input device (e.g., a keyboard, a mouse, a microphone, a switch, a button, a sensor, etc.) that accepts input from the outside. The output device C6 is an output device (e.g., a display, a speaker, an LED lamp, etc.) that performs output to the outside. Note that the input device C5 and the output device C6 may be integrated into one structure (e.g., a touch panel).

In addition, each device, such as the processor C1 and the memory C2, is connected by a bus C7 for communicating information. The bus C7 may be configured as a single bus, or may be configured with different buses between the devices.

The processing steps, sequences, flow charts, etc. of each aspect/embodiment described in this disclosure may be reordered unless inconsistent. For example, the methods described in this disclosure present elements of various steps using an example order and are not limited to the particular order presented.

The input and output information may be stored in a specific location (e.g., memory) or may be managed using a management table. The input and output information may be overwritten, updated, or added to. The output information may be deleted. The input information may be transmitted to another device.

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

Each aspect/embodiment described in this disclosure may be used alone, in combination, or switched depending on the execution. In addition, notification of specific information (e.g., notification that "X is the case") is not limited to being done explicitly, but may be done implicitly (e.g., not notifying the specific information).

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.

Software, instructions, information, etc. may also be transmitted and received via 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, Digital Subscriber Line (DSL)), and/or wireless technologies (such as infrared, microwave), then these wired and/or wireless technologies are included within the definition of a transmission medium.

The information, signals, etc. described in this disclosure may be represented using any 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.

As used in this disclosure, the terms "system" and "network" are used interchangeably.

As used in this disclosure, the terms "determining" and "determining" may encompass a wide variety of actions. "Determining" and "determining" may include, for example, judging, calculating, computing, processing, deriving, investigating, looking up, searching, inquiring (e.g., searching in a table, database, or other data structure), ascertaining, and deeming something to be "determined" or "determined." Furthermore, "judgment" and "decision" may include regarding receiving (e.g., receiving information), transmitting (e.g., transmitting information), input, output, and accessing (e.g., accessing data in a memory) as having been "judged" or "decided". Furthermore, "judgment" and "decision" may include regarding resolving, selecting, choosing, establishing, comparing, and the like as having been "judged" or "decided". In other words, "judgment" and "decision" may include regarding some action as having been "judged" or "decided". Furthermore, "judgment" may be interpreted as "assuming", "expecting", "considering", and the like.

The terms "connected", "coupled", or any variation thereof, refer to any direct or indirect connection or coupling between two or more elements, and may include the presence of one or more intermediate elements between two elements that are "connected" or "coupled" to each other. The coupling or connection between elements may be physical, logical, or a combination thereof. For example, "connected" may be read as "access". As used in this disclosure, two elements may be considered to be "connected" or "coupled" to each other using at least one of one or more wires, cables, and printed electrical connections, as well as electromagnetic energy having wavelengths in the radio frequency range, microwave range, and light (both visible and invisible) range, as some non-limiting and non-exhaustive examples.

As used in this disclosure, the phrase "based on" does not mean "based only on," unless expressly stated otherwise. In other words, the phrase "based on" means both "based only on" and "based at least on."

When the terms "include," "including," and variations thereof are used in this disclosure, these terms are intended to be inclusive, similar to the term "comprising." Additionally, the term "or," as used in this disclosure, is not intended to be an exclusive or.

In this disclosure, where articles have been added through translation, such as a, an, and the in English, this disclosure may include that the noun following these articles is in the plural form.

In this disclosure, the term "A and B are different" may mean "A and B are different from each other." The term may also mean "A and B are each different from C." Terms such as "separate" and "combined" may also be interpreted in the same way as "different."

[Additional Notes]
An application log analysis device according to one embodiment has a terminal information acquisition unit that acquires, from a plurality of user terminals, information identifying the user terminal, a camera app usage log including date and time information, and location information identifying the location where the camera app was used for each of the usage logs; a filtering unit that filters each of the camera app usage logs associated with the location information related to the plurality of user terminals acquired by the terminal information acquisition unit to filter camera app usage logs that satisfy specific conditions based on terminal-related information that is information related to each of the plurality of user terminals; and a POI estimation unit that estimates a POI suitable for using the camera app by classifying the camera app usage logs associated with location information extracted by the filtering unit based on the location information and the date and time information.

