Recommendation Systems

Abstract

[Problem] To provide a system with which it is possible to improve the possibility of being updated to user information (user profile) that conforms to a person's likes and tastes. [Solution] This system comprises: a user information acquisition unit for acquiring user information that indicates a user's interest in content being provided; an output unit for outputting request information that requests, from the user, input of information relating to content for which the information amount is relatively less than that of an established comparison object, from among information for each item of content recorded in a recording medium as the user information; and an update unit for updating the user information on the basis of information inputted by the user in response to the request information.

Description

[Technical field]

[0001] The present invention relates to a recommendation system. [Background technology]

[0002] Conventionally, a technology is known that recommends products and the like to individual users by referring to user information (also called user profile) on a WEB site such as a shopping site. For example, Patent Document 1 discloses a configuration in which a profile of a target user is estimated using the profiles of other users and the profile of the target user is updated. In addition, Patent Document 2 discloses a technology for refining one's own profile by using the profiles of other users who have similar interests. [Prior art documents] [Patent documents]

[0003] [Patent Document 1] WO 03 / 017137 [Patent Document 2] JP 2000-132559 A Summary of the Invention [Problem to be solved by the invention]

[0004] In the configurations of Patent Document 1 and Patent Document 2, the hobbies and preferences of the target user are inferred using other people's user profiles (i.e., other people's history and other people's user information), so there is a possibility that the profile may not be updated to reflect the user's own hobbies and preferences.

[0005] The present invention has been made in consideration of the above-mentioned problems, and aims to provide a system that can increase the likelihood that user information (i.e., user profile) will be updated to match the individual's hobbies and preferences. [Means for solving the problem]

[0006] In order to achieve the above objective, the recommendation system includes a user information acquisition unit that acquires user information indicating a user's interest in the content to be provided, an output unit that outputs request information that requests the user to input information about content that has a relatively small amount of information compared to a predetermined comparison target, from the information for each piece of content recorded on a recording medium as the user information, and an update unit that updates the user information based on the information input by the user in response to the request information.

[0007] That is, the recommendation system requests the user to input information about content that has a relatively small amount of information compared to a predetermined comparison target in the user information, and updates the user information based on the information input by the user in response to the request. In other words, it is possible to increase information that reflects the user's preferences with respect to information about content that does not satisfy the criteria. [Brief description of the drawings]

[0008] [Figure 1] FIG. 1 is a block diagram showing a configuration of a recommendation system. [Diagram 2] FIG. 2A is a diagram showing an example of content information, and FIG. 2B is a diagram showing an example of user information. [Diagram 3] 13 is a flowchart illustrating an example of a user information update process. [Figure 4] FIG. 11 is a diagram showing another example of user information. [Diagram 5] 13 is a flowchart showing another example of a user information update process. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0009] Here, the embodiments of the present invention will be described in the following order. (1) Configuration of recommendation system: (2) User information update process: (3) Other embodiments:

[0010] (1) Configuration of recommendation system: FIG. 1 is a block diagram showing the configuration of a recommendation system 10 according to the present invention. In this embodiment, the recommendation system 10 is configured as a server computer, and recommends content to a user of a navigation system 50 as a client. There may be multiple clients of the recommendation system 10, but FIG. 1 shows one of them. The content recommended in this embodiment is a facility (POI (Point Of Interest)) that can be set as a destination or the like in the navigation system.

[0011] The recommendation system 10 includes a control unit 20 including a CPU, RAM, ROM, etc., a recording medium 30, and a communication unit 41, and the control unit 20 can execute a program stored in the recording medium 30 or the ROM. In this embodiment, the program that can be executed is a recommendation program 21. The communication unit 41 includes a circuit for communicating with a navigation system 50, and the control unit 20 can communicate with the navigation system 50 by processing the recommendation program 21.

[0012] The navigation system 50 is a device equipped with a navigation function to a destination. The navigation system 50 may be configured as an in-vehicle device, or may be configured as a portable terminal such as a tablet or a smartphone. The navigation system 50 includes a communication unit 50a, a GNSS receiving unit 50b, a control unit 50c, and a user I / F unit 50d. The control unit 50c includes a CPU, a ROM, a RAM, and a recording medium, and the CPU can execute various programs including a navigation program recorded in the ROM or the recording medium.

[0013] The communication unit 50a includes a communication circuit for wireless communication with other devices. The navigation system 50 can communicate with the recommendation system 10 via the communication unit 50a. The user I / F unit 50d is an interface unit through which the user inputs instructions and provides various information to the user. The user I / F unit 50d includes a touch panel display, switches, speakers, etc. (not shown). That is, the user I / F unit 50d includes an output unit for images and sounds and an input unit for user instructions.

[0014] The GNSS receiver 50b is a device that receives signals from the Global Navigation Satellite System. The GNSS receiver 50b receives radio waves from navigation satellites and outputs a signal for calculating the position of the navigation system 50 via an interface (not shown). By executing a navigation program, the control unit 50c acquires this signal and acquires the position of the navigation system 50. The control unit 50c also acquires time from a clock unit (not shown). Based on the position and time, the control unit 50c estimates that the user of the navigation system 50 is staying at the position when the position of the navigation system 50 remains within a predetermined range for a predetermined time or more. The control unit 50c identifies a facility that exists at the position based on map information. The map information may be recorded in a recording medium of the navigation system 50, or may be configured to be acquired from a navigation server (not shown). In any case, the map information includes at least facility data including the position, name, facility ID, telephone number, facility genre, etc. of a facility that can be a POI. The control unit 50c associates the facility ID of the facility where the user is estimated to have visited or stayed with the user ID, and transmits the facility ID to the recommendation system 10.

[0015] Furthermore, when a user operates the user I / F unit 50d to input a review of a facility that the user has visited into the navigation system 50, the control unit 50c transmits the review data to the recommendation system 10 in association with the user ID. In this embodiment, the review data includes at least text data. Of course, image data taken during the user's stay at the facility may also be included in the review data. Furthermore, when a user operates the user I / F unit 50d to input a favorite facility genre, the control unit 50c transmits the user's favorite facility genre to the recommendation system 10 in association with the user ID.

[0016] When a preset condition for presenting recommended facilities is met in the navigation system 50 (for example, when the user performs a specific operation such as a facility search for setting a destination), the control unit 50c transmits a recommendation request to the recommendation system 10 together with the user ID. In this embodiment, the recommendation request includes a target area, and the recommendation system 10 acquires facilities in the target area as recommended facilities. The control unit 50c displays the facilities returned from the recommendation system 10 in response to the recommendation request on the display unit of the user I / F unit 50d, or outputs them by voice. The preset condition may be a specific operation such as a facility search for setting a destination performed by the user, a preset time has passed, or a preset distance has been traveled. The target area may be an area specified by the user, or a preset range including the current location.

