JP2023127911A - Sense-of-values estimation system and method - Google Patents

Abstract

To define the sense of values specific to an application field such as tourism and estimate the sense of values specific to the application field.SOLUTION: One aspect of the present invention is a sense-of-values estimation system including: an estimation content provision unit configured to provide plural pieces of estimation contents including images and catch phrases specific to an application field; an operation acceptance unit configured to accept a user's operation to select at least one of the plural pieces of estimation contents; an estimation unit configured to estimate user's sense of values defined by a database management unit based on the accepted operation; an information acquisition unit configured to acquire information regarding a user's scene or situation in the application field; and an action acquisition unit configured to acquire user's action regarding the content. When the estimation unit acquires the scene or situation in the application field and the user's action regarding the content, the estimation unit updates the user's sense of values classified into the scene or situation in the application field based on the values of the content corresponding to the action.SELECTED DRAWING: Figure 1

Description

The present invention relates to a value estimation system and method.

Conventionally, techniques for providing content based on user attributes are known. This type of technology is described in, for example, Patent Document 1. The interest analysis device described in Patent Document 1 is described to calculate the importance of content based on user attribute information and interest information indicating that the user is interested in the content. In addition, in order to support the user's travel, the travel support system described in Patent Document 2 uses input information including condition information and user attribute information from the user terminal to determine the route and usage information regarding the user's trip. It is stated that a draft plan should be created that includes items on sightseeing spots, means of transportation, and routes.

Japanese Patent Application Publication No. 2021-157206 Japanese Patent Application Publication No. 2016-139263

The technology described in Patent Document 1 mentioned above calculates the importance of content based on user attribute information, and the technology described in Patent Document 2 calculates the importance of content based on user attribute information. It is stated that a draft should be prepared. However, since the user's attributes such as interests or values change depending on the purpose of the action, the scene, and the situation, there is a problem that the user's attributes cannot be accurately specified. Furthermore, even if various types of processing are performed to identify the user's attributes, there is a problem in that the user's various values cannot be accurately estimated.

The present invention has been made in view of the above-mentioned problems, and is a values estimation system capable of defining values specific to an application field such as tourism, and estimating values specific to the application field. , and methods.

(1) One aspect of the present invention includes a database management unit that manages a database in which user values corresponding to an application field are classified by scene or situation, and an estimation including an image and a catchphrase specific to the application field. an estimation content providing section that provides a plurality of estimation contents; an operation reception section that receives a user's operation to select at least one of the plurality of estimation contents; an estimation unit that estimates the values of the user defined by the database management unit; an information acquisition unit that acquires information regarding the user's scene or situation in the application field; and a behavior acquisition unit that acquires the user's behavior regarding the content. , and the estimating unit is configured to calculate, when the scene or situation in the application field acquired by the information acquisition unit and the behavior of the user with respect to the content acquired by the behavior acquisition unit, the estimated value in the acquired application field. This is a values estimation system that updates the user's values classified into scenes or situations based on the values of content corresponding to the behavior.

(2) One aspect of the present invention is the values estimation system described above, in which the types of the actions include viewing the content, adding the content to favorites or a plan, and the estimation unit is configured to include scene or The user's values classified into situations may be updated based on the values of the content and the weight corresponding to the type of behavior.

(3) In one aspect of the present invention, the user's values include multiple types of values, weights are set for each of the multiple types of user's values, and the estimation unit classifies the values into scenes or situations. The values of the user may be updated based on weights set to the type of values of the user, values of the content, and weights set to the type of behavior.

(4) One aspect of the present invention is the values estimation system described above, in which the values of a plurality of personas are estimated based on the similarity between the values updated by the estimation unit and the values corresponding to the personas. The persona estimation unit may include a persona estimating unit that estimates a persona with the highest degree of similarity.

(5) In one aspect of the present invention, it is assumed that the sense of value has a probability distribution of a beta distribution represented by two statistical values, and the estimation unit uses the two statistical values to estimate the sense of value. may be estimated, the two statistical values may be updated based on the operation received by the operation reception unit, and the values may be updated using the updated two statistical values.

(6) One aspect of the present invention includes the step of providing a plurality of estimation contents including images and catchphrases specific to an application field, and the step of providing a user's operation to select at least one of the plurality of estimation contents. a step of receiving, and a step of estimating, based on the operation, the values of the user defined by a database management unit that manages a database in which values of the user corresponding to the field of application are classified by scene or situation; a step of acquiring information about a user's scene or situation in the field, a step of acquiring the user's behavior with respect to the content, and a step of acquiring the user's behavior with respect to the content and the scene or the situation in the application field; The method includes the step of updating the user's values classified into scenes or situations in an application field based on the values of content corresponding to the action.

According to one aspect of the present invention, values specific to an application field such as tourism can be defined, and values specific to the application field can be estimated.

