JP2020004210A - Estimation device, estimation method and estimation program - Google Patents

Abstract

To provide a device, a method, and a program for improving an estimation accuracy of a destination.SOLUTION: When a user retrieves a location of a certain facility and information (e.g., evaluation or menu of facility) related to this facility, or when the user posts a message to go to the facility, there is a possibility that the user will start traveling with the facility as a destination. Further, when the user retrieves information about a certain facility while on the move, or posts that this user is traveling to that facility, there is a possibility that the user is traveling with the facility as a destination. In an early destination prediction that predicts the destination of the user after the user starts moving, an information providing apparatus 10 estimates a destination to be visited by the user from an activity history on the Internet such as a retrieve query history and a posting history. As a result of such processing, since the information providing apparatus 10 can estimate the destination that the user visits for the first time, it is possible to improve an estimation accuracy of the destination.SELECTED DRAWING: Figure 1

Description

  The present invention relates to an estimation device, an estimation method, and an estimation program.

  2. Description of the Related Art Conventionally, a technique for predicting a purpose to which a user has moved is known. As an example of such a technique, there is known a technique of learning a feature of position history data relating to a movement of a user and estimating a destination of the user and a route to the destination based on the learned feature. .

JP 2011-252844 A

  However, in the related art, there is a possibility that the destination cannot be accurately estimated.

  For example, in the above-described related art, since only features of the user's position history data are learned, it is difficult to estimate a place that the user is going to visit for the first time as a destination.

  The present application has been made in view of the above, and has as its object to improve the accuracy of estimating a destination.

  The estimating device according to the present application is an acquisition unit that acquires a user's action history on a network, and when it is estimated that the user has started moving, based on the action history, determines the destination of the movement. And an estimating unit for estimating.

  According to an aspect of the embodiment, the accuracy of destination estimation is improved.

FIG. 1 is a diagram illustrating an outline of a process executed by the information providing apparatus according to the embodiment. FIG. 2 is a diagram illustrating a configuration example of the information providing apparatus according to the embodiment. FIG. 3 is a diagram illustrating an example of information registered in the position history database according to the embodiment. FIG. 4 is a diagram illustrating an example of information registered in the action history database according to the embodiment. FIG. 5 is a flowchart illustrating an example of a flow of an estimation process and a proposal process performed by the information providing apparatus according to the embodiment. FIG. 6 is a diagram illustrating an example of a hardware configuration.

  Hereinafter, an estimation device, an estimation method, an estimation program, a proposal device, a proposal method, and a mode for implementing the proposal program (hereinafter, referred to as “embodiments”) according to the present application will be described in detail with reference to the drawings. Note that the estimation device, the estimation method, the estimation program, the proposal device, the proposal method, and the proposal program according to the present application are not limited by this embodiment. In addition, the embodiments can be appropriately combined within a range that does not contradict processing contents. In the following embodiments, the same portions are denoted by the same reference numerals, and redundant description will be omitted.

  Further, in the following description, when the user starts moving, the information providing device 10 which is an example of the estimating device and the proposal device estimates a destination such as a destination facility or place, and an estimation process. An example of the providing process for providing information according to the result of the process will be described, but the embodiment is not limited to the following description. For example, the information providing apparatus 10 may execute a guidance process of searching for a route from the current position of the user to the destination based on the estimation result, and providing the search result to the user. Further, the information providing apparatus 10 may execute various guidance processes when the user moves by car or bicycle, when the user is walking, or when public transportation is adopted, for example.

[Embodiment]
[1. Outline of information providing device]
First, an outline of a process executed by the information providing apparatus 10 will be described with reference to FIG. FIG. 1 is a diagram illustrating an outline of a process executed by the information providing apparatus according to the embodiment. In FIG. 1, the information providing apparatus 10 is an information processing apparatus that executes an estimation process and a proposal process described below, and is realized by, for example, a server device or a cloud system.

  The information providing apparatus 10 can communicate with the terminal device 100 and the log server 200 via a network N such as the Internet (see, for example, FIG. 2). The information providing device 10 may be able to communicate with an arbitrary number of terminal devices 100 and log servers 200.

  The terminal device 100 is a mobile terminal such as a smartphone, a tablet terminal or a PDA (Personal Digital Assistant), or a terminal device such as a notebook PC (Personal Computer), and includes a mobile communication network and a wireless LAN (Local Area Network). A terminal device that can communicate with an arbitrary server via the network N.

  In addition, the terminal device 100 has a positioning function of specifying the position of the terminal device 100 (hereinafter, referred to as “current location”) at predetermined time intervals using a satellite positioning system such as a GPS (Global Positioning System). . Then, the terminal device 100 displays and specifies an image such as a map included in the route information on a liquid crystal screen, electroluminescence, an LED (Light Emitting Diode) screen or the like (hereinafter, simply referred to as “screen”). The current location is displayed on the map each time. In addition, the terminal device 100 displays a left turn or a right turn, a change in a lane to be used, an estimated time of arrival at a destination, or the like, or a speaker (for example, using Output from a speaker or the like on which a person rides).

  For example, the terminal device 100 acquires position information indicating the current location at predetermined time intervals, and transmits the acquired position information to the log server 200. Further, when the terminal device 100 receives the information on the destination estimated by the information providing device 10 by the estimation process, the terminal server 100 sends the destination to a guidance server (not shown) that executes the guidance process in cooperation with the terminal device 100, Send location information. Then, the terminal device 100 realizes the guidance process by providing the user with the information distributed from the guidance server and indicating the route from the current location to the destination.

  Further, the terminal device 100 has a function of executing various processes via the Internet. For example, the terminal device 100 includes a web search, a map search, a search for a transaction target in an online shopping mall, a purchase of a transaction target in an online shopping mall, an exhibition at an online auction, a bid and a successful bid, transmission and reception of mail and messages, and SNS (Social) Networking Service) can provide any service via the Internet, such as posting information, searching for the location of a predetermined facility, and the like.

  The log server 200 is an information processing device that collects various logs, and is realized by a server device, a cloud system, or the like. For example, the log server 200 receives position information from the terminal device 100 at predetermined time intervals, and holds the received position information as a position history. In addition, the log server 200 collects, from a server device that provides various services via the Internet, an action history that is a history of the user using the terminal device 100 on the Internet.

  For example, the log server 200 includes a search query and a search date and time input by a user in a web search and a map search, a search query and a search date and time of a transaction target, information on a transaction target that the user has purchased, bid on or made a successful bid for, an email or a message. History of the user, the date and time of the posting, the content of the posting, information on the facility whose location was searched, facilities such as hotels and restaurants reserved by the user, the date and time of the reservation, etc. Collect the action history indicating the history of. In addition, the log server 200 collects, for example, an action history indicating a history of reservations of stores and public transportation. Note that the above-described action history is merely an example, and the log server 200 collects an action history having arbitrary contents.

[1-1. Overview of estimation processing)
Here, in the related art, a destination to which the user moves is estimated based on the position history of the user. For example, in the related art, there is a technology that employs a hypothesis (destination) having the highest likelihood based on the position information of the user. However, in such a technique, it is difficult to estimate a place where a user is going to visit for the first time, and therefore, there is a possibility that a destination cannot be accurately estimated. Therefore, the information providing device 10 executes the following estimation processing. First, the information providing apparatus 10 acquires a user's action history on the network. Then, when it is estimated that the user has started moving, the information providing apparatus 10 estimates the destination based on the action history.

  For example, if a user searches for the location of a facility and information related to the facility (for example, evaluation of the facility, menu, etc.), or submits a message to go to that facility, As you could start moving. Also, when the user searches for information about a facility that is moving or posts to the effect that the user is moving to that facility, there is a high possibility that the user has moved to that facility. In other words, when the user starts moving to a destination, the destination can be estimated from the user's activity history on the Internet.

  Therefore, the information providing apparatus 10 estimates a destination to be visited by the user based on the behavior history of the user. That is, the information providing apparatus 10 considers the history of the user search, the electronic commerce, and the like in the early destination prediction after the start of the movement. For example, the information providing apparatus 10 uses a log of a net action such as a search related to a place (fishing, izakaya, etc.), EC (Electronic Commerce), reservation, and the like. In other words, the information providing apparatus 10 uses a search history of a user, a log related to online shopping, and the like in order to improve the accuracy of early destination prediction.

  More specifically, the information providing apparatus 10 is configured to provide the user with an early destination prediction that predicts the destination of the user after the user starts moving, based on a history of search queries, postings, and the like on the Internet. Estimate the destination to visit. As a result of such processing, the information providing apparatus 10 can estimate the destination that the user visits for the first time, and thus can improve the estimation accuracy of the destination.

[1-2. About the outline of the provision process]
Further, in the related art, only a route to a destination is estimated, and there is no known technology of effective sales promotion that affects the destination itself for a user. Therefore, the information providing device 10 executes the following proposal processing. First, the information providing apparatus 10 acquires a user's action history on the network. Then, when it is estimated that the user has started moving to the predetermined destination, the information providing apparatus 10 proposes a destination different from the predetermined destination based on the action history.

  That is, the information providing device 10 proposes another destination related to the destination predicted after the start of the movement, taking into account the history of the net search. In other words, the information providing apparatus 10 distributes an advertisement that changes the original destination from the destination prediction and search logs. For example, when the user intends to move to the destination #A, the information providing apparatus 10 proposes not the information of the destination #A but the destination #B. To give a more specific example, the information providing apparatus 10 outputs information that suggests a change in the destination that the user wants to visit, such as “Do you want to move to destination #B?”. As a result of such processing, the information providing apparatus 10 can prompt the user to change the destination, and thus can provide effective information to the user.

