Wearing comfort prediction system and wearing comfort prediction program

Abstract

To provide a wearing comfort prediction system, a wearing comfort prediction method, and a wearing comfort prediction program that can predict an appropriate wearing comfort prediction value.SOLUTION: A wearing comfort prediction system 10 includes a storage unit 11 that stores wearing comfort history data of the clothes for a user, and a prediction unit 12 that predicts wearing comfort, on the basis of the weight of the user measured at the time of predicting the wearing comfort and the wearing comfort history data. The wearing comfort history data includes one or more data sets for each date and time. The data set includes a wearing date on which the clothes are worn, the weight on the wearing date, and wearing comfort information indicating the wearing comfort at the wearing time. The prediction unit 12 extracts a data set including the weight equal to or close to the measured weight from the wearing comfort history data by comparison processing of comparing the measured weight with the weight of the data set, and sets the wearing comfort of the extracted data set as a wearing comfort prediction value of the clothes for the user.SELECTED DRAWING: Figure 1

Description

Technical Field

[0001] The present disclosure relates to a comfort prediction system, a comfort prediction method, and a comfort prediction program.

Background Art

[0002] A system for supporting clothing coordination is known (for example, Patent Document 1). The system of Patent Document 1 includes a raw life log analysis unit and a recommendation logic unit. The raw life log analysis unit receives a log consisting of the going-out location, the temperature, humidity, weather, and clothing comfort recorded by the mobile communication means, and converts it into structured data. The recommendation logic unit searches for clothing that meets the user's desires. The recommendation logic unit calculates the similarity between the dressing life log and the user request information and recommends appropriate clothing for the user's request information.

Prior Art Documents

Patent Documents

[0003]

Patent Document 1

Summary of the Invention

Problems to be Solved by the Invention

[0004] By the way, the comfort of clothing changes daily. It is difficult to provide clothing that is comfortable to wear when putting on clothes. From this perspective, there is room for improvement in systems, methods, and programs for providing comfortable clothing.

Means for Solving the Problems

[0005] (1) The comfort prediction system for solving the above problems is a comfort prediction system for predicting the comfort of clothing, comprising a storage unit for storing the comfort history data of the clothing for the user, and a prediction unit for predicting the comfort based on the measured weight of the user at the time of comfort prediction and the comfort history data. The comfort history data includes one or more data sets for each date and time. The data set includes the wearing date of wearing the clothing, the weight on the wearing date, and comfort information indicating the comfort of the clothing at the time of wearing. The prediction unit extracts, by a comparison process of comparing the measured weight with the weight of the data set, the data set including the weight equal to or close to the measured weight from the comfort history data, and sets the comfort of the extracted data set as the comfort prediction value of the clothing for the user. According to this configuration, since the comfort prediction value is set based on the comfort when the weight is equal to or close to the measured weight, appropriate comfort can be provided.

[0006] (2) In the comfort prediction system of the building described in (1) above, when the prediction unit detects only one data set including the weight equal to the measured weight by the comparison process, the prediction unit sets the comfort of the data set detected by the comparison process as the comfort prediction value of the clothing for the user. According to this configuration, appropriate comfort can be provided.

[0007] (3) In the comfort prediction system of the building described in (1) or (2) above, when the prediction unit detects a plurality of data sets including the weight equal to the measured weight by the comparison process, the prediction unit sets the comfort with the highest difficulty of wearing among the comforts of each of the plurality of data sets as the comfort prediction value of the clothing for the user. According to this configuration, appropriate comfort can be provided.

[0008] (4) In the comfort prediction system for any one of the buildings (1) to (3) above, when the prediction unit determines, by the comparison process, that there is no dataset including a weight equal to the measured weight, the comfort of the dataset including the weight closest to the measured weight is set as the comfort prediction value of the clothing for the user. According to this configuration, appropriate comfort can be provided.

