JP5809124B2 - Information processing system, information processing method, and program - Google Patents

Description

  Some embodiments according to the present invention relate to an information processing system, an information processing method, and a program.

  In recent years, it has become easier to compare and examine a large number of products related to the same category that the user desires to purchase due to the improvement of the processing capability of the information processing apparatus and the spread of the Internet. However, if there are too many products, the amount of information increases too much, and eventually it may be difficult for the user to determine which product is most suitable for him.

  Therefore, Patent Document 1 discloses a compatible product recommendation system and the like that can present a real product that is optimal to the user after detecting the potential needs of the user using a technique called conjoint analysis. In this system, a user's needs are detected by performing conjoint analysis based on order information obtained by rearranging virtual product cards relating to virtual products prepared in advance in a preferred order by a user who is a subject.

Japanese Patent No. 4317588

  However, when trying to grasp the potential needs of users based on virtual products prepared in advance as described in Patent Document 1, what values are set in advance for the attributes of these virtual products. Depending on what, the needs of the user who can be grasped change.

  Some aspects of the present invention have been made in view of the above-described problems. An information processing system and an information processing method capable of providing appropriate product information after appropriately grasping different needs for each user One of the purposes is to provide a program.

  An information processing system according to the present invention includes a registration unit that receives registration of a plurality of product information in which attribute values are set for a plurality of attributes, and a first that receives input of user preference information for each of the plurality of attributes. A second input unit that receives a user input for ordering a plurality of virtual products having at least some attribute values different from each other based on the preference information; and the plurality of the plurality of attributes Based on the order for the virtual product, the calculation unit for obtaining the relative importance between the plurality of attributes in the user, the plurality of product information, attribute values of the plurality of attributes each product information has, An evaluation unit that evaluates each based on relative importance between a plurality of attributes, and an output unit that outputs the plurality of product information based on an evaluation result by the evaluation unit .

  An information processing method according to the present invention includes a step of receiving registration of a plurality of product information in which attribute values are set for a plurality of attributes, a step of receiving input of user preference information for each of the plurality of attributes, Based on the preference information, the step of receiving user input for ordering a plurality of virtual products in which at least some of the attribute values of the plurality of attributes are different from each other, and based on the order for the virtual product, Determining the relative importance between the plurality of attributes; and the plurality of product information, the attribute values of the plurality of attributes included in each product information, and the relative importance between the plurality of attributes. Each of the information processing apparatus performs an evaluation step based on the evaluation result and a step of outputting the plurality of product information based on the evaluation result. Method.

  The program according to the present invention includes a process of receiving registration of a plurality of product information in which attribute values are set for a plurality of attributes, a process of receiving input of user preference information for each of the plurality of attributes, and the preference Based on the information, a process for receiving user input for ordering a plurality of virtual products having different attribute values of at least some of the plurality of attributes, and the plurality of the users in the user based on the order for the virtual products The relative importance between the attributes of the plurality of product information based on the attribute values of the plurality of attributes of each product information and the relative importance between the plurality of attributes A program for causing a computer to execute a process of evaluating each of the processes and a process of outputting the plurality of pieces of product information based on the result of the evaluation.

  In the present invention, “part”, “means”, “apparatus”, and “system” do not simply mean physical means, but “part”, “means”, “apparatus”, “system”. This includes the case where the functions possessed by "are realized by software. Further, even if the functions of one “unit”, “means”, “apparatus”, and “system” are realized by two or more physical means or devices, two or more “parts” or “means”, The functions of “device” and “system” may be realized by a single physical means or device.

  According to the present invention, an information processing system, an information processing method, and a program capable of providing appropriate product information after appropriately grasping different needs for each user based on different preferences for each user are provided. can do.

It is a block diagram which shows the function structure of the system which concerns on embodiment of this invention. It is a figure which shows the specific example of property information. 3 is a flowchart for explaining a processing flow of the information processing apparatus illustrated in FIG. 2. It is a figure for demonstrating the specific example of the setting method of a parameter. It is a figure for demonstrating the specific example of the setting method of a parameter. It is a figure for demonstrating the specific example of the setting method of a parameter. It is a figure for demonstrating the specific example of a conjoint investigation. It is a figure for demonstrating the specific example of a conjoint investigation. It is a figure for demonstrating the example of the specific method of a conjoint analysis. It is a figure which shows the specific example of the result of a conjoint analysis. It is a figure which shows the specific example of the presentation method of property information. It is a block diagram which shows the structure of the hardware which can mount the information processing apparatus shown in FIG.

  Embodiments of the present invention will be described below. In the following description and the description of the drawings to be referred to, the same or similar components are denoted by the same or similar reference numerals.

(Embodiment)
1 to 12 are diagrams for explaining the embodiment. Hereinafter, embodiments will be described along the following flow with reference to these drawings. First, “1” describes the functional configuration of the entire system including the information processing server according to the embodiment. Then, after describing the outline of the processing flow of the information processing server in “2”, the details of each processing will be described in “3” with specific examples. “4” indicates a specific example of a hardware configuration capable of mounting the information processing server. Finally, the effects and the like according to the present embodiment will be described after “5”.

