Automobile recommendation system

Abstract

To recommend an automobile by estimating a difference / change in user's needs depending on a difference in an attribute of a user or by estimating an individual user's value system. An automobile recommendation system 1 that recommends an automobile to a user, including: a user-based value system calculation unit 14 that calculates an index indicating a value system of the user from a difference between a characteristic of a vehicle type requested by each of users belonging to an attribute same as or similar to an attribute of the user at a predetermined point of time and a characteristic of a vehicle type requested by the user at the predetermined point of time; and a current user-requested vehicle type characteristic calculation unit 15 that adds a characteristic of a vehicle type requested by each of users belonging to an attribute same as or similar to a current attribute of the user to the index indicating the value system of the user to calculate a characteristic of a vehicle type currently requested by the user as a characteristic of a vehicle type to be recommended to the user.

Description

AUTOMOBILE RECOMMENDATION SYSTEM

[0001] The present invention relates to a technique for recommending a product to an individual user, and particularly relates to an effective technique when being applied to an automobile recommendation system that recommends an automobile.

[0002] In selling products, a mechanism for recommending an appropriate product in accordance with user's needs or value system has been studied. However, usually, the user's needs or value system cannot be directly observed unless the user is asked through a questionnaire or the like.

[0003] In contrast, for example, JP 2022-90717 A (Patent Literature 1) discloses that a content (need) of a characteristic desired by a user for a predetermined characteristic item of a product (for example, an automobile) is estimated on the basis of attribute information of the user, and the estimated need is compared with an actual content of the characteristic for each product to selectively propose a product that meets the user's need.

[0004] JP 2022-90717 A

[0005] According to the related art, it is possible to estimate a need on the basis of an attribute of a user and recommend a product matching the need to the user.

[0006] In this case, in order to estimate the need (tendency) from the attribute of the user, for example, a user group having the same or similar attribute is segmented, and individual needs of these users are averaged. However, it is not necessarily appropriate to recommend the same product to users having the same or similar attribute. Even if the users have the same attribute, they have different value systems (preferences) individually, so that it is required to recommend a product matching their value system. However, usually, such value systems cannot be directly observed unless the user is asked.

[0007] Therefore, an object of the present invention is to provide an automobile recommendation system that recommends a product (particularly, an automobile) by estimating a difference / change in needs depending on a difference in an attribute of a user or by estimating an individual user's value system.

[0008] The above-described and other objects and novel features of the present invention will become apparent from the description herein and the accompanying drawings.

[0009] A representative embodiment of the invention disclosed in the present application will be briefly outlined as follows.

[0010] An automobile recommendation system that is a representative embodiment of the present invention is an automobile recommendation system that recommends an automobile to a user, and includes: a user-based value system calculation unit that calculates an index indicating a value system of the user from a difference between a characteristic of a vehicle type requested by each of users belonging to an attribute same as or similar to an attribute of the user at a predetermined point of time and a characteristic of a vehicle type requested by the user at the predetermined point of time; and a current user-requested vehicle type characteristic calculation unit that adds a characteristic of a vehicle type requested by each of users belonging to an attribute same as or similar to a current attribute of the user to the index indicating the value system of the user to calculate a characteristic of a vehicle type currently requested by the user as a characteristic of a vehicle type to be recommended to the user.

[0011] An effect of the representative embodiment of the invention disclosed in the present application will be briefly outlined as follows.

[0012] That is, according to a representative embodiment of the present invention, it is possible to recommend an automobile by estimating a difference / change in needs depending on a difference in an attribute of a user or by estimating an individual user's value system.

