JP6767952B2 - Estimator, estimation method and estimation program - Google Patents

Description

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

Conventionally, there are known techniques for providing various types of information to users. As an example of such a technology, a technology that selects information that is presumed to be of interest to the user based on various attributes such as the user's hobbies, tastes, gender, etc., and provides the selected information to the user. Are known.

Japanese Unexamined Patent Publication No. 2014-194710

However, in the above-mentioned prior art, it may not be possible to appropriately identify the information presumed to be of interest to the user.

For example, in the above-mentioned prior art, the information to be provided is only selected according to the attributes of the users registered in advance in the database. Therefore, if the user's attribute is not registered in the database, the information presumed to be of interest to the user cannot be properly registered.

The present application has been made in view of the above, and an object of the present application is to improve the estimation accuracy of user attributes.

The estimation device according to the present application is a specific unit that identifies a user's visit target based on action schedule information indicating the user's future behavior, and the user based on the visit target specified by the specific unit. It is characterized by having an estimation unit for estimating the attributes of.

According to one aspect of the embodiment, the estimation accuracy of the user's attribute can be improved.

FIG. 1 is a diagram showing an example of a specific process executed by the information providing device according to the embodiment. FIG. 2 is a diagram showing a configuration example of the information providing device according to the embodiment. FIG. 3 is a diagram showing an example of information registered in the user database according to the embodiment. FIG. 4 is a diagram showing an example of information registered in the route search log database according to the embodiment. FIG. 5 is a diagram showing an example of information registered in the visited database according to the embodiment. FIG. 6 is a diagram illustrating a flow of estimation processing executed by the information providing device according to the embodiment. FIG. 7 is a flowchart showing an example of a flow of estimation processing executed by the information providing device according to the embodiment. FIG. 8 is a diagram showing an example of a hardware configuration.

Hereinafter, a mode for carrying out the estimation device, the estimation method, and the estimation program according to the present application (hereinafter, referred to as “the embodiment”) will be described in detail with reference to the drawings. Note that this embodiment does not limit the estimation device, estimation method, and estimation program according to the present application. Further, in each of the following embodiments, the same parts are designated by the same reference numerals, and duplicate description is omitted.

[Embodiment]
[1-1. An example of an information providing device]
First, an example of a specific process executed by an information providing device, which is an example of the specific device, will be described with reference to FIG. FIG. 1 is a diagram showing an example of a specific process executed by the information providing device according to the embodiment. In FIG. 1, a terminal device 101 used by each user U1 and U2 (hereinafter, may be collectively referred to as “user U”) via a predetermined network N (see, for example, FIG. 2) such as the Internet. , 102 (hereinafter, may be collectively referred to as “terminal device 100”), the search server 200, the SNS (Social Networking Service) server 300, and the information providing device 10 can communicate with each other.

The terminal device 100 is a smart device such as a smartphone or tablet, and is a mobile terminal device capable of communicating with an arbitrary server device via a wireless communication network such as 3G (3rd Generation) or LTE (Long Term Evolution). is there. The terminal device 100 may be an information processing device such as a desktop PC (Personal Computer) or a notebook PC as well as a smart device.

Further, the terminal device 100 may have a function of specifying the current position by using a positioning system such as GPS (Global Positioning System) and outputting position information indicating the specified position. Further, the terminal device 100 may have a function of communicating with various wearable devices worn by the user U and acquiring various biological information such as the pulse, blood pressure, exercise amount, and brain wave of the user U. .. The terminal device 100 executes various processes that can be executed by a general mobile device, such as a search using the Internet, a route search, a purchase in an electronic shopping street, a use of an online auction, and a reproduction of music contents and video contents. It shall be possible.

The search server 200 is an information processing device that provides a service related to route search to the user U, and is realized by, for example, a server device or a cloud system. For example, the search server 200 accepts the departure point and the arrival point from the terminal device 100 as a route search search query. In such a case, the search server 200 searches for the movement route from the departure place to the arrival place, the means of transportation (train, bus, taxi, walking, etc.), the required time, the estimated time of arrival, etc. To provide.

Here, the search server 200 can perform a route search from an arbitrary departure point to an arbitrary arrival point. For example, the search server 200 adopts not only important points of public transportation such as bus stops and stations, but also landmarks and addresses of hotels and various facilities, places selected by map contents, etc. as departure and arrival places. be able to. Further, the search server 200 may adopt, for example, the current position acquired by the terminal device 100 using GPS as the starting point. In the following description, as an example of the route search executed by the search server 200, an example of accepting the designation of a station as a departure place or an arrival place will be described.

In addition, the search server 200 can accept the designation of the departure date and time and the arrival date and time. For example, when the search server 200 accepts the designation of the departure place and the arrival place together with the departure date and time, the movement route when departing from the designated departure place to the arrival place after the designated departure date and time. Or search for estimated time of arrival. In addition, when the search server 200 receives the designation of the departure place and the arrival place together with the arrival date and time, the search server 200 sets the movement route and the departure date and time so as to arrive at the designated arrival place before the designated arrival date and time. Search for.

In addition to the above-mentioned processing, the search server 200 can realize an arbitrary route search. Further, it is assumed that the search server 200 holds the history of the route search by the user U as a log. For example, the search server 200 holds a log in which a search query such as a departure place, an arrival place, a departure date / time, an arrival date / time, etc. received from the terminal device 100 is associated with a route search result for the search query.

The SNS server 300 is an information processing device such as an SNS that manages and provides posted information posted on the Internet by the user U, and is realized by a server device, a cloud system, or the like.

The information providing device 10 is an information processing device that executes an estimation process described later, and is realized by, for example, a server device or a cloud system. For example, the information providing device 10 acquires a log of a search query held by the search server 200, and executes the estimation process according to the invention by using the acquired log.

[1-2. About the outline of estimation processing]
Here, it can be said that the route search log in which a future date and time is set as a departure date and time or an arrival date and time to some extent is information indicating the future action of the user U, that is, action schedule information. By analyzing such action schedule information, it is conceivable to predict what kind of user will visit what kind of visit target.

Here, the visit target is a concept that includes not only geographical targets such as stations, places, and various facilities, but also various events such as live performances and events. For example, the information providing device 10 may specify the holding of various events, the opening of a new store, or the like as an event. For example, it is extended to the effect that a character appearance event has occurred in a game application or the like using location information. It is possible to specify as an event that a location-linked event is occurring in an application using reality, free distribution of products, information on the appearance of celebrities, and the like. That is, the information providing device 10 may target any target as a visit target regardless of whether it is virtual or real.

Here, facilities appearing in content such as animation and movies are expected to be frequently visited by users who like the content. In addition, users who like an artist are more likely to participate in live performances and events in which that artist appears. In addition, users who are interested in gourmet food are likely to visit newly opened restaurants and popular restaurants. In this way, it is considered that the visit target visited by the user has an attribute corresponding to the hobby and taste of the user who visits the visit target. In addition, the visit target, such as a store popular with women and a store popular with men in their thirties, has attributes corresponding to various attributes such as demographic attributes and psychographic attributes as well as the hobbies and tastes of users. Conceivable.

On the other hand, the user U who actually visited the visited target is considered to have a stronger motivation for the visited target than the user U who does not visit the visited target. Therefore, it is considered that the user U who actually visited the visited target is likely to have an attribute corresponding to the attribute of the visited target.

Therefore, the information providing device 10 estimates the attributes of the user U by executing the following estimation processing. First, the information providing device 10 identifies the visit target of the user U based on the action schedule information indicating the future action of the user U. Then, the information providing device 10 estimates the attribute of the user U based on the specified visit target.

For example, the information providing device 10 identifies a user's visit target based on a search query for searching a movement route from a predetermined departure place to a predetermined arrival place at a predetermined date and time. Then, the information providing device 10 estimates the attributes possessed by the user based on the attributes given in advance to the specified visit target. For example, when the idol "A" performs an event at the date and time # 1 at the place # 1, the information providing device 10 determines that the idol "A" is preferred for the event and the attribute of the user who prefers the idol "A". Correspond. Then, the information providing device 10 estimates that when a certain user U searches for a route to the place # 1 at the date and time # 1, the user U participates in the event of the idle "A", and the user. As the attribute of U, the attribute of preferring the idol "A" and the attribute of the user who prefers the idol "A" are given.

