User Behavior Support Device, Method, and Program

Abstract

An aspect of this invention acquires a notification log that represents an output result of notification information when behavior modification of a user is assisted by outputting the notification information for encouraging the user to modify the behavior, acquires an access log that represents a result of a response to the notification information from the user, and additionally acquires a behavior log that represents a behavior result of the user. In addition, the aspect of the invention selects, on the basis of the acquired notification log and access log, an access log from the user to at least a plurality of pieces of past notification information including the notification information output immediately before the behavior of the user, and estimates, on the basis of the behavior log of the user and the selected access log, user characteristics that represent a tendency pertaining to the user behavior.

Description

Technical Field

[0001] One aspect of the present invention relates to a user behavior support device that provides support for promoting changes in user behavior, for example, and a user behavior support method and program executed by this device.

Background Art

[0002] Conventionally, by using an information communication terminal such as a smartphone used by a user, an access log to a web service or application by the user is collected, and based on the collected access log and a behavior log representing the behavior content of the user, a technology for estimating user characteristics representing the tendency related to the user's behavior has been proposed (see, for example, Patent Document 1). When using this technology, for example, when providing a service that promotes changes in user behavior, it becomes possible to generate and present an appropriate message or the like according to the characteristics for each user, and a high effect can be expected in promoting behavior changes.

Prior Art Documents

Patent Documents

[0003]

Patent Document 1

Summary of the Invention

Problems to be Solved by the Invention

[0004] However, generally, this type of conventional technology estimates user characteristics based on the relevance between the behavior log at the time when the user acts and the content of the access log immediately before that. For this reason, the estimation result of user characteristics only reflects a temporary relevance between the user's behavior content and access to web services or the like, so the estimation accuracy may be low.

[0005] This invention has been made paying attention to the above circumstances, and aims to provide a technology that can further improve the estimation accuracy of user characteristics related to behavior.

Means for Solving the Problem

[0006] In order to solve the above problems, one aspect of the user behavior support device or user behavior support method according to the present invention is to obtain a notification log representing the output result of the notification information when supporting the behavior change of the user by outputting notification information for prompting the behavior change to the user, and obtain an access log representing the response result from the user to the notification information, and further obtain a behavior log representing the behavior result of the user. Then, based on the obtained notification log and access log, select access logs from the user for at least a plurality of past notification information including the notification information output immediately before the behavior of the user, and estimate user characteristics representing the tendency related to the behavior of the user based on the selected access logs and the behavior log of the user.

Effect of the Invention

[0007] According to one aspect of the present invention, user characteristics representing the tendency related to the above behavior of the user are estimated based on access logs from the user for at least a plurality of past notification information including the notification information output immediately before the behavior of the user and the behavior log of the user. Therefore, it is possible to more accurately estimate the behavior characteristics of the user compared to the case of using only the access log immediately before the behavior of the user.

[0008] That is, according to one aspect of the present invention, it is possible to provide a technology that enables further improvement in the estimation accuracy of user characteristics related to behavior.

Brief Description of the Drawings

[0009]

Figure 1

Figure 2

Figure 3

Figure 4

Figure 5

Figure 6

Figure 7

Figure 8

Figure 9

Figure 10

Figure 11

Figure 12

Figure 13

DETAILED DESCRIPTION OF THE INVENTION

[0010] Hereinafter, embodiments of the present invention will be described with reference to the drawings.

[0011] [One Embodiment] (Example of Configuration) (1) System FIG. 1 is a diagram showing the overall configuration of a system including a server device SV that functions as a user behavior support device according to an embodiment of the present invention.

[0012] In the system of one embodiment, a plurality of user terminals UT owned by users can be connected to the server device SV via a network NW.

[0013] The network NW includes, for example, a wide area network centered on the Internet and an access network for accessing this wide area network. As the access network, for example, a public data communication network using wired or wireless, or a LAN (Local Area Network) using wired or wireless is used.