FIG. 1 is a block diagram illustrating a configuration example of a suggestion system according to an embodiment. It is a figure for explaining an example of scene information in the suggestion system of an embodiment. It is a figure for explaining another example of scene information in the suggestion system of an embodiment. FIG. 3 is a diagram illustrating an example of the relationship between input data and estimation targets. It is a figure showing an example of the whole processing of the suggestion system in an embodiment. It is a figure which shows an example of the content for estimation. FIG. 3 is a diagram illustrating an example of a use case, questions, options, and estimated values of the information providing service in the embodiment. FIG. 3 is a diagram for explaining an example of a process of updating a user's values based on user behavior. FIG. 4 is a diagram for explaining an example of a process of updating values by applying the concept of Bayesian estimation. FIG. 7 is a diagram for explaining an example of the number of times n of user actions for updating values. It is a figure which shows the relationship between the values of a user to be updated, examples of values, operations and processes, and weights. FIG. 3 is a diagram for explaining persona estimation in the embodiment. FIG. 3 is a diagram for explaining a process of calculating the degree of similarity between a user's values and a persona's values. It is a figure for explaining the process (step ST14) of updating values and estimating a persona by the suggestion device 200 in the embodiment. It is a figure which shows the relationship between input data, the addition rule of a suggestion score, and the type of suggestion. It is a figure which shows an example of the process which performs the suggestion in embodiment. It is a figure showing an example of calculation processing of a suggestion score. It is a figure which shows an example of a content list screen. FIG. 6 is a diagram for explaining an example of a suggestion score addition rule based on a target user, a scene and/or situation, and a persona.

Hereinafter, a value estimation system and method to which the present invention is applied will be explained with reference to the drawings. The values estimation system and method to which the present invention is applied are realized, for example, by a suggestion system that estimates a user's values and suggests content to the user based on the estimated values. For example, the suggestion system of the embodiment makes suggestions based on the user's values in response to a search request based on the user's operation. Suggestion is a process of providing content including a list of content to the user. As a result, the suggestion system realizes an information provision service that is highly convenient for users. Note that in the following embodiments, the field of application of the suggestion system is sightseeing and walking around town, but it is not limited thereto, and can be applied to any field such as shopping.

FIG. 1 is a block diagram showing an example of the configuration of a suggestion system according to an embodiment. The suggestion system includes, for example, a user terminal device 100, a suggestion device 200, a content providing device 300, and an environmental information providing device 400. The user terminal device 100, the content providing device 300, the content providing device 300, and the environmental information providing device 400 are connected to a communication network, for example. Each device connected to the communication network includes a communication interface such as a NIC (Network Interface Card) or a wireless communication module (not shown in FIG. 1). Communication networks include, for example, the Internet, WAN (Wide Area Network), LAN (Local Area Network), cellular network, and the like.

The user terminal device 100 is, for example, a portable terminal device such as a smartphone or a tablet terminal. In the user terminal device 100, a UA (User Agent) such as a browser or an application program is activated. The UA is, for example, an application for receiving an information providing service provided by the suggestion device 200. The user terminal device 100 generates a search request including operation information based on the user's operation, location information of the user terminal device 100 using GPS (Global Positioning System, Global Positioning Satellite), and a query, and sends the search request to the suggestion device 200 or the content. The information is transmitted to the providing device 300. The user terminal device 100 performs display processing, operation acceptance processing, etc. using the content list information provided from the suggestion device 200 and the content received from the content providing device 300.

The suggestion device 200 is, for example, a server device that performs processing for an information providing service. The suggestion device 200 includes, for example, an operation reception section 210, an action history acquisition section 220, a scene/situation acquisition section 230, a user attribute estimation section 240, a statistical data acquisition section 250, a suggestion section 260, and a values database. It includes a management section 270, a persona attribute management section 272, and an estimation content storage section 280. Functional units such as the operation reception unit 210, the action history acquisition unit 220, the scene/situation acquisition unit 230, the user attribute determination unit 240, the statistical data acquisition unit 250, and the statistical data acquisition unit 250 are implemented by, for example, a CPU (Central Processing Unit). This is achieved by a processor executing a program stored in a program memory. In addition, some or all of these functional units may be implemented using LSI (Large Scale Integration), ASIC (Application Specific Integrated Circuit), or FPGA (Field-Programmable Gate Array). ), etc., or may be realized by hardware such as It may also be realized by software and hardware working together.

The operation reception unit 210 acquires operation information from the user terminal device 100. The behavior history acquisition unit 220 acquires behavior history information related to the user's values, such as operation information, user location information, content viewing history, and favorite registration. The scene/situation acquisition unit 230 acquires scene information representing a scene that affects the user's sense of values and/or situation information representing a situation. The user attribute estimation unit 240 estimates values and/or persona as user attributes. The user attribute estimation section 240 includes a values estimation section 242 and a persona estimation section 244, but is not limited thereto, and may not include the persona estimation section 244. User attributes include, for example, in addition to fixed user information (address, gender, age, etc.), values and personas that change depending on the scene and/or situation. The user's values are, for example, tastes or preferences in tourism. A persona is a customer model that receives information provision services. The statistical data acquisition unit 250 acquires statistical data regarding content. The statistical data is, for example, the popularity of the content, but is not limited to this, and may be any information that serves as an index for selecting the content, such as the number of content views, quality, number of external links, freshness, keywords, etc. The suggestion unit 260 suggests content to the user in response to a search request obtained from the user, using a content suggestion score calculated based on values and a content suggestion score calculated based on statistical data.