[1-3. About an example of estimation processing]
Hereinafter, an example of the estimation process executed by the information providing device 10 will be described with reference to FIG. For example, the search server 200 accepts the designation of the departure place and the arrival place as a search query for the route search (step S1). More specifically, the search server 200 accepts a search query including a user ID (Identifier) that identifies a station as a departure place, a station as a destination, and a user U. Here, the search server 200 may accept a search query including a departure date and time or an arrival date and time. In such a case, the search server 200 performs a route search based on the content of the search query and provides the search result to the terminal device 100 (step S2). Further, the search server 200 holds a route search log in which the departure place and the arrival place received as the search query are associated with the arrival date and time provided as the search result, that is, the arrival date and time at the arrival place in the search result. ..

On the other hand, the information providing device 10 acquires the route search log as the action schedule information at a predetermined timing (for example, at regular intervals) (step S3). Then, the information providing device 10 executes the following specific processing for each station that is the destination. Further, the information providing device 10 acquires various posted information from the SNS server 300 (step S4).

Subsequently, the information providing device 10 specifies an event that occurs, a place where the event occurs, and a date and time when the event occurs as a visit target (step S5). For example, the information providing device 10 specifies various events, venues, dates and times, etc., based on the posted information acquired from the SNS server 300. It should be noted that such a specific process can be realized by, for example, various character analysis techniques. The visit target may be specified, for example, from news delivered on a news site or the like, or may be specified based on event information managed on a ticket sales site. Further, the identification of the visit target may be manually registered in advance.

For example, in the example shown in FIG. 1, the information providing device 10 has a place "place # 1" where the event to be visited is held, an event "event # 1" to be held, and a date and time "date and time # 1" to be held. To identify. In such a case, the information providing device 10 identifies a nearby station "station D" which is a station located in the vicinity of the place "place # 1". When the location "location # 1" indicates a predetermined facility, the information providing device 10 may specify a station registered in advance as the nearest station of the facility as a nearby station. Then, the information providing device 10 identifies the place "place # 1", the event "event # 1", the date and time "date and time # 1", and the nearby station "station D" as the visit target ID "VTID # 1". Is associated with and held.

In addition, different events may be held at the same facility in the morning and afternoon. Therefore, as described above, the information providing device 10 holds the date and time of each visit in association with each other. However, the embodiments are not limited to this. For example, when the same event is continuously held in a certain facility, the information management device 10 does not have to hold the holding date and time, and may hold only the date when the event is held, for example.

Subsequently, the information providing device 10 analyzes the route search log, identifies the user who visits each visit target, and assigns the attribute corresponding to the attribute of the specified user to each visit target (step). S6). For example, an event in which many users who like the idol "A" will participate may have an attribute corresponding to the idol "A". Therefore, the information providing device 10 estimates the visit target visited by each user U from the route search log of each user U.

For example, the information providing device 10 specifies a visit target to be held in the vicinity of the arrival place set by the user U1 as a visit target to be visited by the user U1 at the arrival date and time set by the user U1. Further, the information providing device 10 identifies a visit target that each user U will visit at each date and time based on the route search log of each user U. Then, the information providing device 10 determines the attribute to be given to the visit target based on the commonality of the attribute given to the user U who will visit a certain visit target.

For example, when the information providing device 10 is given attribute information such as "attribute # 1" to many users U who visit the visit target indicated by the visit target ID "VTID # 1", the visit target ID "VTID" Attribute information "attribute # 1" is given to "# 1". Further, for example, when the information providing device 10 is given attribute information such as "attribute # 2" to many users U who visit the visit target indicated by the visit target ID "VTID # 2", the visit target ID Attribute information "attribute # 2" is given to "VTID # 2".

In the example shown in FIG. 1, the user (for example, user U1) indicated by the user ID "U1" performs a route search with "station D" as the destination at "date and time # 1". In such a case, it is considered that the user U1 is trying to visit "event # 1" which is performed in the vicinity of "station D" at "date and time # 1". Therefore, the information providing device 10 assigns the attribute "attribute # 1" of the user U1 to the "event # 1". For example, the information providing device 10 holds the "attribute # 1" associated with the visit target ID "VTID # 1" of the "event # 1".

Further, in the example shown in FIG. 1, the user (for example, user U2) indicated by the user ID "U2" performs a route search with "station D" as the arrival place at "date and time # 2". In such a case, it is considered that the user U2 is trying to visit "event # 2" which is performed in the vicinity of "station D" at "date and time # 2". Therefore, the information providing device 10 assigns the attribute "attribute # 2" of the user U1 to the "event # 2". For example, the information providing device 10 holds the "attribute # 2" associated with the visit target ID "VTID # 2" of the "event # 2". Further, the information providing device 10 executes the above-described processing for each specified visit target, and assigns an attribute corresponding to the attribute of the user who visits the visit target to each visit target.

Note that the information providing device 10 may not execute step S6 and may add attribute information set by a predetermined operator to each visit target. Further, the information providing device 10 may specify the name of the idol who appears from the name of the event itself to be visited (for example, the name of the event) and give the name of the specified idol as attribute information. .. Further, for example, when the event is the screening of a movie, the information providing device 10 may set the performer, director, series name, etc. of the movie as attributes. That is, if the information providing device 10 assigns the attribute of the visit target itself or the attribute of the user who visits the visit target to the visit target, the information providing device 10 may assign an attribute indicating an arbitrary content by an arbitrary method. ..

Subsequently, the information providing device 10 identifies the visit target to be visited by the user from the route search log of the user to be processed (step S7). For example, the information providing device 10 selects a user to which attribute information is not assigned (for example, user U3) as a user to be processed, and extracts a route search log input by the selected user U3. .. Then, the information providing device 10 estimates the visit target to be visited by the user U by using the route search log.

For example, the information providing device 10 is held at "date and time # 1" in the vicinity of "station D" when the user U3 is performing a route search with "station D" as the destination at "date and time # 1". Search for events. Here, in the example shown in FIG. 1, the event indicated by the visit target ID "VTID # 1" is scheduled to be held at "date and time # 1" in the vicinity of "station D". Therefore, the information providing device 10 specifies that the user U3 visits the visit target indicated by the visit target ID “VTID # 1”. That is, the information providing device 10 specifies the place to be visited by the user U and the date and time to visit the place based on the action schedule information, and is an event associated with the specified place, which is the specified date and time. The event that occurs in is specified as an event in which the user U participates.

Then, when the attribute given to the visit target is not given to the user to be processed, the information providing device 10 gives the attribute corresponding to the specified visit target (step S8). For example, the information providing device 10 assigns the attribute "attribute # 1" given to the visit target indicated by the visit target ID "VTID # 1" to the user U3. After that, the information providing device 10 provides information according to the attribute (step S9). For example, the information providing device 10 provides the user U with various advertising contents selected according to the attributes and recommended information of the transaction target selected according to the attributes.

As described above, the information providing device 10 is, for example, based on the attributes of the user U (that is, the visiting user) who is different from the user U3 and who visits the same visit target as the user U3. , Estimate the attributes of user U3. For example, the information providing device 10 identifies the visit target of the user U1 based on the action schedule information (that is, the route search log) indicating the future action of the user U1, and uses it for the specified visit target. An attribute corresponding to the attribute of the person U1 is given. Then, the information providing device 10 identifies the visit target of the user U3 based on the action schedule information of the user U3, and sets the attributes of the user U3 based on the attributes given to the specified visit target. presume.

More specifically, the information providing device 10 is a search query input by the user U1 and is based on a search query for searching a movement route from a predetermined departure place to a predetermined arrival place at a predetermined date and time. Therefore, the visit target of the user U1 is specified, and the attribute of the user U1 is given to the specified visit target. Then, the information providing device 10 is a search query input by the user U3, and is used based on a search query for searching a movement route from a predetermined departure place to a predetermined arrival place at a predetermined date and time. The visit target of the person U3 is specified, and the attribute given to the specified visit target is given to the user U3.

In this way, the information providing device 10 sets the attributes to be given to the user U, centering on the visit target visited by each user U. As a result, the information providing device 10 imparts the attribute of the user U who performs the act of visiting the visit target, that is, the user U who has a strong motivation for the visit target, to the visit target, and similarly to the visit target. For users with strong motivation, give attributes corresponding to the visit target. Therefore, the information providing device 10 can give each user a more accurate attribute.