According to the above application log analysis device, the POI is estimated by filtering each of the camera application usage logs associated with location information acquired from multiple user terminals based on device-related information, which is information about each of the multiple user terminals, and then classifying the logs based on the location information and date and time information. As described above, the POI is estimated using the camera application usage logs after filtering based on the device-related information about each of the multiple user terminals, so that areas suitable for photography can be estimated based on the application usage logs acquired from the multiple user terminals. In particular, since the POI is estimated after filtering to remove camera application usage logs that may become noise in estimating the POI, a location more suitable as a POI can be estimated.

The POI estimation unit may estimate, as a POI, an area in which there are many usage logs of the camera app during a specific period.

Areas with many camera app usage logs can be assumed to be POIs suitable for camera app use for more users. Therefore, by estimating the above areas as POIs, it is possible to estimate areas suitable for photography.

The POI estimation unit may estimate, as a POI, an area in which the ratio of the number of camera app usage logs to the population in the area during a specific period is high.

The number of camera app usage logs can vary depending on the attractiveness of the POI as well as the influence of the population of the area. Therefore, by configuring the system to estimate an area as a POI based on the ratio of the number of camera app usage logs to the population in the area during a specific period, it is possible to estimate the POI without being influenced by differences in population.

The device-related information may be location information of the user device during times when the camera app is not being used, and the device information acquisition unit may acquire location information of the user device other than the location where the camera app is being used from the multiple user devices. The filtering unit may identify the user's area of activity from the location information of the user device during times when the camera app is not being used, and if the location information associated with the camera app usage log matches the user's everyday life area as the user's area of activity, determine that the camera app usage log is not to be used for estimating a POI in the POI estimation unit.

The camera app usage log in the user's daily life area may be related to personal photography and may therefore not be suitable for estimating a POI that may be presented to other users. Therefore, as described above, the user's area of activity of the user's terminal is identified based on the location information of the user terminal during times when the camera app is not being used, and the camera app usage log in a location that coincides with the user's daily life area as the user's area of activity is determined not to be used for estimating a POI, thereby preventing a POI from being estimated based on the camera app usage log in the daily life area. Note that the location information used to identify the area of activity may be combined with the location information of the user terminal during times when the camera app is being used.

The device-related information may be a usage log of an application other than the camera app on the user device, the device information acquisition unit may acquire the usage log of the application other than the camera app from the multiple user devices, and the filtering unit may extract the usage log of the camera app to be used for estimating the POI in the POI estimation unit based on the type of application used during a predetermined time before and after the use of the camera app, which is identified based on the usage log of the application other than the camera app.

When a camera app is used on a user terminal, it is expected that it will be linked to other applications. In addition, if the use of the camera app is based on a particular hobby, etc., it is also expected that a related app will be used before and after the use. Therefore, as described above, by extracting the camera app usage log used to estimate the POI in the POI estimation unit based on the type of application used during a specified period before and after the use of the camera app on the user terminal, it is possible to estimate the POI from the camera app usage log where a related application was used before and after, and therefore it is possible to estimate a POI that has a strong relationship with the related application.

The device-related information may be information that identifies a segment related to a user of the user device, and the filtering unit may extract a usage log of the camera app in the user device owned by a user who matches a specific segment as the usage log of the camera app to be used for estimating a POI in the POI estimation unit.

As described above, by using information that identifies a segment related to a user to extract the usage log of the camera app used to estimate the POI in the POI estimation unit, it becomes possible to estimate the POI that is related to the user's segment.

The device-related information may be information that identifies the type of camera app in the user device, and the device information acquisition unit may acquire information that identifies the type of camera app used for each camera app usage log from the multiple user devices, and the filtering unit may extract a specific type of camera app usage log from the camera app usage logs in the user device as the camera app usage log to be used for POI estimation in the POI estimation unit.

Depending on the type, there may be camera apps that are specialized for specific uses. Therefore, as described above, by configuring the POI estimation unit to extract the usage log of the camera app used to estimate the POI based on the type of camera app, it becomes possible to estimate the POI associated with a specific camera app.

The system may further include a POI category estimation unit that assigns a category related to the type of POI to the POI estimated by the POI estimation unit, and the POI category estimation unit may assign a category to the POI based on the terminal-related information used for filtering in the filtering unit.

As described above, when assigning a category related to the type of POI to an estimated POI, a category based on the device-related information used for filtering in the filtering unit is assigned, making it possible to assign a category based on the device-related information.