[0017] The recommendation system 10 acquires the data transmitted from the navigation system 50 via the communication unit 41. The recording medium 30 of the recommendation system 10 stores map information 30a, content information 30b, and user information 30c.

[0018] The map information 30a includes node data indicating the positions of nodes corresponding to the endpoints of road sections, shape interpolation point data indicating the positions of shape interpolation points for identifying the shape of roads between nodes, link data indicating links connecting nodes, etc. Here, the nodes correspond to intersections, and the links correspond to road sections from intersection to intersection. The map information 30a is used for identifying the position of a vehicle, searching for a route to a destination, route guidance, etc. The map information 30a further includes facility information. The facility information includes a facility ID for identifying a facility, facility coordinates, facility genre (e.g., restaurant, convenience store, etc.), etc.

[0019] The content information 30b is data in which each piece of content, that is, in this embodiment, the facility ID of each facility, is associated with various information related to the facility. FIG. 2A is a diagram showing an example of the configuration of the content information 30b. The content information 30b includes multiple records corresponding to each of the multiple facilities (in this embodiment, eating and drinking facilities). The records correspond to each row shown in FIG. 2A. In each record, a facility ID, facility coordinates, facility genre (category), number of stores, review data, total number of reviews, and keywords (attributes) are associated with each other.

[0020] The facility ID is information for identifying a facility, and corresponds to the facility ID in the map information 30a. The facility coordinates are information indicating the location coordinates of the providing facility, and correspond to the facility coordinates included in the map information 30a. The facility genre is information indicating the type of product provided by the providing facility. In this embodiment, the facility genre includes Italian, French, Chinese, hamburger, and ramen as food genres. However, the facility genre is not limited to this, and the definition of the facility genre is not limited to the embodiment. As another example, it may be indicated by a larger classification (or a finer classification) such as fast food and dessert. In other words, the facility genre may have a hierarchical structure.

[0021] The number of stores is the number of stores corresponding to each facility genre. For example, when the facility genre is Italian restaurants, the number of stores is 50. The review data is data indicating the user's impressions of the facility and evaluations based on scores, etc., and text data and scores associated with the user ID of the user who wrote the review are stored. In FIG. 2A, the review data includes reviews for each facility ID, while the number of stores is associated with a common value for facilities of the same genre. The review data is transmitted to the recommendation system 10 from the navigation system 50 and various other information terminals in association with the user ID.

[0022] The total number of reviews is the cumulative number of reviews from each user, and it can be said that the more reviews a facility has, the more well-known the facility is. Each user can post only one review for each facility. Keywords are terms preset by the facility (e.g., close to the station, reasonable, hideaway, etc.) or terms extracted from the user's review data (e.g., good value for money, etc.). Although not shown in the configuration example of FIG. 2A, the content information 30b may also include the name, address, telephone number, etc. of the facility in association with the facility ID.

[0023] The user information 30c is data in which various information related to a user is associated with a user ID, and is also called a user profile. FIG. 2B is a diagram showing an example of the configuration of the user information 30c. The user information 30c shown here records data on multiple users (user information 1, user information 2, ...), and here, information on a specific target user (user information 1) is described as an example. The user information 30c includes multiple records corresponding to multiple facility genres (in this embodiment, the genre of dining facilities). The records correspond to each row shown in FIG. 2B. In each record, an interest level indicating the user's interest is associated with an interest attribute (keyword). That is, in this embodiment, the content to be provided is information related to the facility, and the user information 30c indicates an interest in the facility, which is the content.

[0024] The interest level indicates whether or not there is an interest in a facility genre, that is, it is an index showing the user's preference or the user's interest level in the corresponding facility genre in multiple stages. For example, in the example shown in FIG. 2B, the interest level in Italian food is shown as "○", the interest level in Chinese food is shown as "△", and the interest level in ramen is shown as "◎". Of the symbols indicating the interest level, "◎" indicates the highest interest, followed by "○" and then "△", in that order. Although not shown here, the interest level is lowest when the symbol is "×". Furthermore, the symbol "-" for hamburgers and French cuisine means that there is no data yet for these hamburgers and French cuisine (i.e., it is unknown).

[0025] The attribute of interest indicates an attribute in which the user is interested, and for example, the user's favorite facility information (store name) in which the user is interested, and keywords in which the user is interested (for example, close to a station, high rating, etc.) are recorded. In this embodiment, the interest level and the attribute of interest are recorded based on information transmitted from the navigation system 50. Note that the attribute of interest may also include facilities that the user has used or visited in the past, facilities registered in favorites, and facilities for which the user has created review data, etc., associated with the user ID.

[0026] The recommendation program 21 executed by the control unit 20 of the recommendation system 10 is a program for causing the computer of the recommendation system 10 to function as a user information acquisition unit 21a, an output unit 21b, and an update unit 21c. In other words, in the following, the processes described as being performed by the user information acquisition unit 21a, the output unit 21b, and the update unit 21c are processes executed by the control unit 20 (CPU).

[0027] The user information acquisition unit 21a acquires user information indicating the user's interest in the content to be provided. Specifically, the user information acquisition unit 21a acquires user information indicating the user's interest in a dining facility, which is an example of the content to be provided. The content to be provided is information to be provided to the user or other devices in any form. In addition, in this embodiment, the user information 30c is updated based on the user information 30c for each user. For this reason, the user information acquisition unit 21a identifies the user whose user information is to be updated by the user ID, and acquires the user information 30c associated with the user ID. That is, the user information acquisition unit 21a acquires the interest level and the interest attribute associated with the user ID. In addition, the user information acquisition unit 21a may acquire the facility genre or keyword of interest input by the user as the interest attribute, or may automatically acquire the interest attribute from the history of the user's actions transmitted from the navigation system 50, for example. For example, if the user has used or visited a facility of the same facility genre more than a predetermined number of times in the past, the facility genre may be acquired as the user's interest attribute.

[0028] The output unit 21b outputs request information that requests the user to input information about content that does not satisfy a predetermined criterion in the user information. The predetermined criterion is a criterion for identifying user information to be updated, and in this embodiment, is a criterion for identifying content in the user information 30c that is deficient in the amount of information. The lack of information is determined, for example, based on whether the amount of information is relatively small compared to a default comparison target. The default comparison target for determining the lack of information may be specified based on the content information 30b or the user information 30c.

