JP2018077737A - Information processing apparatus, information processing system, information processing method and information processing program - Google Patents

Abstract

PROBLEM TO BE SOLVED: To evaluate satisfaction of a user who uses a store.SOLUTION: A service server device 2 has: a feature quantity calculation part 23 which calculates feature quantity for every evaluation item about a state based on sensing data acquired by sensing of the state in a store; and an index calculation part 24 which calculates an index showing a result obtained by presuming subjective sensation of a user about utilization of the store based on a difference between the feature quantity calculated about a state when the user visits the store and a model constructed from learning data accumulated according to store utilization of the user.SELECTED DRAWING: Figure 2

Description

  The present invention relates to an information processing apparatus, an information processing system, an information processing method, and an information processing program.

  There is a known service for issuing coupons that include benefits such as discounts in the provision of food and drinks at information sites that provide information on stores such as restaurants, etc. Yes. With this service, it is possible to promote the use of a store where information is posted, and a user who obtains a coupon and comes to the store can receive a privilege. In order to promote the visit by the use of a coupon, a technology of a coupon providing server that can set the contents of a privilege according to the user's preference has been proposed.

  In addition, a store may grasp the degree of satisfaction of the user due to the use of the store and apply it to the service so that the user who visits the store with the access to the information site repeatedly uses the store in the future. In order to make use of information related to customer satisfaction with services at stores in order to improve services and the like, a technique that employs a questionnaire system has been proposed as an example of a technique that collects information related to satisfaction. In the questionnaire method, the customer answers a subjective evaluation according to the question of the questionnaire.

JP 2014-178724 A JP 2007-156762 A JP 2002-123586 A JP 2002-197359 A

  When a user who browses an information site visits a store, the user is satisfied with the store if the state in the store deviates from the assumption based on information on the information site due to circumstances such as congestion in the store I sometimes feel disappointed. For such a user, in order to make up for the discouragement caused by such divergence and to encourage the visit to the store again, it is conceivable to set the content of the privilege presented on the coupon according to the degree of discouragement.

  However, in the case of the above-described technique for collecting information about satisfaction, whether or not information about the subjective feeling of the user with respect to the state in the store can be obtained depends on the answer to the questionnaire by the user. It is desired that information about the user's subjective feeling can be obtained without the user's actions such as answering a questionnaire. Furthermore, the above-described technology regarding the setting of the content of the privilege does not include the assumption that the evaluation of the user's satisfaction by using the store is reflected in the content of the privilege.

  In one aspect, an object of the present invention is to provide an information processing apparatus, an information processing system, an information processing method, and an information processing program capable of evaluating the satisfaction of a user who uses a store.

  In one aspect, the information processing apparatus includes: a feature amount calculation unit that calculates a feature amount for each evaluation item for the state based on sensing data acquired by sensing the state in the store; and the user visits the store Based on the difference between the feature amount calculated for the state at the time and the model constructed from the learning data accumulated according to the user's store use, the user's subjective And an index calculation unit that calculates an index representing a result of guessing a sense.

  It is possible to evaluate the satisfaction of the user who uses the store.

FIG. 1 is a diagram illustrating the configuration of the information processing system according to the embodiment. FIG. 2 is a block diagram illustrating a functional configuration of the service server apparatus according to the embodiment. FIG. 3 is a block diagram illustrating an example of the store server apparatus. FIG. 4 is a block diagram illustrating an example of a mobile terminal device. FIG. 5 is a diagram illustrating details of a functional configuration of the service server apparatus according to the embodiment. FIG. 6 is a diagram illustrating an example of learning data. FIG. 7 is an explanatory diagram illustrating an example of a user experience model. FIG. 8 is a diagram (1) illustrating a control sequence of information processing according to the embodiment. FIG. 9 is a diagram (2) illustrating the control sequence of the information processing according to the embodiment. FIG. 10 is a flowchart illustrating an example of the procedure of index calculation processing. FIG. 11 is a diagram illustrating a first example of the relationship between the feature amount and the index. FIG. 12 is a diagram illustrating a second example of the relationship between the feature amount and the index. FIG. 13 is an explanatory diagram illustrating an example of a method for calculating the degree of disappointment. FIG. 14 is an explanatory diagram illustrating an example of a method for calculating the degree of fun. FIG. 15 is an explanatory diagram illustrating an example of a calculation method of the brightness. FIG. 16 is a diagram illustrating an example of the content of coupon information. FIG. 17 is a diagram illustrating an example of a screen on which a coupon is displayed. FIG. 18 is a flowchart illustrating an example of the procedure of the model learning process. FIG. 19 is a diagram illustrating a hardware configuration example of a computer that executes an information processing program according to the embodiment.

  Hereinafter, an information processing apparatus, an information processing system, an information processing method, and an information processing program according to the present application will be described with reference to the accompanying drawings. Note that this embodiment does not limit the disclosed technology. Each embodiment can be appropriately combined within a range in which processing contents are not contradictory.

[System configuration]
FIG. 1 is a diagram illustrating the configuration of the information processing system according to the embodiment. A provider of an information providing service by the information processing system 1 operates an information site on which store information is posted and provides information to the user. The provider of the information providing service may be simply referred to as “provider”. The user is an individual user who receives information provided by the provider. Stores are stores that have requested providers to post store information, and include various stores such as restaurants that provide food and drinks, and stores that sell products. In addition, the operator of the store server device 3 may be referred to as “a store clerk”. “Clerk etc.” may include any person involved in store sales.

  A store clerk issues a coupon to a user via an information site which a provider manages. In the coupon, the contents of the privilege given when the store is used are displayed. The service server device 2 is based on the difference between the model constructed from the learning data accumulated according to the user's experience of using the store and the feature amount based on the sensing data when the user visits the store. An index representing the result of inferring a subjective feeling is calculated. The service server device 2 sets the content of the privilege displayed on the coupon based on the calculated index.

  The information processing system 1 includes a service server device 2, a store server device 3, and a mobile terminal device 4. The service server device 2, the store server device 3, and the mobile terminal device 4 are connected to each other via a network 5. The network 5 can employ any type of communication network such as the Internet, a LAN (Local Area Network), and a VPN (Virtual Private Network) regardless of wired or wireless.

  The service server apparatus 2 that is an information processing apparatus is a computer that provides an information providing service. The service server device 2 provides the information posted on the information site to the user and accepts a request for information posting from the store. As an embodiment, the service server device 2 may be implemented as a Web server that realizes an information providing service, or may be implemented as a cloud that provides an information service by outsourcing. The service server device 2 can also be implemented by preinstalling or installing an information providing program provided as package software or online software on a desired computer.