The values database management unit 270, the persona attribute management unit 272, and the estimation content storage unit 280 are configured using, for example, an HDD (Hard Disc Drive), a flash memory, an EEPROM (Electrically Erasable Programmable Read Only Memory), or a ROM (Reasonable Memory). dOnly Memory) , RAM (Random Access Memory), or a hybrid storage device using a plurality of these. The values database management section 270, the persona attribute management section 272, and the estimation content storage section 280 may be realized by an external storage device that can be accessed via various networks. An example of an external storage device is a NAS (Network Attached Storage) device. The values database management unit 270 manages a database in which the user's values corresponding to the user's purpose of action are defined in association with scenes or situations. Furthermore, the values database management unit 270 stores the estimation results of values. The persona attribute management unit 272 stores information for estimating a persona and a persona estimation result. The estimation content storage unit 280 manages the attributes of the persona corresponding to the user.

The content providing device 300 is, for example, a server device that provides various types of content. The content providing device 300 provides content related to, for example, restaurants, stores, public facilities (parks, etc.), transportation facilities, and commercial facilities at a travel destination. The content providing device 300 stores, in the content storage unit 300A, content to be provided including, for example, advertising information of a restaurant, coupon information, a URL of a restaurant, and the like.

The environmental information providing device 400 is a server device that distributes environmental information that changes from moment to moment, such as weather information, traffic information, and disaster information. The environmental information providing device 400 periodically transmits environmental information based on a request from the suggestion device 200, for example. Note that the environmental information may include user information such as current location information and electronic money balance information.

FIG. 2 is a diagram for explaining an example of scene information in the suggestion system of the embodiment. The scene information is, for example, information classified into 5Ws including why, what, with whom, when, and where. In other words, the scene information is information representing the main purpose, action, companion, objective time, and objective location. The scene information is estimated by the scene/situation acquisition unit 230.

FIG. 3 is a diagram for explaining another example of scene information in the suggestion system of the embodiment. For example, values classified into why, what, and when based on keywords extracted from travel magazines may be further classified by scene. Values for travel plans include, for example, healing, beauty, classics, and learning. Values related to individual spots (content) include high class, casual, etc. Time or seasonal values include seasons, time zones, and the like.

FIG. 4 is a diagram illustrating an example of the relationship between input data and estimation targets. The suggestion system acquires various input data in order to infer various values, personas, scenes, situations, etc. from the input data.
The scene/situation acquisition unit 230 may estimate the scene based on, for example, the location of the user terminal device 100 using GPS, the schedule input to the scheduler, the travel plan based on the operation information, the companion, and the purpose. The scene/situation acquisition unit 230 may estimate the situation based on weather data, celestial data, operation data and operation data of transportation means, and congestion data acquired from the environmental information providing device 400.
The values estimation unit 242 estimates the user's original values based on the user's behavior, such as a content viewing log, a content registration log in favorites, an external link in the content, or a visit log to an actual tourist spot. presume. The values estimating unit 242 may estimate the user's original values based on selection results based on the determination content and statistical data of the content in addition to the user's behavior.
In the suggestion system, a scene/situation acquisition unit 230 estimates a scene/situation based on the user's behavior acquired by the behavior history acquisition unit 220, and a values estimation unit 242 estimates values according to the scene/situation. It's fine. The values estimating unit 242 may estimate values according to the scene/situation based on the selection result of the content for estimation and the statistical data of the content in addition to the user's behavior.
The values estimation unit 242 may estimate the current values based on the selection result of the content for estimation and the statistical data of the content. The persona estimation unit 244 may estimate the persona based on the user's behavior, the selection result of the content for estimation, and the input result of the persona based on the user's operation.
The suggestion system may create a customer journey map based on the companion and purpose input by the user, and estimate the user's behavioral purpose based on the customer journey map, the user's behavior, and the travel plan.

FIG. 5 is a diagram illustrating an example of the overall processing of the suggestion system in the embodiment.
First, when the user terminal device 100 installs or starts an application for receiving an information providing service, for example, it transmits a connection request S10 to the suggestion device 200. The suggestion device 200 transmits estimation content S11 to the user terminal device 100 in response to the connection request S10. The user terminal device 100 receives a selection operation for the estimation content S11, and transmits operation information S12 to the suggestion device 200. The suggestion device 200 estimates the user's values based on the operation information S12 based on the selection operation, and adds the estimated values to the user information (step ST10). That is, the suggestion device 200 updates the user attributes in the values database.

Thereafter, the user terminal device 100 transmits various types of operation information S13a, position information S13b, or scene/situation information S13c to the suggestion device 200. In response, the suggestion device 200 updates the values based on the transmitted information 13 (step ST11). Further, the environmental information providing device 400 transmits environmental information S14 regarding the scene/situation to the suggestion device 200 when a condition such as when a preset timing arrives or when the environmental information changes is satisfied. In response, the suggestion device 200 updates the values based on the transmitted information S14 (step ST12).

Thereafter, the user terminal device 100 transmits a search request S15 including a query for searching for content related to tourism to the suggestion device 200. In response to the search request S15, the suggestion device 200 acquires the content data 16a and/or the statistical data 16b from the content providing device 300, and calculates a suggestion score based on the current user's values and the statistical data 16b (step ST13). ). The suggestion device 200 transmits suggestion data S17 for displaying a content list to the user terminal device 100 based on the calculated suggestion score. The user terminal device 100 receives the suggestion data S17 and displays a content list. When the user terminal device 100 receives an operation resulting from displaying a list of contents, it transmits operation information S18 to the suggestion device 200. Operations resulting from displaying a content list include, for example, viewing content based on selecting content included in the content list, adding content to favorites, and adding content to a plan. The suggestion device 200 updates the values based on the transmitted operation information S18 (step ST14).