(1 System functional configuration)
(1.1 System overview)
FIG. 1 is a diagram illustrating a schematic configuration of a system 1 including an information processing server 100 which is an embodiment of the information processing system. As shown in FIG. 1, the system 1 includes an information processing server 100, terminals S1 to Sn (hereinafter also collectively referred to as terminals S) of those who want to sell the property, and terminals B1 to Bm (hereinafter also referred to as terminals S) Hereinafter, it is also collectively referred to as terminal B.). The information processing server 100, the terminal S, and the terminal B can communicate with each other via the network N. Specific examples of the network N include the Internet, an internal network of a real estate agent that manages the information processing server 100, and the like.

  The system 1 provides information on real estate properties registered by applicants for sale in accordance with the preferences of each purchase applicant. More specifically, the sales applicant registers the property information of the property to be sold in the property information database (DB) 110 managed by the information processing server 100 by operating the terminal S. The purchase applicant can operate the terminal B to set parameters and evaluate virtual properties (parameter settings and virtual properties to enable acquisition of property information that matches the preferences of each purchase applicant. Details of the evaluation will be described later). The information processing server 100 transmits suitable property information that matches the preference of the purchase applicant to the terminal B based on the parameter setting and property evaluation from the purchase requester.

  Here, the terminal S or B is not necessarily operated by the person who wants to sell the property or the person who wants to purchase the property. For example, it may be a real estate agent or a real estate agent who has received a request from a sales applicant or purchase applicant. In the following description, it is assumed that the terminal S or B is operated by a person who wants to sell or purchase a property.

  In the present embodiment, a case where processing is performed on a real estate property, particularly a property for sale, will be described as an example, but the present invention is not limited to this. For example, real estate properties may be handled as rental properties, and article information other than real estate and other job offer information (these can be collectively referred to as “property information”) are processed. It is also possible to do. In this embodiment, the case where the system 1 targets both newly built properties and second-hand properties will be described as an example. However, the present invention is not limited to this, and only one of the properties is handled as a target. It is also possible to do.

  As described above, the information processing server 100 receives registration of property information to be sold from the terminal S of the applicant for sale, and performs parameter setting indicating the purchaser's preference and evaluation of the virtual property from the terminal B of the purchase applicant. Operation information is received. In addition, the information processing server 100 transmits information on a property to be recommended to the terminal B of the purchase applicant. Details of the configuration of the information processing server 100 will be described later in “1.2”.

  The terminal S is one or more information processing terminals operated by a person who wants to sell, and the terminal B is operated by a person who wants to purchase. The terminals S and B can be realized as various information processing apparatuses such as a personal computer, a mobile phone (whether it is a so-called feature phone or a smartphone), and a tablet terminal. It is also conceivable that one information processing terminal operates as the terminal S of the applicant for sale and the terminal B of the applicant for purchase. Furthermore, it is conceivable to realize at least one of the terminal S and the terminal B as one information processing apparatus combined with the information processing server 100.

(1.2 Configuration of information processing server)
Hereinafter, the functional configuration of the information processing server 100 will be described. As shown in FIG. 1, the information processing server 100 includes a property information DB 110, a GUI (Graphical User Interface) providing unit 120, a conjoint analysis unit 130, and a property evaluation unit 140.

  The property information DB 110 is a database for managing property information registered by a person who wants to sell by operating the terminal S, and is stored in, for example, an HDD (Hard Disk Drive) or an SSD (Solid State Drive) not shown.

FIG. 2 shows a specific example of property information registered in the property information DB 110. As shown in FIG. 2, the property information includes various attribute information for each segment such as basic information, location, building, residence, management, and image. For example, the segment “basic information” has attributes such as “XXXXX”, “201”, and “Tokyo XX ward XX- □” for the attributes such as the building name, room number, and location address. The values for the segment “location” are “X line”, “X station”, “5 minutes”, “Type 2 middle- and high-rise residences” for the attributes such as the railway line, station name, number of minutes on foot, and restricted area. An attribute value such as “dedicated area” is set. In the segment “building”, attribute values such as “used”, “January 2000”, and “A” are set for each attribute such as new construction category, date of construction, and seller brand evaluation. . For the segment “Residence”, “Second floor”, “Corner”, “*” for each attribute such as location floor, location, exclusive area, daily evaluation, indoor space expanse evaluation, and outdoor view release evaluation. * M ² ”,“ A ”,“ B ”,“ C ”
Each attribute value is set. In addition, the segment “management” includes information on management associations, parking lots, and bicycle parking lots, and the segment “image” includes photographs inside and outside the building (not shown).

  Note that the property information may include attribute information other than that illustrated in FIG. 2, or a part of the attribute information illustrated in FIG. 2 may be omitted.