[1-3. Example of estimation processing performed by information providing apparatus]
Next, an example of a flow of an estimation process executed by the information providing apparatus 10 will be described with reference to FIG. First, the log server 200 acquires the position information and the action history of the user who uses the terminal device 100 (Step S1). For example, the log server 200 acquires, at predetermined time intervals, position information indicating the current location measured by the terminal device 100 using the GPS at predetermined time intervals (for example, one minute intervals) from the terminal device 100. In addition, the log server 200 acquires an action history indicating an action on the Internet of the user who uses the terminal device 100 from a server device that provides various services to the user via the Internet.

  On the other hand, the information providing device 10 acquires a position history and an action history, which are history of position information, from the log server 200 (step S2). Then, when the terminal device starts moving, the information providing device 10 starts early destination guidance (step S3). For example, the information providing apparatus 10 determines whether or not the user has moved from a predetermined base based on the position history acquired within a predetermined period (for example, the latest several minutes), and determines that the user has moved. If determined, early destination guidance is started. To give a more specific example, the information providing apparatus 10 estimates that when the user has moved from the area registered as the base to the outside of the area, the user has started moving to some destination. And start early destination guidance. The base of the user may be registered in advance or may be estimated from a pattern of the position history of the user. Further, a plurality of such bases may be set.

  Subsequently, the information providing apparatus 10 estimates a destination from the position history (step S4). For example, the information providing apparatus 10 specifies the position history when the user has departed from the same base as at the time of estimation in the same past time zone as at the time of estimation. Subsequently, the information providing apparatus 10 estimates the destination of the user based on the specified position history. For example, in the example illustrated in FIG. 1, the information providing apparatus 10 estimates a cafe #A, a supermarket #B, a home improvement center #C, and a pub #D as destinations of the user. The information providing apparatus 10 also estimates a route to each destination, as indicated by a dotted line in FIG. Note that any known technique can be adopted for estimating the destination based on such a position history.

  Subsequently, the information providing apparatus 10 narrows down the destination from the action history (Step S5). For example, if the information providing apparatus 10 is searching for the name and address of the cafe #A as a search query immediately before departure, and if the posting is related to the cafe #A, the cafe #A Raise the value of the destination score indicating the possibility of the destination. Also, for example, if the information providing apparatus 10 has posted about a product sold in the supermarket #B or the home center #C immediately before departure, the destination score value of the supermarket #B or the home center #C To rise. Note that the information providing apparatus 10 may, for example, search for a product sold in the home center #C or a product related to the product in the online shopping mall, or purchase the product in the home center #C. May be increased. On the other hand, for example, the information providing apparatus 10 lowers the destination score of the izakaya #D when the search or the posting related to the izakaya #D is not performed immediately before the departure. Then, for example, when the score of the pub #D falls below a predetermined threshold, the information providing apparatus 10 excludes the pub #D from the destination as shown by STU1 in FIG.

  That is, the information providing apparatus 10 narrows down the destination based on the action history at the time of departure. For example, the information providing apparatus 10 attaches more importance to an action history in a time zone closer to the time of departure (for example, up to 10 minutes before), and regarding an action history in a time zone farther from the time of departure (for example, one day before), The weighting according to the period elapsed from the behavior indicated by the behavior history to the departure may be considered so as not to be considered.

  Subsequently, the information providing apparatus 10 repeatedly executes the estimation process until the destination score exceeds a predetermined threshold (Step S6). More specifically, the information providing apparatus 10 repeats the narrowing down of the destination based on the position history and the action history after the user leaves until the destination score of any candidate exceeds a predetermined threshold. Execute.

  For example, when the user moves to the position P1, the user will be off the route to the cafe #A. Therefore, the information providing apparatus 10 decreases the destination score of the cafe #A and increases the destination scores of the supermarket #B and the home center #C. Here, even if the user moves to the position P2, the destination scores of the supermarket #B and the home improvement center #C do not seemingly change. However, for example, in the section AP1 from the position P1 to the position P2, when the user performs a position search for the super #B, that is, when the user performs an action on the Internet related to the super #B, The possibility that the user has moved to supermarket #B is higher than the possibility that the user has moved to home improvement center #C. Thus, in the section AP1, when the user performs an action related to the supermarket #B, the information providing apparatus 10 increases the destination score of the supermarket #B and decreases the destination score of the home center #C.

  Here, the information providing apparatus 10 may consider a weight according to the distance to each destination. For example, when the information providing apparatus 10 has performed a search related to Super # B at a position closer to Super # B, the information providing apparatus 10 may correct the destination score of Super # B to a higher value. To give a more specific example, when the information providing apparatus 10 searches for the super #B in the section AP1, the information providing apparatus 10 increases the destination score of the super #B by 10 points, and increases the destination AP2 closer to the super #B. In the case where the search for Super # B is performed, the destination score of Super # B may be increased by 20 points.

  Then, for example, when the destination score of the super #B exceeds the predetermined threshold, the information providing apparatus 10 estimates the super #B as the destination and distributes information according to the estimated destination (step S10). S7). For example, the information providing apparatus 10 determines that the possibility that the user has moved to the supermarket #B exceeds 60%, or that the difference between the destination score of the supermarket #B and the destination score of another destination is a predetermined value. When the threshold value is exceeded, the information on super #B is provided to the terminal device 100. To give a more specific example, the information providing apparatus 10 may provide a route from the current position to the supermarket #B, a predicted arrival time, congestion information and weather information on the route, information of a sale and a coupon in the supermarket #B, and the like. I do.

  As described above, the information providing apparatus 10 estimates the destination based on not only the position history but also the user's action history. For this reason, the information providing apparatus 10 can accurately estimate the destination of the user.

[1-4. Example of providing process performed by information providing device]
Next, an example of a flow of a providing process performed by the information providing apparatus 10 will be described with reference to FIG. First, the information providing apparatus 10 estimates a user's destination by the above-described estimation processing. Then, when the information providing apparatus 10 estimates the destination of the user, the information providing apparatus 10 estimates another destination based on the action history of the user (step S8).

  For example, if the user has performed a web search for products sold at the supermarket #B and the home center #C before or during the movement, the destination score based on the search history is the supermarket #B and the home center #C. Is likely to be about the same. On the other hand, when the user has a history of visiting the supermarket #B and no history of visiting the home improvement center #C, the destination score based on the location history is higher for the supermarket #B. As a result, in early destination guidance based on the position history and the action history, information on the supermarket #B is provided to the user. However, in consideration of the search history, even if the user is guided to the home improvement center #C, since the product desired by the user is sold, the possibility that the user's impression is deteriorated is low.

  In addition, some users may start moving with no particular purpose. For example, some users may be trying to visit a store they visited in the past on a holiday. If such a user is guided to another store instead of a store that has been visited in the past, it is possible that the user can cause a new purchasing behavior. In addition, when the user intends to move to a predetermined store with a purpose, when providing information of a gas station affiliated with the store (for example, a gas station providing a coupon of the predetermined store) , The user can be guided to the gas station.

  Therefore, the information providing apparatus 10 calculates the destination score of each destination based on the action history and the position history, as in the estimation processing. Then, the information providing apparatus 10 selects a destination whose calculated destination score satisfies a predetermined condition as a guidance destination. For example, the information providing apparatus 10 selects the destination having the second highest score value as the guidance destination. Then, the information providing device 10 proposes the selected destination to the user (step S9). For example, the information providing apparatus 10 provides a coupon that can be used at the destination of the guide, along with the location of the destination of the guide, the name of the store, and the like, and is a product sold at the destination of the guide. Provides information on products for which users have searched the web in the past. As a result, the information providing apparatus 10 can guide the user from the original destination Super #B to the home center #C.

[2. Example of functional configuration of information providing device]
Hereinafter, the information providing apparatus 10 that performs the above-described estimation processing and providing processing will be described in detail. First, an example of a functional configuration of the information providing apparatus 10 will be described. FIG. 2 is a diagram illustrating a configuration example of the information providing apparatus according to the embodiment. As shown in FIG. 2, the information providing device 10 includes a communication unit 20, a storage unit 30, and a control unit 40.

  The communication unit 20 is realized by, for example, a NIC (Network Interface Card). The communication unit 20 is connected to the network N by wire or wirelessly, and transmits and receives information to and from an arbitrary device.

  The storage unit 30 is realized by, for example, a semiconductor memory element such as a random access memory (RAM) or a flash memory, or a storage device such as a hard disk or an optical disk. The storage unit 30 stores a position history database 31 and an action history database 32.

  A position history of the user is registered in the position history database 31. For example, FIG. 3 is a diagram illustrating an example of information registered in the position history database according to the embodiment. As shown in FIG. 3, information having items such as “user ID (Identifier)”, “location history ID”, “date and time”, and “location information” is registered in the location history database.

  Here, the “user ID” is an identifier for identifying the user or the terminal device 100 used by the user. The “position history ID” is an identifier for identifying the position history. The “date and time” is information indicating the date and time when the terminal device 100 has acquired the associated position information. The “location information” is the location information acquired by the terminal device 100.

  In the example shown in FIG. 3, conceptual values such as “user # 1”, “position history # 1-1”, “date and time # 1-1”, and “position # 1-1” are described. Actually, a numerical value and a character string for identifying the user and the position history, a numerical value indicating the date and time, a value indicating the position of the terminal device 100 (for example, a numerical value obtained by GPS), and the like are registered. In addition to the information shown in FIG. 3, arbitrary information on the position history may be registered in the position history database 31.