[0029] Since the content information 30b is defined by associating various information with each content, if the content information 30b has a large amount of information, the content is highly known, and the amount of information of the well-known content is likely to be relatively large in the user information 30c as well. Therefore, the amount of information of the content recorded in the content information 30b can be used as a criterion for whether the amount of information of the content in the user information 30c is insufficient. In this case, for example, if the amount of information of the content recorded in the content information 30b is not a predetermined ratio (e.g., 10%) or more of the amount of information of the content recorded in the content information 30b, it can be considered that the user information 30c of the content is insufficient.

[0030] On the other hand, when the amount of information of a particular content is relatively large in the user information 30c, but the amount of information of other content is relatively small, it may be difficult to determine the user's preference based on the latter content. Therefore, for example, a predetermined threshold may be used as a standard, and the amount of information of each content recorded in the user information 30c may be compared with the predetermined threshold, and content with less information than the predetermined threshold may be deemed to be insufficient. Alternatively, the amount of information of each content recorded in the user information 30c may be compared, and the content with the least amount of information or a predetermined number of contents in order of least amount of information may be deemed to be insufficient.

[0031] In any case, a criterion for identifying the content with a lacking amount of information is determined in advance, and the output unit 21b acquires the lacking content (information-lacking area) in the user information 30c based on the criterion. Then, the output unit 21b outputs a request for the user to input information regarding the lacking information. At that time, the output unit 21b asks the user a question regarding the content of the lacking information.

[0032] The update unit 21c updates the user information 30c based on the information input by the user in response to the request information from the output unit 21b. That is, the update unit 21c updates the user information 30c (i.e., the user profile) using the information input by the user. If the answer input by the user to the request from the output unit 21b is a positive answer, the update unit 21c updates the information on the interest level and interest attribute regarding the content to which the answer was obtained. The interest level can be indicated in multiple stages as described above. Therefore, if the answer from the user is a positive answer, the interest level in the corresponding facility genre is increased, for example, to "◎" or "○", and the information is updated as the user information. Similarly, if the answer input by the user to the request from the output unit 21b is a negative answer, the interest level in the corresponding facility genre is decreased, for example, to "△" or "×", and the information is updated as the user information. In addition, the update unit 21c accepts an answer to a question from the output unit 21b and updates the information based on the answer.

[0033] When updating the user information 30c, the control unit 20 updates the user information using a profile of another person similar to the target user in a conventionally known system (for example, Patent Document 1 and Patent Document 2 mentioned above). On the other hand, when updating the user information of the target user using a profile of another person, the user information may not be updated to match the hobbies and preferences of the target user. Therefore, in this embodiment, when the control unit 20 updates the user information 30c, the user information is updated to match the hobbies and preferences of the target user himself / herself.

[0034] Specifically, the control unit 20 acquires user information of the target user from the user information 30c, and ascertains the missing information from the acquired user information. Here, the control unit 20 ascertains the information that is missing in the user information of the target user with respect to the content information 30b. Specifically, the control unit 20 compares the content information 30b with the user information 30c, and ascertains the information that is missing (or the missing area) in the user information 30c of the target user.

[0035] 2A and 2B, the content information 30b in FIG. 2A describes information on contents corresponding to Italian, French, Chinese food, hamburgers, and ramen. In contrast, the user information (user information 1 in this embodiment) of the target user in FIG. 2B describes information on contents corresponding to Italian, Chinese food, and ramen, but does not describe information on the contents of French and hamburgers. That is, in the user information 1, there is no information on whether the target user is interested in French and hamburgers, and therefore information on French and hamburgers is insufficient. That is, it is unclear whether the target user is interested in French and hamburgers. In other words, French and hamburgers are areas of insufficient information in the user information 1. When making such a judgment, the criterion for whether there is insufficient information is whether the amount of information for each content recorded in the content information 30b is less than a predetermined ratio. Therefore, in addition to simply not having data like French and hamburgers in FIG. 2B, if data less than a predetermined ratio is recorded in the user information 30c with respect to the number of target facility information (number of stores) recorded in the content information 30b, it is judged that the criterion is not met. On the other hand, for example, when a predetermined ratio or more of data is recorded in the user information 30c with respect to the number of pieces of facility information of interest recorded in the content information 30b, it is determined that the criterion is met.

[0036] Next, the control unit 20 extracts items (stores) that can be output to the user from the user's current location or current time zone from the missing information. That is, the control unit 20 extracts stores recommended from the missing information of hamburgers and French food to the target user's current location. Alternatively, the control unit 20 extracts stores recommended from the missing information of hamburgers and French food to the target user's current time zone. Specifically, the control unit 20 extracts stores within a predetermined distance from the user's current location or stores that can provide hamburgers or French food in the current time zone (e.g., lunchtime) from the facilities that provide hamburgers and the facilities that provide French food in the content information 30b.

[0037] Next, the control unit 20 asks the user a question about the content of the extracted store. For example, in this embodiment, it is assumed that a store called "ABC Burger" is extracted as a store that serves hamburgers as a result of extracting stores under the above-mentioned condition of "the user's current location or current time zone among the missing information". In this case, the control unit 20 asks the target user whether or not he / she is interested in the content (i.e., facility genre, category) of hamburgers among the missing information. That is, for content that does not satisfy a predetermined standard as the user information 30c, output is performed to request the user to input information on whether or not he / she is interested in the content (in other words, whether or not he / she is interested). Note that this question (output) is output to the navigation system 50 via the communication unit 41. Then, when the communication unit 50a of the navigation system 50 acquires the question, the question is output to the user I / F unit 50d under the control of the control unit 50c. When it is assumed that the navigation system 50 is mounted on a vehicle, it is preferable that the output to the user I / F unit 50d is output from a speaker from the viewpoint of the operability of the vehicle. Of course, the information may be output in various other ways, such as from a display, or from both a speaker and a display.

[0038] Next, the control unit 20 judges whether there is an answer from the user. That is, the control unit 20 judges whether there is an answer to the question about the missing information to the user. That is, in the above example, the control unit 20 judges whether there is any feedback from the user, such as whether the user is interested or not interested in the question "Are you interested in hamburgers?". The answer from the user is transmitted from the navigation system 50 to the recommendation system 10. In addition, the user may answer the above-mentioned selection of "interested" or "not interested" by voice input, or by input by operation on the user I / F unit 50d. Alternatively, the user may answer by input by some kind of gesture such as a gesture.