[0009] (5) In the comfort prediction system for any one of the buildings (1) to (4) above, when the prediction unit determines, by the comparison process, that there is no dataset including a weight equal to the measured weight, and determines that for all of the datasets, the weight of the dataset is greater than the measured weight, or determines that for all of the datasets, the weight of the dataset is less than the measured weight, the comfort of the dataset including the weight closest to the measured weight is set as the comfort prediction value of the clothing for the user. According to this configuration, appropriate comfort can be provided.

[0010] (6) In the comfort prediction system for any one of the buildings (1) to (5) above, when the prediction unit determines that there is no dataset including a weight equal to the measured weight, and determines, by the comparison process, that there are datasets including weights greater than the measured weight and datasets including weights less than the measured weight in the comfort history data, and the comfort of the dataset including a weight greater than and closest to the measured weight is equal to the comfort of the dataset including a weight less than and closest to the measured weight, the comfort of the dataset including a weight greater than and closest to the measured weight, or the comfort of the dataset including a weight less than and closest to the measured weight is set as the comfort prediction value of the clothing for the user. According to this configuration, appropriate comfort can be provided.

[0011] (7) In the comfort prediction system for any one of the buildings (1) to (6) above, when the prediction unit determines that there is no dataset including a weight equal to the measured weight, and when, by the comparison process, it is determined that there are datasets including weights greater than the measured weight and datasets including weights smaller than the measured weight in the comfort history data, and when the comfort of the dataset including a weight greater than the measured weight and closest to the measured weight is different from the comfort of the dataset including a weight smaller than the measured weight and closest to the measured weight, the comfort of the dataset including the weight closest to the measured weight is set as the comfort prediction value for the clothing for the user. According to this configuration, appropriate comfort can be provided.

[0012] (8) In the comfort prediction system for any one of the buildings (1) to (7) above, when the prediction unit determines that there is no dataset including a weight equal to the measured weight, and when, by the comparison process, it is determined that there are datasets including weights greater than the measured weight and datasets including weights smaller than the measured weight in the comfort history data, and when the dataset including a weight greater than the measured weight and closest to the measured weight is defined as the first dataset, and the dataset including a weight smaller than the measured weight and closest to the measured weight is defined as the second dataset, and when the comfort of the first dataset is different from the comfort of the second dataset, and when the difference between the weight of the first dataset and the measured weight is equal to the difference between the weight of the second dataset and the measured weight, the comfort with the highest discomfort level among the comfort of the first dataset and the comfort of the second dataset is set as the comfort prediction value for the clothing for the user. According to this configuration, appropriate comfort can be provided.

[0013] (9) In the comfort prediction system for any one of the buildings (1) to (8) above, the dataset further includes the comfort prediction value and prediction accuracy information, and the prediction accuracy information has, as information, a match value indicating that the comfort prediction value matches the comfort level, and a non-match value indicating that the comfort prediction value does not match the comfort level. According to this configuration, the validity of the comfort prediction value can be evaluated.

[0014] (10) In the comfort prediction system for the building of (9) above, further provided is a validity evaluation unit that evaluates the validity of the comfort prediction value for predicting the comfort level. The storage unit stores a database including the comfort history data of each of a plurality of clothes. When the prediction unit predicts the comfort prediction value of the clothes, the validity evaluation unit determines the validity of the comfort prediction value based on whether there is information indicating that the prediction accuracy information is a non-match value from the past comfort history data within a predetermined period before the prediction date including the prediction time for one or more other clothes different from the clothes. According to this configuration, the validity of the comfort prediction value can be evaluated.

[0015] (11) In the comfort prediction system for the building of (10) above, when the prediction unit predicts the comfort prediction value of the clothes, if there is no information indicating that the prediction accuracy information is a non-match value in the past comfort history data within a predetermined period from the prediction date for one or more of the other clothes, the validity evaluation unit determines that the comfort prediction value is valid and determines the comfort prediction value. According to this configuration, a high-accuracy comfort prediction value can be provided.