  The store server device 3 is a computer installed in each store that requested the provider to post store information. The store server device 3 transmits store information input by a store clerk or the like to the service server device 2. The store information includes coupon information and category information in addition to basic information about the store such as an address and a telephone number.

  The coupon information is information indicating the content of a coupon issued via the information site, and includes the content of the privilege when the user uses the store, conditions for receiving the privilege, and the like. The category information indicates the category of the store registered by the store. A category is a keyword for classifying stores according to their characteristics. For example, “Calm”, “Quiet”, “Buzzy”, “Bright”, etc. represent the state and atmosphere of the store in an easy-to-understand manner. To do. The category may include keywords related to service at the store, such as “short waiting time”. Categories are also used to search for stores on information sites. In addition, the store server device 3 transmits sensing data acquired by sensing the state of the store to the service server device 2.

  The user carries the mobile terminal device 4 when visiting the store. The mobile terminal device 4 receives information provided from the service server device 2 and transmits data related to the user to the service server device 2. The mobile terminal device 4 transmits sensing data acquired by sensing the user's behavior or state to the service server device 2. As an embodiment, the mobile terminal device 4 can employ a mobile communication terminal such as a smart phone, a mobile phone or a PHS (Personal Handyphone System), a tablet terminal, a slate terminal, or the like. For example, a user application provided by an information service provider is installed in the mobile terminal device 4.

[Configuration of Service Server Device 2]
FIG. 2 is a block diagram illustrating a functional configuration of the service server apparatus according to the embodiment. The service server device 2 includes a communication unit 11, a control unit 12, and a storage unit 13. The communication unit 11 includes a store I / F (interface) unit 21 and a user terminal I / F unit 22. The store I / F unit 21 is an interface that performs communication control with the store server device 3. The store I / F unit 21 receives store information and sensing data transmitted from the store server device 3. Further, the store I / F unit 21 transmits the store information registered in the service server device 2 to the store server device 3 in response to a request from the store server device 3. In addition, the store I / F unit 21 may transmit various notifications from the provider to the store to the store server device 3.

  The user terminal I / F unit 22 is an interface that performs communication control with the mobile terminal device 4. The user terminal I / F unit 22 transmits store information to the mobile terminal device 4 and receives sensing data from the mobile terminal device 4. In addition, the user terminal I / F unit 22 may transmit various notifications from the provider to the user to the mobile terminal device 4.

  The control unit 12 controls the entire service server device 2. The control unit 12 processes information and data input from the communication unit 11. The control unit 12 includes a feature amount calculation unit 23, an index calculation unit 24, a coupon creation unit 25, and a model learning unit 26. The feature quantity calculation unit 23 is based on the sensing data for the store and the user, and the feature quantity (first feature quantity) for each evaluation item for the state in the store and the evaluation item for the user's action or state Each feature amount (second feature amount) is calculated.

  The index calculation unit 24 calculates the index based on the difference between the feature amount when the user visits the store and the user experience model. The index represents the result of inferring the user's subjective feeling about the use of the store. The user experience model is a model constructed from learning data accumulated according to the user's experience of using the store. The coupon creation unit 25 creates a coupon issued to the user. The coupon creation unit 25 sets the content of the privilege displayed on the coupon according to the index calculated by the index calculation unit 24. The model learning unit 26 constructs a user experience model from the learning data to which the feature amount calculated by the feature amount calculation unit 23 is added.

  The storage unit 13 includes a store information storage unit 27, a user information storage unit 28, and a sensing data storage unit 29. The store information storage unit 27 is an area in which store information for each store is stored. The store information stored in the store information storage unit 27 includes coupon information and category information transmitted from the store server device 3. The store server device 3 can arbitrarily update the coupon information and category information stored in the store information storage unit 27 by accessing the service server device 2.

  The user information storage unit 28 is an area in which user information for each user is stored. The user information stored in the user information storage unit 28 includes a user experience model. The user experience model includes a category model and a store model. The store model is a model (first model) constructed from learning data for each store. The category model is a model (second model) constructed from learning data for each category. In addition, learning data is stored in the user information storage unit 28. The sensing data storage unit 29 is an area in which sensing data transmitted from the mobile terminal device 4 is accumulated. In the sensing data storage unit 29, the sensing data transmitted from the store server device 3 may also be temporarily stored.

[Configuration of store server device 3]
FIG. 3 is a block diagram illustrating an example of the store server apparatus. FIG. 3 shows the store server device 3 and the multi-sensor 6 connected to the store server device 3. The multi-sensor 6 is installed in the store and senses the state in the store. The store server device 3 and the multi-sensor 6 can be connected in any form, wired or wireless. For example, when the store server device 3 and the multi-sensor 6 are connected by wire, they are connected via USB (Universal Serial Bus) or HDMI (registered trademark) (High-Definition Multimedia Interface). When the store server device 3 and the multi-sensor 6 are connected wirelessly, in addition to short-range wireless communication such as BLE (Bluetooth (registered trademark) Low Energy), a wireless LAN (Local Area) such as Wi-Fi (registered trademark) is used. Network).

  The multisensor 6 includes, for example, a plurality of sensing devices such as a camera that shoots the inside of a store, a volume meter that measures volume, an illuminance meter that measures illuminance, and a thermometer that measures temperature. The multi-sensor 6 senses a state for several items in the store environment. In addition, you may change the sensing device with which the multi sensor 6 is provided suitably according to the object of sensing. For example, when smoke due to smoking is to be sensed, the multi-sensor 6 includes a floating particle sensor. The multi-sensor 6 may be provided to the store by the provider of the information providing service, or may be obtained by the store from other than the provider.

  The store server device 3 includes a communication unit 31, a control unit 32, a storage unit 33, and an input unit 34. The communication unit 31 includes a service server I / F unit 35 and a sensing data acquisition unit 36. The service server I / F unit 35 is an interface that performs communication control with the service server device 2. The service server I / F unit 35 transmits the store information input from the input unit 34 and the sensing data stored in the storage unit 33 to the service server device 2. Further, the service server I / F unit 35 receives store information registered in the service server device 2. The sensing data acquisition unit 36 acquires sensing data from the multisensor 6.

  The control unit 32 controls the store server device 3 as a whole. The control unit 32 processes information and data input from the communication unit 31 and the input unit 34. The input unit 34 receives an input operation to the store server device 3 by a store clerk or the like. The input unit 34 is a pointing device, a keyboard, or the like. The storage unit 33 stores coupon information and category information input from the input unit 34. In addition, sensing data acquired from the multisensor 6 is accumulated in the storage unit 33.