The GUI providing unit 120 provides a user interface for transmitting and receiving information to and from the terminals S and B. In the present embodiment, the GUI providing unit 120 includes a property registration unit 121, a parameter setting unit 123, a conjoint survey unit 125, and a property presentation unit 127, as shown in FIG.
The property registration unit 121 provides a user interface for accepting registration of property information whose specific example is shown in FIG. 2 from the terminal S1 of the sales applicant.

  The parameter setting unit 123 generates a GUI for receiving setting of parameters (preference information) indicating the user's preference from the user who desires to purchase on the screens illustrated in specific examples in FIGS. 4 to 6 described later. While transmitting to the terminal B, the parameter is received from the terminal B. At this time, for example, the user is allowed to have an ideal range (for example, “A”) and allowance for various items (attributes) such as an area, a cityscape, and a distance from a station associated with each piece of property information. It is possible to set a possible range (for example, “B”), an unacceptable range (NG range), and the like. Of these, properties that fall within the ideal range and allowable range are subject to the conjoint analysis described below. Further, property information having at least one parameter in an unacceptable range is excluded from the object of recommendation (not evaluated) to the user by screening.

  The conjoint research unit 125 presents virtual property information for investigating the preferences of the user from the purchaser user on the screens shown in specific examples in FIGS. 7 and 8 to be described later, and the virtual property information. The operation input for ordering is received.

  Here, the virtual property information includes virtual parameters in which parameters set as “ideal range” and “allowable range” by parameter setting in the parameter setting unit 123 are set as attribute values. This is property information. The virtual property information does not need to be property information of a real property registered in the property information DB 110, and can be created virtually in order to grasp the user's preference. Each virtual property information has at least some attribute information values (attribute values) different from each other. The conjoint analysis unit 130 determines which attribute has a greater meaning for the user than other attributes when the user ranks the virtual property information having different attribute information. It can be understood by joint analysis.

  Note that the conjoint research unit 125 may create virtual property information for all combinations of attribute information, but if implemented in such a manner, the number of virtual property information becomes enormous, so that the user appropriately There is a higher chance that you will not be able to rank. Therefore, in this embodiment, the number of virtual properties to which the user ranks is reduced by a method using an orthogonal table of all concept methods (full profile method).

  The property presentation unit 127 outputs (transmits) each property information registered in the property information DB 110 to the terminal B based on the result evaluated by the property evaluation unit 140 according to the preference of the user who wants to purchase. Thereby, the user can suitably obtain property information in the order according to his / her preference. A specific example of the property presentation method by the property presentation unit 127 will be described later with reference to FIG.

  The conjoint analysis unit 130 performs conjoint analysis that analyzes the user's preference based on the result of the conjoint survey performed by the conjoint survey unit 125 to rank the virtual properties. In the conjoint analysis, it is possible to obtain the relative importance with respect to the attributes of each property information. There are various conjoint analysis methods such as least square analysis (multiple regression analysis), monotone regression analysis, logit model, probit model, and the like. Here, the case of using least square analysis will be described as an example. Details of the conjoint analysis performed by the conjoint analysis unit 130 will be described later with reference to FIG.

  The property evaluation unit 140 evaluates the attribute value of the attribute of each property information according to the relative importance between the attributes performed by the conjoint analysis unit 130, so that the user's preference degree for each property information Is calculated. A specific example of the property evaluation method by the property evaluation unit 140 will be described later with reference to FIG.

(2 Process flow)
Hereinafter, the processing flow of the information processing server 100 will be described with reference to FIG. FIG. 3 is a flowchart showing a flow of processing leading to property presentation in the information processing server 100. In the flowchart of FIG. 3, processing related to property registration is omitted.

  Here, each processing step, which will be described later, can be executed in any order or in parallel as long as there is no contradiction in processing contents, and other steps are added between the processing steps. Also good. Further, a step described as a single step for convenience can be executed by being divided into a plurality of steps, and a step described as being divided into a plurality of steps for convenience can be executed as one step.

  First, the information processing server 100 inputs a parameter (attribute value) from the terminal B by providing the terminal B with a user interface for parameter setting in response to a request from the terminal B of the user who is a purchase applicant. (S301). As described above, the parameter (attribute value) received here is an ideal range (for example, “A” for various items (attributes) associated with the property information, such as the area, cityscape, and distance from the station. ”), An allowable range (for example,“ B ”), and an unacceptable (NG range). Details of specific examples of parameters that can be set by the user will be described with reference to FIGS.

  The conjoint survey unit 125 performs the conjoint survey for the user based on the parameters received from the user in S301 in two stages (basic conjoint survey (S303 and S305) and premium conjoint survey (S307 and S309). Call it separately.) The basic conjoint survey and the premium conjoint survey have different attribute groups for performing user preference surveys. In this embodiment, in the basic conjoint survey, area, cityscape, station distance (minutes walk from the station), building age (or information on whether it is new or used), size, sunlight, and positional relationship (for example, in an apartment) Information such as whether it is a corner room, whether it is the top floor, whether it is the second floor or higher, etc.), and price information is the object of the survey. On the other hand, the premium conjoint survey covers area, seller brand (brand evaluation), view, expanse of space (whether the space feels wide), a sense of grade, and price.