  For example, in the example shown in FIG. 3, the position history database 31 stores the user ID “user # 1”, the position history ID “position history # 1-1”, the date and time “date and time # 1-1”, and the position information. “Position # 1-1” is registered in association with it. Such information is sent from the user indicated by the user ID “user # 1” or the terminal device 100 used by the user as the position history indicated by the position history ID “position history # 1-1” as the date and time “date and time”. The position information acquired at the date and time indicated by “# 1-1” is the position information “position # 1-1”.

  Returning to FIG. 2, the description will be continued. The behavior history of the user is registered in the behavior history database 32. For example, FIG. 4 is a diagram illustrating an example of information registered in the action history database according to the embodiment. As shown in FIG. 4, information having items such as “user ID”, “attribute information”, “action history ID”, “date and time”, and “action history” is registered in the action history database.

  Here, the “attribute information” is attribute information indicating the attribute of the user indicated by the associated “user ID” or the user of the terminal device 100 indicated by the “user ID”. Various attribute information such as attributes and psychographic attributes. The “action history ID” is an identifier for identifying the action history. The “date and time” is information indicating the date and time when the action on the Internet indicated by the associated action history was performed. The “action history” is an action history indicating an action performed by the user on the Internet.

  In the example shown in FIG. 4, conceptual values such as “attribute # 1”, “history # 1-1”, and “action history # 1-1” are described. A character string, a numerical value and a character string for identifying the action history, a character string indicating the content of the action history, and the like are registered. For example, in the action history database 32, various internets such as web search and map search queries, search queries in online shopping malls, information on purchased, bid or successful bid products, and information posted on the SNS are stored as the action history. Information indicating the above action and the content of the action is registered. Further, in addition to the information shown in FIG. 4, any information regarding the action history may be registered in the action history database 32.

  For example, in the example shown in FIG. 4, the action history database 32 stores the user ID “user # 1”, the attribute “attribute # 1”, the action history ID “history # 1-1”, the date and time “date and time # 1- 1 and an action history “action # 1-1” are registered in association with each other. In such information, the attribute of the user indicated by the user ID “user # 1” or the user using the terminal device 100 is the attribute indicated by the attribute information “attribute # 1”, and the action history ID “history # 1” The action history indicated by “1-1” indicates that the performed action acquired at the date and time indicated by date and time “date and time # 1-1” is the action indicated by action history “action history # 1-1”.

  Returning to FIG. 2, the description will be continued. The control unit 40 is a controller. For example, various programs stored in a storage device inside the information providing apparatus 10 by a processor such as a CPU (Central Processing Unit) and an MPU (Micro Processing Unit) are stored in a RAM or the like. As a work area. The control unit 40 is a controller, and may be realized by an integrated circuit such as an ASIC (Application Specific Integrated Circuit) or an FPGA (Field Programmable Gate Array). As illustrated in FIG. 2, the control unit 40 includes an acquisition unit 51, a movement determination unit 52, an estimation unit 53, a provision opportunity determination unit 54, a proposal unit 55, a provision unit 56, and a setting unit 57.

  The processing units 51 to 57 included in the control unit 40 realize and execute the functions and operations (for example, FIG. 1) of the estimation processing and the provision processing described below, but these are for explanation. It is a functional unit arranged in the same manner, and it does not matter whether it matches actual hardware elements or software modules. That is, the information providing apparatus 10 may implement and execute the evaluation process and the guidance process in arbitrary functional units as long as the following functions and actions of the evaluation process and the guidance process can be realized and executed.

[3. Example of operation and effect of processing executed by information providing apparatus]
Hereinafter, the contents of the estimation processing and the proposal processing executed / realized by each of the processing units 51 to 547 will be described using the flowchart illustrated in FIG. 5. FIG. 5 is a flowchart illustrating an example of a flow of an estimation process and a proposal process performed by the information providing apparatus according to the embodiment.

  First, the acquisition unit 51 acquires a position history (step S101). Further, the acquisition unit 51 acquires a search history (Step S102). For example, the acquisition unit 51 accesses the log server 200 and acquires an action history indicating information collected by the log server 200 and indicating a user's action on the network. Further, the acquisition unit 51 accesses the log server 200 and acquires the position history of the user. Then, the acquisition unit 51 registers the action history and the position history in each of the databases 31 and 32. Note that the acquisition unit 51 acquires the action history and the position history at predetermined time intervals.

  The movement determination unit 52 determines whether the user has started moving (Step S103). For example, the movement determination unit 52 refers to the position history of the user and determines whether or not the user has started moving from the base based on the position information acquired within a predetermined period (for example, within three minutes of the process). Is determined. Then, when the movement determining unit 52 does not determine that the user has started moving (Step S103: No), the acquiring unit 51 executes Step S101.

  On the other hand, when the movement determination unit 52 determines that the user has started moving (Step S103: Yes), the estimation unit 53 estimates a destination candidate based on the position history. Then, the estimation unit 53 calculates a destination score with each candidate as a destination based on the position history and the search history (step S105). That is, when it is estimated that the user has started moving, the estimating unit 53 executes an estimation process for estimating the destination of the movement based on the action history. More specifically, the estimation unit 53 estimates a destination based on the position history and the action history. For example, the estimation unit 53 estimates a destination candidate based on the position history, and estimates a destination from the candidates based on the action history.

  For example, the estimating unit 53 estimates, as a destination candidate, a facility or an area visited by a user when the user departs the base in the same time zone or the same day of the week, based on past positional information. The estimating unit 53 then determines, for each destination candidate, the commonality between the date and time when the destination was visited and the estimation time, the commonality of the weather, the period elapsed since the previous visit to each destination, and the like. Based on this, a destination score indicating the possibility that each destination candidate can be a destination is calculated. That is, the estimation unit 53 calculates the destination candidates and the destination score of each candidate based on the position history.

  Subsequently, the estimation unit 53 corrects the destination score of each candidate based on the action history. For example, the estimating unit 53 identifies an activity on the Internet performed within a predetermined period (for example, within 6 hours) from the time of estimation, and calculates a destination score of a destination candidate having a relationship with the identified activity. to add. For example, the estimating unit 53 specifies a destination candidate that sells a product for which the user has performed a web search, and increases the measured destination score of the candidate. That is, the estimation unit 53 specifies a candidate having a higher relationship with the action history among the destination candidates.

  Further, the estimating unit 53 specifies an action history related to the destination candidate from the action history, and estimates the destination from the candidates based on the specified action history. For example, the estimating unit 53 extracts an action history related to the destination candidate from the action history for each destination candidate estimated based on the position history. For example, the specifying unit 53 refers to the action history database 32, and searches the action history for various types of search using a search query having relevance to the destination candidate, and for a transaction target having relevance to the destination candidate. It is determined whether purchase, exhibition, bidding, bidding, or the like has been performed. It should be noted that the action history having a relationship with such a destination candidate may be extracted on a rule basis, or may be extracted by various relationship identification techniques.

  The estimating unit 53 may specify the date and time when the action indicated by the action history was performed, and may estimate the destination based on the action history and the specified date and time. For example, the estimating unit 53 increases the value of the destination score by the first value for the destination candidate related to the search query of various searches performed within one hour from the estimation time, and increases the destination score by 1 from the estimation time. For destination candidates related to search queries of various searches performed more than a time ago, the value of the destination score may be increased by a second value lower than the first value. That is, the estimating unit 53 may estimate the destination with more emphasis on more recent activities on the Internet.

  Further, the estimating unit 53 may estimate the destination according to, for example, a time zone in which the action indicated by the action history is performed. For example, the estimation unit 53 may change the value of the destination score according to whether the action indicated by the action history is performed at night or during the day.

  Here, if the user searches for a map or the Internet while traveling to any destination, the search is more relevant to the destination than the search before departure to the destination. It is considered highly likely. Therefore, the estimating unit 53 may give more importance to the behavior on the Internet during the movement than the behavior on the Internet before the movement. For example, the estimating unit 53 increases the destination score of the destination candidate having a relevance with the action before the user is determined to have started moving by the movement determining unit 52 by the first value. The destination score of a destination candidate having a relevance to an action after the user is determined to have started moving by the movement determining unit 52 is a third value higher than the first value. May be raised only. Further, the estimating unit 53 may estimate the destination according to the type of the action history. For example, the estimation unit 53 may calculate a score of a candidate for each destination by giving importance to a search query in a map search over a search query in a web search.

  In addition, the estimation unit 53 may correct the value of the destination score of each candidate in consideration of the weight based on the various elements described above. For example, the estimating unit 53 may determine the content of the action on the Internet associated with each destination candidate, the date and time when the action was performed, whether the action was performed before or after the movement, and the like. Accordingly, the value of the destination score based on the action history may be calculated, and the sum of the calculated value of the destination score and the value of the destination score based on the position history may be calculated. The estimating unit 53 may set a predetermined weight for each of the value of the destination score based on the position history and the value of the destination score based on the action history. It should be noted that, as described later, various other elements can be employed as the elements and the weight values when calculating the destination score.