[0014] The user terminals UT1 to UTn are composed of information communication terminals such as smartphones, tablet terminals, and notebook personal computers. The user terminals UT1 to UTn have communication means for transmitting and receiving information data to and from the server device SV, and a behavior support application for receiving a behavior support service by the server device SV.

[0015] When the behavior support application receives notification information for behavior support from the server device SV, it once stores the notification information, and then performs a process of displaying it on a display device according to a user operation, and when the user performs an operation for confirmation on the displayed notification information, it stores the operation result as an access log, senses the user's behavior state, creates and stores a behavior log including the sensing result, and has a function of performing a process of transmitting the notification log, access log, and behavior log to the server device SV in response to a request from the server device SV or autonomously.

[0016] Note that, as the communication means with the server device SV, for example, a browser, a mailer, or an SNS (Social Network Service) is used.

[0017] (2) Server device SV FIG. 2 and FIG. 3 are block diagrams showing the hardware configuration and software configuration of the server device SV, respectively.

[0018] The server device SV consists of, for example, a server computer arranged on the Web or in the cloud, and includes a control unit 1 that uses a hardware processor such as a Central Processing Unit (CPU). Then, a storage unit having a program storage unit 2 and a data storage unit 3 and a communication I / F unit 4 are connected to the control unit 1 via a bus 5. Note that the server device SV may be connected to, for example, a local network of a company or organization.

[0019] Under the control of the control unit 1, the communication I / F unit 4 uses a communication protocol defined by the network NW to transmit and receive data to and from the user terminals UT1 to UTn, respectively.

[0020] The program storage unit 2 is configured by combining, for example, a non-volatile memory such as an HDD (Hard Disk Drive) or SSD (Solid State Drive) that can be written to and read from at any time as a storage medium and a non-volatile memory such as a ROM (Read Only Memory). In addition to middleware such as an OS (Operating System), it stores various programs necessary to execute various control processes according to an embodiment of the present invention.

[0021] The data storage unit 3 is, for example, a combination of a non-volatile memory such as an HDD or SSD that can be written to and read from at any time as a storage medium and a volatile memory such as a RAM (Random Access Memory). As the main storage area necessary to implement an embodiment of the present invention, it includes a notification log storage unit 31, an access log storage unit 32, an action log storage unit 33, an action model storage unit 34, and a user characteristic / intervention information storage unit 35.

[0022] The notification log storage unit 31 is used to store, in association with the user terminals UT1 to UTn or the identification information of the users (hereinafter referred to as user IDs), the transmission results of the notification information for each of the user terminals UT1 to UTn, that is, the notification logs indicating the notification results for the users.

[0023] The access log storage unit 32 is used to store, in association with the user ID, the access logs obtained from the user terminals UT1 to UTn and indicating the confirmation results of the users for the above notification information.

[0024] FIG. 6 shows an example of the configurations of the notification log and the access log. In this example, both the notification log and the access log include the generation date and time, the user ID, and the identification information of the notification or access.

[0025] The behavior log storage unit 33 is used to store, in association with the user ID, the behavior logs obtained from the user terminals UT1 to UTn and including the sensing results of the users' behaviors. FIG. 7 shows an example of the configuration of the behavior log. In this example, the behavior log associates the behavior start date and time and the user ID with the values representing the goals and achievements.

[0026] The behavior model storage unit 34 is used to store, for example, a logistic regression analysis formula for creating a behavior model, and to temporarily store the data for creating the behavior model generated from the above notification log, access log, and behavior log.

[0027] The user characteristic / intervention information storage unit 35 is used to store, in association with each of a plurality of assumed behavior patterns of the user, the user characteristics corresponding to the behavior pattern and the intervention information corresponding to the user characteristics. The intervention information includes, for example, the information specifying the number of notification times or the notification timing of the notification information preset corresponding to the above plurality of user characteristics, and the message representing the notification content.