[Configuration of Mobile Terminal Device 4]
FIG. 4 is a block diagram illustrating an example of a mobile terminal device. FIG. 4 shows the mobile terminal device 4 and the wearable sensor 7 connected to the mobile terminal device 4. The wearable sensor 7 is attached to the user's body and senses the user's movement or body state. The portable terminal device 4 and the wearable sensor 7 can be connected in any form such as USB or BLE regardless of wired or wireless.

  The wearable sensor 7 includes, for example, a sensing device such as an acceleration sensor that measures the moving speed of the user. Note that the sensing device provided in the wearable sensor 7 may be changed as appropriate according to the sensing target. For example, when the user's pulse is a sensing target, the wearable sensor 7 includes a pulse sensor. Sensing about the user's behavior or state may be performed by a function provided in the mobile terminal device 4. For example, the mobile terminal device 4 acquires position information indicating the current position of the mobile terminal device 4 as sensing data about the user's behavior using the GPS function. When the sensing target does not require sensing by the wearable sensor 7, the mobile terminal device 4 may omit the connection of the wearable sensor 7.

  The mobile terminal device 4 includes a communication unit 41, a control unit 42, a storage unit 43, and a display / input unit 44. The communication unit 41 includes a service server I / F unit 45 and a sensing data acquisition unit 46. The service server I / F unit 45 is an interface that performs communication control with the service server device 2. The service server I / F unit 45 receives information from the service server device 2, and transmits sensing data about the user's action or state and information input from the display / input unit 44 to the service server device 2. . The sensing data acquisition unit 46 acquires sensing data from the wearable sensor 7.

  The control unit 42 controls the entire mobile terminal device 4. The control unit 42 processes information and data input from the communication unit 41 and the display / input unit 44. The display / input unit 44 receives an input operation to the mobile terminal device 4 by the user. The display / input unit 44 displays a screen corresponding to the input operation and a screen corresponding to the information transmitted from the service server device 2. Information on the information site and coupons issued from the store to the user via the information site are displayed on the display / input unit 44. For example, the display / input unit 44 includes a touch panel as a pointing device and a liquid crystal panel as a display device. The storage unit 43 stores applications. In an environment where an environment in which an application operates is prepared in the mobile terminal device 4, the mobile terminal device 4 receives information from the service server device 2 and transmits sensing data to the service server device 2 through an information providing service. And do.

  FIG. 5 is a diagram illustrating details of a functional configuration of the service server apparatus according to the embodiment. The store I / F unit 21 includes a sensing data acquisition unit 51 and a store information acquisition unit 52. The sensing data acquisition unit 51 acquires the sensing data transmitted from the store server device 3. The sensing data acquired by the sensing data acquisition unit 51 is input to the feature amount calculation unit 23 directly or through temporary storage in the sensing data storage unit 29. The store information acquisition unit 52 acquires store information transmitted from the store server device 3. Store information acquired by the store information acquisition unit 52 is stored in the store information storage unit 27.

  The user terminal I / F unit 22 includes a sensing data acquisition unit 53 and a service usage determination unit 54. The sensing data acquisition unit 53 acquires user sensing data transmitted from the mobile terminal device 4. Sensing data acquired by the sensing data acquisition unit 53 is accumulated in the sensing data storage unit 29. The service usage determination unit 54 determines the start of service usage by the user in response to the notification from the mobile terminal device 4.

  The feature amount calculation unit 23 calculates a feature amount for each evaluation item for the state of the store based on the sensing data acquired by the sensing data acquisition unit 51. “Waiting time”, “degree of congestion”, “noisiness”, “illuminance”, and “temperature” are examples of evaluation items regarding the state of a store. The “waiting time” is an item indicating the time that the user waits in the store, and for example, an elapsed time from when the user enters the store until food or the like is provided is a feature amount. For example, the feature amount calculation unit 23 analyzes the image data obtained by photographing with the camera in the multi-sensor 6 so as to grasp the time when the user enters the store and the time when the food or drink is provided. The “congestion degree” is an item indicating a congestion state when the user stays in the store, and for example, an average of the number of customers during the stay of the user is used as a feature amount. The feature amount calculation unit 23 obtains the number of customers by analyzing the image data.

  “Noisiness” is an item indicating the amount of noise when the user stays in the store, and the average of the measurement results obtained by the volume meter in the multi-sensor 6 is used as the feature amount. “Illuminance” is an item indicating the brightness in the store, and the average of the measurement results obtained by the illuminometer in the multi-sensor 6 is used as the feature amount. “Temperature” is an item indicating the room temperature of the store, and an average of measurement results obtained by the thermometer in the multi-sensor 6 is used as a feature amount. The evaluation item may include any item regarding the state of the store. The evaluation item for the state of the store may include, for example, the concentration of smoke due to smoking.

  The feature amount calculation unit 23 calculates a feature amount for each evaluation item regarding the user's action or state based on the sensing data read from the sensing data storage unit 29. "Departure degree" and "visit frequency" are examples of evaluation items for user behavior. “Departure degree” is an item indicating the degree to which the user has departed before visiting the store. For example, the number of staying points is the number of points where the user stopped moving until the user visited the store. Is a feature amount. The feature amount calculation unit 23 obtains the number of staying points based on the result of monitoring the position information of the mobile terminal device 4 and the measurement result of the acceleration sensor in the wearable sensor 7. The “visit frequency” is an item indicating the frequency at which the user visits the store. For example, the cumulative number of visits to the store by the user is used as the feature amount. The feature amount of “visit frequency” may be a value that takes into account the interval between visits. The evaluation item may include any item regarding the user's action or state. The evaluation items regarding the user's condition may include, for example, the number of pulses.

  The index calculation unit 24 calculates “disappointment”, “enjoyment”, and “degree”, which are indexes representing the result of estimating the user's subjective feeling about using the store. The “disappointment degree” indicates the degree of disappointment, which is one of the subjective feelings of the user by using the store. “Degree of fun” is one of the subjective feelings of the user with respect to the store, and indicates the degree that the user is looking forward to using the store to be visited. “Degree” is one of the subjective feelings of the user with respect to the store, and indicates the degree to which the user particularly likes the store from other stores based on the experience of using the store. The index calculation unit 24 calculates each index based on the feature amount calculated by the feature amount calculation unit 23 and the user experience model read from the user information storage unit 28.