  The provision opportunity determination unit 54 determines whether the destination score satisfies a predetermined provision condition (step S106). That is, the providing opportunity determination unit 54 determines whether the probability that the estimated destination candidate is the destination exceeds a predetermined threshold. For example, when the value of the destination score of any candidate exceeds a predetermined threshold, or when the difference from the destination score of another candidate exceeds a predetermined threshold, the providing opportunity determination unit 54 It is determined that the destination score satisfies a predetermined provision condition. When it is determined that the destination score does not satisfy the predetermined proposal condition (Step S106: No), the estimating unit 53 executes Step S104 again.

  That is, when the user starts moving, the above-described estimating unit 53 repeatedly calculates the destination score of each destination candidate until the value of the destination score satisfies the predetermined condition. Then, the estimation unit 53 estimates a destination candidate whose destination score value satisfies a predetermined condition as a destination. In other words, the estimation unit 53 calculates, for each destination candidate, the probability that the user is moving to the destination candidate, and sets a candidate whose calculated accuracy satisfies a predetermined condition as a destination candidate. . As described above, when the user starts moving, the estimation unit 53 repeatedly updates the destination score of each destination candidate based on the user's action history, thereby allowing the user to visit. Is estimated.

  Here, the estimation unit 53 calculates a destination score of each destination candidate based on the position history. At this time, the estimation unit 53 specifies the position indicated by the latest position information and the distance to each destination candidate. Then, a destination score corresponding to the specified distance may be calculated. For example, as the user approaches a destination candidate, the estimation unit 53 may increase the value of the destination score of the candidate.

  On the other hand, when it is determined that the destination score satisfies the predetermined provision condition (Step S106: Yes), the provision opportunity determination unit 54 determines whether the current position of the user satisfies the predetermined proposal condition (Step S106). Step S107). Then, when it is determined that the current position of the user satisfies the predetermined proposal condition (Step S107: Yes), the suggestion unit 55 is a candidate different from the destination candidate having the highest destination score, and Are proposed as destinations that satisfy the condition (step S108). That is, when it is estimated that the user has started moving to the predetermined destination, the suggestion unit 55 determines a destination different from the predetermined destination based on the action history (hereinafter referred to as “proposed location”). May be described).

  For example, the suggestion unit 55 proposes a proposed location having a predetermined association with a predetermined destination among the destinations based on the action history. In addition, the suggestion unit 55 proposes a proposed location whose relevance with the action history exceeds a predetermined threshold. On the other hand, when it is determined that the providing position of the user does not satisfy the predetermined proposal condition (step S107: No), the providing unit 56 provides information on the destination having the highest score to the user (step S109). ). That is, the suggesting unit 55 and the providing unit 56 provide information on the destination appropriately estimated based on the action history or propose information on the proposed location.

  For example, in the example shown in FIG. 1, a case will be described where the score of supermarket #B exceeds a predetermined threshold, and then the score of home improvement center #C increases. For example, when the user has departed from the base and is moving in the section AP1, the distance to the supermarket #B or the home center #C is short, and therefore, the information providing apparatus 10 sets the score of the supermarket #B or the home center #C. Gradually rise. Here, when the information about the home center #C is provided to the user when the user moves to the position P2, the user who has moved to the supermarket #B may be guided to the home center #C. . On the other hand, for example, when the user moves to the position P3 in front of the supermarket #B, even if the user provides information on the home center #C, the user may not be able to be guided to the home center #C. Is high.

  Therefore, the information providing apparatus 10 estimates the distance between the position of the user and each of the destination candidates. If the user is located within a predetermined range from any of the destination candidates, the information providing apparatus 10 has the highest score. The information of the destination candidate having a high value of, that is, the information of the estimated destination is provided. On the other hand, when the user is not located within a predetermined range from each destination candidate, the information providing apparatus 10 provides information for guiding the user to a proposed location different from the estimated destination. I do.

  For example, when the score of the supermarket #B is the highest, but the score of the home center #C exceeds a predetermined threshold, and the user is offered the home center #C, there is a possibility that the user can be guided to the home center #C. Is high. Therefore, when the user is not located near the supermarket #B or the home center #C, the information providing apparatus 10 is a destination other than the destination estimated from the destination score, and the destination score is equal to the predetermined threshold. Is proposed to the user as a destination. For example, the information providing apparatus 10 proposes a home improvement center #C to a user heading for a supermarket #B. As a result, the information providing apparatus 10 can realize user guidance. When the user has moved to the position P3 in front of the supermarket #B, the information providing apparatus 10 transmits information about the supermarket #B instead of information for guiding the user to the home center #C. Provide to users.

  As described above, when it is estimated that the user has started moving, the estimating unit 53 estimates a plurality of destinations based on the action history and estimates the probability that the user is moving to each destination. I do. Then, the suggestion unit 55 selects a destination to be proposed from destinations other than the destination with the highest accuracy among the destinations. For example, the providing opportunity determination unit 54 determines whether the plurality of probabilities estimated by the estimation unit 53 satisfy a predetermined condition, and the suggestion unit 55 determines whether the probabilities satisfy the predetermined condition. And propose a destination different from the destination to which the user is traveling.

  When the user has moved to the vicinity of any of the destination candidates, the score value of the candidate is estimated to exceed a predetermined threshold. Therefore, the information providing device 10 determines whether the distance between the user and the destination candidate satisfies a predetermined condition, but not whether the score of the destination candidate satisfies a predetermined condition. May be performed. For example, the providing opportunity determination unit 54 determines whether the highest probability among the certainties is equal to or less than the first threshold. Then, when it is determined that the highest certainty among the certainties is equal to or less than the first threshold, the suggestion unit 55 selects one of the destinations whose estimated certainty is equal to or larger than the second threshold that is lower than the first threshold. Is proposed as the proposed destination. On the other hand, when the highest probability among the probabilities exceeds the first threshold, the providing unit 56 provides the user with information on the destination corresponding to the probabilities exceeding the first threshold.

  Returning to FIG. 5, when the information on the destination and the proposed location is provided to the user, the setting unit 57 sets a reward (step S110). For example, when the information on the proposed location is proposed to the user, the setting unit 57 sets a first reward associated with the distribution of the information. For example, the setting unit 57 sets, as the first reward, a probability that the user moves to the proposed location, that is, a reward according to the destination score of the proposed location. For example, the setting unit 57 sets a higher reward as the accuracy is lower. When the user has moved to the proposed destination, the setting unit 57 sets a second reward associated with the guidance of the user.

  For example, in the example shown in FIG. 1, it is assumed that the user is traveling with Super #B as the destination. In such a case, it is assumed that the estimating unit 53 calculates the highest score for the supermarket #B and the second highest score for the home improvement center #C. In such a case, the suggesting unit 55 or the providing unit 56 provides information regarding the supermarket #B or proposes the home center #C to the user.

  In such a case, for example, when the proposal of the home center #C is made while the user is in the section AP1, the setting unit 57 determines that the information for guiding the user to the home center #C has been delivered. Set rewards. That is, the setting unit 57 sets an advertisement fee for advertising the home center #C to the user. In addition, as shown in STU2 of FIG. 1, for example, the setting unit 57 allows the user to go to the home center #C after proposing the home center #C even though the super #B has a higher score. When the user moves, it is assumed that the user has been guided, and a reward is set according to the user's guidance. For example, the setting unit 57 sets an advertisement fee and a reward for the home center #C.

  On the other hand, when the user is in the section AP2, the position of the user approaches the supermarket #B, so that the score of the supermarket #B increases and the score of the home improvement center #C decreases. If the proposal of the home improvement center #C is made in such a state, the impression of the user may be deteriorated. However, when the home center #C desires to guide the user who visits the supermarket #B, it is considered that there is a demand for the proposal of the home center #C even if the score of the home center #C is low.

  Therefore, the setting unit 47 sets a higher reward when the proposal of the home center #C is made to the user in the section AP2 than when the proposal of the home center #C is made to the user in the section AP1. Make the settings for Similarly, the setting unit 47 is more expensive when the proposal of the home center #C is made to the user in the section AP3 than when the proposal of the home center #C is made to the user in the section AP2. Set rewards. In addition, as shown in STU3 of FIG. 1, when the user moves to the home center #C at the position P3, the setting unit 47 sends the user from the position P2 as a reward associated with the guidance of the user. Set a higher reward than when you are guided to #C.

  It should be noted that an arbitrary setting can be made in advance in what state the guidance to the home center #C is performed. For example, when the score of the home center #C satisfies a predetermined condition set in advance, the suggestion unit 55 proposes the home center #C to the user. Then, the setting unit 47 may set a reward based on the score of the home center #C and the score range of the supermarket #B when the proposal of the home center #C is made. For example, the setting unit 47 may set various rewards based on the difference between the highest score and the score of the home improvement center #C.

  In addition, when setting various rewards, the setting unit 47 performs the setting of rewards, for example, by executing a payment process in cooperation with the payment server 300 that executes various payment processes. For example, the setting unit 47 notifies the settlement server 300 of a reward amount calculated based on various scores and user guidance, and a billing destination of the reward (that is, a facility serving as a destination or a proposal place). , Set rewards.

  Returning to FIG. 5, the description will be continued. When the reward of the setting is performed, the movement determination unit 52 determines whether or not the user has completed the movement (Step S111). If the user has not finished moving (Step S111: No), the estimating unit 53 executes Step S104. On the other hand, when it is determined that the user has finished moving (Step S111: Yes), the information providing apparatus 10 ends the estimation processing and the proposal processing.

[4. Modification)
The information providing apparatus 10 according to the above-described embodiment may be implemented in various different forms other than the above-described embodiment. Therefore, another embodiment of the information providing apparatus 10 will be described below.