[0039] Next, when the user answers the question about the missing information, the control unit 20 stores the answer input by the user as user information. That is, the control unit 20 learns the user's answer to the question of the output unit 21b in the control unit 20 and updates it as user information. In other words, the control unit 20 supplements the missing information in the user information 30c. In this embodiment, the update unit 21c in the control unit 20 accepts the answer to the question about hamburgers (are you interested in hamburgers) which is the missing information in this embodiment as input data to the user information 1. When the answer input by the user to the request from the output unit 21b is an affirmative answer, the update unit 21c updates the information on the interest level and interest attribute regarding the hamburger to which the answer was obtained. The interest level can be expressed in multiple stages as described above. Therefore, when the answer from the user to the hamburger is an affirmative answer, the interest level in the hamburger becomes "◎" or "○", and the update unit 21c updates the information as user information. On the other hand, if the answer input by the user is negative, the degree of interest in hamburgers becomes "Δ" or "X", and the update unit 21c updates this information as user information.

[0040] Next, when the user answers the question to the effect that the user is interested, the control unit 20 recommends a store in the content of interest to the user. That is, when the user answers that the user is interested in the missing information, the control unit 20 recommends the store to which the answer was given. In this embodiment, as described above, "hamburger" and "French" are the missing information, and "ABC burger" is extracted in relation to the user's current location, and a question is asked as to whether the user is interested in "hamburger". Therefore, the output unit 21b in the control unit 20 recommends "ABC burger" as a specific item in response to the answer from the user. Also, when the user sends a response based on the recommendation, the update unit 21c in the control unit 20 accepts the response as input data to the user information 30c, that is, updates the user information 30c.

[0041] Next, the operation of this embodiment will be described. As described above, in this embodiment, the user information 30c of the target user is compared with the content information 30b to grasp the missing information. In addition, the user information 30c is updated based on the response of the target user himself / herself as to whether or not he / she is interested in the missing information. Therefore, in the user information 30c of this embodiment, the missing information can be supplemented, and user information that matches the user's hobbies and preferences can be constructed. In other words, compared to the conventional systems described in Patent Documents 1 and 2 above, where user information is updated based on other people's user information, more accurate user information can be constructed.

[0042] In addition, specific restaurants in the dining facility are recommended to the user based on the updated user information, allowing the user to select a restaurant that matches his or her own hobbies and preferences.

[0043] Furthermore, in this embodiment, when the recommendation system 10 asks the navigation system 50 a question about missing information, a speaker or a touch panel display is used as a user interface. In particular, when a speaker is used as a user interface, questions are output by voice, and therefore, for example, when the navigation system 50 is mounted on a vehicle, passengers can ask and answer questions by voice, which can reduce or avoid disturbance to driving, etc., and improve driving safety.

[0044] Furthermore, in conventionally known systems, content and items are recommended using user information of similar other users, so the target user is more likely to receive suggestions for known content and items. On the other hand, in this embodiment, recommendations for missing information are made based on the target user's own input, so the target user can receive unknown suggestions that interest him / her, compared to when content and items are recommended using user information of similar other users. In other words, the target user can receive new and unexpected content or items that interest him / her.

[0045] (2) User information update process: Next, the user information update process executed by the control unit 20 will be described. Fig. 3 is a flowchart showing an example of the user information update process executed by the control unit 20. In this embodiment, the user information update process is started, for example, when the navigation system 50 is started. Alternatively, in the case where the navigation system 50 is mounted on a vehicle, the process is started when the accessory power supply or the ignition power supply of the vehicle is turned on, etc.

[0046] When the user information update process is started, first, the control unit 20 acquires user information of the target user from the user information 30c, and grasps the missing information from the acquired user information (step S1). This is a step for grasping the missing information when updating the user information of the target user. Specifically, the control unit 20 compares the user information 30c with the content information 30b, and grasps the user information 30c that is missing from the content information 30b. In addition to judging whether there is missing information in the user information 30c when there is no information corresponding to the content information 30b, it is also judged that there is missing information when there is corresponding information but the amount of information does not meet a predetermined standard. In the example of FIG. 2A and FIG. 2B, as described above, in the user information 1 of the target user, it is unclear whether the target user is interested in facilities that provide hamburgers and French cuisine, and the interest attribute is also unclear. Therefore, in the example of FIG. 2A and FIG. 2B, hamburgers and French cuisine correspond to the missing information of the target user.

[0047] Next, the control unit 20 extracts items (stores) that can be output to the user from the user's current location or current time zone from the missing information (step S2). That is, the control unit 20 extracts stores that can be recommended from the target user's current location or current time zone for hamburgers and French cuisine, which are the missing information. Specifically, the control unit 20 acquires the target user's current location, i.e., the location of the navigation system 50, and extracts a store corresponding to the target user's acquired current location from the facility coordinates corresponding to the facility ID. In this embodiment, the store extracted in this way is assumed to be "ABC Burger."

[0048] Next, the control unit 20 asks the user a question about the content of the extracted item (step S3). In this embodiment, "ABC Burger" is extracted in step S2. That is, since the facility genre of the content of the extracted item (ABC Burger) is hamburgers, the target user is asked whether or not he is interested in hamburgers. In other words, the control unit 20 requests the target user to input information about the content that does not meet the criteria (missing information). This request (i.e., output to the user) is output to the user I / F unit 50d, such as a speaker of the navigation system 50, for example.

[0049] Next, the control unit 20 judges whether there is a response from the user (step S4). That is, the control unit 20 judges whether there is a response to the request for the target user to input information about content that does not satisfy the criteria. Specifically, the control unit 20 judges whether there is a response from the target user to the question "Are you interested in hamburgers?" In other words, it judges whether there is any feedback from the target user. If the judgment in step S4 is negative, that is, if there is no response from the target user, the control unit 20 returns without executing the subsequent processing.

[0050] On the contrary, if the answer is affirmative in step S4, that is, if there is an answer from the target user, the content of the answer is stored as the user information of the target user, and the user information 30c is updated (step S5). That is, the control unit 20 accepts the answer from the target user as input data from the target user, and updates the user information 30c. In this embodiment, if there is an answer such as "interested" or "not interested" in hamburgers, the control unit 20 accepts either answer as input data of the target user, and updates the information as the user information 30c. That is, in this embodiment, the user information 30c is updated based on the input data of the target user himself. As a result, at least a part of the missing information is updated. The routine in the flowchart of FIG. 3 is repeatedly executed at predetermined short intervals at least until the missing information is completed.

[0051] In addition, in step S5, if the target user answers in step S4 and the answer is positive (i.e., the target user is interested in hamburgers), a corresponding content item may be recommended to the target user. That is, in this embodiment, "ABC Burger" may be recommended to the target user. In addition, when recommending ABC Burger, facility information and routes for ABC Burger may also be displayed on the display of the navigation system 50.