[0013] Fig. 1 is a diagram illustrating an outline of a configuration example of an automobile recommendation system that is an embodiment of the present invention.Figs. 2(a) and 2(b) are diagrams illustrating an outline of an example of a concept of estimating user's needs or a user's value system to recommend an automobile in an embodiment of the present invention.Fig. 3 is a diagram illustrating an outline of an example of a data configuration of a vehicle-type-based performance specification in an embodiment of the present invention.Fig. 4 is a diagram illustrating an outline of an example of a data configuration of a vehicle-type-based characteristic value in an embodiment of the present invention.Fig. 5 is a diagram illustrating an outline of an example of a data configuration of a user's own vehicle type in an embodiment of the present invention.Fig. 6 is a diagram illustrating an outline of an example of a data configuration of a user's attribute at the start of ownership in an embodiment of the present invention.Fig. 7 is a diagram illustrating an outline of an example of a data configuration of a user-based requested vehicle type characteristic in an embodiment of the present invention.Fig. 8 is a diagram illustrating an outline of an example of a data configuration of a user's-attribute-based requested vehicle type characteristic in an embodiment of the present invention.Fig. 9 is a diagram illustrating an outline of an example of a data configuration of a user-based value system in an embodiment of the present invention.Fig. 10 is a diagram illustrating an outline of an example of a data configuration of a current user's characteristic in an embodiment of the present invention.Fig. 11 is a diagram illustrating an outline of an example of a data configuration of a current user-requested vehicle type characteristic in an embodiment of the present invention.Fig. 12 is a diagram illustrating an outline of an example of a data configuration of a recommended vehicle type in an embodiment of the present invention.

[0014] Hereinafter, embodiments of the present invention will be described in detail with reference to the accompanying drawings. In all the drawings for explaining the embodiments, the same parts are denoted by the same reference numerals in principle, and duplicated descriptions thereof will be omitted. Meanwhile, a component denoted by a reference numeral described with reference to a certain drawing may be mentioned again with the same reference numeral in descriptions with reference to another drawing in which the component is not illustrated.

[0015] As described above, it is usually impossible to directly observe needs depending on a difference in an attribute of a user and an individual user's value system. This is because the “needs” and the individual “value system” require some definitions, and even if they could be defined, it would usually be impossible to directly observe them unless directly asking the user.

[0016] On the other hand, in an automobile recommendation system according to an embodiment of the present invention, when recommending an automobile to be purchased to a user, it is considered that user's needs or value system is reflected in a performance index of a vehicle type owned by the user. That is, since it is considered that the user owns the automobile that matches his / her value system, it is considered that the user's needs or value system can be guessed based on the performance index of the currently owned vehicle type.

[0017] In addition, it is considered that the needs depending on the difference in an attribute of a user can be obtained as a representative value (an average value, a median value, and the like) of the performance index of the automobile requested by the user group having the same or similar attribute (hereinafter simply collectively referred to as “same attribute” in some cases), and an individual user's value system can be evaluated as a difference between the performance index requested by the user himself / herself and the representative value mentioned above (for example, “this user places more importance on fuel consumption than general users in their thirties” or the like). Then, it is considered that the user's value system is derived from something inherent in the user, and it is considered that the user's value system does not change greatly even if the life stage changes.

[0018] Figs. 2 is a diagram illustrating an outline of an example of a concept of estimating user's needs or the user's value system to recommend the automobile in an embodiment of the present invention. Fig. 2(a) illustrates an example of estimating the user's value system. The characteristic items (four of “characteristic 1” to “characteristic 4” in the example of Fig. 2) of an automobile are arranged on the horizontal axis of the graph, and the vertical axis indicates the intensity of the characteristic for each characteristic item. For each characteristic item, the hollowed graph indicates the vehicle type characteristic (representative value) requested by the user group having the same attribute as that of the target user at the point of time (start of ownership) when the user started to own the automobile that is currently owned by the user. The shaded graph indicates the vehicle type characteristic requested personally by the user at the start of ownership.