[0016] (12) In the comfort prediction system for the building of (10) or (11) above, when the prediction unit predicts the comfort prediction value of the clothes, if there is information indicating that the prediction accuracy information is a non-match value from the past comfort history data within a predetermined period from the prediction date for one or more of the other clothes, the validity evaluation unit determines that the comfort prediction value is not valid. According to this configuration, it is possible to deal with an invalid comfort prediction value.

[0017] (13) In the comfort prediction system for a building as described in (12) above, the system further includes a correction unit that corrects the comfort prediction value. When the validity evaluation unit determines that the comfort prediction value is not valid, the correction unit corrects the comfort prediction value based on a comparison between a first weight deviation degree and a second weight deviation degree. The first weight deviation degree is the difference between the weight included in the dataset of the other clothing item used for predicting the comfort of the other clothing item and the weight of the user on the day when the comfort of the other clothing item was predicted, for the other clothing item related to the dataset in which the prediction accuracy information is a mismatch value. The second weight deviation degree is the difference between the weight included in the dataset of the clothing item used for predicting the comfort of the clothing item and the weight of the user on the day when the comfort of the clothing item is predicted. According to this configuration, when it is determined that the comfort prediction value is not valid, the comfort prediction value is corrected, so a highly accurate comfort prediction value can be provided.

[0018] (14) In the comfort prediction system for a building as described in (13) above, when the second weight deviation degree is greater than the first weight deviation degree, the correction unit corrects the comfort prediction value toward the difficult-to-wear side. According to this configuration, a highly accurate comfort prediction value can be provided.

[0019] (15) In the comfort prediction system for a building according to any one of (1) to (14) above, the system further includes a selection unit that selects, from a plurality of clothing items, the clothing item with high comfort. The prediction unit predicts the comfort prediction value for the user for each of the plurality of clothing items, and the selection unit selects one or more of the recommended clothing items based on the comfort prediction values of each of the plurality of clothing items. According to this configuration, one or more clothing items can be provided as clothing items with high comfort.

[0020] (16) The method for predicting wearing comfort that solves the above problems is a method for predicting the wearing comfort of clothing, and includes a preparation step of preparing wearing comfort history data of the clothing for a user, and a prediction step of predicting the wearing comfort based on the measured weight of the user at the time of predicting the wearing comfort and the wearing comfort history data. The wearing comfort history data includes one or more data sets for each date and time. The data set includes the wearing date when the clothing was worn, the weight on the wearing date, and wearing comfort information indicating the wearing comfort of the clothing at the time of wearing. In the prediction step, by a comparison process of comparing the measured weight with the weight of the data set, a data set including a weight equal to or close to the measured weight is extracted from the wearing comfort history data, and the wearing comfort of the extracted data set is used as the predicted value of the wearing comfort of the clothing for the user.

[0021] (17) The program for predicting wearing comfort that solves the above problems is a program for predicting the wearing comfort of clothing, and includes a preparation step of causing a computer to prepare wearing comfort history data of the clothing for a user, and a prediction step of causing the computer to predict the wearing comfort based on the measured weight of the user at the time of predicting the wearing comfort and the wearing comfort history data. The wearing comfort history data includes one or more data sets for each date and time. The data set includes the wearing date when the clothing was worn, the weight on the wearing date, and wearing comfort information indicating the wearing comfort of the clothing at the time of wearing. In the prediction step, by a comparison process of comparing the measured weight with the weight of the data set, the computer is caused to extract a data set including a weight equal to or close to the measured weight from the wearing comfort history data, and the wearing comfort of the extracted data set is used as the predicted value of the wearing comfort of the clothing for the user.

Advantages of the Invention

[0022] According to the wearing comfort prediction system, the wearing comfort prediction method, and the wearing comfort prediction program of the present disclosure, an appropriate predicted value of wearing comfort can be provided.

Brief Description of the Drawings

[0023]

Figure 1

Figure 2

Figure 3

Figure 4

Figure 5

Figure 6

Figure 7

Figure 8

Figure 9

Figure 10

Figure 11

Figure 12

Figure 13

Figure 14

Figure 15

Figure 16

Figure 17

Figure 18

Figure 19

Figure 20

Mode for Carrying Out the Invention

[0024] The comfort prediction system 10 of this embodiment will be described with reference to FIGS. 1 to 20. The comfort prediction system 10 predicts the comfort of clothing 7. The comfort prediction system 10 is used when the user selects clothing 7. The comfort prediction system 10 provides clothing 7 with good comfort for the user.