  The index calculation unit 24 obtains a discouragement degree for each evaluation item from the feature amount for each evaluation item regarding the state in the store. The index calculation unit 24 calculates a total discouragement degree that integrates the discouragement degree calculated for each evaluation item. In addition, the index calculation unit 24 obtains the degree of fun and the degree of evaluation for each evaluation item from the feature amount for each evaluation item regarding the user's behavior. The index calculation unit 24 calculates a total enjoyment degree by integrating the enjoyment degrees calculated for each evaluation item. The index calculation unit 24 calculates a total temperature obtained by integrating the temperatures calculated for each evaluation item. As described above, the index calculation unit 24 uses the discouragement degree that is an index (first index) based on the feature amount for the state in the store (first index) and the index (first index) based on the feature amount about the action or state of the user visiting the store. 2), the degree of fun and the degree of delight are calculated.

  The coupon creation unit 25 updates the contents of the coupon information read from the store information storage unit 27 in accordance with the respective indexes of discouragement, fun level, and jealousy calculated by the index calculation unit 24. The service server device 2 transmits the coupon created by the coupon creation unit 25 from the user terminal I / F unit 22 to the mobile terminal device 4. When the service server I / F unit 45 of the mobile terminal device 4 receives the coupon, the display / input unit 44 displays the coupon.

  The model learning unit 26 constructs a model stored in the user information storage unit 28. The model learning unit 26 adds the feature amount calculated by the feature amount calculation unit 23 to the learning data read from the user information storage unit 28, and constructs a category model and a store model based on the learning data. The model learning unit 26 stores the constructed category model, store model, and learning data in the user information storage unit 28.

  FIG. 6 is a diagram illustrating an example of learning data. The learning data is data generated by collecting the feature amount calculated for the store state when the user visits the store and the feature amount calculated for the user's behavior or state. The learning data represents a history of the store state experienced by the user in the use of the store and the user's own behavior or state. The learning data is stored in a table in the user information storage unit 28. The feature amount is written for each row with a store name and date visited by the user. The number of lines in which the feature amount is written increases every time the user uses the store. In FIG. 6, as examples of the feature values written in the table, “waiting time” and “noisy” feature items as evaluation items for the store state, and “detour” as evaluation items for the user's behavior are shown. The characteristic amount of “degree” and “visit frequency” is shown.

  FIG. 7 is an explanatory diagram illustrating an example of a user experience model. The category model stored in the user information storage unit 28 includes a model constructed for each category. The “category A model” is a category model constructed from the learning data of the store where the category “A” is registered. Each of the categories “A”, “B”,... Represents a category such as “There is calm” or “Lively” stored as category information in the store information storage unit 27 of the service server device 2. The “category A model” is a representation of the store state experienced by the user and the user's behavior or state as a probability distribution in the store in which the category “A” is registered. The “category A model” includes a model for each evaluation item regarding the state of the store and a model for each evaluation item regarding the user's behavior or state. Models of categories other than the category “A” are also constructed in the same manner as the “category A model”.

  The store model stored in the user information storage unit 28 includes a model constructed for each store. For example, the “model of the store α” is a store model constructed from the learning data of the store “α”. Stores “α”, “β”,... Represent stores in which feature amounts are written in the learning data. The “model of the store α” represents the store state experienced by the user at the store “α” and the user's behavior or state as a probability distribution. The “model of the store α” includes a model for each evaluation item regarding the state of the store and a model for each evaluation item regarding the user's behavior or state. Store models other than the store “α” are also constructed in the same manner as the “model of store α”.

[Process flow]
Next, a processing flow of the information processing system 1 according to the present embodiment will be described. 8 and 9 are diagrams illustrating a control sequence of information processing according to the embodiment. The store server device 3 accepts input of store information by a store clerk or the like (step S1). For example, the store clerk accesses the information site of the provider and inputs store information according to a predetermined format on the information site. A store clerk or the like inputs category information by selecting a category considered as a feature of the store from a list of selectable categories. Moreover, a salesclerk etc. inputs coupon information, such as the content of the privilege displayed on a coupon, and the conditions for receiving a privilege. For example, the input coupon information includes the content of the basic privilege and the content of the privilege added according to the calculated index.

  The store server device 3 transmits the input store information to the service server device 2 (step S2). When the service server device 2 receives the store information from the store server device 3, the service server device 2 stores the store information in the store information storage unit 27 (step S3). Thereby, the store information input by the store clerk or the like is registered in the service server device 2.

  When the user uses the information site service, the mobile terminal device 4 activates the application in accordance with the operation by the user (step S4). When the mobile terminal device 4 activates the application, it notifies the service server device 2 of the start of service use (step S5). The mobile terminal device 4 may notify the start of service use, for example, when the mobile terminal device 4 accesses the information site instead of starting the application. The service usage determination unit 54 of the service server device 2 determines that service usage by the notification source user has started in response to the notification from the mobile terminal device 4.

  When the service server device 2 receives the service use start notification, the service server device 2 transmits a list of stores in which the store information is registered in the service server device 2 to the mobile terminal device 4 (step S6). The mobile terminal device 4 displays a store list on the display / input unit 44. The service server device 2 may transmit information regarding the state of the store, for example, the current congestion status, to the mobile terminal device 4 together with the list of stores. The display / input unit 44 may display information such as the current congestion status for each store along with the list of stores.

  The user selects a store from the displayed list of stores. The portable terminal device 4 accepts selection of a store by the user (step S7). The portable terminal device 4 transmits the store selection result to the service server device 2 (step S8). The service server device 2 transmits a sensing start request to the store server device 3 of the selected store (step S9). When the store server device 3 receives the sensing start request, the store server device 3 starts sensing the state in the store by the multisensor 6. Moreover, the shop server apparatus 3 stores the sensing data acquired by sensing in the memory | storage part 33 (step S10).

  In addition, the store server apparatus 3 is good also as making sensing always perform in the business hours of a store irrespective of the notification of a sensing start request | requirement. In this case, transmission of a sensing start request from the service server device 2 to the store server device 3 is omitted. The state of the store is sensed at a preset interval, and sensing data is stored every time sensing is performed. The sensing of the state in the store may be performed at least between the time when the user enters the store and the time when the user leaves the store. The service server device 2 may appropriately acquire sensing data of the state in the store from the store server device 3, and display the current state in the store on the information site. Moreover, the service server apparatus 2 may transmit the coupon according to the coupon information stored in the shop information storage part 27 to the portable terminal device 4 of the user who accessed the information site.

  When the mobile terminal device 4 transmits the store selection result (step S8), the mobile terminal device 4 starts sensing the user's behavior (step S11). The portable terminal device 4 transmits the sensing data acquired by sensing to the service server device 2 (step S12). When receiving the sensing data, the service server device 2 stores the sensing data in the sensing data storage unit 29 (step S13). In the period from when the store is selected until the user enters the selected store, the user's behavior is sensed at a preset interval. Sensing data is stored in the sensing data storage unit 29 every time sensing is performed.