[0019] In the present embodiment, in this graph, it is considered that the difference between the hollowed graph and the shaded graph for each characteristic item, that is, the difference between the representative value (tendency) of the user group having the same attribute and the characteristic value requested personally by the target user indicates the value system of the user. In the example of Fig. 2(a), there are differences (black arrows in the graph) in the “characteristic 1” and the “characteristic 3”. It means that these differences indicate the value system of the user. The reason why the start of ownership of the automobile that is currently owned by the user is used here as the reference is that there is an opportunity to obtain, through sales activities, various procedures, and the like, a variety of information such as the attribute of the user at the start of ownership, but the start of ownership is not necessarily used as the reference, and another point of time can be used as the reference.

[0020] Fig. 2(b) illustrates an example of obtaining the characteristics of a vehicle type to be recommended to the user at the current time on the basis of the user's value system obtained in Fig. 2(a). The hollowed graph indicates the vehicle type characteristic (representative value) requested by the user group having the same attribute as that of the current user. The intensity of the characteristic of each characteristic item is different from that of the hollowed graph of Fig. 2(a), indicating a difference or a change in an attribute such as an age (for example, a person in his / her twenties will step into his / her thirties, and may have different needs because of a change in life stage or the like).

[0021] In the present embodiment, it is possible to obtain the characteristic of the recommended vehicle type (shaded graph) reflecting the user's value system by reflecting the user's value system (black arrows) obtained in Fig. 2(a) on this hollowed graph (vehicle type characteristic requested by the user group having the same attribute). In addition, a vehicle type actually having a characteristic close to (similar to) the characteristic of the recommended vehicle type obtained here is a vehicle type to be recommended.

[0022] In this way, by dividing the tendency of each attribute of the user and the individual user's value system, it is possible to recommend an automobile corresponding to the change in the life stage. For example, it is possible to make recommendations which can be interpreted as follows: ”Since the age group has changed, a car such as the model X is preferable as the car to drive next from the general tendency of the user. However, the user focuses on environmental performance, and thus recommendation is made to offer a car with relatively high fuel efficiency”; “recommendation is made to offer the model Y if the user who owned the model X while living in Tokyo moves to Hokkaido”; “recommendation is made to offer the model Y when the user who owned the model X in a single household gets married and lives with a spouse”; “recommendation is made to offer a vehicle type more focusing on a spatial property than the user group having the same attribute requires, to a user interested in camping”; and so on.

[0023] Fig. 1 is a diagram illustrating an outline of a configuration example of an automobile recommendation system that is an embodiment of the present invention. An automobile recommendation system 1 is an information processing system that is constituted by, for example, a server device, a virtual server constructed on a cloud computing service, an information processing terminal, and the like, and achieves various functions related to recommendation of an automobile with a central processing unit (CPU) (not illustrated) executing middleware such as an operating system (OS), a database management system (DBMS), and a web server program loaded on a memory from a recording device such as a hard disk drive (HDD) or a solid state drive (SSD), and software operating on the middleware.

[0024] The automobile recommendation system 1 includes, for example, individual units such as a vehicle type characteristic calculation unit 11, a user-requested vehicle type characteristic calculation unit 12, a user's-attribute-based requested vehicle type characteristic calculation unit 13, a user-based value system calculation unit 14, a current user-requested vehicle type characteristic calculation unit 15, and a recommended vehicle type calculation unit 16 which are implemented as software. In addition, the automobile recommendation system 1 includes individual data stores such as a vehicle-type-based performance specification 21, a user's own vehicle type 22, a user's attribute at the start of ownership 23, a current user's attribute 24, a vehicle-type-based characteristic value 31, a user-based requested vehicle type characteristic 32, a user's-attribute-based requested vehicle type characteristic 33, a user-based value system 34, a current user-requested vehicle type characteristic 35, a recommended vehicle type 36, and the like, which are configured by a database, a file, and the like.

[0025] In the processing of recommending an automobile to a user, first, the vehicle type characteristic calculation unit 11 calculates a characteristic value of each vehicle type of respective manufacturers in advance. Here, data of performance specification of each vehicle type held in the vehicle-type-based performance specification 21 is input to calculate a characteristic value for each vehicle type on the basis of a predetermined model obtained by machine learning and output the result as the vehicle-type-based characteristic value 31. A tool, a library, and the like for machine learning are not particularly limited, and available tools and libraries can be appropriately used.