[0028] The control unit 1 includes, as a processing function according to an embodiment of the present invention, a notification information transmission processing unit 11, an access log acquisition processing unit 12, an action log acquisition processing unit 13, an action model creation processing unit 14, a user characteristic identification processing unit 15, and a notification information generation processing unit 16. Among these, the action model creation processing unit 14 and the user characteristic identification processing unit 15 constitute an estimation processing unit for user characteristics. The above processing units 11 to 16 are all realized by causing the hardware processor of the control unit 1 to execute an application program stored in the program storage unit 2.

[0029] During the action support period for the user, the notification information transmission processing unit 11 performs a process of transmitting the notification information generated by the notification information generation processing unit 16, which will be described later, from the communication I / F unit 4 to the user terminals UT1 to UTn of the target user. Also, each time the notification information transmission processing unit 11 transmits the above notification information, it generates a notification log representing the notification result, and performs a process of storing the generated notification log in the notification log storage unit 31 together with the user ID.

[0030] The access log acquisition processing unit 12 acquires access logs from the user terminals UT1 to UTn corresponding to each user every time a preset period elapses during the action support period, and performs a process of storing the acquired access logs in the access log storage unit 32 together with the user ID of the transmission source.

[0031] The action log acquisition processing unit 13 acquires action logs from the user terminals UT1 to UTn, and performs a process of storing the acquired action logs in the action log storage unit 33 in association with the user ID of the transmission source. Note that the acquisition process of the action logs may be performed independently of the acquisition process of the above access logs, or may be performed simultaneously or synchronously.

[0032] The action model creation processing unit 14 reads, for each user, the notification log, access log, and action log acquired during the target period from the above-mentioned notification log storage unit 31, access log storage unit 32, and action log storage unit 33, respectively. Then, the action model creation processing unit 14 generates data for creating an action model based on the read logs, and creates an action model based on this data for creating an action model. And, a process of determining the action pattern of the target user is performed based on the created action model. Note that a specific example of the process by this action model creation processing unit 14 will be described in the operation example.

[0033] Based on the action pattern of the user determined by the above-mentioned action model creation processing unit 14, the user characteristic identification processing unit 15 searches for user characteristics and intervention information corresponding to the action pattern from the user characteristic · intervention information storage unit 35. And, a process of providing the searched user characteristics and intervention information to the notification information generation processing unit 16 is performed.

[0034] First, at the start of action support, the notification information generation processing unit 16 generates general-purpose notification information for each action type, and performs a process of causing the notification information transmission processing unit 11 to transmit the generated general-purpose notification information to the user terminals UT1 to UTn. Also, after the start of action support, the notification information generation processing unit 16 updates the general-purpose notification information to notification information corresponding to the user characteristics of the user and reflecting the intervention information based on the user characteristics and intervention information specified by the above-mentioned user characteristic identification processing unit 15, and performs a process of causing the notification information transmission processing unit 11 to transmit the updated notification information.

[0035] (Operation example) Next, an operation example of the server device SV configured as described above will be described. Here, a case of supporting the action of a user who tries to execute "walking" for action modification will be described as an example, but the action may be other actions such as muscle training and exercise for physical strength enhancement.

[0036] FIG. 4 is a flowchart showing an example of a processing procedure and processing content of the action support process executed by the control unit 1 of the server device SV.

[0037] (1) Operations at the beginning of support For example, assume that a start request for action support is transmitted from the user terminal UT1. In this case, when the control unit 1 of the server device SV confirms the reception of the start request for the action support in step S10, it proceeds to step S11, where it starts the transmission process of the notification information.

[0038] That is, the control unit 1 of the server device SV first generates general notification information corresponding to, for example, the type of action "walking" reported in advance from the user terminal UT1 under the control of the notification information generation processing unit 16. The general notification information includes, for example, a general notification message prompting walking and information specifying the transmission timing.

[0039] Subsequently, the control unit 1 of the server device SV transmits the generated notification information from the communication I / F unit 4 to the user terminal UT1 under the control of the notification information transmission processing unit 11. In the transmission of the above notification information, substantially the notification message included in the notification information is transmitted at the timing specified by the transmission timing specification information included in the notification information. Therefore, in the following description, the notification information will be replaced with the notification message for explanation.