  The mobile terminal device 4 detects that the user has entered the selected store because the current position of the mobile terminal device 4 matches the location of the selected store (step S14). The mobile terminal device 4 notifies the service server device 2 that the user has entered the store (step S15). The service server device 2 identifies the user from the notification source mobile terminal device 4 and recognizes that the user has entered the store. In addition, it may replace with the detection based on the positional information on the portable terminal device 4, and may detect a user's entrance with the sensor installed, for example in the entrance / exit of the store. Such a sensor identifies the user who entered the store from information transmitted from the mobile terminal device 4. In this case, the store server device 3 notifies the service server device 2 that the user has entered the store.

  The feature amount calculation unit 23 of the service server device 2 reads the sensing data accumulated in the sensing data storage unit 29 (step S16). The feature amount calculation unit 23 calculates a feature amount based on the read sensing data (step S17). As an example, the feature amount calculation unit 23 calculates a feature amount regarding the user's behavior during a period from when the store selection result is transmitted (step S8) to when the store is entered (step S14). The index calculation unit 24 calculates an index based on the calculated feature amount (step S17). As an example, the index calculation unit 24 calculates the degree of fun and the degree of index, which are indexes, based on the difference between the feature amount of the evaluation item for the user's behavior and the user experience model.

  The coupon creation unit 25 updates the content of the coupon information read from the store information storage unit 27 according to the calculated index, and creates a coupon (step S18). The service server device 2 transmits the created coupon to the mobile terminal device 4 (step S19). The mobile terminal device 4 displays the coupon on the display / input unit 44 (step S20).

  The mobile terminal device 4 detects the exit of the user from the store when the current position of the mobile terminal device 4 leaves the location of the store (step S21). The mobile terminal device 4 notifies the service server device 2 that the user has left the store (step S22). The service server device 2 identifies the user from the communication source mobile terminal device 4 and recognizes that the user has left. In addition, you may detect a user's leaving | exiting with the sensor installed in the entrance / exit of the store similarly to entrance. In this case, the store server device 3 notifies the service server device 2 that the user has left the store.

  The service server device 2 transmits a data transmission request to the store server device 3 (step S23). Upon receiving the data transmission request, the store server device 3 transmits the sensing data stored in the storage unit 33 to the service server device 2 (step S24). When the service server device 2 receives the sensing data, the feature amount calculation unit 23 calculates a feature amount based on the sensing data (step S25). As an example, the feature amount calculation unit 23 calculates a feature amount for a state in the store in a period from when the user enters the store to when the user leaves the store. The index calculation unit 24 calculates an index based on the calculated feature amount (step S25). As an example, the index calculation unit 24 calculates a discouragement degree that is an index based on a difference between a feature amount of an evaluation item for a state in a store and a user experience model.

  The coupon creation unit 25 updates the content of the coupon information read from the store information storage unit 27 according to the calculated index, and creates a coupon (step S26). The service server device 2 transmits the created coupon to the mobile terminal device 4 (step S27). The mobile terminal device 4 displays the coupon on the display / input unit 44 (step S28).

  The model learning unit 26 adds the feature amount calculated in step S17 and step S25 to the learning data stored in the user information storage unit 28. The model learning unit 26 performs model learning for constructing a category model and a store model based on the learning data (step S29). The model learning unit 26 stores the constructed category model, store model, and learning data in the user information storage unit 28. As a result, the information processing system 1 ends the processes shown in FIGS.

  FIG. 10 is a flowchart illustrating an example of the procedure of index calculation processing. The feature amount calculation unit 23 acquires sensing data based on sensing of the state in the store and sensing data based on sensing of the user's behavior or state (step S41). The feature amount calculation unit 23 calculates a feature amount for each evaluation item for the state in the store and a feature amount for each evaluation item for the user's action or state based on the respective sensing data (step S42). .

  The index calculation unit 24 determines whether or not the cumulative number of visits of the user to the store visited by the user this time is greater than or equal to a threshold value (step S43). When it is determined that the cumulative number of visits is greater than or equal to the threshold (Yes at Step S43), the index calculation unit 24 calculates a discouragement degree, a fun degree, and a jealousy degree for each evaluation item (Step S44). Furthermore, the index calculation unit 24 calculates the total disappointment, fun, and jealousy by integrating the evaluation items (step S45).

  On the other hand, when it is determined that the cumulative number of visits is less than the threshold (No at Step S43), the index calculation unit 24 skips the calculation of the degree of accuracy for each evaluation item and the total degree of visit. The index calculation unit 24 calculates the disappointment level and the enjoyment level for each evaluation item (step S46). Further, the index calculation unit 24 calculates the total discouragement and enjoyment that integrates the evaluation items (step S47). Thereby, the service server apparatus 2 complete | finishes the process shown in FIG.

  Note that the service server device 2 calculates the index by the process shown in FIG. 10 at least once for each use of the store by the user. One index calculation is performed, for example, when the user leaves the store. In the example of the control sequence shown in FIGS. 8 and 9, the index calculation is performed at two times when the user enters the store and when the user leaves the store.

  FIG. 11 is a diagram illustrating a first example of the relationship between the feature amount and the index. For example, the index calculation unit 24 converts the feature amount for each evaluation item into the disappointment level for each evaluation item based on the relationship according to the first example. The index calculation unit 24 calculates a discouragement degree, which is an index indicating a difference between the feature quantity x and the model, by using the following formula (1), with the feature quantity for each evaluation item calculated by the feature quantity calculation unit 23 as x. To do. The term erf represents a Gaussian error function. μ represents an average value of feature quantities in each probability distribution of the category model and the store model. σ represents the standard deviation of the feature amount in each probability distribution of the category model and the store model.

  The relationship between the feature quantity and the index shown in FIG. 11 represents an example in which the index becomes higher as the feature quantity x is lower from μ. According to this example, the index when the feature amount x is μ or more is 0%, the index when the value is −σ away from μ is 50%, and the index when the feature amount x is −2σ or more away from μ. 100% is set. Note that the relationship used for the conversion from the feature amount to the index may be appropriately changed according to the characteristics of the evaluation item. For the conversion from the feature quantity to the index, for example, a relationship set so that the index becomes higher as the feature quantity x increases from μ may be used.

  FIG. 12 is a diagram illustrating a second example of the relationship between the feature amount and the index. For example, based on the relationship according to the second example, the index calculation unit 24 converts the feature amount for each evaluation item into a degree of fun for each evaluation item and a degree of accuracy for each evaluation item. Assuming that the feature amount for each evaluation item calculated by the feature amount calculation unit 23 is x, the index calculation unit 24 uses the following equation (2) to calculate the degree of fun and 贔 屓 that are indexes indicating the difference between the feature amount x and the model. Calculate the degree.