[1-4. About attributes given to users]
In the above-mentioned example, the information providing device 10 imparts the attribute given to the visit target to the user U who visits the visit target. However, the embodiments are not limited to this. For example, the information providing device 10 may give the user U different attributes according to the attributes given to the visit target. For example, when the visit target is given the attribute "idol A", the information providing device 10 may give the user U an attribute such as "like idol A".

Further, the information providing device 10 identifies the attributes corresponding to both the attributes given to the user U and the attributes given to the visit target visited by the user U, and uses the specified attributes. It may be newly given to the person U. For example, the information providing device 10 may estimate more detailed attributes than the attributes of the user U, based on the attributes of the user U and the attributes of the visit target.

To give a more specific example, the information providing device 10 is given an attribute of "group B", which is an idle group, to the user U1, and the visit target visited by the user U1 is "group B". When the attribute of "idol C" belonging to "" is given, it is estimated that the user U1 likes "idol C" belonging to "group B", and the attribute of "idol C" is given to the user U1. May be given. That is, the information providing device 10 estimates more detailed attributes than the attributes given to the user U1 according to the attributes of the visit target visited by the user U1, and provides the estimated attributes to the user U1. You may. In other words, the information providing device 10 may refine the attributes of the user U1 based on the attributes of the user U1 and the attributes of the visit target.

Further, the information providing device 10 may estimate the attributes of the user U based on the attributes of a plurality of visited targets. For example, the information providing device 10 may specify a plurality of visit targets visited by the user U1 and estimate the attributes of the user U1 based on the commonality of the plurality of visit targets. For example, the information providing device 10 identifies a plurality of visit targets visited by the user U1, and when the specified plurality of visit targets are given an attribute such as "shrines and temples" in common, " An attribute such as "I like shrines and temples" may be given to the user U1.

Further, for example, in the information providing device 10, when the first event visited by the user U1 is given an attribute such as "idle D" and the second event is given an attribute such as "idle E", " The group to which "idol D" and "idol E" belong (for example, "group C") is specified. Then, the information providing device 10 may give the specified "group C" to the user U1 together with the "idol D" and the "idol E". That is, the information providing device 10 may specify a common item for a plurality of visited objects visited by the user U, and may give the user U an attribute related to the specified item. Here, the common items are not only the attributes that are commonly assigned, but also other attributes that can be inferred from the attributes that are assigned to each visit target (for example, attributes that are higher-level conceptualized for each attribute). , An attribute commonly associated with each attribute) is also included.

Further, the information providing device 10 may estimate the attribute of the user U and the probability that the user U has the attribute based on the visit target. For example, the information providing device 10 holds an attribute for the user U in association with a certainty indicating the certainty that the user U has the attribute. Then, when the information providing device 10 specifies the visit target to be visited by the user U, the value of the accuracy of the attribute corresponding to the specified visit target may be increased by a predetermined value. Further, the information providing device 10 may provide selected information according to the accuracy of each attribute. Further, the information providing device 10 may give the attribute to the user U when the number of visits to the visit target corresponding to a certain attribute exceeds a predetermined threshold value.

Here, the information providing device 10 determines whether or not the user U has visited the specified visit target based on the history of the location information of the user U, and if so, is based on the visit target. , The attribute of the user U may be estimated. For example, the information providing device 10 may increase the accuracy of the attribute corresponding to the visited target actually visited.

[1-5. About the attributes given to the visit target]
Here, when the information providing device 10 assigns an attribute to the visited target, the commonality of the user U who visits the visited target, the number of times each user U has visited, and the accuracy of the attribute of each user U. Attribute may be added based on the above. For example, when the number of users U having a predetermined attribute exceeds a predetermined threshold among the users U who visit the visit target, the information providing device 10 assigns the predetermined attribute to the visit target. May be good.

In addition, the information providing device 10 visits the predetermined attribute when the number of users U whose visit target exceeds the predetermined threshold value among the users U who visit the visit target. May be given to. Further, the information providing device 10 may set a plurality of attributes and the probability that the visited target has each attribute based on the attributes of the user U who visits the visited target. For example, it is assumed that 50 users U having the attribute # 1 visit and 30 users U having the attribute # 2 visit the visit target # 1. In such a case, the information providing device 10 may set the attribute # 1 and its accuracy “50” and the attribute # 2 and its accuracy “30” for the visit target # 1. Then, the information providing device 10 assigns the attribute # 1 and its probability "50" and the attribute # 2 and the attribute corresponding to the probability "30" to the other user U who has visited the visit target # 1. May be good.

Further, the information providing device 10 may assign the predetermined attribute to the visit target when the total accuracy of the predetermined attribute exceeds the predetermined threshold value or when the average value exceeds the predetermined threshold value. .. Further, the information providing device 10 may impart the predetermined attribute to the visit target to the visit target visited by the user having the predetermined attribute a predetermined number of times.

Further, the information providing device 10 may give an attribute to the visited target based on arbitrary information of the user U who visits the visited target. For example, when the user purchases many products related to the idol "X", the information providing device 10 may give an attribute indicating the idol "X" to the visit target. Further, when the user frequently uses the luxury hotel, the information providing device 10 may give an attribute indicating the luxury hotel to the visit target.

Further, for example, when the user frequently inputs a search query related to a predetermined soccer team "Y", the information providing device 10 may assign an attribute indicating the soccer team "Y" to the visit target. Good. Further, the information providing device 10 may give an attribute indicating the restaurant to the visit target when the cause user frequently posts a character string or a photograph relating to the restaurant. In addition, the information providing device 10 identifies the amount of exercise of each user from biological information, and when the amount of exercise exceeds a predetermined threshold value, visits an attribute indicating a user with a large amount of exercise, such as opening a new gym or a marathon event. It may be given to the target.

Even if the information providing device 10 specifies the difference between the information of the user who visits the visit target and the information of the user who does not visit the visit target, and assigns the attribute corresponding to the specified difference to the visit target. Good. For example, the information providing device 10 indicates "fishing" when the information of the user who visits the visited target contains more information about "fishing" than the information of the user who does not visit the visited target. The attribute may be given to the visit target.

It should be noted that various difference extraction techniques for extracting differences in user attributes and various histories can be adopted for such differences in information. For example, the information providing device 10 may purely extract the difference in information, and the tendency of the information of the visiting user and the tendency of the information of another user (hereinafter, referred to as "general user"). Or the tendency of the visiting user's information as seen by the general user may be extracted as the difference. Extraction of such differences is realized by various models such as SVM (Support Vector Machine) and DNN (Deep Neural Network) that have learned the characteristics of information useful for identifying the causative event among user information. You may.

[1-6. About the visit target]
Here, the information providing device 10 may set an arbitrary visit target with an arbitrary particle size. For example, the information providing device 10 may set a visit target regardless of the date and time. More specifically, the information providing device 10 may specify the place visited by the user U based on the action schedule information, and estimate the attribute of the user U based on the attribute corresponding to the place. .. For example, the information providing device 10 may give the attribute of "movie M" to the user U who frequently searches for a route with the location of a certain movie M as the destination.

Further, the information providing device 10 may specify an event in which the user U participates based on the action schedule information, and estimate the user's attribute based on the attribute corresponding to the specified event. Further, the information providing device 10 may specify a place to be visited by the user U based on the action schedule information, and may specify an event associated with the specified place as an event in which the user U participates. For example, the information providing device 10 is an event held at a predetermined place or various places, and corresponds to the event for a user U who visits an event having the same name (for example, a live tour). Attributes may be added.

[1-7. About action schedule information]
In the above-mentioned example, the information providing device 10 uses the search query of the route search as the action schedule information. However, the embodiments are not limited to this. That is, the information providing device 10 can adopt arbitrary information as action schedule information as long as it is information suggesting the location of the user at a certain date and time in the future.

For example, the information providing device 10 specifies a place where the user U visits and a date and time when the user U visits from the contents of an e-mail including various reservation contents such as a hotel reservation and a train reservation made by the user U. .. In such a case, the information providing device 10 may specify the visit target of the user U on the assumption that the user U visits the specified place from the e-mail at the specified date and time.

In addition, the information providing device 10 predicts the future behavior of the user U based on the posted information posted on the Internet by the user U such as SNS, and uses it based on the action schedule information indicating the predicted behavior. The visit target of the person U may be specified. For example, the information providing device 10 specifies the place where the user U visits and the date and time when the user U visits from the posted information, and specifies the visit target as if the user U visits the specified place at the specified date and time. You may. Further, when the terminal device 100 has the function of a contactless IC (Integrated Circuit) card system, the information providing device 10 acquires the ticket information held in the terminal device 100 and uses the acquired ticket information. The place where the person U visits and the date and time when the person U visits the place may be specified. Further, the information providing device 10 may specify the place and the date and time when the user U visits from the credit card usage history and the like.