[0052] In this manner, in the present embodiment, the user information 30c is updated based on the input data of the target user himself / herself. Therefore, compared to a case where the user information 30c of the target user is updated using the user information of another person, the user information 30c is more in line with the hobbies and preferences of the target user himself / herself.

[0053] (3) Other embodiments: Next, other embodiments will be described. In the above embodiment, when updating the user information 30c, the control unit 20 acquires the missing information from the user information of the target user himself / herself, and asks the user whether or not he / she is interested in the content of the missing information. Then, the control unit 20 updates the user information based on the user's answer to the question. On the other hand, if the user information can be updated to match the hobbies and preferences of the target user himself / herself, the acquisition of the missing information is not limited to being based only on the user information of the user, but may also include user information of other people. For example, the control unit 20 may update the user information of the target user by considering user information of other people similar to the target user in addition to the user information of the target user himself / herself. A specific description will be given below.

[0054] First, the user information 30c records data on a plurality of users as described above, and includes user information on other people (hereinafter, referred to as similar users) who are similar to the target user. Here, the similar user means a user who has a similar interest in content to the target user. The similarity of interest can be evaluated for at least one of parameters indicating interest, such as the interest level and the attribute of interest, and when the degree of match of interest is evaluated by the parameter indicating interest, it can be said that the interests are similar if the degree of match is equal to or greater than a threshold. The degree of match may be defined by various methods. For example, when the interest level and the attribute of interest match between the target user and the similar user, the degree of match is 100%. In addition, the degree of match may be specified by the ratio of the number of times that the interest level and the attribute of interest match between the target user and the similar user. When there is an item in which the interest level and the attribute of interest are evaluated with multi-level values, when the deviation of the values ​​between the target user and the similar user is equal to or less than a threshold, the interest in the item may be evaluated as being similar.

[0055] FIG. 4 is an example of user information 30c including the user (user information 1) of the target user and user information (user information 2) of a similar user. User information 1 is the same as the user information described in FIG. 2B. In the example shown in FIG. 4, the user information of the similar user has a degree of interest in the facility genre "ramen" of "◎", which is the same as the target user's degree of interest in "ramen". Similarly, the user information of the similar user has a degree of interest in the facility genre "Chinese food" of "△", which is the same as the target user's degree of interest in "Chinese food". Note that for other facility genres, there is no information on the degree of interest and attribute of interest for at least one of the target user and the similar user. Thus, in the example shown in FIG. 4, the interest levels of the users shown in user information 2 match in 2 / 3 of the facility genres ("Italian", "ramen" and "Chinese food") in which the target user is interested. For this reason, in this example, user information 2 is assumed to be user information similar to user information 1.

[0056] Although the example shown in FIG. 4 illustrates an example in which the user information of one similar user is included, the number of user information of similar users included in the user information 30c is not limited to one and may be multiple.

[0057] Next, the user information update process executed by the control unit 20 will be described. Fig. 5 is a flowchart showing an example of the user information update process, and the flowchart shown in Fig. 5 differs from the flowchart in Fig. 3 in that it includes a process based on the user information of a similar user. Note that, like the example in Fig. 3, the user information update process shown in Fig. 5 is started, for example, when the navigation system 50 is started, or when the accessory power source or ignition power source of the vehicle equipped with the navigation system 50 is turned on. The contents of the process will be specifically described below.

[0058] When the user information update process is started, first, the control unit 20 calculates a target to be recommended to the target user (step S10). In this embodiment, any of the information recorded as the content information 30b is calculated as the content to be recommended. In addition, in this embodiment, since the aim is to update the user information 30c to make up for the lack of the user information 30c, the control unit 20 calculates the content corresponding to the missing information that is missing in the user information 30c. Specifically, the control unit 20 uses the function of the user information acquisition unit 21a to compare the user information 30c and the content information 30b and acquires the missing information that is missing in the user information 30c of the target user with respect to the content information 30b. The content information 30b is information generated based on the user information of each user, and information about French and hamburgers is recorded in the content information 30b shown in FIG. 2A. In contrast, information about French and hamburgers is not recorded in the user information 1 of the target user shown in FIG. 4, that is, information is missing. In addition, the determination of whether there is missing information may be made not only when there is no information corresponding to the content information 30b, but also when there is corresponding information but the amount of information does not meet a predetermined standard. In the example shown in FIG. 4, as described above, in the user information 1 of the target user, it is unclear whether the target user is interested in facilities that provide hamburgers and French cuisine, and the interest attribute is also unclear. In other words, in the example of FIG. 4, "hamburgers and French cuisine" corresponds to the missing information for the target user.

[0059] Then, when the control unit 20 acquires the missing information, the control unit 20 calculates a target to be recommended based on the acquired missing information by using the function of the output unit 21b. As described above, in the example of FIG. 4, "hamburger and French" corresponds to the missing information. Therefore, in this step S10, the control unit 20 calculates (in other words, extracts) hamburger and French as targets to be recommended to the target user.

[0060] After calculating the target to be recommended to the target user, the control unit 20 advances the process to step S11 and step S17. That is, in the embodiment shown in FIG. 5, after the target to be recommended is calculated, two processes are executed in parallel. Here, the processes after step S11 (steps S11 to S16) will be explained first, and after explaining the processes after step S11, the processes after step S17 (steps S17 to S18) will be explained. In addition, in the processes after step S11, the content calculated in step S10 is used to create a question for the target user. In the processes after step S17, the content calculated in step S10 is used for the recommendation.

[0061] In step S11, the control unit 20 extracts items (stores) that can be output to the target user from the current location of the target user from the missing information (step S11). That is, the control unit 20 extracts stores that can be recommended from the current location of the target user for the missing information of hamburger and French, using the function of the output unit 21b. Specifically, the control unit 20 acquires the current location of the target user, that is, the location of the navigation system 50. Then, the control unit 20 refers to the content information 30b and extracts stores corresponding to the acquired current location of the target user (for example, stores within a radius of 5 km) from the facility coordinates corresponding to the facility ID. In the embodiment shown in FIG. 5, it is assumed that the stores extracted in this way are "ABC Burger" and "XYZ French".