[0026] Fig. 3 is a diagram illustrating an outline of an example of a data configuration of the vehicle-type-based performance specification 21 in an embodiment of the present invention. The vehicle-type-based performance specification 21 is a table that holds the data of the performance index and the performance specification of each vehicle type for each manufacturer, and includes, for example, respective key items of a manufacturer, a body shape, a vehicle type name, and a model year (items marked with * in the drawing. The same applies hereinafter), and respective items of performance specification such as a new car price, a horsepower, a fuel consumption, and a total length / total width / total height. The items of the performance specification are not limited to these items, and other items may be included. The data of each item can be automatically or manually collected from various media such as a website or a catalog of each automobile manufacturer and commercially available data.

[0027] Fig. 4 is a diagram illustrating an outline of an example of a data configuration of the vehicle-type-based characteristic value 31 in an embodiment of the present invention. The vehicle-type-based characteristic value 31 is a table that holds the characteristic value calculated for each vehicle type by the vehicle type characteristic calculation unit 11 on the basis of the data of the vehicle-type-based performance specification 21, and includes, for example, respective key items including the manufacturer, the body shape, the vehicle type name, and the model year, and respective items including characteristic values which are a characteristic 1, a characteristic 2, a characteristic 3, ... and so on. In the present embodiment, values of the respective items of the characteristic 1, the characteristic 2, the characteristic 3, ... are calculated on the basis of a predetermined model obtained by machine learning. Therefore, meanings of the respective items and numerical values are in a black box and basically cannot be interpreted by a human. The number of items is also not particularly limited, and is determined according to the design of the model.

[0028] Returning to Fig. 1, next, the user-requested vehicle type characteristic calculation unit 12 receives data of the current vehicle type owned by each user held in the user's own vehicle type 22 and data of the attribute information of the user at the start of ownership held in the user's attribute at the start of ownership 23, calculates what characteristic the owned vehicle type has for each user, that is, what vehicle type characteristic each user requests, with reference to the vehicle-type-based characteristic value 31 described above, and outputs the result as the user-based requested vehicle type characteristic 32.

[0029] Fig. 5 is a diagram illustrating an outline of an example of a data configuration of the user's own vehicle type 22 in an embodiment of the present invention. The user's own vehicle type 22 is a table that holds the information of the vehicle type owned by each user, and includes, for example, a key item including a user ID, and respective items including a manufacturer of the owned vehicle type, a body shape of the owned vehicle type, an owned vehicle type name, a model year of the owned vehicle type, and so on, that specify a vehicle type.

[0030] Fig. 6 is a diagram illustrating an outline of an example of a data configuration of the user's attribute at the start of ownership 23 in an embodiment of the present invention. The user's attribute at the start of ownership 23 is a table that holds the information of the attribute of the user at the time when each user starts to own that vehicle type, and includes, for example, a key item including the user ID and respective items related to the attribute of the user including gender, a prefecture at the start of ownership, an age category at the start of ownership, and so on. The items related to the attribute of the user are not limited thereto, and may have other attribute items such as occupation, family structure, and hobby, for example.

[0031] In addition, Fig. 7 is a diagram illustrating an outline of an example of a data configuration of the user-based requested vehicle type characteristic 32 in an embodiment of the present invention. The user-based requested vehicle type characteristic 32 is a table that holds the information of the attribute of each user and the information of the characteristic of the owned vehicle, that is, the information of what characteristic each user requested to the vehicle type when the user started to own that vehicle type, and includes, for example, a key item including the user ID, respective items related to the attribute of the user including gender, the prefecture at the start of ownership, the age category at the start of ownership, and so on, respective items including the manufacturer of the owned vehicle type, the body shape of the owned vehicle type, the owned vehicle type name, the model year of the owned vehicle type, and so on, that specify a vehicle type, and respective items including characteristic values which are the characteristic 1, the characteristic 2, the characteristic 3, ... and so on. The contents of the respective items described above are the same as the corresponding items in the user's own vehicle type 22, the user's attribute at the start of ownership 23, and the vehicle-type-based characteristic value 31 that are to be inputs.