  The relationship between the feature quantity and the index shown in FIG. 12 represents an example in which the index has a higher value as the feature quantity x is higher. According to this example, the index when the feature amount x is μ is set to 50%, the index when it is equal to or larger than μ + σ is 100%, and the index when it is equal to or smaller than μ−σ is set to 0%. Note that the relationship used for the conversion from the feature amount to the index may be appropriately changed according to the characteristics of the evaluation item. For the conversion from the feature quantity to the index, for example, a relationship set so that the index becomes a lower value as the feature quantity x is higher may be used.

  FIG. 13 is an explanatory diagram illustrating an example of a method for calculating the degree of disappointment. For example, when a user uses a store α in which the category “is calm” is used, the “noisiness” of the store α is compared to the model of the evaluation item “noisy” in the category model of “with calm”. "Is markedly high. In this case, since the state of the store α is greatly different from the “calm” atmosphere that is assumed in the user's experience, it can be assumed that the user is likely to be disappointed about the use of the store α. . Alternatively, the same estimation can be made when the evaluation item “degree of congestion” is remarkably high compared to the assumed “calm” atmosphere. Furthermore, if the evaluation item “illuminance” of the store where the category “bright” is registered is markedly low, the evaluation item “noisy” of the store where the category “lively” is registered is markedly low and quiet It is possible to guess that there is a high possibility of being discouraged.

  For example, it is assumed that this “waiting time” is significantly longer than the model of the evaluation item “waiting time” in the store model of the store α that is always used by the user. In this case, since the waiting time of the store α is longer than usual, it can be estimated that the user is likely to be disappointed about the use of the store α. In addition, if the evaluation items “degree of congestion”, “noisiness” or “temperature” are remarkably high and the environment in the store is very different from the usual state, the possibility of being discouraged is high. Guessing is possible.

  The index calculation unit 24 calculates a total discouragement degree that integrates the discouragement degree for each evaluation item by obtaining an average of the discouragement degree calculated for each evaluation item. For example, the index calculation unit 24 calculates the total disappointment by dividing the total value of the disappointment for each evaluation item by the number of evaluation items for which the disappointment is calculated. The index calculation unit 24 may obtain a weighted average that gives a weight to each evaluation item instead of the addition average.

  For example, the evaluation items “noisiness” and “congestion degree” in the category model “with calmness” and the disappointment degrees for the evaluation item “waiting time” in the store model are calculated as 20%, 40%, and 50%, respectively. Suppose that The index calculation unit 24 calculates 25%, which is the result of the weighted average of the disappointment level for each evaluation item, as the total disappointment level.

  FIG. 14 is an explanatory diagram illustrating an example of a method for calculating the degree of fun. For example, when the user who goes to the store α has a smaller “degree of detour” than when the user has visited the store α so far, it can be estimated that the user is looking forward to visiting the store α. In addition, when the “visit frequency” is high, it can be estimated that there is a high possibility of looking forward to visiting the store α.

  The index calculation unit 24 calculates a total enjoyment degree in which the enjoyment degrees for each evaluation item are integrated by obtaining an average of the enjoyment degrees calculated for each evaluation item. For example, the index calculation unit 24 calculates the total enjoyment by dividing the total value of the enjoyment for each evaluation item by the number of evaluation items for which the enjoyment is calculated. The index calculation unit 24 may obtain a weighted average instead of the addition average.

  For example, it is assumed that the degree of fun for the evaluation items “departure degree” and “visit frequency” in the store model is calculated as 80% and 70%. The index calculation unit 24 calculates 75%, which is the result of the addition average of the enjoyment levels for each evaluation item, as the total enjoyment level. It should be noted that the user may have been looking forward to visiting the store α even when the “departure degree” of the store α is small and the “visit frequency” is high among the stores in which the common category is registered. I can guess it is expensive. The index calculation unit 24 may calculate the total enjoyment using not only the evaluation items in the store model but also the enjoyment of the evaluation items in the category model.

  FIG. 15 is an explanatory diagram illustrating an example of a calculation method of the brightness. For example, if the “visit frequency” of the store α is high among the stores in which the common category is registered, it can be estimated that the user is highly likely to have the store α as a habit. In addition, when the “departure degree” is small, it can be estimated that there is a high possibility that the store α is a habit.

  The index calculation unit 24 calculates a total temperature obtained by integrating the temperatures for each evaluation item by obtaining an average of the temperatures calculated for each evaluation item. For example, the index calculation unit 24 calculates the total accuracy by dividing the total value of the accuracy for each evaluation item by the number of evaluation items for which the accuracy is calculated. The index calculation unit 24 may obtain a weighted average instead of the addition average.

  For example, it is assumed that the degrees of evaluation items “visit frequency” and “degree of detour” in the category model are calculated as 65% and 55%, respectively. The index calculation unit 24 calculates 60%, which is the result of the average addition of the degrees for each evaluation item, as the total degree. In addition, the index calculation unit 24 may calculate the total degree of temperature using the degree of the evaluation item in the store model as well as the evaluation item in the category model.

  FIG. 16 is a diagram illustrating an example of the content of coupon information. The coupon information includes the contents of the privilege set for each numerical range of the discouragement degree, the pleasure degree, and the jealousy degree. The content of the privilege is input from the store server device 3 by a store clerk or the like and stored in the store information storage unit 27 of the service server device 2. As for the discouragement degree, a privilege is set so that the higher the discouragement degree, the more advantageous to the user. As a result, when the state of the store visited deviates from the assumption of the user and the user is disappointed, the disappointment is compensated by the privilege and the user is prompted to visit the store again. In the example shown in FIG. 16, “3% discount”, “5% discount” and “7% discount” representing discounts from accounting are given as benefits for the numerical range of discouragement of 10% or more, 20% or more, and 30% or more. Are set respectively. The higher the discouragement, the higher the discount rate is set.

  As for the degree of fun, a privilege is set so that the lower the degree of fun, the more advantageous to the user. As a result, the user who has a low degree of pleasure in visiting the store is encouraged to visit the store. As for the temperature, a privilege is set so that the higher the temperature, the more advantageous to the user. As a result, the user who already likes the store is encouraged to continue to use the store in the future. In the example shown in FIG. 16, a privilege of “large service” is set for a fun degree of 70% or less, and a “5% discount” privilege is set for a degree of fun of 80% or more. The privilege content included in the coupon information may be set for each index, and can be arbitrarily set by a store clerk or the like. The coupon creation unit 25 updates the content of the privilege from the basic coupon when the index of the numerical range in which the content of the privilege is set is calculated. The service server device 2 transmits the coupon with the updated benefit content to the mobile terminal device 4. On the other hand, the coupon creation unit 25 does not update the content of the privilege from the basic coupon when the index of the numerical range in which the content of the privilege is set is not calculated.