[4-1. Variation of estimation process)
Hereinafter, variations of the estimation process performed by the information providing apparatus 10 will be described.

[4-1-1. About score calculation)
In the example described above, the information providing apparatus 10 estimates destination candidates based on the position history, and narrows down the destination candidates based on the action history. However, embodiments are not limited to this. For example, the information providing apparatus 10 calculates the destination score of the destination candidate by calculating the sum of the value of the destination score based on the position history and the value of the destination score based on the action history. May be. The information providing apparatus 10 may set a predetermined weight for each destination score.

  Further, the information providing apparatus 10 may individually set the weight value for each user. For example, the information providing apparatus 10 specifies the history of the destination that the user has actually visited from the base based on the location history, and indicates the strength of the relationship between the history of the specified destination and the activity history. Calculate the relevance score. On the other hand, the information providing apparatus 10 calculates the commonality of the time zone and the date and time when the user visits each destination as a positional relationship score.

  When the value of the positional relationship score is higher than the value of the behavioral relationship score, the information providing apparatus 10 attaches importance to the position history rather than the activity history, and sets the value of the positional relationship score to the value of the behavioral relationship score. If it is lower, the action history may be more important than the position history. For example, when the value of the positional relationship score is higher than the value of the behavioral relationship score, the information providing device 10 may compare the score calculated based on the activity history with the score calculated based on the location history. The destination score of the destination candidate may be calculated by multiplying the high value coefficient. In addition, for example, when the value of the behavioral relationship score is higher than the value of the positional relationship score, the information providing apparatus 10 compares the score calculated based on the location history with the score calculated based on the behavior history. A destination score of a destination candidate may be calculated by integrating coefficients having lower values.

  In addition, the information providing apparatus 10 may individually use the values of the score based on the position history and the score based on the action history. For example, the information providing apparatus 10 may estimate the destination using only the score based on the action history. Further, the information providing apparatus 10 estimates the probability that the candidate estimated based on the position history is the destination and the probability that the candidate estimated based on the action history is the destination, and obtains a higher value accuracy. , The destination may be estimated.

  For example, when the value of the score based on the position history for the first destination is higher than the value of the score based on the action history, the information providing apparatus 10 adopts the score based on the position history as the score of the first destination. . On the other hand, when the value of the score based on the position history for the second destination is lower than the value of the score based on the action history, the information providing apparatus 10 adopts the score based on the action history as the score of the second destination. . Then, the information providing apparatus 10 may estimate the destination by comparing the score of the first destination and the score of the second destination.

  In addition, the information providing apparatus 10 may correct the weight of the term for calculating the score based on the position history based on the action history. For example, the information providing apparatus 10 generates a polynomial for calculating a score value based on the position history. Such a polynomial can be realized by, for example, a linear programming method. Then, the information providing apparatus 10 may specify an action history related to each item, and correct the value of each item based on the specified action history.

[4-1-2. Consideration of user attributes)
The information providing apparatus 10 may calculate the scores of the candidates for each destination according to the demographic attribute and the psychographic attribute of the user. More specifically, the information providing apparatus 10 may change the weight based on the position history and the weight based on the action history when estimating the destination according to the attribute information. The information providing apparatus 10 may calculate a score value based on the relationship between the user attribute information and the action history. For example, if the information providing apparatus 10 has performed a web search that is highly relevant to the age of the user immediately before departure, the information providing apparatus 10 determines that there is a high possibility of moving to a destination that is highly relevant to the web search, and determines that the score is high. The value may be calculated.

[4-1-3. Regarding business hours)
The information providing apparatus 10 may estimate the destination in consideration of the business hours of the destination candidate. For example, the information providing apparatus 10 may estimate destination candidates based on the action history, and may estimate the destination from the candidates based on a time zone in which the estimated candidates can be used. For example, the information providing apparatus 10 calculates the score of each destination candidate based on the position history and the action history, and collects a time zone (for example, business hours) in which each destination candidate can be used. Further, a predicted time at which the user arrives at each destination candidate is estimated. Then, the information providing apparatus 10 may decrease the score value when the estimated predicted time is not included in the available time zone.

[4-1-4. Consideration of user's state during action)
Further, the information providing apparatus 10 may estimate the state of the user when the action indicated by the action history has been performed, and may estimate the destination based on the action history and the estimated state. For example, based on the position history, the information providing apparatus 10 estimates the movement state of the user when the user is performing the action indicated by the action history. To give a more specific example, the information providing apparatus 10 estimates whether or not the user is moving and the means of moving the user. For example, when the moving speed of the user exceeds a predetermined threshold and the position of the user exists on the track, the information providing apparatus 10 estimates that the user has moved by train. Further, for example, when the user is traveling on a road at a predetermined traveling speed, the information providing apparatus 10 estimates that the user is traveling by car. That is, the information providing apparatus 10 estimates the movement mode of the user (ie, the movement context) when the action indicated by the action history is performed, based on the position history.

  Then, the information providing apparatus 10 changes the contribution ratio of the action indicated by the action history to the score according to the estimated movement mode. For example, the information providing apparatus 10 may reduce the contribution rate to the score for various search histories when the user is on the train. To give a more specific example, the information providing apparatus 10 determines that the user is on a train when a web search related to the home center #C is being performed. If the value of the score of the home center #C is increased by 5 points and it is estimated that the user is walking when a web search related to the home center #C is performed, the score value of the home center #C is increased. You may increase by 10 points.

  Further, the information providing apparatus 10 may change the contribution rate given to the score based on the terminal used by the user when performing the action indicated by the action history. For example, the information providing apparatus 10 may increase the score value of the home center #C by 20 points when performing a web search having a connection with the home center #C via the navigation system of the automobile. In other words, the information providing apparatus 10 may place more importance on activities on the Internet performed via the terminal device that are highly relevant to the mode of movement than activities performed via the terminal device 100.

  In addition, the information providing apparatus 10 may determine whether the context of the user's movement after starting the early destination guidance and the context of the user's movement when performing the action indicated by the action history match. The contribution rate given to the score may be changed. That is, the information providing apparatus 10 may estimate the destination based on the commonality between the current movement mode of the user and the estimated movement mode and the action history.

  For example, the information providing apparatus 10 estimates the movement mode of the user as the current movement mode from the position history after the start of the early destination guidance. Subsequently, the information providing apparatus 10 estimates an action history having relevance to each destination, and estimates a past movement mode of a user's movement mode when performing the action indicated by each action history. Then, for example, the information providing apparatus 10 sets a larger value when the current movement mode and the past movement mode match than when the current movement mode and the past movement mode do not match. Add to the score.

  A more specific example will be described. For example, the information providing apparatus 10 estimates that the user is moving by train as the current movement mode. The information providing apparatus 10 also indicates from the movement history and the action history that the user has performed a web search for the supermarket #B while moving on foot, and that the home center #C while the user is moving by train. It is estimated that the web search was performed. In such a case, the information providing apparatus 10 increases the score of the supermarket #B by 10 points and increases the score of the home improvement center #C by 20 points. That is, the information providing apparatus 10 changes the contribution ratio of the score based on the action history based on the commonality between the context of the action performed when the action indicated by the action history is performed and the context of the current movement.

[4-1-5. About the consideration of the weather)
Further, the information providing apparatus 10 specifies the weather when the action indicated by the action history is performed, and estimates the destination based on the action history and the specified weather (for example, weather or temperature). Is also good. For example, the information providing apparatus 10 changes the contribution ratio of the score based on the action history according to the weather when the action indicated by the action history is performed. In addition, for example, the information providing apparatus 10 determines the contribution ratio of the score by the action history according to the commonality between the weather when the action indicated by the action history is performed and the weather after the early destination guidance is started. May be changed.

  For example, the information providing apparatus 10 specifies the weather at the time of performing the action indicated by the action history, and when the specified weather is sunny, the contribution rate to the score of the destination is higher than when the weather is rainy. May be raised. In addition, when the specified weather and the weather at the time of starting the early destination guidance are common, the information providing apparatus 10 may increase the contribution rate to the action history score as compared with the case where the weather is not common.

[4-1-6. About use of model)
In addition, the information providing apparatus 10 may estimate the destination using various models. For example, the information providing apparatus 10 determines the relationship between the history of the visited place visited by the user and the behavior history of the user, which is the behavior history before the user visits the visited place. The destination may be estimated from the action history using the learned learning model. Such a model may be, for example, a DNN (Deep Neural Network), an LSTM (Long Short-Term Memory) convolutional neural network, or a recursive neural network. The model may be a combination of the functions of the convolutional neural network and the recursive neural network. Further, the information providing apparatus 10 may use an arbitrary regression model such as Support Vector Regression.

  For example, the information providing apparatus 10 specifies the destination visited by the user based on the position history, and specifies the date and time when the user has moved to each destination and the movement mode. Subsequently, the information providing apparatus 10 specifies the action history within a predetermined period from the specified date and time, and specifies the weather when the action indicated by the specified action history is performed. Then, the information providing apparatus 10 sets each of the destinations so that the score of the predetermined destination becomes highest when the movement mode, the content of the action history, and the weather at the time of each action history are input. Learning of the model that calculates the score of the ground. In addition, the information providing apparatus 10 may employ any information other than the above as input.

  Note that the information providing apparatus 10 may estimate the contribution rate of each input in such learning, and exclude inputs from which the estimated contribution rate falls below a predetermined threshold from learning targets. For example, the information providing apparatus 10 compares the accuracy of the model when the weather is input with the accuracy of the model when the weather is not input. If there is not much difference between the respective accuracy, the information providing device 10 determines the weather. It may be excluded from the input.