[0032] Returning to Fig. 1, next, the user's-attribute-based requested vehicle type characteristic calculation unit 13 receives the user-based requested vehicle type characteristic 32 described above, performs totalization / statistical processing in units of attributes of a user, and outputs the result as the user's-attribute-based requested vehicle type characteristic 33.

[0033] Fig. 8 is a diagram illustrating an outline of an example of a data configuration of the user's-attribute-based requested vehicle type characteristic 33 in an embodiment of the present invention. The user's-attribute-based requested vehicle type characteristic 33 for each user attribute is a table that holds the data obtained by performing totalization / statistical processing on the characteristics of the vehicle type held by each user in units of the attributes of the user at the start of ownership, that is, the information related to needs (tendencies) depending on differences in the attribute of the user, and includes, for example, key items including gender, the prefecture at the start of ownership, the age category at the start of ownership, and so on, and respective items including characteristic values which are the characteristic 1, the characteristic 2, the characteristic 3, ... and so on.

[0034] The present embodiment focuses on gender, a prefecture, and an age category (for example, “twenties”, “thirties”, “forties”, ..., and so on) as an attribute of a user in the totalization unit, but the present invention is not limited to this. A residential area of a user may be divided not into prefectures, but into narrower areas, or conversely divided by wider areas such as Kanto, Tohoku, Chubu, ..., and so on. Other attribute items such as occupation and family structure may be used. As the values of the respective items including the characteristic 1, the characteristic 2, the characteristic 3, ..., for example, average values derived from totalization in units of attributes of a user may be used, or another statistical index such as a median value may be used.

[0035] Returning to Fig. 1, next, the user-based value system calculation unit 14 receives the user-based requested vehicle type characteristic 32 and the user's-attribute-based requested vehicle type characteristic 33 described above, calculates the value system of each user (it is more accurately “index indicating value system”, but is simply referred to as “value system”) and outputs the result as the user-based value system 34.

[0036] Fig. 9 is a diagram illustrating an outline of an example of a data configuration of the user-based value system 34 in an embodiment of the present invention. The user-based value system 34 is a table that holds the information of the difference between the value system of each user, that is, the representative value of characteristics of the vehicle type required by a user group having the same attribute as that of the user, and a characteristic of the vehicle type required by the user, and includes, for example, a key item including the user ID, and respective items including characteristic values which are the characteristic 1, the characteristic 2, the characteristic 3, ... and so on. Here, as illustrated in Fig. 2(a), the values of the respective items including the characteristic 1, the characteristic 2, the characteristic 3, ... are differences between the values of the respective items including the characteristic 1, the characteristic 2, the characteristic 3, ... in the user's-attribute-based requested vehicle type characteristic 33 described above (needs (tendencies) based on the attribute of the user at the start of ownership) and the values of the respective items including the characteristic 1, the characteristic 2, the characteristic 3, ... in the user-based requested vehicle type characteristic 32 (vehicle type characteristic requested by the user at the start of ownership).

[0037] Returning to Fig. 1, next, the current user-requested vehicle type characteristic calculation unit 15 receives the user-based value system 34 and the data of the current user's attribute 24 described above, calculates information of the characteristic of the vehicle type requested by each current user, that is, information of the characteristic of the vehicle type to be recommended to the current user, with reference to the user's-attribute-based requested vehicle type characteristic 33 described above, and outputs the result as the current user-requested vehicle type characteristic 35.