Here, the information providing device 10 may adopt the information input by the user U as the action schedule information, and is the information input by another user and indicates the future action of the user U. May be adopted as action schedule information. For example, the information providing device 10 predicts congestion at a predetermined place based on the action schedule information input by the user U, and based on the information of the user who has input the action schedule information related to the predetermined place. , You may specify the visit target.

That is, the information providing device 10 identifies the visit target that the user U will visit in the future based on the place, the date and time, etc. that the user U will visit in the future, and assigns the attribute corresponding to the specified visit target to the user. .. By using such action schedule information, for example, even if the position information of the user U cannot be acquired, the attribute of the user U can be estimated by using the result of the route search or the like.

[1-8. About processing considering congestion]
Here, when many users U try to visit the same event, it is considered that congestion occurs at the place where the event occurs or at a nearby station. When a user U visits an event that causes such congestion, it is highly likely that the user U has an attribute corresponding to the event. Therefore, the information providing device 10 may execute the estimation process according to the congestion.

For example, action schedule information indicating the behavior of the user U in the future, such as a route search, can be an index of congestion that will occur in the future. Therefore, the information providing device 10 predicts congestion at a predetermined place based on the action schedule information indicating the future action of the user U. Then, when it is predicted that congestion will occur, the information providing device 10 identifies an event that causes congestion based on the user's information associated with the action schedule information related to the predetermined place, and the event. On the other hand, an attribute corresponding to the attribute of the user associated with the action schedule information related to the predetermined place may be added.

For example, the information providing device 10 predicts that if the number of route search logs with a predetermined station as an arrival point exceeds a predetermined threshold value at a predetermined date and time, congestion will occur at that station. In such a case, the information providing device 10 identifies the user U who has performed a route search with the predetermined station as the arrival place at the predetermined date and time as the cause user.

Subsequently, the information providing device 10 acquires various histories as information on the cause user. For example, the information providing device 10 acquires the history of the search by the cause user using the Internet, the history of the route search, and the like as the search history. Here, the search using the Internet is a concept that includes not only a search of various web pages but also a search within various service domains such as a shopping search for searching a transaction target in an electronic shopping street and a bulletin board search. is there. Further, the information providing device 10 acquires a position history, which is a history of the position information acquired by the terminal device 100 used by the cause user. In addition, the information providing device 10 includes the contents and price of the transaction target purchased by the cause user in the electronic shopping district, the contents and price of the transaction target sold or purchased at the auction, and the transaction target purchased using a credit card or electronic money. Acquire various purchase histories such as the contents and prices of.

Further, the information providing device 10 randomly selects a predetermined number of users as general users from the users U other than the cause user among the users U. In addition, the information providing device 10 acquires various histories of general users as general user information. Then, the information providing device 10 identifies the difference between the information of the cause user and the information of the general user. That is, the information providing device 10 identifies information unique to the cause user.

Then, the information providing device 10 identifies the event related to the specified difference as the cause of the abnormal congestion. For example, the information providing device 10 identifies an event (hereinafter, referred to as “scheduled event”) scheduled to occur at a date and time when abnormal congestion is predicted to occur in the vicinity of a place where abnormal congestion is predicted to occur. To do. Then, the information providing device 10 identifies an event whose relevance to the specified difference exceeds a predetermined threshold value among the specified scheduled events.

For example, in the information providing device 10, when the cause user purchases more products related to the idol "X" than the general user, an event in which the idol "X" appears among the scheduled events, etc. , An event that is highly related to the idol "X" may be specified. In addition, when the cause user uses the luxury hotel more frequently than the general user, the information providing device 10 is related to the luxury hotel, such as a new opening event of the luxury hotel, among the scheduled events. High events may be identified.

Further, for example, in the information providing device 10, if the cause user frequently inputs a search query related to the predetermined soccer team "Y" as compared with the general user, the soccer team "Y" participates. Among the scheduled events such as a match, an event highly related to the soccer team "Y" may be selected. In addition, the information providing device 10 is related to the restaurant among the scheduled events such as the rotation of a new restaurant when the cause user frequently posts a character string or a photo related to the restaurant as compared with the general user. Highly probable events may be selected. In addition, the information providing device 10 identifies the amount of exercise of each user from biological information, and when the cause user has a large amount of exercise compared to a general user, a scheduled event such as opening a new gym or a marathon event. Of these, an event that is highly relevant to a user with a large amount of exercise may be selected.

Then, the information providing device 10 assigns an attribute corresponding to the attribute given to the cause user U to the selected event. For example, the information providing device 10 may assign an attribute commonly assigned to a majority of users U among the cause users U to a selected event.

[1-9. About processing considering abnormal congestion]
Further, the information providing device 10 may execute an estimation process in consideration of abnormal congestion. For example, when sudden congestion occurs due to a certain event, that is, abnormal congestion occurs, the user who causes the abnormal congestion can be regarded as a user who has a strong relationship with the event. it can. Therefore, the information providing device 10 predicts abnormal congestion, and when it is predicted that abnormal congestion will occur, the user who causes the abnormal congestion, that is, the cause utilization, is used for the event that causes the abnormal congestion. Attributes may be added according to the attributes of the person.

For example, the information providing device 10 is based on the difference between the action schedule information indicating the user's behavior at a predetermined place on a predetermined date and time and the action schedule information indicating the average user's behavior at the predetermined place. Congestion at a predetermined place on a predetermined date and time may be predicted. To give a more specific example, the information providing device 10 extracts all the logs whose arrival place is "station D" from the route search logs. Then, the information providing device 10 counts the number of logs including the arrival date and time in the predetermined period (for example, one hour). That is, assuming that each user U acts according to the result of the route search, the information providing device 10 coefficients the number of users who visit the station D for each predetermined period.

Then, the information providing device 10 calculates an average pattern of the number of users visiting the station D based on the counting result. For example, the information providing device 10 calculates the average value of the number of users visiting the station D for each time zone on weekdays, and uses the calculated average value pattern as the averaging pattern. The information providing device 10 also calculates the averaging pattern for each time zone on Saturday and for each time zone on Sunday.

Then, the information providing device 10 searches for a date and time when the number of users deviates from the averaging pattern. For example, the information providing device 10 calculates the number of users who use the station D for each time zone for each day. Then, for example, the information providing device 10 predicts that abnormal congestion will occur when the number of users using the station D in a certain time zone is larger than the number of users indicated by the averaging pattern by a predetermined threshold value or more. ..

That is, the information providing device 10 is based on the difference between the action schedule information indicating the user's behavior at a predetermined place on a predetermined date and time and the action schedule information indicating the average user's behavior at the predetermined place. Congestion at a predetermined place on a predetermined date and time may be predicted. More specifically, the information providing device 10 has an average number of action schedule information indicating the user's behavior at a predetermined place and an action schedule information indicating the user's behavior at a predetermined place on a predetermined date and time. Congestion of a predetermined place at a predetermined date and time may be predicted based on the comparison result with the number.

[1-10. Regarding the provision of information]
Further, the information providing device 10 may provide the user U with arbitrary information according to the specified attribute. For example, the information providing device 10 indicates that the user U to which the attribute indicating the idol "X" is given has a live performance of the idol "X", goods and live information of the idol "X", and the idol "X". Information such as the location and time of abnormal congestion caused by the live performance may be provided.

[2. Configuration of information providing device]
Hereinafter, an example of the functional configuration of the information providing device 10 that realizes the above-mentioned learning process will be described. FIG. 2 is a diagram showing a configuration example of the information providing device 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) or the like. Then, the communication unit 20 is connected to the network N by wire or wirelessly, and transmits / receives information to / from the terminal device 100 and the search server 200.

The storage unit 30 is realized by, for example, a semiconductor memory element such as a RAM (Random Access Memory) or a flash memory (Flash Memory), or a storage device such as a hard disk or an optical disk. In addition, the storage unit 30 stores the user database 31, the route search log database 32, and the visited database 33 (hereinafter, may be referred to as “each database 31 to 33”).