[4-1-7. Regarding consideration of road conditions)
In addition, the information providing apparatus 10 may estimate the destination based on the situation of the moving route from the user's current location to the destination candidate. For example, when a traveling route to a certain destination is mixed, the information providing apparatus 10 may lower the score of the destination.

  Further, the information providing apparatus 10 specifies the direction of the candidate based on the user in the moving route from the current position to the candidate of the destination as the state of the moving route, and calculates the score of the destination based on the specified direction. May be calculated. For example, the information providing apparatus 10 estimates the movement mode of the user, and when it is estimated that the user is moving to the destination by car, the destination is determined in the route to the destination in the moving direction. To determine if it is on the right or left. Then, when the destination is on the right side, that is, when it is necessary to make a right turn to reach the destination, the information providing apparatus 10 may decrease the score value of the destination. When the median strip exists on the road in front of the destination, the information providing apparatus 10 may decrease the score value of the destination.

  The information providing apparatus 10 may specify the driving characteristics of the user as the attribute information of the user, and may estimate the destination based on the specified driving characteristics and the road conditions. For example, the information providing apparatus 10 estimates the specific driving of the user by using various estimation techniques for estimating the driving characteristics of various users (for example, whether or not the user is good at driving). Then, when the driving characteristics of the user are high, the information providing apparatus 10 calculates the score value regardless of the direction of the destination with respect to the moving direction. On the other hand, when the driving characteristics of the user are low, the information providing apparatus 10 may change the score value depending on whether the destination is on the right side or the left side in the moving direction.

  Further, the information providing apparatus 10 may learn a model for calculating a score based on such driving characteristics and the position of the destination. Further, the information providing apparatus 10 may change the weight of the action history based on such model parameters.

[4-2. Variations of proposal processing)
Hereinafter, variations of the proposal processing executed by the information providing apparatus 10 will be described.

[4-2-1. About proposal of stopover place)
For example, when the information providing apparatus 10 proposes a proposed location to a user, the information providing apparatus 10 may propose a relay point and then propose a destination. For example, the information providing apparatus 10 further obtains position information indicating the current position of the user, selects a proposed location based on the action history, and selects a relay point from the current position to the proposed location, After proposing the selected relay point, the selected proposed point may be proposed. Here, the information providing apparatus 10 may select a relay point having relevance to the proposed point.

  For example, in FIG. 1, it is assumed that the home center #C is selected as the proposed location while the user is moving to the destination Super #B by car. Here, when a gas station exists on the route from the current location of the user to the home center #C, the information providing apparatus 10 may guide the user to the gas center and then to the home center #C. As such a gas station, a gas station affiliated with the home improvement center #C may be proposed.

  For example, the information providing apparatus 10 proposes, to the user, a gas station that provides coupons that can be used at the home center #C instead of the home center #C, which is the proposed location, as a relay point. Then, when the user visits the relay point, the information providing apparatus 10 may provide information of various benefits provided to the user at the proposed location, such as a coupon available at the home center #C. . As described above, since the information providing apparatus 10 guides the user step by step, the information providing apparatus 10 can sequentially guide the user to the provided value.

[4-2-2. About proposed place)
In addition, the information providing apparatus 10 may select any destination as a proposed location. For example, the information providing apparatus 10 may select the destination having the second highest score as the proposed location. In addition, the information providing apparatus 10 receives in advance the setting of the advertisement fee from the facility that can be each destination, and proposes the destination with the highest advertisement fee among the destination candidates estimated from the position history and the action history. It may be selected as the ground. That is, the information providing apparatus 10 may propose a proposed location in a bid format. In addition, the information providing apparatus 10 calculates a score of each destination candidate based on the advertisement fee in addition to the position history and the action history, and determines a destination candidate whose calculated score value exceeds a predetermined threshold. You may choose as a proposal place.

  Further, the information providing apparatus 10 may select a proposed location related to the destination. For example, the information providing apparatus 10 may select another destination belonging to the same category as the destination as the proposed location. More specifically, when the destination is a supermarket, the information providing apparatus 10 may select another supermarket as the proposed location. Alternatively, the information providing apparatus 10 may specify the average consumption amount at the destination, and select a proposed location whose average consumption amount is approximately the same as the specified consumption amount. That is, the information providing apparatus 10 may cause the user to go to a destination competition. Further, the category of the destination and the category of the proposed location may be the same or different. The category of the destination and the proposed location may be a continuous (predetermined order) category such as the order of consumption or restriction.

  In addition, the information providing apparatus 10 may estimate the purpose of the user heading to the destination, and may propose a proposed location that achieves the same purpose as the estimated purpose but is different. For example, when there are a plurality of pools at positions at the same distance from the user but in different directions, the information providing apparatus 10 specifies a pool to be a destination based on the position history. Then, the information providing apparatus 10 may set a pool different from the pool specified as the destination as the proposed location. Further, for example, the information providing apparatus 10 specifies a pool having many gas stations and restaurants on a route from pools different from the pool specified as the destination, and provides the specified pool to the user as a proposed location. May suggest.

  That is, the information providing apparatus 10 distributes the information of the proposed location, that is, the advertisement of the proposed location, aiming at the information that it is moving (early destination prediction). Since the route changes with the movement of the person, the destination and the proposed location change, and the candidates for the advertisement for distribution can be narrowed. Therefore, the information providing apparatus 10 distributes information of a proposal target having the highest appealing effect (that is, good) while narrowing due to route restrictions and the like. That is, the information providing apparatus 10 can narrow down the categories due to the early destination prediction.

  Here, the information providing apparatus 10 may estimate an unknown association (that is, serendipity) existing between the destination and the proposed location. For example, the information providing apparatus 10 estimates a destination and proposes a proposed location randomly selected at a predetermined ratio to the user. Then, when the user moves to the proposed location, the information providing apparatus 10 may estimate the commonality or relevance between the attribute of the destination and the attribute of the proposed location as unknown relevance.

[4-2-3. Proposal timing)
Here, the information providing apparatus 10 may make a proposal of a proposed location at an arbitrary timing. For example, among the candidates for each destination, a candidate for a proposed location may be proposed from candidates whose score value exceeds a predetermined threshold. In other words, the information providing apparatus 10 selects, from the candidates for the destination, a candidate that is presumed to be of some interest to the user or a candidate to which the user is likely to be moving to some extent as the proposed location. May be.

  Further, the information providing apparatus 10 selects a proposed location to be proposed to the user, and provides information on the proposed location at a timing corresponding to a reward amount set in advance for distribution of the information on the selected proposed location. May be. For example, when a higher reward amount is set, the information providing apparatus 10 may provide information on the proposed location at a faster timing. In addition, when the amount of reward higher than the predetermined threshold is set, the information providing apparatus 10 selects the proposed location even if the score is lower than the predetermined threshold, and provides the user with May be suggested. For example, when the amount of reward of the home center #C exceeds a predetermined threshold, the information providing apparatus 10 sends the home center #C to the user even if the user moves to the vicinity of the supermarket #B. May suggest.

  Further, the information providing apparatus 10 may select a proposed place to be proposed to the user, and distribute information on the selected proposed place at a timing according to the attribute of the user. For example, when the age of the user exceeds a predetermined threshold, the information providing apparatus 10 may propose a proposed location at a faster timing.

  Further, the information providing apparatus 10 may select a proposed location to be proposed to the user based on the action history and the distance from the destination, and suggest the selected proposed location. For example, when the supermarket #B is the destination, the information providing apparatus 10 specifies the distance between the supermarket #B and the home center #C. If the distance between the supermarket #B and the home center #C is within a predetermined threshold, the information providing apparatus 10 selects the home center #C as a proposed location, and if the distance exceeds the predetermined threshold, The home center #C may be excluded from the proposed location.

  In addition, the information providing apparatus 10 may select the proposed location and set the timing in consideration of various moving costs such as a distance between the destination and the proposed location and a traveling time. For example, the information providing apparatus 10 may select the proposed location so that the travel cost from the current location to the destination and the travel cost from the current location to the proposed location are substantially the same. In addition, the information providing apparatus 10 may provide the information of the proposed location to the user at the timing when the respective moving costs become substantially equal.

  Further, for example, the information providing apparatus 10 determines whether each destination candidate is suitable as a proposed location separately from a score (hereinafter, referred to as “destination score”) indicating a possibility that the user will move. The proposed location score shown may be calculated, and the proposed location may be selected based on the calculated score. For example, the information providing apparatus 10 may calculate a proposed location score based on the action history, and may set a candidate for a destination where the calculated proposed location score exceeds a predetermined threshold as a proposed location. In addition, when the value of the proposed location score exceeds a predetermined threshold, the information providing apparatus 10 may provide the information of the proposed location even if there is no relevance to the purpose.

  Further, when the value of the proposed location score exceeds a predetermined threshold, the information providing apparatus 10 may provide the information of the proposed location at a timing when the user leaves the base. In addition, the information providing apparatus 10 may balance the value of each score and the cost of moving to the proposed location, and provide the information of the proposed location at a timing estimated to be appropriate.

  Note that any timing can be adopted as the timing at which the information providing apparatus 10 provides the information of the proposed location to the user. For example, the information providing apparatus 10 may suggest the information of the proposed location in a range in the middle of the route where the user reaches the destination, and does not have to propose the information of the proposed location in a range close to the destination. good.