[0038] Fig. 10 is a diagram illustrating an outline of an example of a data configuration of the current user's characteristic 24 in an embodiment of the present invention. The current user's characteristic 24 is a table that holds the current attribute information of each user, and includes, for example, a key item including the user ID, and respective items related to the attribute of the user including gender, a current age, and a current prefecture. Similarly to the above description, the attribute items are not limited to those illustrated in the example of Fig. 10, and other attribute items such as occupation and family structure may be used, or the residential area may be grasped with sections different from the prefectures.

[0039] Fig. 11 is a diagram illustrating an outline of an example of a data configuration of the current user-requested vehicle type characteristic 35 in an embodiment of the present invention. The current user-requested vehicle type characteristic 35 is a table that holds the information of the current attribute of each user and the information of the characteristic of the requested vehicle type, that is, information on the vehicle type requested currently by each user, and includes, for example, a key item including the user ID, respective items related to the attribute of the user including gender, the current age, and the current prefecture, and respective items of characteristic values which are the characteristic 1, the characteristic 2, the characteristic 3, ... and so on.

[0040] Here, as illustrated in Fig. 2(b), the values of the respective items including the characteristic 1, the characteristic 2, the characteristic 3, ... are obtained by adding the values of the respective items including the characteristic 1, the characteristic 2, the characteristic 3, ... (value system of the user) in the user-based value system 34 of the target user to the values of the respective items including the characteristic 1, the characteristic 2, the characteristic 3, ... (needs (tendencies) based on the current attribute of the user) obtained from the user's-attribute requested vehicle type characteristic 33 based on the current attribute information of the target user. That is, the values of the respective items including the characteristic 1, the characteristic 2, the characteristic 3, ... indicate the characteristic of the vehicle type to be recommended to the user.

[0041] Returning to Fig. 1, next, the recommended vehicle type calculation unit 16 receives the current user-requested vehicle type characteristic 35, calculates a candidate of specific vehicle types to be recommended with reference to the vehicle-type-based characteristic value 31, and outputs the result as the specific recommended vehicle type 36 with reference to the vehicle-type-based performance specification 21. Here, one or more vehicle types having characteristic values similar to the values of the respective items including the characteristic 1, the characteristic 2, the characteristic 3, ... in the current user-requested vehicle type characteristic 35 are extracted from the vehicle-type-based characteristic value 31 and are appointed as the candidates of the specific vehicle types to be recommended. A method of determining similarity is not particularly limited, and for example, an appropriate method can be used in which the values of the respective items including the characteristic 1, the characteristic 2, the characteristic 3, ... are vectorized as elements, and the results are evaluated as vector similarity.

[0042] Fig. 12 is a diagram illustrating an outline of an example of a data configuration of the recommended vehicle type 36 in an embodiment of the present invention. The recommended vehicle type 36 is a table that holds the vehicle type to be recommended to each user and the information of the characteristic of the vehicle type, and includes, for example, key items including the user ID, the manufacturer, the body shape, the vehicle type name, and the model year, and respective items of performance specifications including the new car price, the horsepower, the fuel consumption, and the total length / total width / total height.

[0043] Note that the examples of the respective data configurations illustrated in Figs. 3 to 12 illustrate configurations (views) of logical tables, and in practice, for example, a plurality of logical tables may be normalized in a database and implemented as one or a plurality of physical tables.

[0044] In addition, the present embodiment considers an example of a case where each user currently owns an automobile, that is, a case where the user purchases a new automobile, but the present invention is not limited to this. As long as the information for grasping the characteristic of the vehicle type requested by the user can be acquired, it is not essential that there is an automobile currently owned, and the present invention can also be applied to a case of newly purchasing an automobile. For example, in a case where it is grasped that a specific vehicle type is intensively browsed from a web browsing history or the like of the user, the characteristic of the vehicle type may be treated as the characteristic of the vehicle type requested by the user.

[0045] As described above, with the automobile recommendation system 1 according to an embodiment of the present invention, it is possible to recommend the automobile by estimating the difference / change in needs depending on the difference in the attribute of the user or by estimating the individual user's value system, on the basis of a performance index of the vehicle type owned by the user. In addition, in this way, by dividing the tendency of each attribute of the user and the individual user's value system, it is possible to recommend the automobile corresponding to the change in the life stage.