[0062] Next, the control unit 20 extracts items (stores) that can be output to the target user from the current location of the target user from the missing information in the user information of the similar user (step S12). That is, the control unit 20 extracts stores that can be output to the target user from the user information of the similar user by using the function of the output unit 21b. Specifically, the control unit 20 refers to the user information 30c by using the function of the user information acquisition unit 21a to acquire the user information of the similar user (i.e., the user information 2 described in the example of FIG. 4). Then, the control unit 20 compares the acquired user information of the similar user with the content information 30b to acquire the missing information that is missing in the user information of the similar user with respect to the content information 30b. It is to be noted that the judgment as to whether or not there is missing information is made when the amount of information of the corresponding content does not satisfy a predetermined standard. In the example shown in FIG. 4, in the user information 2 of the similar user, it is unclear whether or not there is interest in facilities that provide Italian and French cuisine, and the interest attribute is also unclear. That is, in the example of FIG. 4, Italian and French cuisine correspond to the missing information of the similar user.

[0063] Then, when the control unit 20 acquires the missing information in the user information of the similar user, the control unit 20 uses the function of the output unit 21b to extract restaurants that can be output to the target user from the current location of the target user from the missing information of the similar user. That is, the control unit 20 extracts restaurants that can be recommended from the current location of the target user, targeting Italian and French restaurants that are the missing information of the similar user. Specifically, the control unit 20 refers to the content information 30b, and extracts restaurants that correspond to the current location of the target user acquired in step S11 (for example, restaurants within a 5 km radius) from the facility coordinates corresponding to the facility ID. In the embodiment shown in FIG. 5, the restaurant extracted in this way is assumed to be "XYZ French."

[0064] Next, the control unit 20 extracts (generates) content corresponding to the attribute of interest based on the missing information extracted in steps S11 and S12 (step S13). Specifically, the control unit 20 extracts missing information corresponding to both the missing information in the user information of the target user extracted in step S11 and the missing information in the user information of the similar user extracted in step S12 by the function of the output unit 21b. Then, the control unit 20 extracts one of the multiple attributes of interest with reference to the user information 30c in FIG. 4, associates the missing information corresponding to both the target user and the similar user with the extracted attribute of interest, and generates content corresponding to the attribute of interest. More specifically, the control unit 20 extracts information that is missing in the user information of both the target user and the similar user (i.e., information about content that does not meet the criteria as an amount of information) from the items extracted in steps S11 and S12 with reference to the user information 30c. As described above, "ABC Burger" and "XYZ French" are extracted as the missing information of the target user, and "XYZ French" is extracted as the missing information of the similar user. Referring to the user information 30c in FIG. 4, the user information 2 of the similar user records information on the interest level and interest attribute for the facility genre "hamburger". Since it is assumed that the similar user has similar interests to the target user, the interest level of the target user in "hamburger" is currently unknown, but it can be estimated from the interest level of the similar user that the interest level of the target user is similar. In other words, since the interest level of the similar user in "hamburger" is "◎", it is also possible to estimate that the interest level of the target user in "hamburger" is "◎" or "○". By making such an estimation, the control unit 20 considers that the information to be collected preferentially is the content that is missing in the user information of both the target user and the similar user. As a result, it is possible to preferentially collect information that cannot be estimated from the user information 30c of both the target user and the similar user. In the above example, "French" is extracted as content that is lacking in the user information of both the target user and the similar users, and is regarded as content to be collected with priority.Then, the control unit 20 refers to the user information 30c in FIG. 4, for example, associates the interest attribute "close to the station" with "French food" and extracts "French food close to the station" as content corresponding to the interest attribute.

[0065] Here, the information output by the control unit 20 using the function of the output unit 21b may include information such as distance information such as closeness to a station, facility genre, facility name, facility price range, and facility reviews, and includes at least one of these pieces of information. The facility genre is, for example, a food genre such as "French" or "hamburger", but is not limited to this and may be a larger classification such as fast food. The facility name is the name of the facility, such as "XYZ French" in the above example. The facility price range is a conceptual one such as "reasonable" or a specific amount such as "5000 yen to 6000 yen". In addition, the facility review may be a specific evaluation score such as "highly rated", "high score", or "4 or more out of 5 points". These pieces of information output by the control unit 20 correspond to "information related to the content".

[0066] In addition, considering that the information output by the control unit 20 is recommended to the target user and that the user information of the target user is updated, it is preferable that the information corresponds to either the interest attribute of the target user (user information 1) or the interest attribute of a similar user (user information 2), but it does not necessarily have to be information recorded here. For example, it may be information indicating a time period such as "lunch" or "dinner" that is not recorded in the user information shown in FIG. 4.

[0067] Next, the control unit 20 asks the target user a question about the item of the extracted content (step S14). Specifically, the control unit 20, by using the function of the output unit 21b, asks the target user a question about the item "XYZ French" of the content "French restaurants near stations" extracted in step S13. For example, the control unit 20 asks the target user a question about whether or not the user is interested in French food, such as "What do you think of XYZ French (are you interested)?". In addition, in the case where the "XYZ French" store has information about the store (such as information about a specialty menu), the control unit 20 may ask a question about information about the store (such as "How about the specialty XX quiche?"). In this way, the control unit 20, by using the function of the output unit 21b, asks the target user a question about the missing information that does not meet the information amount criteria for both the target user and similar users. Then, the control unit 20 requests the target user to input an answer to the question. The request (ie, output) to the target user is output to a user I / F unit 50d such as a speaker or a display unit of the navigation system 50, for example.

[0068] Next, the control unit 20 judges whether there is a response from the target user (step S15). That is, the control unit 20 judges whether there is a response to the request for input of information on content that does not satisfy the criteria for the target user by using the function of the output unit 21b. Specifically, the control unit 20 judges whether there is a response from the target user, such as interest or disinterest, to the question, for example, "How is XYZ French?" by using the function of the output unit 21b. The response from the target user is transmitted from the navigation system 50 to the recommendation system 10. In addition, the user may answer the above-mentioned selection of "interested" or "disinterested" by voice input, or may answer by input operation to the user I / F unit 50d. This input operation may be an input operation such as inputting specific text text or touching a display such as "good / bad". Alternatively, the response may be an input by some kind of gesture such as a gesture. If the negative judgment is made in this step S14, that is, if there is no response from the target user, the control unit 20 returns without executing the subsequent processing. On the other hand, if the target user responds and thus the answer is affirmative in step S15, the process proceeds to step S16.