Hereinafter, an example of information registered in each database 31 to 33 will be described with reference to FIGS. 3 to 5. First, an example of information registered in the user database 31 will be described with reference to FIG. FIG. 3 is a diagram showing an example of information registered in the user database according to the embodiment. As shown in FIG. 3, the user database 31 has items such as "user ID", "attribute information", "search history", "purchase history", "location history", and "biological information log". Information is registered. In addition to the information shown in FIG. 3, arbitrary information regarding the user U (for example, history of posted information) may be registered in the user database 31.

Here, the "user ID" is an identifier for identifying the user U. Further, the "attribute information" is information indicating various attributes such as the name, gender, age, hobbies and tastes of the user U, and the place of residence. Further, the "search history" is a history of search queries input by the user U, and may include not only the history of search queries for searches using the Internet but also the history of search queries for route searches. .. The "purchase history" is a history of purchases made by the user U through an electronic shopping street or an auction. Further, the "position history" is a history of position information acquired by the terminal device 100 used by the user U. The "biological information log" is a history of biometric information of the user U.

For example, in the example shown in FIG. 3, in the user database 31, the user ID "U1", the attribute information "attribute information # U1", the search history "search history # U1", the purchase history "purchase history # U1", The position history "position history # U1" and the biometric information log "biological information # U1" are registered in association with each other. In such information, the attribute information of the user U1 indicated by the user ID "U1" is "attribute information # U1", the search history is "search history # U1", and the purchase history is "purchase history # U1". , The position history is "position history # U1", and the history of biometric information is "biological information # U1".

In the example shown in FIG. 3, conceptual values such as "attribute information # U1" are described, but in reality, the user database 31 contains the attributes of the user U, the history of search queries, and purchases. Character strings, numerical values, etc. indicating history, position history, biometric information history, etc. will be registered.

Next, an example of the information registered in the route search log database 32 will be described with reference to FIG. FIG. 4 is a diagram showing an example of information registered in the route search log database according to the embodiment. As shown in FIG. 4, the route search log database 32 contains the "log ID", "departure date and time", "departure place", "arrival date and time", "arrival place", "user ID", and "search date and time". Information having an item such as "" is registered. In addition to the information shown in FIG. 4, various information related to the route search may be registered in the route search log database 32.

Here, the "log ID" is an identifier that identifies the route search log. The "departure date and time" is the date and time set as the departure date and time in the route search result. The "departure place" is a departure place designated by the user U in the route search. The "arrival date and time" is the date and time set as the arrival date and time in the route search result. Further, the "arrival place" is the arrival place designated by the user U in the route search. The "user ID" is an identifier of the user U who has performed a route search. The "search date and time" is information indicating the date and time when the route search was performed.

For example, in the example shown in FIG. 4, the route search log database 32 contains the log ID "Q1", the departure date and time "2017/9/12/8: 00", the departure place "station E", and the arrival date and time "2017/9". Information such as "/ 12/9: 00", the arrival place "Station D", the user ID "U1", and the search date and time "2017/8/31/10: 00" is registered. Such information is the route from the departure place "station E" to the arrival place "station D" by the user U1 indicated by the user ID "U1" at the search date and time "2017/8/31/10: 00". The search is performed, and it is shown that the departure date and time in the search result of the route search is "2017/9/12/8: 00" and the arrival date and time is "2017/9/12/9: 00".

Next, an example of the information registered in the visited database 33 will be described with reference to FIG. FIG. 5 is a diagram showing an example of information registered in the visited database according to the embodiment. As shown in FIG. 5, information having items such as "visit target ID", "location", "neighborhood station", "event", "date and time", and "attribute information" is registered in the visit target database 33. Will be done. In addition to the information shown in FIG. 5, various information related to the visited target may be registered in the visited target database 33.

Here, the "visit target ID" is an identifier that identifies the visit target. Further, the "location" is information indicating a location where an event occurs. Further, the "neighborhood station" is information indicating a station located in the vicinity of the place where the event occurs. The "date and time of the event" is information indicating the date and time when the event is performed. Further, the "attribute information" is information indicating the attributes given to the visit target.

For example, in the example shown in FIG. 5, the visit target database 33 contains the visit target ID “VTID # 1”, the place “place # 1”, the nearby station “station D”, the event “event # 1”, and the date and time “date and time”. "# 1" and the attribute information "N / A" are registered in association with each other. Such information indicates that the event "event # 1" is held at the place "place # 1" at the date and time "date and time # 1" as the visit target indicated by the visit target ID "VTID # 1". Further, such information indicates that a nearby station "station D" exists in the vicinity of the place "place # 1". In addition, such information indicates that the attribute has not yet been registered for the visit target indicated by the visit target ID “VTID # 1”.

In the example shown in FIG. 5, conceptual values such as "place # 1", "event # 1", "date and time # 1", and "attribute # 2" are described in the visited database 33. Actually, a character string indicating the place where the event is held, a character string indicating the event name, the date and time when the event is held, an attribute, and the like are registered.

Returning to FIG. 2, the explanation will be continued. The control unit 40 is a controller, and for example, various programs stored in a storage device inside the information providing device 10 by a processor such as a CPU (Central Processing Unit) or an MPU (Micro Processing Unit) are stored in a RAM or the like. Is realized by executing as a work area. Further, the control unit 40 is a controller, and may be realized by, for example, an integrated circuit such as an ASIC (Application Specific Integrated Circuit) or an FPGA (Field Programmable Gate Array).

As shown in FIG. 2, the control unit 40 includes an acquisition unit 41, a prediction unit 42, a grant unit 43, a specific unit 44, an estimation unit 45, and a providing unit 46. The acquisition unit 41 acquires action schedule information indicating the future behavior of the user. For example, the acquisition unit 41 acquires the route search log including the route search search query input by the user U from the search server 200 as the action schedule information. Then, the acquisition unit 41 registers the acquired route search log in the route search log database 32.

In addition, the acquisition unit 41 acquires a log of posted information from the SNS server 300. Then, the acquisition unit 41 identifies the place to be visited, the event, the date and time when the event occurs, and the like by using various character analysis techniques, and registers the information about the specified visit target in the visit target database 33. In addition, the visit target information input by a predetermined administrator may be registered in the visit target database 33.

The prediction unit 42 predicts congestion at a predetermined location based on the action schedule information. For example, the prediction unit 42 predicts congestion at a predetermined location based on the action schedule information input by the user U. Further, for example, the prediction unit 42 predicts congestion at a predetermined place based on a search query for searching a movement route from a predetermined departure place to a predetermined arrival place at a predetermined date and time as action schedule information. .. More specifically, the prediction unit 42 predicts congestion at a predetermined destination included in the search query based on the search query.

For example, the prediction unit 42 determines, based on the difference between the action schedule information indicating the user's behavior at a predetermined place on a predetermined date and time and the action schedule information indicating the average user's behavior at the predetermined place. Predict congestion at a given date and time at the location of. To give a more specific example, the prediction unit 42 has an average number of action schedule information indicating the user's behavior at a predetermined place and an action schedule indicating the user's behavior at a predetermined place on a predetermined date and time. Based on the comparison result with the number of information, the congestion of a predetermined place at a predetermined date and time is predicted. For example, the prediction unit 42 determines that the number of action schedule information indicating the user's behavior at a predetermined place on a predetermined date and time is larger than the average number of action schedule information indicating the user's behavior at the predetermined place. If it is more than the threshold value, it is predicted that congestion will occur at a predetermined place at a predetermined date and time.

For example, the prediction unit 42 selects a station to be predicted for congestion, and extracts a route search log with the selected station as the destination from the route search log database 32. Then, the prediction unit 42 predicts the congestion at the selected station by using the extracted log. For example, the prediction unit 42 estimates the average number of users in each time zone of the selected station using the extracted log. For example, the prediction unit 42 counts the number of logs whose arrival date and time are included in that time zone for each time zone, and counts the number of counted logs from the oldest arrival date and time included in the extracted log to the latest arrival date and time. Estimate the average number of users in each time zone of the selected station by dividing by the number of days. Then, the prediction unit 42 generates an averaging pattern based on the estimated number of users.