  Thus, the reason why the conjoint survey is a two-stage survey is to reduce the number of comparisons of users in each survey. As the number of attributes to be investigated increases, the number of virtual properties that the user should rank increases exponentially. However, as described above, if the number of virtual properties increases, the user will be confused, and there is a high possibility that the user will not be able to properly rank. Therefore, in this embodiment, by dividing the attribute into two groups, the number of virtual properties required for the conjoint survey is reduced, and as a result, the user is easily ordered.

  In addition, among the attributes to be surveyed in the basic conjoint survey and the premium conjoint survey, the size and price are commonly surveyed items. This is to provide the same parameters for the basic conjoint survey and the premium conjoint survey, thereby clarifying the relationship between both attributes and integrating the results of both surveys. This integration of survey results will be described later with reference to FIGS.

  After the conjoint survey for the user by the conjoint survey unit 125 is completed, the conjoint analysis unit 130 performs a conjoint analysis for evaluating the relative importance between the attributes of the user (S311). In the conjoint analysis, a scale (corresponding to an evaluation value and corresponding to “Utility” described later) of how much the user evaluates when each attribute that each property information can have takes which attribute value is calculated. The

  The property registration unit 121 evaluates each property information registered in the property information DB 110 according to the scale obtained as a result of the conjoint survey by the conjoint analysis unit 130 (S313). More specifically, first, in the parameter setting of S301, all properties having attribute values in a range set as unacceptable (NG) by the user are screened (excluded). Next, by evaluating the remaining property information according to the scale obtained by the conjoint analysis with respect to the attribute values possessed by each, the evaluation value for each property information (estimated user preference) Is calculated. In the present embodiment, the property evaluation value calculated by the property evaluation unit 140 corresponds to the sum of evaluation values related to attribute values set in each property information.

  When all the processing for the property information registered in the property information DB 110 is completed, the property presentation unit 127 transmits (outputs) the property information to the terminal B operated by the user in the order of preference (S315). ). Thereby, the user can obtain a list of suitable property information in accordance with his / her preference.

  Thereafter, if a parameter change is input from the user (Yes in S317), the GUI providing unit 120 determines whether the parameter change is accompanied by a large change in the user's preference ( If the parameter change is large (Yes in S319), the process is repeated from S303. If the parameter change is small (No in S319), the process is repeated from S313. The case where the parameter change is large may be, for example, a case where the change range of the parameter set by the user exceeds a threshold value, or a case where the numerical values of a plurality of parameters are changed.

  As a general rule, if the parameters are changed, it is considered that it is more accurate to restart the conjoint survey. However, since the conjoint survey imposes a load on the user, Thus, when the change is small, the user load can be reduced by omitting the conjoint survey.

(3 Specific examples of processing)
Hereinafter, a specific example of processing of the information processing server 100 will be described with reference to FIGS. 3 to 11.

(3.1 Parameter setting)
First, a specific example of a method for receiving parameter settings from a user will be described with reference to FIGS. 4 to 6 are diagrams illustrating specific examples of the user interface provided by the parameter setting unit 123 to the terminal B when receiving parameter settings from the user. Hereinafter, it demonstrates in order.

(3.1.1 Area setting)
FIG. 4 is a diagram illustrating a specific example of a screen for the user to set an area related to the target property. In the screen of FIG. 4, a line 41 and a line 43 (also referred to as drawing lines 41 and 43) are input by the user using a mouse or the like, and each is an ideal area of the property that the user wants to search for. (A in the figure) is received from the user by a line 41, and an allowable area (B in the figure) is received by a line 43.

  The area setting method is not limited to this, and various methods are conceivable. For example, after designating a route, it may be possible to designate a method such as designating the number of minutes on foot from the station on the route, designating a school district, or designating with the strength of the ground.

(3.1.2 Cityscape setting)
FIG. 5 is a diagram illustrating a specific example of a screen for the user to set a cityscape related to the target property. In the example of FIG. 5, whether the user is ideal (“A” in the figure), acceptable (“B” in the figure), or unacceptable (“NG” in the figure) for each use area defined by the City Planning Law. ]) And a description area 53 for explaining each use area.

  In the setting area 51, as described above, an ideal area (A) can be allowed for each use area such as a first-class low-rise residential area, a second-type low-rise residential area, a first-type middle- and high-rise residential area, and the like. The user's preference information can be registered when the user inputs “◯” in either the region (B) or the unacceptable region (NG).

  In the explanation area 53, for each use area in the City Planning Act, such as a first-class low-rise residential area, a second-class low-rise residential area, a first-class middle- and high-rise residential area, An explanatory note is written. The user can register user preference information in the setting area 51 while referring to the description.