FIG. 6 is a diagram showing an example of estimation content. The estimation content includes a plurality of sets of tourism-specific images and tourism-specific catchphrases. When the user terminal device 100 receives the estimation content, it displays a values estimation screen. The values estimation screen includes four sets of images, a catchphrase corresponding to each image, and a message asking, "What do you like?" The image and catch phrase included in the estimation content are images and catch phrases corresponding to the attributes of the content. When the user terminal device 100 receives an operation to select one of the four sets, it transmits selection result information indicating the selected image and catchphrase to the suggestion device 200. The suggestion device 200 estimates the user's values corresponding to the attributes of the content based on the selection result information, and transmits content list information to the user terminal device 100 based on the estimated values. As a result, the user terminal device 100 displays a recommended spot list screen including a list of contents. The user terminal device 100 may transmit filter information such as categories and tourist spots to the suggestion device 200, and the suggestion device 200 may transmit list information of contents matching the filter information to the user terminal device 100. Note that the estimation content may include an image and a catchphrase that update the values of the model plan specific to tourism.

Although the values estimation screen shown in FIG. 6 displays four sets of images and catchphrases, the present invention is not limited to this, and eight sets may be displayed at once. Furthermore, the sets of images and catchphrases to be displayed at once may be displayed in a random order or position without setting a priority or ranking. Furthermore, each set of images and catchphrases that are displayed at one time should be of the same size and aligned on the screen so that the user can see them all at once, thereby maintaining fairness between the images. is desirable.

FIG. 7 is a diagram illustrating an example of a use case, questions, options, and estimated values of the information providing service in the embodiment. For example, when searching for a travel plan, the suggestion device 200 displays a values estimation screen that includes a question message asking about the overall theme of the trip, and an image and catchphrase representing the theme of the trip as options. Estimation content is provided to the user terminal device 100. The suggestion device 200 can estimate the user's values, such as the values of the travel plan and the values of the spots, based on the operation information for selecting the option.

FIG. 8 is a diagram for explaining an example of a process of updating a user's values based on user behavior. As shown in FIG. 8, the suggestion device 200 of the embodiment updates the user's values corresponding to the attributes of the content when an operation (behavior) is performed. When an operation such as viewing content is performed on content belonging to "local" and "traditional", the suggestion device 200 increases the value representing the user's original values corresponding to "local" and "traditional". Update. On the other hand, since no operation has been performed on the content belonging to "Photogenic", the suggestion device 200 updates the numerical value representing the user's original values corresponding to "Photogenic" to a lower value. Thereby, the suggestion device 200 can more accurately capture the user's sense of values by reflecting the user's operation on the user's sense of values estimated using the estimation content.

The suggestion device 200 may update the values at the timing when it detects an action on the content (browsing, adding to favorites, adding to plan). The suggestion device 200 updates the values at each of the timings when content is viewed, added to favorites, and added to a plan, for example. If the same content is viewed, added to favorites, or added to a plan, the values will be updated three times. Furthermore, when the same content is viewed three times, the sense of values is also updated three times. Note that even if the content is deleted after it has been added to favorites and added to the plan, the values do not need to be updated.

After updating the values based on the behavior toward the content, the suggestion device 200 presents the estimated content to the user, and further updates the updated values based on the behavior toward the content based on the estimated content selected by the user. You can update it. For example, the suggestion device 200 updates the values based on the behavior toward the content when starting to use the information service, and then uses the estimation content to update the values calculated by the suggestion device 200. It is possible to get closer to the user's actual values.

The suggestion device 200 may change the weight (degree of influence) for updating values based on the type of behavior. That is, the suggestion device 200 may set the weight of the action. For example, it is assumed that adding content to favorites captures the user's preferences more than viewing content. Therefore, the suggestion device 200 may update the numerical value representing values more widely when adding content to favorites than when viewing content. The suggestion device 200 updates numerical values representing values using the following formula, for example.
Post-update values = (1-weight) x pre-update values + weight x content attributes For example, the suggestion device 200 calculates (1-0.2) x pre-update values + 0.2 for content added to the plan. The following calculation is performed: x content attribute. For example, the suggestion device 200 performs a calculation of (1-0.2) x pre-update values + 0.2 x content attributes for the content added to favorites. For example, the suggestion device 200 performs the following calculation on the viewed content: (1-0.1) x values before update + 0.1 x content attributes. Thereby, the suggestion device 200 can set the weight for adding a plan and adding to favorites to a higher value than the weight for viewing content.

FIG. 9 is a diagram for explaining an example of a process of updating values by applying the concept of Bayesian estimation. The suggestion device 200 may estimate values by applying the concept of Bayesian estimation. Assuming that the sense of value has a beta distribution probability distribution expressed by two statistical values based on user operations, the suggestion device 200 estimates the sense of value using the two statistical values. The suggestion device 200 updates two statistical values based on a user's operation, and updates values using the updated two statistical values.