  In addition, the information providing apparatus 10 may change the selection of the proposed location and the proposed timing based on the evaluation (review) of the user for each destination candidate. For example, the information providing apparatus 10 may select a candidate for a destination having a high evaluation of the user as a proposed location, and may provide information of the proposed location at an earlier timing as the evaluation of the user is higher.

[4-2-4. About use of reaction of user]
In addition, the information providing apparatus 10 may select a proposed location using a reaction of the user. For example, the information providing apparatus 10 may select a new proposed location to the user based on a reaction of the user when the proposed location is proposed to the user. For example, when the information providing apparatus 10 proposes the home center #C to the user as the proposed location, the user is friendly to the proposal based on the voice generated by the user, the direction in which the user has moved, and the like. Is estimated. Then, when it is estimated that the user is not favorable to the proposal, the information providing apparatus 10 may select a new proposed place other than the home center #C and provide the selected proposed place.

  Further, the information providing apparatus 10 may calculate the proposed location score based on the history of the reaction of the user. For example, the information providing apparatus 10 increases the destination score and the proposed location score for a proposed location to which the user has responded favorably, when the proposed location has been selected as a destination candidate in the subsequent estimation processing. For a proposed location having a negative response, a destination score or a proposed location score may be reduced when the proposed destination is determined as a destination candidate in the subsequent estimation processing. In addition, the information providing apparatus 10 may calculate the destination score or the proposed destination score based on the history of the user's reaction, or may modify a model for performing such calculation. That is, the information providing apparatus 10 may correct a candidate for early purpose prediction based on a user's reaction (a statement that did not go there, a facial expression, and the like). As a result of such processing, the information providing apparatus 10 can specify, for example, a category to which the user wants to go (that is, a category of a destination).

[4-2-5. About incentives)
Further, the information providing apparatus 10 may set a reward (incentive) according to the timing of providing the information of the proposed location. For example, the information providing apparatus 10 may set a higher reward as the difference between the highest destination score and the destination score of the proposed location is larger. That is, the information providing apparatus 10 may calculate the amount of reward based on the score balance of each destination candidate. In addition, the information providing apparatus 10 may set a reward according to the attribute of the user as the proposal destination. For example, the information providing apparatus 10 provides information on a proposed location to a user who frequently searches for a predetermined product, rather than providing information on the proposed location to other users. A higher reward may be set. Such a reward may be calculated, for example, by a predetermined calculation formula using the value of the user attribute, the provision timing, or the value of the destination score as a parameter, or may be set according to a preset pattern.

  Further, the information providing apparatus 10 may change the timing of providing the information on the proposed location according to a preset reward amount. For example, the information providing apparatus 10 calculates the score balance of the candidate of each destination when providing the information of the proposed location based on the reward amount, and calculates the score balance of the candidate of the destination in the estimation process. If the balance is reached, information on the proposed location may be provided.

[4-3. Individual execution of estimation processing and proposal processing)
Note that the information providing apparatus 10 may individually execute the above-described estimation processing and proposal processing. For example, the information providing apparatus 10 performs only the process of estimating the destination of the user based on the behavior history of the user and providing the user with information on the estimated destination by the above-described estimation process. Is also good. At this time, the information providing apparatus 10 may change the timing at which the information about each destination is provided according to the amount of reward set in advance for each destination.

  Further, the information providing apparatus 10 may, for example, estimate a destination based on the position history and provide information of a proposed location different from the estimated destination to the user. For example, the information providing apparatus 10 may provide information on a proposed location having a relevance to the estimated destination or information on a proposed location located around the estimated destination. At this time, the information providing apparatus 10 proposes a candidate of a proposed location that is higher in a preset reward amount and a candidate of a proposed location that is more compatible with a user attribute, among candidates of a proposed location located in Shuichi of the destination. It may be selected as the ground.

[4-4. About guidance processing)
The information providing apparatus 10 may execute the above-described estimation processing and proposal processing regardless of the user's movement mode. For example, the information providing apparatus 10 can be used via an arbitrary route, such as when walking on foot, when traveling by train, when traveling via a road such as a car, bicycle, or motorcycle, or when combining various transportation means. The above-described evaluation processing may be applied to early destination guidance when a person moves.

[4-5. About the route)
The information providing apparatus 10 can set an arbitrary section as a route. That is, the information providing apparatus 10 may set an arbitrary position as the point A or the point B. For example, the information providing apparatus 10 may set a route from a predetermined intersection to another intersection, or may set a section from an arbitrary point to a certain point as a route. Further, the information providing apparatus 10 may set a range to be a movement route based on a unit of a route used in various kinds of guidance.

[4-6. Others)
Further, among the processes described in the above embodiment, all or some of the processes described as being performed automatically can be manually performed, and conversely, the processes described as being performed manually can be performed. Can be automatically or completely performed by a known method. In addition, the processing procedure, specific names, and information including various data and parameters shown in the above document and drawings can be arbitrarily changed unless otherwise specified. For example, the various information shown in each drawing is not limited to the information shown.

  Each component of each device illustrated is a functional concept and does not necessarily need to be physically configured as illustrated. In other words, the specific form of distribution / integration of each device is not limited to the illustrated one, and all or a part of the distribution / integration may be functionally or physically distributed / arbitrarily in arbitrary units according to various loads and usage conditions. Can be integrated and configured.

  Further, the respective embodiments described above can be appropriately combined within a range that does not contradict processing contents.

[4-7. program〕
Further, the information providing apparatus 10 according to the above-described embodiment is realized by, for example, a computer 1000 having a configuration as shown in FIG. FIG. 6 is a diagram illustrating an example of a hardware configuration. The computer 1000 is connected to an output device 1010 and an input device 1020, and a form in which a computing device 1030, a primary storage device 1040, a secondary storage device 1050, an output IF (Interface) 1060, an input IF 1070, and a network IF 1080 are connected by a bus 1090. Having.

  The arithmetic device 1030 operates based on a program stored in the primary storage device 1040 or the secondary storage device 1050, a program read from the input device 1020, or the like, and executes various processes. The primary storage device 1040 is a memory device such as a RAM that temporarily stores data used by the arithmetic device 1030 for various calculations. The secondary storage device 1050 is a storage device in which data used by the arithmetic device 1030 for various calculations and various databases are registered, such as a ROM (Read Only Memory), a HDD (Hard Disk Drive), and a flash memory. Is realized by:

  The output IF 1060 is an interface for transmitting information to be output to an output device 1010 that outputs various types of information such as a monitor and a printer. For example, a USB (Universal Serial Bus), a DVI (Digital Visual Interface), This is realized by a connector of a standard such as HDMI (registered trademark) (High Definition Multimedia Interface). The input IF 1070 is an interface for receiving information from various input devices 1020 such as a mouse, a keyboard, and a scanner, and is realized by, for example, a USB.

  The input device 1020 includes, for example, an optical recording medium such as a CD (Compact Disc), a DVD (Digital Versatile Disc), and a PD (Phase change rewritable Disk), a magneto-optical recording medium such as an MO (Magneto-Optical disk), and a tape. A device that reads information from a medium, a magnetic recording medium, a semiconductor memory, or the like may be used. The input device 1020 may be an external storage medium such as a USB memory.

  The network IF 1080 receives data from another device via the network N and sends the data to the arithmetic device 1030, and transmits the data generated by the arithmetic device 1030 to the other device via the network N.

  The arithmetic device 1030 controls the output device 1010 and the input device 1020 via the output IF 1060 and the input IF 1070. For example, the arithmetic device 1030 loads a program from the input device 1020 or the secondary storage device 1050 onto the primary storage device 1040, and executes the loaded program.

  For example, when the computer 1000 functions as the information providing device 10, the arithmetic device 1030 of the computer 1000 executes the program or data (for example, the model M1) loaded on the primary storage device 1040, thereby causing the control unit 40 to operate. Implement the function. The arithmetic unit 1030 of the computer 1000 reads and executes these programs or data (for example, the model M1) from the primary storage device 1040. As another example, the arithmetic unit 1030 obtains these programs from another device via the network N. May be.

[5-1. Effect of estimation processing)
As described above, the information providing apparatus 10 acquires the user's action history on the network, and when it is estimated that the user has started moving, estimates the destination of the movement based on the action history. I do. As a result of such processing, the information providing apparatus 10 can improve the estimation accuracy of the destination of the user.

  In addition, the information providing apparatus 10 further acquires the position history of the user, and estimates the destination based on the position history and the action history. For example, the information providing apparatus 10 estimates a destination candidate based on the position history, and estimates a destination from the candidates based on the action history. For example, the information providing apparatus 10 specifies an action history related to a destination candidate from the action history, and estimates a destination from the candidates based on the specified action history. As a result of such processing, the information providing apparatus 10 can further improve the estimation accuracy of the destination.

  Further, the information providing apparatus 10 further acquires attribute information indicating an attribute of the user, and changes a weight based on the position history and a weight based on the action history when estimating the destination according to the attribute information. As described above, since the information providing apparatus 10 can estimate, for each user, a destination to which the user is likely to move, it is possible to further improve the estimation accuracy of the destination.

  Further, the information providing apparatus 10 estimates the probability that the candidate estimated based on the position history is the destination and the probability that the candidate estimated based on the action history is the destination, and obtains a higher value accuracy. The destination is estimated based on. In addition, the information providing apparatus 10 calculates, for each destination candidate, the probability that the user is moving to the destination candidate, and sets a candidate whose calculated accuracy satisfies a predetermined condition as a destination candidate. . Further, the information providing apparatus 10 specifies the distance to each destination candidate based on the position history, and increases the accuracy value as the specified distance becomes narrower. In addition, the information providing apparatus 10 estimates a destination candidate based on the action history, and estimates a destination from the candidates based on a time zone in which the estimated candidates can be used.