(3.1.3 Other parameter settings)
FIG. 6 is a diagram illustrating a specific example of a screen for the user to set other parameters. In the example of the screen of FIG. 6, the distance from the station, the new construction or the age of the building, the area, the position in the building, the daylight, and the price can be set.

  Regarding the distance from the station, the ideal range (A), acceptable range (B), and other (unacceptable range, NG) can be set by inputting a threshold value that delimits these ranges. ing. In terms of age and size, ideal age range (A), acceptable age range (B), and others (unacceptable range) are ideal for age. An area range (A), an allowable exclusive area range (B), and other areas (an unacceptable range) can be set by inputting a threshold value that delimits these ranges.

  As for the position in the building, “the top floor of the 4th floor or higher” for each of the “corner room” (a room where at least two directions are not adjacent to other rooms) and the “medium room” (other than the corner room) , “Middle hierarchy of 4th floor or higher”, “2-3 floor”, “1st floor” is ideal for buyer (A), acceptable (B), unacceptable (NG), It can be set by inputting "○". At this time, evaluation may be performed including frontage information.

  With regard to sunlight, the ideal daylight (in the example of FIG. 6 “generally southward, with almost nothing blocking the sunshine”) and acceptable sunlight (in the example of FIG. 6 “facing east or west, Alternatively, the user can set the conditions for unacceptable sunlight (in the example of FIG. 6, “not so much sun”).

  Regarding the price, you can set the budget range, the price range that is a little cheap and happy, the marginal maximum price range, and other (unacceptable range) by entering each threshold. .

  Note that parameters (attributes) that can be set by the user and items (attributes) to be subjected to conjoint investigation and conjoint analysis described later are not limited to those in the present embodiment. For example, another attribute such as a floor plan may be considered as a survey item.

(3.2 Conjoint Survey)
Next, a specific example of a screen related to the conjoint survey will be described with reference to FIGS. 7 and 8 are diagrams illustrating specific examples of the user interface provided to the terminal B by the conjoint survey unit 125 during the conjoint survey. Hereinafter, it demonstrates in order. In addition to the all-concept method (profile method) ranking method used in this embodiment, there are a number of methods such as a two-factor list method and a paired comparison method. Any method may be used as long as it is an investigation method that finally understands the degree of preference of each attribute of the user.

(3.2.1 Basic conjoint survey)
FIG. 7 is a diagram illustrating a specific example of a screen when performing a basic conjoint survey for a user. In the example of FIG. 7, it relates to a virtual property with attribute values for the area, cityscape, distance from the station, building age, area, sunlight, location in the building, and price, which are attributes related to the basic conjoint survey. Information is shown as card 1 and card 2.

  Here, the attribute values set for each card are based on the parameters input in “3.1” above, and as described above, each virtual property information used for the conjoint survey has an attribute value. At least some of them are different. In the example of FIG. 7, the age (Card 1 is 5 years old, Card 2 is 20 years old), the condition of the sun (Card 1 is less sun-lit, Card 2 is very sunny), the streets (Card 1 is very much Good cityscapes, card 2 is a cityscape with no problem), and other attribute values are the same.

  If the user is a card 1 or a card 2, the user selects which property he prefers. By repeating the ordering of these two virtual properties over and over again, the conjoint survey unit 125 creates a concatenated survey of the user rankings of information related to all virtual properties created for the basic conjoint survey. The part 125 can be obtained.

(3.2.2 Premium conjoint survey)
FIG. 8 is a diagram illustrating a specific example of a screen when a premium conjoint survey is performed on a user. In the example of FIG. 8, the information relating to the size, seller brand, view, space expanse, grade, and price, which are attributes related to the premium conjoint survey, is shown as a card 1 and a card 2.

  As in the case of the basic conjoint survey, each virtual property information shown on each card differs in at least a part of the attribute value. In the example of FIG. 8, the information regarding the size, the view, and the spread of the space is different, and the other attribute values are the same.

  The point that the user repeats the comparison of the two cards, so that the conjoint research unit 125 grasps the order of the users related to all the virtual properties created for the premium conjoint research. It is the same.

(3.3 Conjoint analysis)
Hereinafter, a specific example of conjoint analysis will be described with reference to FIG. Here, a case where least square analysis is used as a conjoint analysis method will be described as an example, but the present invention is not limited to this. Any analysis method may be used as long as it is an analysis method that shows the degree of preference for each attribute of the user.

  Here, the conjoint analysis method is described in an abstract manner. In the example of FIG. 9, eight conjoint investigation cards are created, and attribute values for the attributes of items A to E are set for each card. For example, with respect to item A, as shown in the table on the left side of FIG. 9A, the attribute value “a” is assigned to the cards 1, 2, 3 and 7, and the attribute value “b” is assigned to the cards 5 and 6. An attribute value “c” is set for each of the cards 4 and 8.

  Further, as a result of the conjoint investigation, it is assumed that the user ranks the cards in the order of cards 7, 1, 8, 5, 2, 6, 3, and 4. According to the ranking, these cards are given points from 1 to 8 points.