The suggestion device 200 represents values as a probability distribution from 1 to 0, as shown in the lower rows of FIGS. 9(A) and 9(C). As shown in the lower part of FIG. 9(A), it can be seen that the value x has a high possibility of being 0.2 and has a low possibility of being 1.0. As the number of user actions (number of logs acquired) increases, values x become closer to the user's original values, and the range of estimated values becomes smaller and the probability density becomes higher. In addition, as the number of user actions (number of logs obtained) increases, the update range of values x becomes smaller, so values x are prevented from greatly fluctuating due to the most recent user actions (logs), and gradually the simple average value approach. Thereby, the suggestion device 200 can more accurately capture the user's original values by applying the idea of Bayesian estimation. For example, if the expected value of the probability distribution for selecting "local" content is estimated to be 0.2, and the user views "local" content or adds it to favorites (Figure 9 (B)) , it is estimated that the expected value of the probability distribution in which the user chooses the "local" item is a little higher than 0.2, so the values x can be updated as shown in the lower part of FIG. 9(C).

The suggestion device 200 assumes that the distribution of values x is a beta distribution, and that the distribution of the likelihood of actions with respect to content is a Bernoulli distribution. Since the beta distribution has two parameters α and β, the value x is expressed as α/(α+β). For example, α represents the number of successes, β represents the number of failures, and α and β are based on whether the user views content with specific values. As shown in the upper part of FIG. 9A, for example, when α=1 and β=4, the value x is 1/(1+4)=0.2. Note that α and β are tuned. When the suggestion device 200 receives the user's behavior, the suggestion device 200 updates the values x using the following formula.
α'/(α'+β')=(x+α)/(α+β+n)
However, α'=x+α, β'=n−x+β, where n is the number of actions of the user and x is the number of actions performed with respect to the target content.

FIG. 10 is a diagram for explaining an example of the number of times n of user actions for updating values. The number of user actions n is basically 1 because it is updated for each action, but it is weighted. The suggestion device 200 adjusts the number of user actions n based on the type of action or the values to be updated. For example, when viewing content with attributes of "meal" and "local," the suggestion device 200 sets n to 0.5 in order to update the values for the categories of "meal" and "local," If you have registered content with attributes of "meal" and "local" as favorites, set n to 1.0, and if you have selected content for estimation that has attributes of "meal" and "local", set n to Set to 10.0.

When the suggestion device 200 detects a behavior for a certain category, the suggestion device 200 may also update the genre of the category or the values of the detailed information. At this time, the suggestion device 200 may set n corresponding to the genre of the category or the values of the detailed information to a smaller value than n corresponding to the values of the category. For example, when viewing content, the suggestion device 200 sets n corresponding to the values of the category to 0.5, and sets n corresponding to the values of the genre or detailed information of the category to 0.25. It's fine.

When the suggestion device 200 detects a behavior for a certain category, the suggestion device 200 may also update values other than the category, genres other than the category, or values of detailed information. At this time, the suggestion device 200 replaces n corresponding to values other than the category, genres other than the category, or values of the detailed information with n corresponding to the values of the category and the genre of the category or the values of the detailed information. may be set to a value smaller than n corresponding to . Further, the suggestion device 200 may set n corresponding to a genre other than the category or values of detailed information to a smaller value than n corresponding to values other than the category. For example, when viewing content, the suggestion device 200 sets n corresponding to the values of the category to 0.5, sets n corresponding to the genre of the category or the values of detailed information to 0.25, and sets n corresponding to the values of the category or detailed information to 0.25. n corresponding to the values of the category may be set to 0.1, and n corresponding to the values of genres or detailed information other than the category concerned may be set to 0.05.

Note that when the suggestion device 200 detects a user's selection operation for estimated content for a certain category, it may set n to 0 for values other than the category and genres other than the category. When the suggestion device 200 detects a user's selection operation on estimation content that causes selection of a model plan for a certain category, the suggestion device 200 does not need to set n for the genre or detailed information for the category, and n (=5.0) for the genre or detailed information may be set smaller than n (=10.0) corresponding to the values of the category.

When the suggestion device 200 detects that a model plan or article is viewed, added to favorites, or added to a plan, it updates only the values of the model plan, and updates the values corresponding to the content attribute, the genre, or the detailed information. No need to update. Moreover, when the suggestion device 200 selects a model plan in the estimation content, the suggestion device 200 may update only the values corresponding to the selected content attribute.

The suggestion device 200 sets the number of actions x for the target content as a multiplication value of the content attribute value and n. The content attribute value is a score related to attributes (values) given to the estimation content. If a value corresponding to the attribute of the selected estimation content is set, the attribute-related score is the value of the value, and corresponds to the genre or detailed information that matches the selected estimation content. If the selected estimation content does not match the genre or detailed information, it is set to 0.