[0040] In this example, the transmission timing of the above notification message is set to 0 minutes every hour, but it may be set to an arbitrary time interval shorter than 1 hour, or may be set to an arbitrary time interval longer than 1 hour, for example, every few hours, every day, or every week.

[0041] Each time the notification information transmission processing unit 11 transmits the above notification message, it generates a notification log representing the transmission result of the above notification information in step S13, and stores the generated notification log in the notification log storage unit 31. The notification log includes, for example, the notification date and time, the user ID of the notification destination, and information indicating "notification" as shown in FIG. 6, but may also include other information such as the notification message.

[0042] When the user terminal UT1 receives the above notification message, it temporarily stores the notification message. Then, in this state, when the user performs an operation to view the above notification message, for example, a display operation of the message, the user terminal UT1 displays the above notification message on the display device. Note that the output means of the notification message may not be limited to display, and it may be converted into voice and output. Also, the confirmation of the notification message may be to access the corresponding site by clicking the URL (Uniform Resource Locator) described in the notification message to confirm the content of the message.

[0043] Furthermore, when the user terminal UT1 receives a confirmation operation of the above notification message by the user, it generates an access log representing the result of the confirmation operation and stores the generated access log. The access log includes, for example, the date and time when the confirmation operation was performed, the user ID, and information indicating "access".

[0044] Also, assume that the user walks autonomously after confirming the above notification message. Then, the user terminal UT1 measures the number of steps of the above walk every hour, for example, by the built-in step counting application, and generates and stores an action log including the measured number of steps. As shown in an example in FIG. 7, the action log includes the measurement start date and time set every hour, the user ID, the target value previously set by the user, and further the measured number of steps every hour is described as the actual value.

[0045] Note that the measurement period of the number of steps may be set to any time length other than one hour. Also, the target value and the actual value described in the action log may be other values such as walking time and number of walks instead of the number of steps. Furthermore, the target value does not necessarily have to be set, and only the actual value may be described.

[0046] Similarly, in the user terminal UT1 hereafter, every time the user performs a confirmation operation on the notification message sent from the server device SV, a new access log is generated and saved. Also, every time the user walks, the number of steps is measured, and an action log including the measured number of steps is newly generated and saved.

[0047] (2) Acquisition of Access Log and Action Log The control unit 1 of the server device SV executes a process of acquiring an access log and an action log from the user terminal UT1 at preset intervals for a user during action support. Here, the above period is set to a length such as half a day or one day so that the number of notifications of the notification message includes multiple times, but the length of time can be set arbitrarily.

[0048] The control unit 1 of the server device SV, at each of the above intervals, in step S14, under the control of the access log acquisition processing unit 12, sends an acquisition request to the user terminal UT1. Then, the access log generated and saved during the above period, which is transmitted from the user terminal UT1 in response to the above acquisition request, is received via the communication I / F unit 4, and the received access log is stored in the access log storage unit 32.

[0049] At the same time, the control unit 1 of the server device SV, at each of the above intervals, in step S15, under the control of the action log acquisition processing unit 13, sends an acquisition request to the user terminal UT1. Then, the action log generated and saved during the above period, which is transmitted from the user terminal UT1 in response to the above acquisition request, is received via the communication I / F unit 4, and the received action log is stored in the action log storage unit 33.

[0050] Note that the acquisition requests for the above access logs and behavior logs may be shared one per user terminal. Also, instead of separately acquiring the access logs and behavior logs, they may be received in a multiplexed state in a batch from the user terminal UT1, and after reception, separated and stored in the access log storage unit 32 and the behavior log storage unit 33, respectively. Further, as illustrated in FIG. 6, the access logs may be stored in chronological order in a state of being mixed with the notification logs in the notification log storage unit 31.