(3.3.1 Utility)
Hereinafter, a method for obtaining a relative evaluation value (Utility) for the user with respect to each value (attribute value) of the item A will be described with reference to FIG.

  The conjoint analysis unit 130 obtains the average value of the user's points set on the card having each of the attribute values a, b, and c that can be taken by the item A. For example, with regard to the attribute value a, the attribute value a is set for the cards 1, 2, 3 and 7, and the respective points are 7, 4, 2, and 8. Therefore, the average value is 5.25. It becomes a point. If the average values are similarly calculated for the attribute values b and c, they are 4 points and 3.5 points, respectively.

By adjusting this so that the average value becomes 0, Utility 1 for attribute value a, Utility-0.25 for attribute value b, and Utility-0.75 for attribute value c can be obtained.
This Utility corresponds to an index indicating how much the attribute value of each item (attribute) has an influence on the order of cards. That is, by calculating Utility for the attribute values of all items (attributes), it is possible to obtain the importance of each attribute value in consideration of the relative importance of each attribute.

(3.3.2 Importance)
Next, how to determine the importance indicating the importance of each attribute will be described with reference to FIG.

  For calculating the importance, first, the difference between the maximum value (max) and the minimum value (min) of the utility of each attribute (item A to item E in the example of FIG. 9) is calculated. In the example of FIG. 9, the difference between the maximum value and the minimum value of Utility related to item A is 1.75, item B is 1.5, item C is 3, item D is 1, item E is 2. 5. The importance of each attribute (item) can be calculated as the specific gravity in the total of these values. Since the sum of the differences between the maximum and minimum values for each item is 9.75, the specific gravity of Item A is calculated by dividing 1.75 by 9.75 and rounding off to four decimal places. 17.949. For other items (attributes), Importance can be calculated in the same manner.

(3.3.3 Results of conjoint analysis)
The result of the conjoint analysis performed by the method described in the above “3.5.1” and “3.5.2” will be described with reference to FIG. FIG. 10 is a diagram illustrating a specific example of Utility and Importance obtained as a result of conjoint analysis.

  As can be seen from FIG. 10, for each of the basic conjoint analysis and the premium conjoint analysis, Importance is calculated for each target attribute, and Utility is calculated for each attribute value.

  For example, taking the “area” attribute of a basic conjoint as an example, the “area” attribute has an importance of 17.76 and an attribute value A (corresponding to the area surrounded by the line 41 indicating the interface in FIG. 4). It can be seen that Utility (relative importance of the user with respect to the attribute value A) is 2.375, and Utility (relative importance of the user with respect to the attribute value B) is -2.375.

  Further, the attributes “width” and “price” are included in both the basic conjoint analysis and the premium conjoint analysis as described above. Therefore, by scaling the Utility based on at least one of these attributes, it is possible to make the Utility obtained as a result of the basic conjoint analysis and the Utility obtained as a result of the premium conjoint analysis relative to each other. .

  More specifically, for example, when scaling on the basis of “width”, each utility value of the premium conjoint is divided by 1.5 and multiplied by 0.875 to obtain the basic content. It is possible to scale each Utility of the joint.

(3.4 Evaluation and output of property information)
Finally, a property evaluation and property information presentation method will be described with reference to FIG. FIG. 11 is a diagram illustrating a specific example of a list of property information displayed on the terminal B of the user who is the purchase applicant by the property presenting unit 127 based on the result of the property evaluation unit 140 evaluating.

(3.4.1 Property evaluation)
First, a specific example of the property evaluation method will be described. In the example of FIG. 11, the first property (property name “XX court”) is “A” as the attribute value of the “area” attribute, “A” as the attribute value of the “cityscape” attribute, The distance attribute has an attribute value “A” and the building age attribute “new construction”.

  At this time, in the result of the conjoint analysis shown in FIG. 10, the utility of the attribute value “A” of the “area” attribute is 2.375. Therefore, the property uses this value as the evaluation value for the area attribute. be able to.

  However, actually, in the example of FIG. 11, 2.61 is given as the evaluation value for the area attribute. This is because, in the survey related to ordinary real estate, the user often overestimates the evaluation value of the price based on experience, so by increasing the Utility value other than the price attribute by 10% This is because this is corrected.

That is, the Utility (evaluation value) 2.61 for the attribute value “A” of the “Area” attribute related to the first property is multiplied by 1.1 to Utility 2.375 obtained as a result of the conjoint analysis. Obtained by rounding off. The user's evaluation value (Utility) for the “townscape”, “position”, and “daily” of the property can be similarly calculated.
Such correction is not always necessary, and it is conceivable that correction is not performed. Even in the case of correction, the numerical value and the attribute to be corrected can be changed as appropriate.

  Attributes such as building age, distance from the station, area, and price are parameters that can be expressed continuously in numerical values, so the values obtained as a result of conjoint analysis are the parameters (attribute values) set for the property. Is linearly scaled. Thereby, suitable property evaluation becomes possible.