[0046] Although the invention made by the present inventors has been specifically described on the basis of the embodiments, the present invention is not limited to the embodiments described above, and it goes without saying that various modifications may be made without departing from the gist of the present invention. The embodiments above have been described in detail to explain the present invention in an easy-to-understand manner, and the present invention is not necessarily limited to the embodiments including all the components described. Another component may be added to, deleted from, or replaced with a part of the configuration of each embodiment described above.

[0047] A part or all of the components, functions, processing units, processing procedures, and the like described above may be implemented by hardware by being designed as an integrated circuit, for example. Alternatively, the components, functions, and the like described above may be implemented by software by a processor interpreting and executing programs for implementing the individual functions. Information such as programs, tables, and files for implementing the individual functions may be stored in a recording device such as a memory, a hard disk, or a SSD or in a recording medium such as an integrated circuit (IC) card, a secure digital (SD) card, or a digital versatile disc (DVD).

[0048] Each of the drawings mentioned above illustrates control lines and information lines considered to be necessary for the description, and does not necessarily illustrate all the implemented control lines and information lines. It may be considered that almost all the components are mutually connected in practice.

[0049] The present invention can be used for an automobile recommendation system that recommends an automobile.

[0050] 1 Automobile recommendation system 11 Vehicle type characteristic calculation unit 12 User-requested vehicle type characteristic calculation unit 13 User's-attribute-based requested vehicle type characteristic calculation unit 14 User-based value system calculation unit 15 Current user-requested vehicle type characteristic calculation unit 16 Recommended vehicle type calculation unit 21 Vehicle-type-based performance specification 22 User's own vehicle type 23 User's attribute at the start of ownership 24 Current user's attribute 31 Vehicle-type-based characteristic value 32 User-based requested vehicle type characteristic 33 User's-attribute-based requested vehicle type characteristic 34 User-based value system 35 Current user-requested vehicle type characteristic 36 Recommended vehicle type

Claims

1. An automobile recommendation system that recommends an automobile to a user, comprising: a user-based value system calculation unit that calculates an index indicating a value system of the user from a difference between a characteristic of a vehicle type requested by each of users belonging to an attribute same as or similar to an attribute of the user at a predetermined point of time and a characteristic of a vehicle type requested by the user at the predetermined point of time; and a current user-requested vehicle type characteristic calculation unit that adds a characteristic of a vehicle type requested by each of users belonging to an attribute same as or similar to a current attribute of the user to the index indicating the value system of the user to calculate a characteristic of a vehicle type currently requested by the user as a characteristic of a vehicle type to be recommended to the user.

2. The automobile recommendation system according to claim 1, further comprising a recommended vehicle type calculation unit that appoints a vehicle type having a characteristic value similar to a characteristic of a vehicle type currently requested by the user, the characteristic value being calculated by the current user-requested vehicle type characteristic calculation unit, as a candidate of the vehicle type to be recommended to the user.

3. The automobile recommendation system according to claim 1, further comprising: a user-requested vehicle type characteristic calculation unit that calculates, based on a characteristic of a vehicle type owned by each user and an attribute of each user at a start of ownership of the vehicle type, a characteristic of a vehicle type requested by each user at the start of ownership; and a user's-attribute-based requested vehicle type characteristic calculation unit that calculates a characteristic of a vehicle type requested by each user belonging to a predetermined attribute by totalizing the characteristic of the vehicle type requested by each user at the start of ownership for each attribute of the user, the characteristic being calculated by the user-requested vehicle type characteristic calculation unit.

4. The automobile recommendation system according to claim 1, further comprising: a vehicle type characteristic calculation unit that calculates a characteristic value for a predetermined characteristic item for each vehicle type based on a model created by machine learning from data of performance specifications of each vehicle type.

Images