[0069] In step S16, the control unit 20 records the answer input by the target user in the recording medium 30, and updates the user information of the target user. Specifically, the control unit 20 uses the function of the update unit 21c to link the content of the question made in step S14 with the answer from the target user to the question, and records (i.e., stores) the linked question and answer in the recording medium 30. Then, the control unit 20 uses the function of the update unit 21c to update the user information of the target user (i.e., user information 1). This makes it possible to supplement the missing information in the user information 1. In the embodiment shown in FIG. 5, the control unit 20 accepts the answer to the missing information, "question about French," as input data for the user information 1. If the answer input by the user in response to the request from the output unit 21b is an affirmative answer, the update unit 21c updates the information on the interest level and interest attribute regarding French to which the answer was obtained. The interest level can be indicated in multiple stages as described above. Therefore, if the user's answer to French is an affirmative answer, the interest level for French becomes "◎" or "○", and the update unit 21c updates the information on the interest level as user information. On the other hand, if the answer input by the user is a negative answer, the interest level for French becomes "△" or "×", and the update unit 21c updates the information on the interest level as user information.

[0070] Also, in this step S16, if the target user answers in step S15 and the answer is positive, the corresponding content item may be recommended to the target user. In the embodiment shown in FIG. 5, as described above, from the missing information of both the target user and the similar user, "XYZ French" that is a store that can be output from the current location of the target user is extracted, and a question is asked whether the target user is interested in "XYZ French". Therefore, the control unit 20 recommends "XYZ French" as a specific item in response to the answer from the user by the function of the output unit 21b. Note that in the process of step S13, if a plurality of stores that can be output are extracted (for example, five), the control unit 20 sorts the five extracted stores, for example, in order of proximity to the current location of the target user, or in order of high reviews, and recommends a specific store (item) to the target user.

[0071] Furthermore, when making the recommendation, the control unit 20 may use the function of the output unit 21b to also display facility information and routes of the recommended facility on a display or the like in the navigation system 50. Then, when a response to the recommendation is transmitted from the user, the control unit 20 uses the function of the update unit 21c to accept the response as input data for the user information 1 and update the user information 30c.

[0072] Thus, in the embodiment shown in FIG. 5, the control unit 20 acquires missing information in the user information of both the target user and the similar user. Then, the control unit 20 updates the user information 30c based on the input of the target user himself / herself for the information that is missing for each user. Therefore, compared to a case where the user information 30c of the target user is simply updated using the user information of another person, the user information 30c is in line with the hobbies and preferences of the target user himself / herself. Also, in the embodiment shown in FIG. 5, the area that is missing information of the target user and is also missing information of the similar user is set as the missing information of the target user, so that the user information can be enriched earlier than a case where the user information 30c is updated based on the missing information of only the target user.

[0073] Next, other processes executed in parallel after the above-mentioned step S10 will be described. As described above, in the embodiment shown in Fig. 5, after calculating the target to be recommended to the target user in step S10, the process proceeds to step S11 and step S17. Since the process from step S11 onwards is as described above, the process from step S17 onwards will be described here.

[0074] In step S17, the control unit 20 outputs the recommendation target calculated in step S10. Specifically, the control unit 20, by using the function of the output unit 21b, recommends to the target user at least one of the contents "hamburger, French" that are lacking in the target user calculated in step S10. The recommendation content may be, for example, "How about French (or hamburger)?". The recommendation output to the target user is output to the user I / F unit 50d, such as a speaker or display unit of the navigation system 50.

[0075] Next, the control unit 20 asks a question about the target user's response to the recommendation (step S18). Specifically, the control unit 20 accepts input of a response to the recommendation made in step S17 from the target user through the function of the output unit 21b. The target user responds with "good / bad" or the like by operating the display unit in the user I / F unit 50d, for example. Note that the target user's response may also be a response in the form of text or a response by voice output.

[0076] Then, the control unit 20 asks a specific question in response to the answer of the target user. For example, if the answer of the target user is a positive answer such as "good" or "I'm in the mood for French (or hamburger)", the control unit 20 uses the function of the output unit 21b to ask a question such as "How about XYZ French (or ABC burger)?" After asking the question, the control unit 20 advances the process to step S15.

[0077] On the other hand, if the target user answers negatively, such as "bad," "not in the mood for French food (or hamburgers)," or "not the time to eat," the control unit 20 asks a question such as "what are you in the mood for?" After asking the question, the control unit 20 proceeds to step S15.

[0078] In addition, in this step S18, the control unit 20 may, through the function of the output unit 21b, directly ask a specific question such as "What do you think of the XYZ French (or the ABC Burger)?" even if the target user does not respond with a "good / bad" answer.

[0079] In step S15, the control unit 20 judges whether or not the target user has input a response to the specific question asked in step S18. If the target user has provided some response such as "I want to eat XYZ French (or ABC burger)" or "I'm not in the mood for XYZ French (or ABC burger)", the determination in step S15 is affirmative, and the control unit 20 advances the process to step S16. On the other hand, if the target user has not provided a response to the question asked in step S18, the determination in step S15 is negative, and the control unit 20 returns without executing any further processes.

[0080] In step S16, the control unit 20 records the answer entered by the target user in the recording medium 30, and updates the user information of the target user. Specifically, the control unit 20 uses the function of the update unit 21c to link the content of the question asked in step S18 with the answer from the target user to the question, and records the linked question and answer in the recording medium 30. Then, the control unit 20 uses the function of the update unit 21c to update the user information of the target user (i.e., user information 1).

[0081] 5, an example has been described in which the processing of step S11 and subsequent steps and the processing of step S17 and subsequent steps are performed in parallel after step S10, but these processing steps may be performed independently of each other. Also, in the embodiment of FIG. 5, the processing of calculating the target of recommendation is performed in step S10 as a prerequisite for performing the processing of step S11 and subsequent steps, but when performing the processing of steps S11 to S16, step S10 may be skipped.

[0082] The above embodiment is an example for implementing the present invention, and various other embodiments can be adopted. For example, the recommendation system 10 may be composed of multiple devices (servers, clients, etc.). Some functions of the user information acquisition unit 21a, the output unit 21b, and the update unit 21c constituting the recommendation system 10 may be realized by the control unit 20, and the remaining functions may be realized by the control unit 50c of the navigation system 50. In addition, as long as the above-mentioned control can be executed, some configurations of the embodiment may be omitted, and the order of processing may be changed or omitted.

[0083] In the above embodiment, the control unit 20 extracts a store that can be output from the user's current location or current time zone from among the missing information, using only "ABC Burger" as an example, but there may be multiple items extracted. There may also be multiple contents (i.e., facility genres) of the extracted items. For example, in the example of FIG. 2A and FIG. 2B, in the user information 1, there are hamburgers and French as missing information. In this state, it is assumed that a store that provides hamburgers (ABC Burger) and a store that provides French (XYZ French Restaurant) are extracted from the target user's current location or current time zone. In this case, the control unit 20 requests input from the target user according to the degree of fulfillment of the content (i.e., asks whether the target user is interested in the missing content). Specifically, the control unit 20 requests input for content that is determined to have a relatively high degree of fulfillment, and does not request input for content that is determined to have a relatively low degree of fulfillment. Here, the degree of fulfillment is an index indicating whether the amount of information (amount of information) for a specific content is greater than the amount of information for other content. An example of a parameter for the amount of information is the number of stores. Content with a relatively large number of stores can be said to have a high level of fulfillment, whereas content with a relatively small number of stores can be said to have a low level of fulfillment.