A specific example of calculation for updating values when behavior toward content is detected will be described below.
For example, suppose that values related to "meal-local" are to be updated, and the initial values of parameters regarding the user's "meal-local" are α=1 and β=4. At this time, the value is 1/(1+4)=0.2. The suggestion device 200 determines that when the user views meal content with an attribute value of “meal-local” of “1”, n=0.5, x=0.5,
α'=0.5+1=1.5, β'=0.5-0.5+4=4
Therefore, the updated values are
α'/(α'+β')=1.5/5.5=0.272
becomes. After that, if the user adds meal content with the attribute value of “meal-local” of “1” to the travel plan, n=1, x=1,
α'=1+1.5=2.5, β'=1-1+4=4
Therefore, the updated values are
α'/(α'+β')=2.5/6.5=0.384
becomes. After that, if the user adds meal content whose attribute value of "Meal-Local" is "0" to the travel plan, n = 1, x = 0,
α'=0+2.5=2.5, β'=1-0+4=5
Therefore, the updated values are
α'/(α'+β')=2.5/7.5=0.333
becomes. After that, if the user adds meal content with a local attribute value of "1" other than meals to the travel plan, n = 0.2, x = 0.2, and the value of "meal - local" is changed to n =0.2.

A specific example of calculation for updating values when behavior toward a model plan or article is detected will be described below.
For example, suppose that values related to "appearance (see FIG. 9)" included in the values of the model plan are to be updated, and α=1 and β=4 are set as initial values of parameters related to "appearance" for the user. At this time, the value is 1/(1+4)=0.2. When the user views a model plan or article with an attribute value of "1" for "brilliant", the suggestion device 200 calculates that n=0.5, x=0.5,
α'=0.5+1=1.5, β'=0.5-0.5+4=4
Therefore, the updated values are
α'/(α'+β')=1.5/5.5=0.272
becomes. After that, if the user adds a model plan or article whose attribute value of "brilliant" is "1" to favorites, n = 1, x = 1,
α'=1+1.5=2.5, β'=1-1+4=4
Therefore, the updated values are
α'/(α'+β')=2.5/6.5=0.384
becomes. After that, if the user adds the attribute value of "brilliant" to the favorites of "0", n = 1, x = 0,
α'=0+2.5=2.5, β'=1-0+4=5
Therefore, the updated values are
α'/(α'+β')=2.5/7.5=0.333
becomes. After that, if the user adds something other than the model plan or article to favorites, n=0.2 and x=0.2, but the values are not updated.

FIG. 11 is a diagram showing the relationship among the values of the user to be updated, examples of values, operations and processes, and weights. The suggestion device 200 may set weights on the types of values of users to be updated. The types of user values to be updated include, for example, the values of the main purpose/plan, the values of the spot category, the values of the category priority corresponding to the category, the values of the spot genre, and the values of the spot category. There is value in detailed information. For example, the suggestion device 200 sets the weight of the values of the main purpose/plan and the value of the spot category to "1", and the values of the category priority corresponding to the category, the values of the spot genre, and The value weight of the detailed information corresponding to the category is set to "0.5". When the suggestion device 200 receives operation information for viewing content related to the model plan, adding it to favorites, or adding it to spots in the travel plan, the suggestion device 200 updates the numerical value representing the values with the weight of the main purpose or the values of the plan. . Thereby, the suggestion device 200 can reduce bias in suggestions due to genres or values of detailed information.

The suggestion device 200 may update the numerical value representing the values using both the weight of the action and the weight corresponding to the type of values of the user to be updated.

FIG. 12 is a diagram for explaining persona estimation in the embodiment. The suggestion device 200 estimates one persona among a plurality of personas based on the user's values. The suggestion device 200 may estimate that the person does not correspond to any of the plurality of personas. When setting a persona for a user, the suggestion device 200 updates the user's values based on the user's behavior toward the content, and estimates the persona based on the updated user's values. At this time, the suggestion device 200 calculates the degree of similarity between the user's values and the values set in the persona, and selects the persona with the highest degree of similarity among the personas whose degree of similarity exceeds the threshold as the persona corresponding to the user. presume. For example, assume that the user's values for the updated categories are estimated to be 0.7 for "local," 0.3 for "photogenic," and 0.5 for "traditional." On the other hand, assume that personas A, B, and C also each have numerical values representing their values set as "local," "photogenic," and "traditional." The suggestion device 200 sets the threshold value to 0.5, for example, and estimates persona B with the highest similarity of 0.8.

FIG. 13 is a diagram for explaining the process of calculating the degree of similarity between the user's values and the persona's values. For example, the suggestion device 200 represents persona B's values by performing calculations in which the denominator is the sum of persona B's values and the numerator is the sum of the inner products of the user's values and persona B's values. Calculate numbers. The suggestion device 200 calculates a numerical value representing persona C's values by performing calculations in which the denominator is the sum of persona C's values and the numerator is the sum of the inner products of the user's values and persona C's values. calculate. Thereby, the suggestion device 200 can estimate that persona B's values are similar to the user's values.

FIG. 14 is a diagram for explaining the process of updating values and estimating a persona (step ST14) by the suggestion device 200 in the embodiment. The user terminal device 100 displays the content list screen received from the suggestion device 200 (step ST20). The suggestion device 200 receives, for example, operation information based on an action such as content viewing on a content list, and acquires content data about the content operated with the operation information from the content providing device 300. The suggestion device 200 confirms the values of the acted content (step ST30). Note that the suggestion device 200 may calculate the values of the content from information such as genre included in the content data. Next, the suggestion device 200 calculates the user's values by combining the values of the content, the user's behavior, and the user's values before the update (step ST31), and adds the calculated user values to the user's values after the update. It is updated as the user's sense of values (step ST32). The suggestion device 200 calculates the degree of similarity between the updated user's values and the persona's values (step ST33). The suggestion device 200 estimates a persona that has the highest degree of similarity to the user's values (step ST34).