[0051] (3) Creation of Behavior Model and Judgment of Behavior Pattern When the acquisition of the above access logs and behavior logs is completed, the control unit 1 of the server device SV then executes, under the control of the behavior model creation processing unit 14 in step S16, the creation processing of the behavior model and the judgment processing of the user's behavior pattern using this behavior model as follows.

[0052] FIG. 5 is a flowchart showing the processing procedures and processing contents of the above behavior model creation processing and behavior pattern judgment processing by the behavior model creation processing unit 14.

[0053] (3-1) Extraction of Feature Quantities Representing Behavior Situation The behavior model creation processing unit 14 first obtains, in step S161, feature quantities representing the situation of the user's behavior based on the behavior logs. For example, the target value and the actual value of the number of steps per hour described in the behavior log are compared, and if the actual value reaches the target value, it is determined as "1", and if the actual value does not reach the target value, it is determined as "0". Also, regardless of the target value, if the actual value of the number of steps reaches a preset threshold value, it may be determined as "1", and if the actual value of the number of steps does not reach the threshold value, it may be determined as "0".

[0054] The reason for using the information determined as "1" or "0" as the feature quantity of the action situation is that this example assumes the case of adopting logistic regression analysis as the action model creation method. However, when adopting multiple regression analysis as the action model creation method, instead of determining the actual number of steps as "1" or "0", the actual number of steps may be used as it is or multi-value determination may be performed using a plurality of thresholds, and these may be used as the feature quantity of the action situation.

[0055] (3-2) Extraction of Feature Quantities Representing Access Situation for Notifications Next, in step S162, the action model creation processing unit 14 obtains, based on the notification log and the access log, feature quantities representing the access situation for notifications made during a predetermined period before and after the user's action for these periods.

[0056] For example, the action model creation processing unit 14 determines whether the user has performed a confirmation operation (access) within, for example, 10 minutes from each of the notification immediately before the timing when the user starts walking (the previous notification) and the notification one further before that (the notification two before), and sets it to "1" if accessed and "0" if not. At the same time, for the notification received after the timing when the user starts walking (the next notification), it also determines whether the user has performed a confirmation operation (access) within 10 minutes from that notification, and sets it to "1" if accessed and "0" if not.

[0057] That is, the action model creation processing unit 14 determines the presence or absence of access for each of at least one notification in the past and at least one notification after the start of walking, not limited to the notification immediately before the timing when the user starts walking, and uses the determination result as the feature quantity indicating the user's access situation for the notification.

[0058] Note that, as a feature quantity indicating the access status for a notification, not limited to the determination results of "1" and "0", the number of times a user accesses a notification, the response time of access to the notification timing, etc. may be used. Further, the notifications subject to the determination of the access status are not limited to one before and two before and one after the walking start timing, and may be targeted at more than n notifications before and after the walking start.

[0059] (3-3) Creation of Data for Action Model Creation When the extraction process of each feature quantity of the above action status and access status is completed, the action model creation processing unit 14 then creates data for action model creation in step S163. For example, as shown in FIG. 8, for each date and time, data is created by associating a user ID, a feature quantity representing the achievement status of the action target (a determination value indicating whether the target has been achieved), and a feature quantity representing the access status for each of a plurality of notifications before and after the start of the action (a determination value indicating whether the notification has been accessed). As a result, data representing the time-series characteristics of the achievement status of the action target by the user and the access status for a plurality of notifications is created.

[0060] (3-4) Creation of Action Model Next, in step S164, the action model creation processing unit 14 creates an action model for estimating the tendency of the user's action using the above data for action model creation. In this example, a logistic regression analysis is performed with the determination result of the achievement status of the action target in the above data for action model creation as the objective variable and the determination results of access to the notification one before, access to the notification two before, and access to the notification one after as the explanatory variables.