  FIG. 17 is a diagram illustrating an example of a screen on which a coupon is displayed. The screen is displayed on the display / input unit 44 of the mobile terminal device 4. For example, it is assumed that the index calculation unit 24 calculates a disappointment level of “20% or more” and a pleasure level of “70% or less”. The coupon creation unit 25 creates a coupon including a combination of “5% discount” and “great service”, which is a privilege corresponding to each index, according to the coupon information shown in FIG. In the example illustrated in FIG. 17, the content of the coupon privilege is displayed on the screen on which the search result of the store on the information site is displayed. When the content of the coupon of the store “△△ restaurant” is updated according to the index, the row of the store “△△ restaurant” in the search result after the update indicates “5 % OFF ”and“ Daisen Service ”are displayed. In the example shown in FIG. 17, a word indicating the current congestion status of each store is further displayed.

  The service server device 2 calculates, as an index, the degree of discouragement of the user due to the fact that the store state deviates from the model based on the user's experience. The service server device 2 makes it possible to set the contents of the coupon based on the discouragement degree, thereby making up for discouragement caused by the divergence and prompting the customer to visit the store again. Further, the service server device 2 can set the contents of the coupon based on the degree of enjoyment and the degree of pleasure, thereby prompting the user to visit the store according to the degree of pleasure of visiting the store and the degree of hesitation of the store. be able to. Thereby, use of a store by a user can be promoted.

  FIG. 18 is a flowchart illustrating an example of the procedure of the model learning process. The model learning unit 26 of the service server device 2 acquires the feature amount calculated by the feature amount calculation unit 23 (step S51). The model learning unit 26 includes the acquired feature value in the learning data, and reconstructs the category model and the store model for the evaluation item of the feature value (step S52). The model learning unit 26 determines whether or not the reconstruction of the category model and the store model has been completed for all the evaluation items (step S53).

  When the reconstruction of the category model and the store model for all the evaluation items has not been completed (No at Step S53), the model learning unit 26 is characterized about the evaluation item for which the reconstruction of the category model and the store model has not been completed. An amount is acquired (step S51). When the reconstruction of the category model and the store model is completed for all the evaluation items (Yes at Step S53), the model learning unit 26 ends the process illustrated in FIG. Thereby, the service server apparatus 2 can obtain a model expressing the state of the store experienced by the user and the behavior or state of the user.

  The information processing system 1 may store a user experience model and learning data for each user in the storage unit 43 in the mobile terminal device 4 instead of the storage unit 13 in the service server device 2. The storage unit 43 is provided with an area for storing a user experience model and learning data. The model learning unit 26 of the service server device 2 transmits the constructed category model, store model, and learning data to the mobile terminal device 4. The information processing system 1 may store the sensing data acquired from the multisensor 6 in the storage unit 13 in the service server device 2 instead of the storage unit 33 in the store server device 3. The storage unit 13 is provided with an area in which sensing data for the state in the store is stored. The store server device 3 transmits the sensing data acquired from the multisensor 6 to the service server device 2.

  The service server device 2 is not limited to the one that calculates the index from the feature amount for both the sensing data for the store and the sensing data for the user. The service server device 2 only needs to calculate a feature amount based on at least sensing data regarding the state in the store and calculate an index from the calculated feature amount. Further, the service server device 2 only needs to calculate at least the discouragement degree among the discouragement degree, the enjoyment degree, and the jealousy degree as an index.

[One aspect of effect]
As described above, the service server device 2 according to the present embodiment obtains a feature amount for the state in the store based on the sensing data acquired by sensing the state in the store. The service server device 2 calculates an index that represents a user's subjective feeling estimation result based on the difference between the feature value and the user experience model. The service server device 2 uses the information obtained by sensing the state in the store to calculate the index, thereby obtaining information on the subjective feeling of the user without intervention by the user such as answering a questionnaire. be able to. As a result, the satisfaction level of the user who uses the store can be evaluated.

  The processing units such as the feature amount calculation unit 23, the index calculation unit 24, the coupon creation unit 25, and the model learning unit 26 can be implemented as follows. For example, it can be realized by causing a central processing unit such as a CPU (Central Processing Unit) to develop and execute a process exhibiting the same function as each of the above processing units on a memory. These processing units do not necessarily have to be executed by the central processing unit, but may be executed by an MPU (Micro Processing Unit). Moreover, each said process part is realizable also by a hard wired logic.

  In addition, various semiconductor memory elements such as DRAM (Dynamic Random Access Memory), SRAM (Static Random Access Memory), and flash memory can be adopted as an example of the main storage device that each processing unit refers to as a work area. . In addition, the storage device referred to by each processing unit described above is not necessarily a main storage device, and may be an auxiliary storage device. In this case, an HDD (Hard Disk Drive), an optical disk, an SSD (Solid State Drive), or the like can be employed.

  In addition, each component of each illustrated apparatus does not necessarily have to be physically configured as illustrated. In other words, the specific form of distribution / integration of each device is not limited to that shown in the figure, and all or a part thereof may be functionally or physically distributed or arbitrarily distributed in arbitrary units according to various loads or usage conditions. Can be integrated and configured. For example, the feature amount calculation unit 23, the index calculation unit 24, the coupon creation unit 25, or the model learning unit 26 may be connected as an external device of the service server device 2 via a network. In addition, each device has a feature amount calculation unit 23, an index calculation unit 24, a coupon creation unit 25, or a model learning unit 26. It may be realized.

[Information processing program]
The various processes described in the above embodiments can be realized by executing a program prepared in advance on a computer such as a personal computer or a workstation. In the following, an example of a computer that executes an information processing program having the same function as that of the above-described embodiment will be described with reference to FIG.

  FIG. 19 is a diagram illustrating a hardware configuration example of a computer that executes an information processing program according to the embodiment. As illustrated in FIG. 19, the computer 100 includes a communication unit 110, a ROM 120, a RAM 130, and a CPU 140. These units 110 to 140 are connected via a bus 150.