The suggestion section 260 will be described in detail below.
FIG. 15 is a diagram showing the relationship between input data, suggestion score addition rules, and suggestion types. The suggestion unit 260 uses the user's original values estimated by the values estimation unit 242 as input data to make suggestions that capture the user's original values. The suggestion unit 260 calculates a suggestion score according to a point addition rule based on values that are common to all users based on values that match the scene and/or situation, and uses the calculated suggestion score to calculate a scene that is common to all users. and/or make suggestions that capture the situation. The suggestion unit 260 uses as input data temporary current values (see FIG. 4) based on a filter such as a category or genre specified by the user, and makes suggestions that capture the user's current values. The suggestion unit 260 calculates a suggestion score according to the point addition rule for each persona according to the scene and/or situation, and uses the calculated suggestion score to make suggestions that capture the scene and/or situation for each persona. The suggestion unit 260 calculates a suggestion score for the content included in the content list according to a point addition rule based on statistical data, and uses the calculated suggestion score to perform suggestions based on statistical data.

In addition to or in place of the scene and/or situation, the suggestion unit 260 calculates a suggestion score according to a point addition rule according to values common to all users from values according to the purpose of action, and uses the calculated suggestion score. By doing so, it makes suggestions that capture the common behavioral goals of users. When making suggestions for each persona, the suggestion unit 260 calculates a suggestion score according to the point addition rule for each persona according to the purpose of action in addition to or instead of the scene and/or situation, and uses the calculated suggestion score. , you can make suggestions that capture the behavioral purpose of each persona.

FIG. 16 is a diagram illustrating an example of a process for making suggestions in the embodiment, and FIG. 17 is a diagram illustrating an example of a process for calculating a suggestion score. First, in response to receiving a search request from the user terminal device 100, the suggestion device 200 acquires content data based on the search request, and calculates a suggestion score based on the user's sense of values (step ST40). Note that when a filter such as "meal" is set based on the user's operation, and when a filter such as genre or area is set, the suggestion device 200 excludes content data that does not correspond to the filter. . Further, the suggestion device 200 acquires operation information of the user terminal device 100 and information regarding a scene and/or situation from the environment information providing device 400, acquires content data based on the scene information and/or situation information, and generates a suggestion score. is calculated (step ST41). Further, if a persona is set for the user, the content providing device 300 calculates a suggestion score for each persona (step ST42). The suggestion device 200 calculates a suggestion score based on the statistical data of the content (step ST43). Suggestions based on statistical data are given uniform points regardless of user attributes. The suggestion device 200 adds points to the suggestion score according to the popularity of the content as statistical data, but the statistical data is not limited to the popularity of the content, and may be the popularity of content attributes. , the number of content views, or a value based on information obtained from an information providing device that provides content ranking data.

Note that the suggestion device 200 estimates the user's values, which are used in the process of calculating the suggestion score (step ST40), based on the selection operation of the estimation content and the user's behavior, but is not limited thereto. The suggestion device 200 can calculate a suggestion score based on the statistical values α and β by estimating the user's sense of values from the statistical values α and β as described above.

The suggestion device 200 calculates a final suggestion score by summing up a suggestion score based on the user's values, a suggestion score based on the scene and/or situation, a suggestion score for each persona, and a suggestion score based on statistical data. (Step ST44). The suggestion device 200 creates content list information such that the content with the higher suggestion score is ranked higher on the content list screen according to the final suggestion score (step S45). The suggestion device 200 suggests content by transmitting content list information to the user terminal device 100.
Thereby, the suggestion device 200 can suggest a content list by considering the user's original values, the scene and/or situation, the user's current values, and the persona's values.

FIG. 18 is a diagram showing an example of a content list screen. On the content list screen, the filtered content that matches the search request is displayed in descending order of suggestion score. The suggestion score is calculated based on the user's original values, the scene and/or situation, the user's current values, and the persona's values. The content list screen includes content that matches the genre, content that is suitable for accompanying persons (situations) such as children, and content that is popular (statistical data). It is desirable to display the reason for the usage restriction (scene) on the content list screen. For example, display the current scene or situation in each content, such as not being able to bring children, the weather or temperature being unsuitable (stormy weather, high or low temperatures), or the fact that it is after business hours or on a regular holiday. It's fine.

FIG. 19 is a diagram for explaining an example of a suggestion score addition rule based on a target user, a scene and/or situation, and a persona. The suggestion device 200 stores, for example, the target user, the schedule in the travel plan, the time, the weather, the temperature, the age of the moon, the positioning, the content to be added to points, and the added points in association with each point addition rule number. When the suggestion device 200 receives a search request, the suggestion device 200 determines the target user and persona, the date, time, weather, temperature, moon phase, and positioning in the travel plan, and if the content to be added is included in the content list for the search request, You can add points to the suggestion score. Thereby, the suggestion device 200 can suggest a content list by adding points to the suggestion score based on the target user, scene and/or situation, and persona.