[0061] FIG. 9 shows an example of the execution result of the logistic regression analysis. That is, as a result of the execution of the logistic regression analysis, for each of the intercept and each explanatory variable, a partial regression coefficient (Estimate) indicating the estimated value of the coefficient of the explanatory variable, a standard error (Std. Error) representing the variation of the estimated value of the coefficient, a z-value indicating the result of the hypothesis test of the estimated value of the coefficient, and Pr(>|z|) are calculated respectively. However, the z-value indicates the ratio of the estimated value of the coefficient to the standard error, and Pr(>|z|) indicates the probability that the absolute value of the z-value is larger than the obtained value.

[0062] (3-5) Determination of Behavior Pattern Finally, in step S165, the behavior pattern creation processing unit 14 determines the user's behavior pattern based on the result of the logistic regression analysis. For example, among the intercept and each explanatory variable, those with a probability Pr(>|z|) of 0.05 or less are determined to be significant. Then, the combination of the sign of the partial regression coefficient with the intercept or each explanatory variable determined to be significant is used as the behavior pattern.

[0063] FIG. 10 shows an example of the behavior pattern AP determined as described above. Note that FIG. 10 also shows an example of the user characteristics UC and the intervention information IS corresponding to the behavior pattern AP. The user characteristics UC and the intervention information IS are stored in the user characteristics - intervention information storage unit 35.

[0064] Also, the level for performing the significance determination of the probability Pr(>|z|) may be a value other than 0.05.

[0065] (4) Reading of User Characteristics and Intervention Information When the determination of the behavior pattern AP is completed, the control unit 1 of the server device SV shifts to step S17 shown in FIG. 4. Then, under the control of the user characteristics identification processing unit 15, the user characteristics - intervention information storage unit 35 is searched based on the behavior pattern AP, and thereby the user characteristics UC and the intervention information IS corresponding to the behavior pattern AP are read out.

[0066] As a result, for a user having an action pattern AP that accesses a notification two notifications before starting walking, for example, and does not access the previous notification, as user characteristic UC, information indicating that it is a type with a time lag from access to the notification until starting walking is read out, and as intervention information IS, information recommending setting the subsequent notification timing earlier in consideration of the above time lag, or notifying a reminder to prompt so that the start of walking is not delayed is read out.

[0067] Also, for a user having an action pattern AP that does not access any of the notifications before starting walking and accesses the notifications after finishing walking, as user characteristic US, information indicating that it is a type that checks the result after finishing walking is read out, and as intervention information IS, information recommending sending a notification including a message of appreciation at the next notification timing is read out.

[0068] On the other hand, for a user having an action pattern AP that starts walking without accessing any of the notifications, as user characteristic UC, information indicating that it is a type that starts walking autonomously regardless of the notifications is read out, and as intervention information IS, information recommending sending a notification with an incentive without sending a notification to prompt behavior change hereafter is read out.

[0069] Furthermore, for a user having an action pattern AP that accesses a notification every other time throughout the period before and after starting walking, as shown in FIGS. 11 and 13, for example, as user characteristic UC, as shown in FIG. 12, for example, information indicating that the user accesses a notification every other time and starts walking each time is read out, and as intervention information IS, information recommending making the interval of subsequent notifications every other time is read out.

[0070] (5) Generation of Notification Information When the search for the user characteristics and the intervention information in the server device SV is completed, the control unit 1 then updates, under the control of the notification information generation processing unit 16 in step S18, the notification information to be notified to the user hereafter based on the user characteristics and the intervention information.

[0071] For example, when the user characteristic UC indicates a type in which there is a time lag from access to the notification until the start of walking, the notification information generation processing unit 16 generates notification information including a message prompting the start of walking so as not to be delayed based on the intervention information IS, and timing specification information specifying a notification timing earlier than before.

[0072] Also, for example, when the user characteristic UC indicates a type in which the result is confirmed after the end of walking, the notification information generation processing unit 16 generates notification information including an encouraging message based on the intervention information IS and timing specification information specifying a notification timing earlier than usual.

[0073] On the other hand, for example, when the user characteristic UC indicates a type in which walking is started autonomously regardless of the notification, the notification information generation processing unit 16 changes a message prompting behavior change based on the intervention information IS to a message notifying incentive provision.