  Here, among attribute values to be set in the property information, some attribute values may not be obtained due to reasons such as not being registered by the sales applicant. In such a case, a method of calculating the evaluation value of the property by setting the Utility value related to the attribute to 0 can be considered.

  Also, as a result of the conjoint analysis, when the Utility values for the attribute values “A” and “B” are reversed (that is, the Utility value for the attribute value “A” is greater than the Utility value for “B”). If it is low), it is considered an error in the survey, so it is possible to eliminate such numbers.

  Next, a method for calculating an evaluation value relating to an attribute related to premium conjoint analysis such as “seller” or “view” will be described. The first property has, for example, an attribute value “B” for the “Seller” attribute related to premium conjoint analysis, an attribute value “C” for the “view” attribute, and the like. As described above, the attribute value “B” of the “seller” attribute is −0.25 in the result of the conjoint analysis shown in FIG. 10 (the attribute value “standard” of the attribute “seller brand”). Yes. On the other hand, in order to match the scale of the premium conjoint analysis and the scale of the basic conjoint analysis, after dividing by 1.5 and multiplying by 0.875, the correction of the evaluation scale with the attribute “price” (1 .1), the user's evaluation scale −0.16 for the attribute value “B” of the “Seller” attribute can be obtained. Evaluation values can be similarly calculated for other attributes related to the premium conjoint analysis.

  In this way, the user's evaluation value for the property information is calculated by individually calculating the evaluation value of the user who wants to purchase the parameter (attribute) set for each property information and then adding all the values. A value can be obtained. If the first property in FIG. 11 is “◯ × Coat”, an evaluation value of 3.41 can be obtained.

(3.4.2 Output of property information)
Next, a method for outputting property information will be described. When the property evaluation unit 140 evaluates all property information to be evaluated by the method described in “3.6.1,” the property presentation unit 127 displays a specific example as shown in FIG. A list in which the property information is arranged in descending order of the evaluation value based on the evaluation result is displayed on the terminal B of the user who is a purchase applicant.

  In the example of FIG. 11, for each property information, the parameters (attribute values) set in each property information and evaluation values for the information are included in the list along with information such as location, property name, price, and the like.

(4 hardware configuration)
Hereinafter, an example of a hardware configuration when the information processing server 100 described above is realized by a computer will be described with reference to FIG. Note that the function of the information processing server 100 can be realized by dividing it into a plurality of devices.

  As illustrated in FIG. 12, the information processing server 100 includes a processor 1201, a memory 1203, a storage device 1205, an input interface (I / F) 1207, a data I / F 1209, a communication I / F 1211, and a display device 1213.

  The processor 1201 controls various processes in the information processing server 100 by executing a program stored in the memory 1203. For example, the GUI providing unit 120, the conjoint analysis unit 130, and the property evaluation unit 140 described with reference to FIG. 1 can be realized as a program mainly operating on the processor 1201 after being temporarily stored in the memory 1203.

  The memory 1203 is a storage medium such as a RAM (Random Access Memory). The memory 1203 temporarily stores a program code of a program executed by the processor 1201 and data necessary for executing the program. For example, a stack area necessary for program execution is secured in the storage area of the memory 1203.

  The storage device 1205 is a non-volatile storage medium such as a hard disk or a flash memory. The storage device 1205 stores various data such as an operating system, various programs for realizing the GUI providing unit 120, the conjoint analysis unit 130, and the property evaluation unit 140, and a database for realizing the property information DB 110. . The program and data stored in the storage device 1205 and, if necessary, loaded into the memory 1203 are referred to by the processor 1201.

  The input I / F 1207 is a device for receiving input from the user. Specific examples of the input I / F 1207 include a keyboard, a mouse, a touch panel, and various sensors. The input I / F 1207 may be connected to the information processing server 100 via an interface such as a USB (Universal Serial Bus).

  The data I / F 1209 is a device for inputting data from outside the information processing server 100. Specific examples of the data I / F 1209 include a drive device for reading data stored in various storage media. The data I / F 1209 may be provided outside the information processing server 100. In this case, the data I / F 1209 is connected to the information processing server 100 via an interface such as USB.

  The communication I / F 1211 is a device for data communication with a device external to the information processing server 100 by wire or wireless. For example, communication with the terminal S or the terminal B is performed via the network N whose communication I / F 1211 is the Internet or the like. It is conceivable that the communication I / F 1211 is provided outside the information processing server 100. In that case, the communication I / F 1211 is connected to the information processing server 100 via an interface such as a USB.

  The display device 1213 is a device for displaying various information. Specific examples of the display device 1213 include a liquid crystal display and an organic EL (Electro-Luminescence) display. The display device 1213 may be provided outside the information processing server 100. In that case, the display device 1213 is connected to the information processing server 100 via, for example, a display cable.