The application log analysis device according to one embodiment includes a terminal information acquisition unit that acquires, from a plurality of user terminals, information identifying the user terminal, a camera application usage log including date and time information, and location information identifying the location where the camera application was used for each of the usage logs; a POI estimation unit that estimates a POI suitable for using the camera application by classifying each of the camera application usage logs associated with the location information for the plurality of user terminals acquired by the terminal information acquisition unit based on the location information and the date and time information; and a POI category estimation unit that assigns a category related to the type of POI to the POI estimated by the POI estimation unit based on information related to a POI whose category is known.

According to the above application log analysis device, the POI is estimated by classifying the camera application usage logs associated with the location information acquired from multiple user terminals based on the location information and date and time information. Then, a category related to the type of new POI is assigned based on the information related to the POI whose category is known. With this configuration, it is possible to estimate an area suitable for photography based on the application usage logs acquired from multiple user terminals, and since a category is assigned, the user can easily understand what type of POI each POI is. Furthermore, by assigning categories using information related to POI whose category is known, for example, the same category can be assigned to POIs with similar characteristics, improving the accuracy of category assignment.

Here, the POI estimation unit may estimate, as a POI, an area in which there are many usage logs of the camera app during a specific period.

Areas with many camera app usage logs can be assumed to be POIs suitable for camera app use for more users. Therefore, by estimating the above areas as POIs, it is possible to estimate areas suitable for photography.

The system may further include a memory unit that holds information identifying a POI whose category is known and information indicating changes over time in the usage log of the camera app at the POI, and the POI category estimation unit may calculate changes over time in the usage log of the camera app at the location estimated as the POI, and compare the changes over time in the usage log of the camera app at the POI whose category is known, thereby estimating the category of the location estimated as the POI.

Some POIs have characteristics in the changes over time in the usage logs of the camera app. Therefore, by comparing the changes over time in the usage logs of the camera app at POIs whose categories are known with the changes over time in the usage logs of the camera app at a location estimated as a POI, the category of the POI can be estimated with high accuracy, particularly for POIs whose categories have characteristics in the changes over time.

The POI category estimation unit may calculate a change over time in the usage log of the camera app at the location estimated as the POI, and if the change over time in the usage log of the camera app at a POI whose category is known is similar, the POI whose category is known may be assigned to the POI estimated by the POI estimation unit.

As described above, when the change over time in the usage log of the camera app at a POI whose category is known is similar to the change over time in the usage log of the camera app at a location estimated as a POI, the category can be accurately estimated by assigning the same category as the POI whose category is known.

The system may further include a filtering unit that filters each of the camera app usage logs associated with the location information related to the multiple user terminals acquired by the device information acquisition unit based on device-related information related to each of the multiple user terminals, and the POI estimation unit may classify each of the camera app usage logs associated with the location information related to the multiple user terminals after filtering by the filtering unit based on the location information and the date and time information, thereby estimating a POI suitable for using the camera app.

The POI is estimated by filtering each of the camera app usage logs associated with location information acquired from multiple user devices based on device-related information, which is information about each of the multiple user devices, and then classifying the logs based on the location information and date and time information. With this configuration, the POI is estimated after filtering to remove camera app usage logs that may become noise in the POI estimation, making it possible to estimate a location that is more suitable as a POI.

1... application log analysis device, 11... application usage log receiving unit, 12... application usage log storage unit, 13... location information receiving unit, 14... location information storage unit, 15... filtering unit, 16... auxiliary information storage unit, 17... POI estimation unit, 18... POI information storage unit, 19... POI category estimation unit, 20... category-specified POI information storage unit, 21... output unit, 9... user terminal, 91... user application, 92... image information storage unit, 93... application usage log acquisition unit, 94... application usage log transmission unit, 95... location information acquisition unit, 96... location information transmission unit.

Claims (6)