[0074] Furthermore, for example, when the user characteristic UC indicates a type in which access is made every other time to the notification and walking is started each time, the notification information generation processing unit 16 generates notification information including timing specification information specifying that the notification interval is every other time without changing the content of the message based on the intervention information IS.

[0075] Note that the notification information generation processing unit 16 may also insert a message indicating user characteristics into the notification information so that the user characteristics are notified to the user in the notification information transmission process described later.

[0076] (6) Transmission of Notification Information When the update process of the above notification information is completed, the control unit 1 of the server device SV determines whether an end request for action support has been received from the user in step S19. If action support is still ongoing, it returns to step S12 and continues to transmit the notification information under the control of the notification information transmission processing unit 11.

[0077] For example, based on the notification information newly generated by the above notification information generation processing unit 16, the notification information transmission processing unit 11 transmits the notification message included in the notification information from the communication I / F unit 4 to the user terminal UT1 at the notification timing specified by the timing specification information included in the same notification information.

[0078] Therefore, the user will be notified of appropriate notification messages corresponding to the user characteristics hereafter at appropriate timings or frequencies according to the user characteristics, and thus more effective walking support becomes possible.

[0079] Note that when the notification information transmission processing unit 11 detects that an end request for action support has been received from the user terminal UT1 in step S19, it ends the walking support process for the user. In this case, after the walking support ends, the updated notification information may be saved in the data storage unit 3. By doing so, when the same user requests to resume action support later, it becomes possible to start the notification based on the updated notification information.

[0080] (Function and Effect) As described above, in one embodiment, when the server device SV supports the user's actions by notifying the user of notification information, a notification log representing the notification result of the above notification information is generated and stored, and access logs representing the user's confirmation results for the above notification information and action logs representing the user's action results are acquired from the user terminals UT1 to UTn. Then, based on these logs, an action model representing the relevance between the user's access status to a plurality of pieces of notification information notified in each period before and after the user takes an action and the user's action status is created, and the created action model is used to determine the user's action pattern AP. Subsequently, based on the determined action pattern AP, user characteristics representing the tendency of the user's actions and intervention information corresponding to these user characteristics are read from the user characteristics and intervention information storage unit 35, the above notification information is updated based on the read intervention information, and the updated notification information is notified to the user hereafter.

[0081] Therefore, by comprehensively judging the user's access status to a plurality of pieces of notification information notified in each period before and after the user's action, user characteristics representing the tendency of the user's actions are estimated. For this reason, compared with the case of estimating user characteristics using only the presence or absence of access to the notification immediately before the user's action, it is possible to accurately estimate user characteristics representing the tendency of the user's actions, and based on the estimation result, it is possible to make the content of the message of the notification information, the notification timing, and the number of notifications for the user hereafter more appropriate.

[0082] [Other Embodiments] In the above-described embodiment, the case where an action model created using logistic regression analysis is used to estimate user characteristics has been described as an example, but multiple regression analysis or other machine learning models may be used instead.

[0083] Also, in the above-described embodiment, the case where user characteristics representing the tendency of user behavior are estimated based on the access from the user to a plurality of past notifications including the notification sent immediately before the user's behavior and the access situation of the user to the notification sent after the user's behavior has been described as an example. However, the user characteristics may be estimated based only on the access from the user to a plurality of past notifications including the notification sent immediately before the user's behavior.

[0084] Furthermore, in the above-described embodiment, the case where all the functions of the user behavior support apparatus according to the present invention are provided in the server apparatus SV has been described as an example. However, the present invention is not limited thereto, and all the functions of the user behavior support apparatus according to the present invention may be provided in the user terminals UT1 to UTn, or the functions of the user behavior support apparatus may be distributed between the server apparatus SV and the user terminals UT1 to UTn. In this case, if the creation of the behavior model and the estimation process of the user characteristics are arranged in the server apparatus SV, a reduction in the processing load of the user terminals UT1 to UTn can be expected.