In addition, the prediction unit 42 counts the number of logs whose arrival date and time are included in each time zone of each day, and the number of users in each time zone in the averaging pattern and the counted logs. Compare with numbers. Then, when the number of logs in any time zone of any date and time is larger than the number of users in that time zone in the averaging pattern by a predetermined threshold value or more, the prediction unit 42 determines that date and time. It is estimated that abnormal congestion will occur during the time period. That is, the prediction unit 42 determines the average number of action schedule information indicating the behavior of the user in the predetermined time zone at the predetermined location and the behavior of the user in the predetermined time zone at the predetermined date and time in the predetermined location. Based on the comparison result with the number of action schedule information shown, the congestion of a predetermined place at a predetermined date and time is predicted. The prediction unit 42 predicts the future behavior of the user U based on the posted information posted by the user U, and predicts the congestion at a predetermined place based on the action schedule information indicating the predicted behavior. You may.

The granting unit 43 identifies the visit target of the user U1 based on the action schedule information indicating the future action of the first user (for example, the user U1), and the user U1 refers to the specified visit target. Assign attributes according to the attributes. For example, the granting unit 43 is a search query input by the user U1 and is based on a search query for searching a movement route from a predetermined departure place to a predetermined arrival place at a predetermined date and time. Identify what to visit. Then, the granting unit 43 assigns the attribute of the user U1 to the specified visit target.

For example, when it is predicted that congestion will occur at a predetermined place, the granting unit 43 identifies an event that causes congestion based on the user's information associated with the action schedule information related to the predetermined place. .. More specifically, the granting unit 43 identifies an event that causes congestion based on the information of the user who has input the action schedule information related to the predetermined place. For example, when the action schedule information includes a search query for a route search, the granting unit 43 identifies an event that causes congestion at the destination indicated by the search query. For example, the granting unit 43 may use the history of searches by the user U using the Internet, the history of electronic commerce by the user U, the history of the position of the user U, the attributes of the user U, or the biometric information of the user U. Identify the event that causes congestion based on at least one.

More specifically, when it is predicted that congestion will occur at a predetermined place at a predetermined date and time, the granting unit 43 causes a user who is predicted to visit the predetermined place at a predetermined date and time based on the action schedule information. Estimate as a user and identify the event that causes congestion based on the estimated cause user information. Here, the granting unit 43 may identify an event that causes congestion based on the commonality of the information of the cause user, and the information of the cause user and another user U (that is, general use) may be specified. The event that causes congestion may be identified based on the difference from the information of the person).

For example, when it is predicted that abnormal congestion will occur at station D between about 6:00 am and about 15:00 pm on December 12, 2017, the granting unit 43 will refer to the route search log database 32. Identify the log in which the arrival date and time is included in each time zone from about 6:00 am to about 15:00 pm on December 12, 2017, and the arrival place is "Station D". Subsequently, the granting unit 43 identifies the user ID included in the specified log as the user ID indicating the cause user. That is, the granting unit 43 identifies the user U who is expected to visit the station D between about 6:00 am and about 15:00 pm on December 12, 2017 as the causative user.

Subsequently, the granting unit 43 randomly selects a predetermined number of users U from users other than the cause user, and makes the selected user U a general user. Then, the granting unit 43 acquires the information of the cause user and the information of the general user from the user database 31, and the difference between the acquired information, that is, the information of the cause user and the information of the general user. Identify the difference.

In addition, the granting unit 43 refers to the visited database 33 and extracts events to be held in the vicinity of the station D after 6:00 am on December 12, 2017. In addition, even if the granting unit 43 extracts the events held in the vicinity of the station D from about 6:00 am to about 15:00 pm on December 12, 2017, when abnormal congestion is predicted to occur. Good. In addition, the granting unit 43 is held near the station D from about 6:00 am on December 12, 2017, which is the first time zone when abnormal congestion is predicted to occur, until a predetermined period elapses. Events may be extracted, and in the vicinity of station D from around 15:00 pm on December 12, 2017, which is the last time zone when abnormal congestion is predicted to occur, until a predetermined period elapses. The event to be held may be extracted.

Then, the granting unit 43 calculates the degree of relevance between the difference between the information of the cause user and the information of the general user and the correspondence information of the extracted event, and determines the event in which the calculated relevance exceeds a predetermined threshold value. Identify. For example, when there is a tendency for the causative user to have a predetermined hobby / preference as compared with the general user, the granting unit 43 identifies an event having the highest relevance to the hobby / preference.

Then, the granting unit 43 assigns an attribute corresponding to the attribute of the cause user to the specified event. For example, the granting unit 43 associates the identified event with the attribute information indicating the attribute corresponding to the attribute of the cause user and registers it in the visit target database 33.

In addition, the granting unit 43 may specify an event that causes congestion based on the information acquired within a predetermined period from a predetermined date and time among the information of the cause user. Further, the granting unit 43 may extract information that deviates from the average pattern from the information of the causative user, and identify an event that causes congestion based on the extracted information. Further, the granting unit 43 searches for an event that is highly related to the information of the causative user from the events scheduled to occur at a predetermined date and time in the vicinity of the predetermined place, and the searched event may be used as the cause of congestion. Good.

The identification unit 44 identifies the visit target of the user based on the action schedule information indicating the future action of the user. For example, the specific unit 44 is a search query input by the user U, and is based on a search query for searching a movement route from a predetermined departure place to a predetermined arrival place at a predetermined date and time. Identify the places and events that you visit. More specifically, the specific unit 44 specifies a place visited by the user U and a date and time when the user U visits the place, and is an event associated with the specified place, which occurs at the specified date and time. Is specified as an event in which the user U participates.

For example, the specific unit 44 refers to the user database 31 and selects the user U to be processed. Subsequently, the specific unit 44 extracts the log of the search query input by the selected user U from the route search log database 32. Then, the identification unit 44 identifies the visit target that the user U will visit in the future from the visit target database 33 based on the extracted search query log.

For example, the specific unit 44 refers to the arrival date and time and the arrival place of each extracted log, and sets the referenced arrival place as a nearby station, and visits the event held at the referenced arrival date and time. Specify all from 33. That is, the specific unit 44 specifies a plurality of visit targets visited by the user U. Then, the specific unit 44 outputs the visit target ID of the specified visit target to the estimation unit 45.

The estimation unit 45 estimates the attributes of the user U based on the visit target specified by the specific unit 44. More specifically, the estimation unit 45 estimates the attribute of the user U based on the attribute corresponding to the place or event specified as the visit target by the specific unit 44. For example, the estimation unit 45 refers to all the attribute information associated with the visit target ID received from the specific unit 44, and estimates the attributes of the user U based on the commonality of the referenced attribute information. To give a more specific example, when the estimation unit 44 includes the attribute information indicating the attribute common to the majority of the attribute information among the referenced plurality of attribute information, the user uses the attribute information indicating the attribute information. Estimated as an attribute of U. Then, the estimation unit 44 associates the user U with the estimated attribute and registers it in the user database 31.

The estimation unit 45 may estimate the attribute of the user U and the probability that the user U has the attribute based on the visit target visited by the user U. Further, when the user U visits the specified visit target based on the history of the location information of the user U, the estimation unit 45 may estimate the user's attributes based on the visit target. Good. Further, the estimation unit 45 sets the attributes more detailed than the attributes given to the user U based on the attributes given to the user U to be processed and the specified visit target. It may be estimated as a new attribute of.

Here, by the processing by the prediction unit 42 and the granting unit 43, attributes corresponding to the attributes of the user U who visits the visiting target are assigned to each visiting target. Therefore, the estimation unit 45 estimates the attributes of the user U based on the attributes of the visiting user who is a different user from the user U and visits the specified visit target. In addition, when the predetermined operator sets the attribute for each visit target in advance, the estimation unit 45 estimates the attribute of the user U based on the attribute.

The providing unit 46 provides the user U with information according to the estimated attributes. For example, the providing unit 46 refers to the user database 31 and identifies the attributes assigned to each user U. Then, the providing unit 46 selects the information corresponding to the specified attribute and distributes the selected information to the terminal device 100 of each user U. The providing unit 46 corresponds to a server device that distributes advertisement contents according to the attributes of the user U, a server device that distributes various contents such as news according to the attributes of the user U, and the attributes of the user U. The attribute information registered in the user database 31 may be provided to the recommendation server or the like that recommends the transaction target, and when the user ID is received from these server devices, it is associated with the received user ID. The specified attribute information may be provided.

In the above-mentioned example, the granting unit 43 executes a process of specifying the visit target of the user U1 based on the action schedule information of the user U1. For example, in the above-described example, the grant unit 43 is based on a search query input by the user U1 for searching a movement route from a predetermined departure place to a predetermined arrival place at a predetermined date and time. Therefore, the visit target of the user U1 was identified. Further, in the above-mentioned example, when the prediction unit 42 predicts that congestion will occur at a predetermined place, the granting unit 43 associates it with the action schedule information related to the predetermined place where congestion is predicted to occur. Based on the user's information, the events that cause congestion were identified. However, the embodiments are not limited to this.