  The communication unit 110 is a communication interface that performs communication with the store server device 3 and the mobile terminal device 4. The ROM 120 stores an information processing program 120A that exhibits the same functions as the feature amount calculation unit 23, the index calculation unit 24, the coupon creation unit 25, and the model learning unit 26 described in the embodiment. The CPU 140 reads the information processing program 120A from the ROM 120 and functions on the RAM 130 as the information processing process 130A. The information processing process 130A expands various data in an area allocated to the information processing process 130A in the storage area of the RAM 130, and executes various processes using the expanded various data. For example, the processes shown in FIGS. 10 and 18 are included as an example of the process executed by the information processing process 130A. Note that in the CPU 140, not all the processing units shown in the above embodiments need to operate, and a processing unit corresponding to a process to be executed may be virtually realized.

  The information processing program 120A is not necessarily stored in the ROM 120 from the beginning. The information processing program 120A may be stored in a computer-readable recording medium with a drive (not shown). The recording medium may be, for example, a flexible disk, a so-called FD, a CD-ROM, a DVD disk, a magneto-optical disk, a portable recording medium such as an IC card, or a semiconductor memory such as a flash memory. Alternatively, the information processing program 120A is stored in another computer or server device connected to the computer 100 via a public line, the Internet, a LAN, a WAN, etc., and the computer 100 acquires the information processing program 120A from these. May be executed.

  The following supplementary notes are further disclosed with respect to the embodiments including the above examples.

(Additional remark 1) Based on the sensing data acquired by the sensing of the state in a store, the feature-value calculation part which calculates the feature-value for every evaluation item about the said state,
Use of the store based on the difference between the feature amount calculated for the state when the user visits the store and a model constructed from learning data accumulated according to the store use of the user And an index calculation unit that calculates an index representing a result of estimating the subjective feeling of the user.

(Appendix 2) A category for classifying the store according to characteristics is registered in advance,
The information processing according to claim 1, wherein the model includes a first model constructed from the learning data for each store and a second model constructed from the learning data for each category. apparatus.

(Additional remark 3) The said feature-value calculation part is the 2nd feature-value about the 1st feature-value which is the said feature-value about the said state in the said store, and the user's action or state which visits the said store. Calculate the feature amount,
The index calculation unit calculates a first index that is the index based on the first feature quantity and a second index that is an index based on the second feature quantity. Or the information processing apparatus of 2.

(Supplementary note 4) The information processing apparatus according to supplementary note 3, wherein the first index includes a degree of disappointment indicating a degree of disappointment of the user due to use of the store.

(Supplementary note 5) The information processing apparatus according to supplementary note 3, wherein the second index includes a degree of enjoyment indicating a degree that the user is looking forward to using the store to be visited.

(Supplementary note 6) The information processing apparatus according to supplementary note 3, wherein the second index includes a degree of degree indicating that the user likes the store based on experience using the store.

(Supplementary note 7) The supplementary notes 1 to 6, further comprising a model learning unit that constructs a model from the learning data added each time the feature amount calculated by the feature amount calculation unit uses a store. The information processing apparatus according to any one of the above.

(Additional remark 8) It further has a coupon creation part which creates the coupon issued with respect to the said user,
The information processing apparatus according to any one of appendices 1 to 7, wherein the coupon creation unit sets the content displayed on the coupon according to the index.

(Supplementary note 9) The information processing apparatus according to any one of supplementary notes 1 to 8, wherein the index calculation unit integrates the index calculated for each evaluation item.

(Additional remark 10) The feature-value calculation part which calculates the feature-value for every evaluation item about the said state based on the sensing data acquired by sensing of the state in a store,
Use of the store based on the difference between the feature amount calculated for the state when the user visits the store and a model constructed from learning data accumulated according to the store use of the user And an index calculation unit that calculates an index representing a result of estimating the subjective feeling of the user.

(Additional remark 11) The process which calculates the feature-value for every evaluation item about the said state based on the sensing data acquired by sensing of the state in a store,
Use of the store based on the difference between the feature amount calculated for the state when the user visits the store and a model constructed from learning data accumulated according to the store use of the user And a process of calculating an index representing a result of estimating the subjective feeling of the user.

(Additional remark 12) The process which calculates the feature-value for every evaluation item about the said state based on the sensing data acquired by sensing of the state in a store,
Use of the store based on the difference between the feature amount calculated for the state when the user visits the store and a model constructed from learning data accumulated according to the store use of the user An information processing program causing a computer to execute a process of calculating an index representing a result of estimating a user's subjective feeling.

DESCRIPTION OF SYMBOLS 1 Information processing system 2 Service server apparatus 23 Feature-value calculation part 24 Index calculation part 25 Coupon creation part 26 Model learning part

Claims (9)

Based on sensing data acquired by sensing the state in the store, a feature amount calculation unit that calculates a feature amount for each evaluation item for the state,
Use of the store based on the difference between the feature amount calculated for the state when the user visits the store and a model constructed from learning data accumulated according to the store use of the user And an index calculation unit that calculates an index representing a result of estimating the subjective feeling of the user. Categories for classifying the stores according to characteristics are registered in advance,
The information according to claim 1, wherein the model includes a first model constructed from the learning data for each store and a second model constructed from the learning data for each category. Processing equipment. The feature amount calculation unit includes a first feature amount that is the feature amount for the state in the store, and a second feature amount that is a feature amount for the action or state of the user who visits the store. Calculate
The index calculation unit calculates a first index that is the index based on the first feature quantity and a second index that is an index based on the second feature quantity. The information processing apparatus according to 1 or 2.   4. The apparatus according to claim 1, further comprising a model learning unit that constructs a model from the learning data added each time the feature amount calculated by the feature amount calculation unit uses a store. The information processing apparatus according to item. It further has a coupon creation unit for creating a coupon issued to the user,
The information processing apparatus according to any one of claims 1 to 4, wherein the coupon creation unit sets the content displayed on the coupon according to the index.   The information processing apparatus according to claim 1, wherein the index calculation unit integrates the index calculated for each evaluation item. Based on sensing data acquired by sensing the state in the store, a feature amount calculation unit that calculates a feature amount for each evaluation item for the state,
Use of the store based on the difference between the feature amount calculated for the state when the user visits the store and a model constructed from learning data accumulated according to the store use of the user And an index calculation unit that calculates an index representing a result of estimating the subjective feeling of the user. Based on sensing data acquired by sensing the state in the store, a process for calculating a feature value for each evaluation item for the state,
Use of the store based on the difference between the feature amount calculated for the state when the user visits the store and a model constructed from learning data accumulated according to the store use of the user And a process of calculating an index representing a result of estimating the subjective feeling of the user. Based on sensing data acquired by sensing the state in the store, a process for calculating a feature value for each evaluation item for the state,
Use of the store based on the difference between the feature amount calculated for the state when the user visits the store and a model constructed from learning data accumulated according to the store use of the user An information processing program causing a computer to execute a process of calculating an index representing a result of estimating a user's subjective feeling.

Images