[0084] In the example of FIG. 2A, there are 20 French restaurants and 50 hamburger restaurants. Therefore, hamburgers are more satisfying than French restaurants, and in this case, the control unit 20 requests the target user to input whether or not they are interested in hamburgers, but does not request whether or not they are interested in French restaurants. When requesting the target user to input multiple contents of the missing information, the content with the higher satisfaction level is given priority. Using the example of FIG. 2A described above, when requesting the target user to input whether or not they are interested in both the hamburger and French facility genres, the hamburger, which has the higher satisfaction level as content information, is given priority and the user is requested to input. Then, after the request for input for hamburgers, the control unit 20 requests input for French restaurants.

[0085] The degree of fulfillment may be an index showing whether the amount of information on a certain content is greater than the amount of information on other content. Therefore, the parameter for judging the degree of fulfillment may be determined from, for example, the sum of the total number of reviews, in addition to the number of stores mentioned above. In the example of FIG. 2A, the sum of the total number of reviews is 110 for hamburgers and 60 for French cuisine, and the degree of fulfillment is higher for hamburgers than for French cuisine. Therefore, in this case, the control unit requests the target user to input information about hamburgers (or prioritizes hamburgers).

[0086] In the above embodiment, the control unit 20 may recommend the corresponding content item to the target user when the target user answers whether or not they are interested in the content of the extracted item and the answer is positive (i.e., when the target user is interested in the target content). However, the control unit 20 may directly recommend the corresponding content item to the target user. That is, the control unit 20 may directly recommend ABC burger to the target user without requesting an answer as to whether or not they are interested in the content of the extracted item (e.g., whether or not they are interested in hamburgers). Then, when the control unit 20 directly recommends the extracted item without asking about the content, the control unit 20 accepts the answer about the item as an answer about the content and updates it as user information 30c.

[0087] In the above embodiment, the control unit 20 compares the content information 30b and the user information 30c to grasp the missing information of the user information of the target user, but when grasping "missing information of the user information of the target user", it may be grasped by simply referring to only the user information 30c. In other words, when the missing information can be grasped without comparing with the content information 30b, the missing information may be acquired by referring only to the user information 30c.

[0088] Furthermore, in this embodiment, the control unit 20 requests the user to input information in the user information regarding content that does not satisfy a predetermined standard. However, the "standard" in this case may be specified by a ratio to the above-mentioned predetermined standard, or by setting a standard in advance for each facility genre.

[0089] In the above embodiment, the control unit 20 uses "the user's current location or current time" as a condition when extracting items that can be output to the user from the user's current location or current time zone from among the missing information. However, this condition is not limited as long as it is a condition that identifies the missing information. For example, the user's interest attribute may be used as a condition. That is, in the example of FIG. 2B, the target user has recorded keywords such as "close to the station" and "reasonable" other than the missing facility genre (Italian, ramen, etc.). Therefore, when extracting the missing information items of hamburger and French, items with attributes such as "close to the station" and "reasonable" may be extracted.

[0090] Furthermore, the content to be provided that is recommended by the recommendation system 10 is not limited to restaurants. For example, it includes facilities that can be destinations in the navigation function, such as commercial facilities. The content to be provided may also be related information such as products, car functions (for example, power windows, sunroofs, audio, automatic transmission shift control, etc.), event information, and product development history. The present invention can also be applied to cases other than navigation systems, such as when presenting product or service recommendations on shopping sites or other sites.

[0091] Furthermore, the technique of the present invention can also be applied as a program or method. The above-mentioned system, program, and method may be realized as a single device or may be realized by using parts shared with each part provided in the vehicle, and include various aspects. In addition, they can be modified as appropriate, such as being partly software and partly hardware. Furthermore, the invention can also be realized as a recording medium of a program that controls the system. Of course, the recording medium of the program may be a magnetic recording medium or a semiconductor memory, and any recording medium developed in the future can be considered in the same way. [Explanation of symbols]

[0092] 10... recommendation system, 20... control unit, 21... recommendation program, 21a... user information acquisition unit, 21b... output unit, 21c... update unit, 30... recording medium, 30a... map information, 30b... content information, 30b... user information, 41... communication unit, 50... navigation system, 50a... communication unit, 50b... GNSS receiving unit, 50c... control unit, 50d... user I / F unit

Claims

1. a user information acquiring unit that acquires user information indicating a user's interest in content to be provided; an output unit that outputs request information for requesting the user to input information regarding a content having a relatively small amount of information compared to a predetermined comparison target, among the information for each content recorded on the recording medium as the user information; an update unit that updates the user information based on information input by the user in response to the request information; With when there are a plurality of contents each having a relatively small amount of information, the output unit requests the input in accordance with a degree of enrichment of the plurality of contents. Recommendation system.

2. A user information acquisition unit that acquires user information indicating a user's interest in content to be provided; an output unit that outputs request information for requesting the user to input information regarding a content having a relatively small amount of information compared to a predetermined comparison target, among the information for each content recorded on the recording medium as the user information; an update unit that updates the user information based on information input by the user in response to the request information; With The user information acquisition unit further acquires user information of similar users who have similar interests in the content to the user, the output unit outputs the request information that requests the user to input information about a content in which the amount of information is relatively small in the user information of the user and the amount of information is relatively small in the user information of the similar user. Recommendation system.

3. The request information requested by the output unit includes information on whether the user is interested in the content having a relatively small amount of information, and the content having a relatively small amount of information is determined according to a current location or a current time period of the user. The recommendation system according to claim 1 or 2.

4. The update unit accepts the answer to the question of whether or not the user is interested as input data for the user information. The recommendation system according to claim 3 .

5. The output unit recommends the content having a relatively small amount of information to the user; The update unit accepts a response to the recommendation as input data for the user information. The recommendation system according to claim 1 or 2.

6. the request information includes information related to the content; The information related to the content includes at least any one of a facility genre, a distance to the facility, a facility name, a price range of the facility, and a review of the facility. The recommendation system according to claim 1 or 2.

7. The update unit records the input information on the recording medium. The recommendation system according to claim 1 or 2.

Images