As described above, according to the suggestion system of the embodiment, when a scene or situation in an application field and a user's behavior with respect to content are acquired, the user's values classified into the scene or situation in the acquired application field can be updated based on the values of the content corresponding to the behavior. Thereby, according to the suggestion system of the embodiment, it is possible to provide more preferable content from the viewpoint of user attributes and behavior.

According to the suggestion system of the embodiment, in response to a search request, the content suggestion score calculated based on the user's values estimated by the user attribute estimation unit 240 and the statistical data acquired by the statistical data acquisition unit 250 are used. A suggestion unit 260 is provided that suggests content to the user using the suggestion score of the content calculated by using the suggestion score of the content. Thereby, according to the suggestion system, highly reliable suggestions can be made in consideration of the user's sense of values.

Although each embodiment and each modification example have been described, it is only an example and is not limited to these. For example, each embodiment or each modification example, a part of each embodiment, or each modification example may be described. One aspect of the present invention may be realized by combining a part with one or more other embodiments or one or more other modifications.

Note that a program for executing each process of the user terminal device 100, the suggestion device 200, the content providing device 300, and the environmental information providing device 400 in this embodiment is recorded on a computer-readable recording medium, and the recording medium is The various processes described above regarding the user terminal device 100, the suggestion device 200, the content providing device 300, and the environmental information providing device 400 may be performed by loading and executing the program recorded in the computer system. .

Note that the "computer system" here may include hardware such as an OS and peripheral devices. Furthermore, the term "computer system" includes the homepage providing environment (or display environment) if a WWW system is used. Furthermore, "computer-readable recording media" refers to flexible disks, magneto-optical disks, ROMs, writable non-volatile memories such as flash memory, portable media such as CD-ROMs, hard disks built into computer systems, etc. storage device.

Furthermore, "computer-readable recording medium" refers to volatile memory (for example, DRAM (Dynamic
It also includes those that retain programs for a certain period of time, such as Random Access Memory). Further, the program may be transmitted from a computer system storing the program in a storage device or the like to another computer system via a transmission medium or by a transmission wave in a transmission medium.

Here, the "transmission medium" that transmits the program refers to a medium that has a function of transmitting information, such as a network (communication network) such as the Internet or a communication line (communication line) such as a telephone line. Moreover, the above-mentioned program may be for realizing a part of the above-mentioned functions. Furthermore, it may be a so-called difference file (difference program) that can realize the above-described functions in combination with a program already recorded in the computer system.

Although the embodiment of the present invention has been described above in detail with reference to the drawings, the specific configuration is not limited to this embodiment, and includes designs within the scope of the gist of the present invention.

DESCRIPTION OF SYMBOLS 100... User terminal device, 200... Suggestion device, 210... Operation reception unit, 220... Action history acquisition unit, 230... Scene/situation acquisition unit, 240... User attribute estimation unit, 242... Value estimation unit, 244... Persona estimation Department, 250...Statistical data acquisition section, 260...Suggestion section, 270...Values database management section, 272...Persona attribute management section, 280...Estimation content storage section

Claims (6)

a database management department that manages a database that categorizes user values corresponding to the field of application by scene or situation;
an estimation content providing unit that provides a plurality of estimation contents including images and catchphrases specific to the application field;
an operation reception unit that accepts a user's operation to select at least one of the plurality of estimation contents;
an estimation unit that estimates the user's sense of values defined by the database management unit based on the operation received by the operation reception unit;
an information acquisition unit that acquires information regarding a user's scene or situation in the application field;
a behavior acquisition unit that acquires user behavior with respect to content;
The estimating unit is configured to determine the scene or situation in the application field obtained by the information obtaining unit and the user's behavior regarding the content obtained by the behavior obtaining unit. updating the classified values of the user based on the values of the content corresponding to the behavior;
Value estimation system. The types of actions include viewing the content, adding the content to favorites or plans, and weights are set for each of the plurality of types of actions,
The estimation unit updates the values of the user classified into scenes or situations based on the values of the content and the weight set to the type of behavior.
The value estimation system according to claim 1. The user's values include multiple types of values,
Weights are set for each of multiple types of user values,
The estimating unit calculates the values of the user classified into scenes or situations based on the weights set for the types of values of the users, the values of the content, and the weights set for the types of actions. update,
The value estimation system according to claim 2. From claim 1, further comprising a persona estimating unit that estimates a persona with the highest degree of similarity among a plurality of personas based on the degree of similarity between the values updated by the estimating unit and the values corresponding to the persona. 3. The value estimation system according to any one of 3. Assuming that the values have a probability distribution of a beta distribution expressed by two statistical values based on operations received by the operation reception unit,
The estimation unit estimates the values using the two statistical values, updates the two statistical values based on the operation received by the operation reception unit, and uses the updated two statistical values. The values estimation system according to any one of claims 1 to 4, which updates the values. providing a plurality of estimation contents including images and catchphrases specific to the application field;
accepting a user's operation to select at least one of the plurality of estimation contents;
Based on the operation, estimating the user's values defined by a database management unit that manages a database in which the user's values corresponding to the field of application are classified by scene or situation;
obtaining information about the user's scene or situation in the field of application;
obtaining user behavior regarding the content;
When a scene or situation in the application field and the user's behavior with respect to the content are acquired, the values of the user classified into the scene or situation in the acquired application field are determined as the values of the content corresponding to the behavior. a step of updating based on
Value estimation methods, including:

Images