a terminal information acquisition unit that acquires, from a plurality of user terminals, information for identifying the user terminal, a usage log of a camera app including date and time information, and location information for identifying a location where the camera app was used for each of the usage logs;
a filtering unit that filters the camera app usage logs associated with the position information related to the plurality of user terminals acquired by the terminal information acquisition unit into camera app usage logs that satisfy a specific condition based on terminal related information that is information related to each of the plurality of user terminals;
a POI estimation unit that estimates a POI suitable for use of the camera app by classifying a usage log of the camera app associated with location information extracted by filtering by the filtering unit based on the location information and the date and time information ;
The POI estimation unit is
An area having a large number of usage logs of the camera app in a specific period is estimated as a POI;
A region in which the ratio of the number of usage logs of the camera app to the population in the region is high during a specific period is estimated as a POI.
App log analyzer. a terminal information acquisition unit that acquires, from a plurality of user terminals, information for identifying the user terminal, a usage log of a camera app including date and time information, and location information for identifying a location where the camera app was used for each of the usage logs;
a filtering unit that filters the camera app usage logs associated with the position information related to the plurality of user terminals acquired by the terminal information acquisition unit into camera app usage logs that satisfy a specific condition based on terminal related information that is information related to each of the plurality of user terminals;
a POI estimation unit that estimates a POI suitable for use of the camera app by classifying a usage log of the camera app associated with location information extracted by filtering by the filtering unit based on the location information and the date and time information;
The terminal-related information is location information of the user terminal during a time period when the camera app is not being used,
the terminal information acquisition unit acquires, from the plurality of user terminals, location information of the user terminal during a time period when the camera app is not being used;
The filtering unit identifies a range of activities of a user of the user terminal from position information of the user terminal other than a location where the camera app is used, and when the position information associated with the usage log of the camera app coincides with a daily life area as the user's range of activities, determines not to use the usage log of the camera app for estimating a POI in the POI estimation unit.
App log analyzer. a terminal information acquisition unit that acquires, from a plurality of user terminals, information for identifying the user terminal, a usage log of a camera app including date and time information, and location information for identifying a location where the camera app was used for each of the usage logs;
a filtering unit that filters the camera app usage logs associated with the position information related to the plurality of user terminals acquired by the terminal information acquisition unit into camera app usage logs that satisfy a specific condition based on terminal related information that is information related to each of the plurality of user terminals;
a POI estimation unit that estimates a POI suitable for use of the camera app by classifying a usage log of the camera app associated with location information extracted by filtering by the filtering unit based on the location information and the date and time information;
The terminal-related information is a usage log of an application other than the camera application in the user terminal,
the terminal information acquisition unit acquires usage logs of applications other than the camera app from the plurality of user terminals;
The filtering unit extracts a usage log of the camera app to be used for estimating a POI in the POI estimation unit, based on a type of an application used during a predetermined time before and after the use of the camera app, the type being identified based on a usage log of an application other than the camera app.
App log analyzer. a terminal information acquisition unit that acquires, from a plurality of user terminals, information for identifying the user terminal, a usage log of a camera app including date and time information, and location information for identifying a location where the camera app was used for each of the usage logs;
a filtering unit that filters the camera app usage logs associated with the position information related to the plurality of user terminals acquired by the terminal information acquisition unit into camera app usage logs that satisfy a specific condition based on terminal related information that is information related to each of the plurality of user terminals;
a POI estimation unit that estimates a POI suitable for use of the camera app by classifying a usage log of the camera app associated with location information extracted by filtering by the filtering unit based on the location information and the date and time information;
The terminal-related information is information for identifying a segment related to a user of the user terminal,
The filtering unit extracts a usage log of the camera app in the user terminal owned by a user who matches a specific segment as the usage log of the camera app to be used for estimating a POI in the POI estimation unit.
App log analyzer. a terminal information acquisition unit that acquires, from a plurality of user terminals, information for identifying the user terminal, a usage log of a camera app including date and time information, and location information for identifying a location where the camera app was used for each of the usage logs;
a filtering unit that filters the camera app usage logs associated with the position information related to the plurality of user terminals acquired by the terminal information acquisition unit into camera app usage logs that satisfy a specific condition based on terminal related information that is information related to each of the plurality of user terminals;
a POI estimation unit that estimates a POI suitable for use of the camera app by classifying a usage log of the camera app associated with location information extracted by filtering by the filtering unit based on the location information and the date and time information;
The terminal-related information is information for identifying a type of a camera application in the user terminal,
the terminal information acquisition unit acquires, from the plurality of user terminals, information for identifying a type of camera app used for each of the camera app usage logs;
The filtering unit extracts a usage log of a specific type of camera app from the usage log of the camera app in the user terminal as the usage log of the camera app to be used for estimating a POI in the POI estimation unit.
App log analyzer. The method further includes a POI category estimation unit that assigns a category related to a type of the POI to the POI estimated by the POI estimation unit,
The application log analysis device according to claim 1 , wherein the POI category estimation unit assigns a category to the POI based on the terminal-related information used for filtering in the filtering unit.

Images