  Further, the information providing apparatus 10 estimates the state of the user when the action indicated by the action history has been performed, and estimates the destination based on the action history and the estimated state. In addition, the information providing apparatus 10 estimates, as the state of the user, a movement mode of the user when the action indicated by the action history is performed. In addition, the information providing apparatus 10 estimates the destination based on the commonality between the current movement mode of the user and the estimated movement mode and the action history.

  Further, the information providing apparatus 10 specifies the date and time when the action indicated by the action history was performed, and estimates the destination based on the action history and the specified date and time. Further, the information providing apparatus 10 specifies the weather when the action indicated by the action history is performed, and estimates the destination based on the action history and the specified weather. The information providing apparatus 10 further acquires the current position of the user, and estimates the destination based on the action history and the state of the moving route from the current position to the destination candidate.

  Further, the information providing apparatus 10 specifies, as the state of the moving route, the direction of the candidate based on the user in the moving route from the current position to the destination candidate. In addition, the information providing apparatus 10 estimates destination candidates, and specifies a candidate having a higher relationship with the action history among the estimated candidates as the destination. In addition, the information providing apparatus 10 acquires a search history of the user, a purchase history of the user, or a reservation history of the store by the user as the action history.

  As a result of each process described above, the information providing apparatus 10 can improve the estimation accuracy of the destination of the user in the early destination guidance.

[5-2. Effect of proposal processing)
Further, the information providing apparatus 10 acquires a user's action history on the network. Then, when it is estimated that the user has started moving to the predetermined destination, the information providing apparatus 10 proposes a destination different from the predetermined destination based on the action history. As a result of such processing, the information providing apparatus 10 can guide the user to a destination different from the destination of our office in the early destination guidance, that is, the proposed destination, and And provide effective information.

  In addition, the information providing device 10 proposes a destination having a predetermined association with a predetermined destination among the destinations based on the action history. In addition, when it is estimated that the user has started moving, the information providing apparatus 10 estimates a plurality of destinations based on the action history and estimates the probability that the user has moved to each destination. . Then, the information providing apparatus 10 selects a destination to be proposed from destinations other than the destination with the highest accuracy among the destinations. For this reason, since the information providing apparatus 10 can set a destination that is highly likely to be changed by the user from the destination of the office as a proposed location, it is possible to improve the possibility of guiding the user.

  In addition, the information providing apparatus 10 determines whether or not the plurality of estimated probabilities satisfy a predetermined condition, and proposes a destination when it is determined that the plurality of probabilities satisfy the predetermined condition. For example, the information providing apparatus 10 determines whether the highest probability among the estimated probabilities is equal to or less than a first threshold, and when the highest accuracy is determined to be equal to or less than a first threshold, the accuracy is Any of the destinations that are equal to or greater than a second threshold lower than the first threshold is proposed. Further, the information providing apparatus 10 further acquires position information indicating the current position of the user, and estimates the probability that the user is moving to each destination based on the position information. For this reason, the information providing apparatus 10 can provide the information of the proposed location at a timing that is easy to guide the user, so that the possibility of guiding the user can be improved.

  Further, the information providing apparatus 10 further obtains position information indicating the current position of the user, selects a destination different from the predetermined destination based on the action history, and sets the destination from the current position indicated by the position information. After selecting a relay point up to and proposing the selected relay point, the selected destination is proposed. In addition, the information providing apparatus 10 selects a relay point having relevance to the destination. In addition, when the user visits the relay point, the information providing apparatus 10 provides information of the profit provided to the user at the selected destination. As a result of such processing, the information providing apparatus 10 can guide the user to the proposed location step by step.

  Further, the information providing apparatus 10 selects a destination newly proposed to the user based on a reaction of the user when the destination is proposed to the user. Further, the information providing apparatus 10 proposes a destination whose relevance with the action history exceeds a predetermined threshold. For this reason, since the information providing apparatus 10 can preferentially provide a proposed location to which the user is interested, the possibility of guiding the user can be improved.

  When information on a destination different from the predetermined destination is proposed to the user, the information providing apparatus 10 sets a first reward associated with the distribution of the information. Further, the information providing apparatus 10 sets, as a first reward, a reward according to the likelihood that the user will move to a destination different from the predetermined destination. The information providing apparatus 10 sets a higher reward as the accuracy is lower. In addition, when the user has moved to the proposed destination, the information providing apparatus 10 sets a second reward according to the guidance of the user. Therefore, the information providing apparatus 10 can set a reward for the user's guidance. In addition, the information providing apparatus 10 can use the information on the destination and the proposed location provided in the early destination guidance as a so-called advertisement.

  In addition, the information providing apparatus 10 selects a destination to be proposed to the user, and provides information on the destination at a timing corresponding to a reward amount set in advance for distribution of information on the selected destination. I do. Further, the information providing apparatus 10 selects a destination to be proposed to the user, and distributes information on the selected destination at a timing according to the attribute of the user. Further, the information providing apparatus 10 selects a destination to be proposed to the user based on the action history and the distance from the predetermined destination, and proposes the selected destination. As a result, the information providing apparatus 10 can improve the possibility of guiding the user.

  As described above, some of the embodiments of the present application have been described in detail with reference to the drawings. However, these are exemplifications, and various modifications, The invention can be implemented in other modified forms.

  Further, the above-mentioned “section (section, module, unit)” can be read as “means”, “circuit”, or the like. For example, the detection unit can be read as a detection unit or a detection circuit.

DESCRIPTION OF SYMBOLS 10 Information provision apparatus 20 Communication part 30 Storage part 31 Position history database 32 Action history database 40 Control part 51 Acquisition part 52 Movement determination part 53 Estimation part 54 Provision opportunity determination part 55 Proposal part 56 Provision part 57 Setting part 100 Terminal device 200 Log Server 300 Payment server

Claims (20)

An acquisition unit for acquiring a user's action history on the network;
An estimating unit for estimating a destination of the movement based on the action history when the user is estimated to have started the movement. The acquisition unit further acquires the position history of the user,
The estimating device according to claim 1, wherein the estimating unit estimates the destination based on the position history and the action history. The said estimation part estimates the candidate of the said destination based on the said position history, and estimates the said destination from the said candidate based on the said action history. The said Claim. Estimation device. The estimating unit specifies an action history related to the destination candidate from the action history, and estimates the destination from the candidates based on the specified action history. Item 4. The estimating device according to item 3. The obtaining unit further obtains attribute information indicating an attribute of the user,
The said estimation part changes the weight based on the said position history at the time of estimating the said destination, and the weight based on the said action history according to the said attribute information. The estimating device according to any one of the above. The estimation unit individually estimates the probability that the candidate of the destination estimated based on the position history is the destination and the probability that the candidate estimated based on the action history is the destination, The estimating device according to any one of claims 2 to 5, wherein the destination is estimated based on a higher value accuracy. The estimating unit calculates, for each of the destination candidates, a probability that the user is moving to the destination candidate, and calculates a candidate whose calculated accuracy satisfies a predetermined condition as the destination candidate. The estimation device according to any one of claims 2 to 6, wherein The estimation unit according to claim 7, wherein the estimation unit specifies a distance to each destination candidate based on the position history, and increases the accuracy value as the specified distance decreases. apparatus. The estimation unit estimates the destination candidates based on the action history, and estimates the destination from the candidates based on a time zone in which the estimated candidates can be used. The estimating device according to claim 1. The estimating unit estimates a state of the user when the action indicated by the action history is performed, and estimates the destination based on the action history and the estimated state. The estimation device according to any one of claims 1 to 9. The estimating device according to claim 10, wherein the estimating unit estimates, as the state of the user, a movement mode of the user when the action indicated by the action history is performed. The said estimation part estimates the said destination based on the commonality between the current movement mode of the user and the estimated movement mode, and the action history. An estimation device as described. The estimating unit specifies a date and time when the action indicated by the action history is performed, and estimates the destination based on the action history and the specified date and time. The estimation device according to any one of the above. The said estimation part specifies the weather at the time of the action which the said action history showed being performed, and estimates the said destination based on the said action history and the specified weather. 14. The estimating device according to any one of to 13. The acquisition unit further acquires a current position of the user,
The said estimation part estimates the said destination based on the said action history and the state of the moving route from the present position to the said candidate of a destination. The any one of Claims 1-14 characterized by the above-mentioned. The estimating device according to any one of the above. The said estimation part specifies the direction of the said candidate based on the said user in the moving route from the present position to the candidate of the destination as a state of the said moving route. Estimation device. The said estimation part estimates the candidate of the said destination, and specifies the candidate with a higher relationship with the said action history among the estimated candidates as the said destination. The estimating device according to any one of the above. The said acquisition part acquires the search history of the said user, the purchase history of the said user, or the reservation history of the store by the said user as the said action history. The estimating device according to any one of the above. In an estimation method performed by the estimation device,
An acquisition step of acquiring a user's action history on the network;
An estimating step of estimating a destination of the movement based on the action history when it is estimated that the user has started the movement. An acquisition procedure for acquiring a user's action history on the network,
An estimating program for causing a computer to execute, when it is estimated that the user has started moving, an estimating procedure for estimating a destination of the moving based on the action history.

Images