For example, a process for identifying the visit target of the user U1, a process for identifying an event that causes congestion when congestion is predicted to occur, a process for identifying the user U associated with the specified event, and the like. In order to assign an attribute to the visit target, various specific processes for specifying the visit target and the user U associated with the visit target may be executed by the specific unit 44. That is, the granting unit 43 may execute only the process of assigning attributes to various visit targets according to the attributes of the user U specified as the user U associated with the visit target.

Here, FIG. 6 is a diagram illustrating a flow of estimation processing executed by the information providing device according to the embodiment. For example, in the example shown in FIG. 6, as shown in FIG. 6 (A), the event in which the cause user to which the attribute "idle # 1" is commonly assigned is held at the facility # 1 on the date and time # 1. It is assumed that a route search has been performed in order to visit. In such a case, the information providing device 10 identifies the common attribute "idle # 1" commonly given to the cause user who visits the facility # 1. Then, as shown in FIG. 6B, the information providing device 10 assigns "idle # 1" as an attribute of the facility # 1 at the date and time # 1.

Subsequently, the information providing device 10 identifies the user U who is scheduled to visit the facility # 1 at the date and time # 1, as shown in FIG. 6C. It is presumed that such a user U is interested in "idol # 1" as well as the causative user. Therefore, the information providing device 10 assigns the attribute "idle # 1" given to the facility # 1 to the user U who is scheduled to visit the facility # 1 at the date and time # 1.

Further, when the attribute is given to the user U in this way, the same processing is performed for other facilities (for example, the goods shop of "Idol # 1") that the user U frequently visits. Allows attributes such as "idle # 1" to be propagated. As a result, as shown in FIG. 6D, the information providing device 10 propagates the attribute information of each user around the visit target such as a facility, place, event, etc. The system can be realized.

[3. Information providing device processing flow]
Next, an example of the procedure of the estimation process executed by the information providing device 10 will be described with reference to FIG. 7. FIG. 7 is a flowchart showing an example of a flow of estimation processing executed by the information providing device according to the embodiment.

First, the information providing device 10 acquires a route search log including a destination corresponding to a predetermined visit event (step S101). Subsequently, the information providing device 10 assigns an attribute to the visited event based on the common attribute of the user who visits the visited event at a predetermined date and time (step S102). That is, the information providing device 10 sets the attributes of the visit target at a predetermined date and time. Subsequently, the information providing device 10 acquires the route search log of the user to be processed (step S103). Then, the information providing device 10 identifies the visit target to be visited by the user (step S104), and estimates the attribute of the user to be processed based on the attribute given to the specified visit target (step S105). ). That is, the information providing device 10 is an attribute given to the visit target visited by the user, and estimates the user attribute based on the attribute at the date and time when the user visits the visit target. Then, the information providing device 10 gives the estimated attribute to the user (step S106), and ends the process.

[4. Modification example]
In the above, an example of the estimation process by the information providing device 10 has been described. However, the embodiments are not limited to this. Hereinafter, variations of the estimation process executed by the information providing device 10 will be described.

[4-1. Handling of date and time]
In the above example, the information providing device 10 uses the arrival date and time obtained as a result of the route search to predict the date and time when congestion will occur. However, the embodiments are not limited to this. For example, the information providing device 10 may predict the date and time when congestion will occur by using the arrival date and time input by the user U as a search query, and set the attribute at a predetermined date and time of the event visited by the user U. You may.

Further, the information providing device 10 inputs the distance from the nearby station to the venue of the event and the date and time when the user U will arrive at that station (for example, the arrival date and time obtained as a result of the route search or a search query). The date and time when the user U is expected to arrive at the venue may be estimated based on the date and time of arrival, and the attributes may be set at a predetermined date and time of the visit target based on the estimated date and time. That is, the information providing device 10 sets the attribute of a predetermined date and time for the visit target based on not only the date and time information included in the route search result but also the arrival date and time of the user U estimated from this information. You may predict the date and time when congestion will occur.

[4-2. Device configuration〕
The information providing device 10 may be communicably connected to an arbitrary number of terminal devices 100, or may be communicably connected to an arbitrary number of search servers 200 and SNS servers 300. Further, the information providing device 10 may be able to communicate with an e-mail distribution server or the like. Further, the information providing device 10 may be able to communicate with a distribution server that distributes various contents according to the attributes of the user U.

Further, the information providing device 10 may be realized by a front-end server and a back-end server that executes various processes. In such a case, the providing unit 46 shown in FIG. 2 is arranged on the front-end server, and the acquisition unit 41, the prediction unit 42, the granting unit 43, the specific unit 44, and the estimation unit shown in FIG. 2 are arranged on the back-end server. Part 45 will be included. Further, each of the databases 31 to 33 registered in the storage unit 30 may be managed by an external storage server.

[4-3. Others]
Further, among the processes described in the above-described embodiment, all or a part of the processes described as being automatically performed can be manually performed, or the processes described as being manually performed can be performed. All or part of it can be done automatically by a known method. In addition, the processing procedure, specific name, 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 figure is not limited to the illustrated information.

Further, each component of each of the illustrated devices is a functional concept, and does not necessarily have to be physically configured as shown in the figure. That is, the specific form of distribution / integration of each device is not limited to the one shown in the figure, and all or part of them may be functionally or physically distributed / physically in arbitrary units according to various loads and usage conditions. It can be integrated and configured.

In addition, the above-described embodiments can be appropriately combined as long as the processing contents do not contradict each other.

[5. program〕
Further, the information providing device 10 according to the above-described embodiment is realized by, for example, a computer 1000 having a configuration as shown in FIG. FIG. 8 is a diagram showing an example of a hardware configuration. The computer 1000 is connected to the output device 1010 and the input device 1020, and the arithmetic unit 1030, the primary storage device 1040, the secondary storage device 1050, the output IF (Interface) 1060, the input IF 1070, and the network IF 1080 are connected by the bus 1090. Has.

The arithmetic unit 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 that temporarily stores data used by the arithmetic unit 1030 for various calculations, such as a RAM. Further, the secondary storage device 1050 is a storage device in which data used by the calculation device 1030 for various calculations and various databases are registered, such as a ROM (Read Only Memory), an 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 information such as a monitor and a printer. For example, USB (Universal Serial Bus), DVI (Digital Visual Interface), and the like. It is realized by a standard connector such as HDMI (registered trademark) (High Definition Multimedia Interface). Further, 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, USB.

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

The network IF1080 receives data from another device via the network N and sends it to the arithmetic unit 1030, and also transmits the data generated by the arithmetic unit 1030 to the other device via the network N.

The arithmetic unit 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 unit 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 unit 1030 of the computer 1000 executes the program or data (for example, the processing model M1) loaded on the primary storage device 1040, so that the control unit 40 To realize the function of. The arithmetic unit 1030 of the computer 1000 reads and executes these programs or data (for example, the processing model M1) from the primary storage device 1040, but as another example, these programs are transmitted from another device via the network N. You may get it.

[6. effect〕
As described above, the information providing device 10 identifies the visit target of the user U based on the action schedule information indicating the future action of the user U. Then, the information providing device 10 estimates the attribute of the user U based on the specified visit target. Therefore, the information providing device 10 gives the user U an attribute according to the visit target that is presumed to be more strongly interested in the user U, that is, the visit target that is likely to lead to an action. Can be granted. As a result, the information providing device 10 can improve the estimation accuracy of the user's attributes.

Further, the information providing device 10 specifies a place to be visited by the user U based on the action schedule information, and estimates the attribute of the user U based on the attribute corresponding to the place. Further, the information providing device 10 identifies an event in which the user U participates based on the action schedule information, and estimates the attribute of the user U based on the attribute corresponding to the event. Further, the information providing device 10 specifies a place to be visited by the user U based on the action schedule information, and identifies an event associated with the specified place as an event in which the user U participates. Therefore, the information providing device 10 can estimate the attribute according to the visit target that is likely to lead to the action as the attribute of the user U.