[0085] In addition, regarding the functions, processing procedures, and processing contents of each processing unit of the user behavior support apparatus, and the types of user behavior, etc., various modifications can be made without departing from the gist of the present invention.

[0086] Although the embodiments of the present invention have been described in detail above, the foregoing description is merely illustrative of the present invention in every respect. Needless to say, various improvements and modifications can be made without departing from the scope of the present invention. That is, in practicing the present invention, a specific configuration according to the embodiment may be appropriately adopted.

[0087] In short, the present invention is not limited to the above embodiments as they are, and at the implementation stage, the components can be modified and embodied without departing from the gist thereof. Also, various inventions can be formed by appropriately combining a plurality of components disclosed in the above embodiments. For example, some components may be deleted from all the components shown in the embodiment. Furthermore, components from different embodiments may be appropriately combined.

Explanation of Symbols

[0088] SV…Server device UT1~UTn…User terminals NW…Network 1…Control unit 2…Program storage unit 3…Data storage unit 4…Communication I / F unit 5…Bus 11…Notification information transmission processing unit 12…Access log acquisition processing unit 13…Behavior log acquisition processing unit 14…Behavior model creation processing unit 15…User characteristic identification processing unit 16…Notification information generation processing unit 31…Notification log storage unit 32…Access log storage unit 33…Behavior log storage unit 34…Behavior model storage unit 35…User characteristic / intervention information storage unit

Claims

1. A user behavior support device that supports the modification of a user's behavior by notifying the user of notification information for promoting the modification of the behavior, a notification log acquisition processing unit that acquires a notification log representing the notification result of the notification information, an access log acquisition processing unit that acquires an access log representing the result of a confirmation operation of the user with respect to the notification information, a behavior log acquisition processing unit that acquires a behavior log representing the behavior result of the user, based on the acquired notification log and access log, selects an access log corresponding to the user's access to at least a plurality of past notification information including the notification information notified immediately before the user's behavior, and based on the selected access log and the user's behavior log, an estimation processing unit that estimates user characteristics representing the tendency of the user's behavior A user behavior support device comprising:

2. The user behavior support device according to claim 1, further comprising a notification information change processing unit that changes the notification information to be notified to the user according to the estimated user characteristics.

3. The notification information change processing unit changes at least one of the notification frequency, notification timing, and notification content of the notification information according to the estimated user characteristics. The user behavior support device according to claim 2.

4. The estimation processing unit further selects an access log from the user for the notification information output after the user's behavior in addition to the access log from the user for a plurality of past notification information including the notification information output immediately before the user's behavior, and based on the selected access log and the user's behavior log, estimates the user characteristics representing the tendency of the user's behavior. The user behavior support device according to any one of claims 1 to 3.

5. The estimation processing unit estimates the user characteristics based on the pattern of the presence or absence of the access log from the user for each of the plurality of past notification information and the notification information after the behavior, and the presence or absence of the behavior represented by the behavior log. The user behavior support device according to claim 4.

6. The estimation processing unit estimates the user characteristics by using a regression model in which information representing the action result represented by the action log is used as an objective variable, and information representing each response result of the user to the notification information represented by a plurality of selected access logs is used as an explanatory variable, respectively. The user action support device according to any one of claims 1 to 5.

7. A user action support method executed by an information processing device that supports the action transformation of a user by notifying the user of notification information for promoting action transformation, a process of acquiring a notification log representing the notification result of the notification information; a process of acquiring an access log representing the result of the user's confirmation operation on the notification information; a process of acquiring an action log representing the action result of the user; Based on the acquired notification log and access log, select the access log corresponding to the user's access to at least a plurality of past notification information including the notification information notified immediately before the user's action, and based on the selected access log and the user's action log, a process of estimating user characteristics representing the tendency related to the user's action A user action support method comprising:

8. A program for causing a processor included in the user action support device to execute the processing by each processing unit included in the user action support device according to any one of claims 1 to 6.

Images