(5 Effects of this embodiment)
As described above, the information processing server 100 according to the present embodiment analyzes each user's preference by conjoint analysis, and then matches each user's preference based on the analysis result. Suitable product information (property information) can be provided to the user. At this time, by presenting the property information to the user with an order, the user can see the properties that are preferable to him in order, so that the user can simply compare with the case of screening properties that do not meet the conditions. This makes it possible to reduce the time and effort required to compare properties.

  Also, as shown in FIGS. 4 to 6, the user can set parameters in advance, and performs conjoint analysis and property evaluation according to the parameter settings, so that such parameter setting cannot be performed. In addition, it is possible to respond to the user's preference in detail.

  Furthermore, in real estate transactions, new properties and second-hand properties are often managed separately, but if the method of this embodiment is used, the user compares them according to the same evaluation scale on the same system. This makes it easier to study.

(6 Appendix)
Note that the configurations of the above-described embodiments may be combined or some components may be replaced. The configuration of the present invention is not limited to the above-described embodiment, and various modifications may be made without departing from the scope of the present invention.

  DESCRIPTION OF SYMBOLS 1 ... System, 41, 43 ... Drawing line, 51 ... Setting area, 53 ... Description area, 100 ... Information processing server, 110 ... Property information database, 120 ... GUI Providing unit, 121 ... Property registration unit, 123 ... Parameter setting unit, 125 ... Conjoint investigation unit, 127 ... Property presentation unit, 130 ... Conjoint analysis unit, 140 ... Property Evaluation unit, 1201... Processor, 1203... Memory, 1205... Storage device, 1207... Input interface, 1209.

Claims (8)

A registration unit that receives registration of multiple product information with attribute values set for multiple attributes,
A first input unit that receives user's preference information for each of the plurality of attributes and receives information on one or more threshold values for dividing the attribute value into a plurality of ranges;
A second input unit that receives a user input for ordering a plurality of virtual products in which at least some of the attribute values of the plurality of attributes are different from each other based on the preference information;
Based on the order for the plurality of virtual products, a calculation unit for obtaining relative importance between the plurality of attributes in the user,
An evaluation unit that evaluates each of the plurality of product information based on attribute values of the plurality of attributes of each product information and relative importance between the plurality of attributes;
An output unit that outputs the plurality of product information based on an evaluation result by the evaluation unit ;
The second input unit is divided into a first attribute group and a second attribute group that are a part of the plurality of attributes, respectively, and receives user input for ordering the plurality of virtual product groups,
The first attribute group and the second attribute group include one or more common attributes,
Information processing system. The first input unit receives, as part of the preference information, information related to an attribute value that is not subject to evaluation for the plurality of attributes,
The output unit outputs only product information in which all of the plurality of attributes are within an evaluation target among the plurality of product information.
The information processing system according to claim 1. The second input unit attaches an order of the plurality of virtual products as a whole by receiving a user input for determining the order of some virtual products among the plurality of virtual products a plurality of times.
The information processing system according to claim 1 or 2. The output unit outputs the plurality of product information in descending order of evaluation by the evaluation unit.
The information processing system according to any one of claims 1 to 3 . The product information is a real estate product.
The information processing system according to any one of claims 1 to 4 . The attributes include at least a part of an area, cityscape, distance from the station, age, size, position in the building, sunlight, and price.
The information processing system according to claim 5 . Receiving registration of a plurality of product information each having an attribute value set to a plurality of attributes;
A first input step of receiving user preference information for each of the plurality of attributes and receiving information relating to one or more threshold values for dividing the attribute value into a plurality of ranges;
A second input step of receiving a user input for ordering a plurality of virtual products in which at least some of the attribute values of the plurality of attributes are different from each other based on the preference information;
Obtaining a relative importance between the plurality of attributes in the user based on an order for the virtual product;
Evaluating each of the plurality of product information based on attribute values of the plurality of attributes included in each product information and relative importance between the plurality of attributes;
Based on the results of the evaluation, have the information processing apparatus line and outputting the plurality of product information,
In the second input step, each of the first attribute group and the second attribute group, which are a part of the plurality of attributes, is divided into user attributes for ranking the plurality of virtual commodity groups,
The first attribute group and the second attribute group include one or more common attributes,
Information processing method. Processing to receive registration of multiple product information with attribute values set to multiple attributes,
A first input process for receiving user preference information for each of the plurality of attributes and receiving information relating to one or more threshold values for dividing the attribute value into a plurality of ranges;
A second input process for receiving user input for ordering a plurality of virtual products, each of which has at least some attribute values different from each other based on the preference information;
Based on the order for the virtual product, a process for obtaining a relative importance between the plurality of attributes in the user;
A process for evaluating each of the plurality of product information based on attribute values of the plurality of attributes included in each product information and relative importance between the plurality of attributes;
A program for causing a computer to execute a process of outputting the plurality of product information based on a result of the evaluation ,
In the second input process, each of the plurality of attributes is divided into a first attribute group and a second attribute group, and a user input for ordering the plurality of virtual product groups is received.
The first attribute group and the second attribute group include one or more common attributes,
Program .