Further, the information providing device 10 specifies a place to be visited by the user U and a date and time to visit the place based on the action schedule information, and is an event associated with the specified place, which is a specified date and time. The event that occurs in is specified as an event in which the user U participates. Therefore, the information providing device 10 can be used even when the event visited by the user U is performed for a limited period of time, or when different events are held at different dates and times in the same facility. The attributes of person U can be estimated appropriately.

Further, the information providing device 10 identifies a plurality of visit targets visited by the user U, and estimates the attributes of the user U based on the commonality of the plurality of visited targets. Therefore, the information providing device 10 can further improve the estimation accuracy of the attribute of the user U.

Further, the information providing device 10 estimates the attribute of the user U and the probability that the user U has the attribute based on the visit target. Further, when the user U visits a visit target specified by a specific unit based on the history of the location information of the user U, the information providing device 10 uses the visit target as the user U. Estimate the attributes of. Therefore, the information providing device 10 can more appropriately estimate the attributes of the user U.

Further, the information providing device 10 estimates an attribute more detailed than the attribute possessed by the user U based on the attribute possessed by the user U and the visit target. Therefore, the information providing device 10 can estimate more detailed attributes possessed by the user U.

Further, the information providing device 10 is a user U different from the user U, and estimates the attributes of the user U based on the attributes of the visiting user U who visits the visited target specified by the specific unit. .. For example, the information providing device 10 assigns an attribute corresponding to the attribute of the first user to the visit target of the first user specified based on the action schedule information indicating the future action of the first user. To do. Then, the information providing device 10 identifies the visit target of the second user based on the action schedule information of the second user, and based on the attributes given to the visit target of the second user, Estimate the attributes of the second user. Therefore, the information providing device 10 can appropriately estimate the attributes of the second user.

Further, the information providing device 10 predicts congestion at a predetermined place based on the action schedule information indicating the future action of the user U, and when it is predicted that congestion will occur, the action related to the predetermined place. Based on the information of the user U associated with the schedule information, the event causing the congestion is identified, and the event is responded to the attribute of the user U associated with the action schedule information associated with the predetermined place. Give attributes. Therefore, the information providing device 10 can give more appropriate attributes to the visited target.

Further, the information providing device 10 determines the difference between the action schedule information indicating the behavior of the user U at a predetermined date and time at a predetermined location and the action schedule information indicating the average behavior of the user U at the predetermined location. Based on this, the congestion at a predetermined place on a predetermined date and time is predicted. Therefore, the information providing device 10 can impart more appropriate attributes to an event that causes abnormal congestion that occurs suddenly.

Further, the information providing device 10 estimates the attribute of the user U based on the attribute set in advance for the specified visit target. Therefore, the information providing device 10 can maintain the estimation accuracy of the attribute of the user U.

Further, the information providing device 10 identifies the visit target of the user U based on a search query for searching a movement route from a predetermined departure place to a predetermined arrival place at a predetermined date and time as action schedule information. .. Therefore, the information providing device 10 can appropriately estimate the attributes of the user U even when the position information and the like cannot be collected.

Further, the information providing device 10 provides information according to the estimated attributes. Therefore, the information providing device 10 can realize information distribution according to the attribute of the user U.

Although some of the embodiments of the present application have been described in detail with reference to the drawings, these are examples, and various modifications are made based on the knowledge of those skilled in the art, including the embodiments described in the disclosure column of the invention. It is possible to practice the present invention in other improved forms.

In addition, the above-mentioned "section, module, unit" can be read as "means" or "circuit". For example, the estimation unit can be read as an estimation means or an estimation circuit.

10 Information provider 20 Communication unit 30 Storage unit 31 User database 32 Route search log database 33 Visit target database 40 Control unit 41 Acquisition unit 42 Prediction unit 43 Granting unit 44 Specific unit 45 Estimating unit 46 Providing unit 100 Terminal device 200 Search server 300 SNS server

Claims (11)

Based on the action schedule information indicating the future behavior of a plurality of first users, the place to be visited by each first user is specified at a predetermined date and time, and based on the number of first users visiting each place. , A prediction unit that predicts congestion at each location,
When congestion at a predetermined place is predicted at the predetermined date and time, the first user who visits the predetermined place at the predetermined date and time is specified, and the history of actions on the network of the specified first user. An event whose relationship with is satisfied with a predetermined date and time is specified as an event that causes congestion at the predetermined place at the predetermined date and time, and the specified event is specified at the predetermined date and time. A granting unit that assigns attributes according to the attributes registered in a predetermined storage device in advance in association with the first user who visits the place .
Based on the search query entered by the second user for searching the movement route from the predetermined departure point to the predetermined arrival point, the place to be visited by the second user is specified . A specific part that identifies an event that occurs in the vicinity of the specified location as an event in which the second user participates ,
An estimation device including an estimation unit that estimates the attributes of the second user based on the attributes given to the event specified by the specific unit. Based on the search query , the specific unit identifies a place to be visited by the second user, and identifies an event that causes congestion at the specified place as an event in which the second user participates. The estimation device according to claim 1. The specific unit includes a place to be visited by the second user and a date and time to visit the place based on a search query for searching a movement route from a predetermined departure place to a predetermined arrival place at a predetermined travel date and time. The invention according to claim 1 or 2, wherein the event that causes congestion at the specified place and occurs at the specified date and time is specified as an event in which the second user participates. Estimator. The specific unit identifies a plurality of events in which the second user participates,
The estimation unit is any one of claims 1 to 3, characterized in that the estimation unit estimates the attributes of the second user based on the commonality of the attributes given to the plurality of events. The estimation device described in. The claim is characterized in that the estimation unit estimates the attribute of the second user and the probability that the second user has the attribute based on the attribute given to the event. The estimation device according to any one of 1 to 4. The estimating unit, based on the history of the position information of the second user, if the second user is estimated to have participated in the event specified by the specifying unit, assigned to the event The estimation device according to any one of claims 1 to 5, wherein the attributes of the second user are estimated based on the specified attributes. The estimation unit is characterized in that an attribute more detailed than the attribute possessed by the second user is estimated based on the attribute possessed by the second user and the attribute given to the event. The estimation device according to any one of claims 1 to 6. The prediction unit, the act on schedule information, the number of users visiting the place on the date and time of Jo Tokoro, based on the difference between the number of average user to visit the place estimating apparatus according to any one of claims 1 to 7, characterized in that predicting the congestion of the predetermined time in the predetermined location. The estimation device according to any one of claims 1 to 8 , further comprising a providing unit that provides information according to the attributes estimated by the estimation unit. A specific method performed by a specific device
Based on the action schedule information indicating the future behavior of a plurality of first users, the place to be visited by each first user is specified at a predetermined date and time, and based on the number of first users visiting each place. , Prediction process for predicting congestion at each location,
When congestion at a predetermined place is predicted at the predetermined date and time, the first user who visits the predetermined place at the predetermined date and time is specified, and the history of actions on the network of the specified first user. An event whose relationship with is satisfied with a predetermined date and time is specified as an event that causes congestion at the predetermined place at the predetermined date and time, and the specified event is identified at the predetermined date and time. A granting process of assigning attributes according to the attributes registered in a predetermined storage device in advance in association with the first user who visits the place, and
Based on the search query entered by the second user for searching the travel route from the predetermined departure point to the predetermined arrival point, the place visited by the second user is specified . A specific process that identifies an event that occurs in the vicinity of the specified location as an event in which the second user participates , and
An estimation method including an estimation step of estimating the attributes of the second user based on the attributes given to the event specified by the specific process. Based on the action schedule information indicating the future behavior of a plurality of first users, the place to be visited by each first user is specified at a predetermined date and time, and based on the number of first users visiting each place. , Prediction procedure for predicting congestion at each location,
When congestion at a predetermined place is predicted at the predetermined date and time, the first user who visits the predetermined place at the predetermined date and time is specified, and the history of actions on the network of the specified first user. An event whose relationship with is satisfied with a predetermined date and time is specified as an event that causes congestion at the predetermined place at the predetermined date and time, and the specified event is specified at the predetermined date and time. A granting procedure for assigning attributes according to the attributes registered in a predetermined storage device in advance in association with the first user who visits the place, and
Based on the search query entered by the second user for searching the movement route from the predetermined departure point to the predetermined arrival point, the place to be visited by the second user is specified . A specific procedure for identifying an event that occurs in the vicinity of the specified location as an event in which the second user participates , and
An estimation program for causing a computer to execute an estimation procedure for estimating the attributes of the second user based on the attributes given to the event specified by the specific procedure.

Images