[0025] As shown in FIG. 1, the comfort prediction system 10 is connected to the imaging device 2, the electronic weighing scale 3, the electronic closet 4, and the display device 5 via the network N.

[0026] The fitting room 1 will be described with reference to FIG. 2. The fitting room 1 is configured to be able to use the comfort prediction system 10. The fitting room 1 is provided with the imaging device 2, the electronic weighing scale 3, the electronic closet 4, and the display device 5. The comfort prediction system 10 may be provided in the fitting room 1 or may be provided outside the fitting room 1. The comfort prediction system 10 may be provided on the cloud. The imaging device 2, the electronic weighing scale 3, the electronic closet 4, and the display device 5 are directly or connected to the comfort prediction system 10 via the network N. The imaging device 2, the electronic weighing scale 3, the electronic closet 4, and the display device 5 are configured to be able to exchange information with the comfort prediction system 10.

[0027] The photographing device 2 photographs a person entering the fitting room 1. The photographing device 2 identifies a person by image recognition. A user is registered in the photographing device 2 in advance. The photographing device 2 identifies the user who has entered from among the registered users. The photographing device 2 sends user information regarding the identified user to the comfort prediction system 10. Based on receiving the user information, the comfort prediction system 10 prepares a database of the clothes 7 of the user related to the user information.

[0028] The electronic weighing scale 3 measures the weight of a person getting on the electronic weighing scale 3. The electronic weighing scale 3 sends the measured weight as the measured weight to the comfort prediction system 10. When receiving the measured weight, the comfort prediction system 10 stores it as the measured weight of the user who has entered.

[0029] Referring to FIG. 3, the electronic closet 4 will be described. The electronic closet 4 houses the clothes 7. The clothes 7 housed in the electronic closet 4 have a first electronic tag 7A including first identification information. The electronic closet 4 manages the clothes 7 so that each of the clothes 7 can be identified. The electronic closet 4 includes an electronic hanger 8 and a reader. The electronic hanger 8 has a notification unit 8B including a second electronic tag 8A. The notification unit 8B has a light emitting device or a speaker. The reader reads the second identification information of the electronic hanger 8 and the first identification information of the clothes 7. The reader sends the second identification information and the first identification information as a set to the comfort prediction system 10. For example, when the user hangs the clothes 7 on the electronic hanger 8, the user causes the reader to read the second identification information of the electronic hanger 8 and the first identification information of the clothes 7. In the comfort prediction system 10, the electronic hanger 8 with the second identification information and the clothes 7 with the first identification information are stored as a set of data. Thereby, the electronic hanger 8 and the clothes 7 can be managed as a set. Based on receiving the first identification information or the second identification information from the comfort prediction system 10, the electronic closet 4 generates a notification sound or notification light for the electronic hanger 8 corresponding to the second identification information. In this way, the comfort prediction system 10 can present a specific piece of clothing 7 to the user by outputting the first identification information or the second identification information.

[0030] The display device 5 displays an image of the clothing 7 and information about the clothing 7. The display device 5 has a display unit corresponding to the size of a person. The display device 5 is also called a smart mirror. The display device 5 acquires the image of the clothing 7 and the information about the clothing 7 from the comfort prediction system 10, and displays the acquired image of the clothing 7 and the information about the clothing 7. The display device 5 receives an input from a person. In one example, the display unit of the display device 5 has a touch panel. The display device 5 replaces the image of the clothing 7 displayed on the display unit by a person's operation. Specifically, the display device 5 sends a signal based on a person's operation to the comfort prediction system 10, and thereby acquires the information about the next clothing 7 prepared in the comfort prediction system 10.

[0031] As shown in FIG. 1, the comfort prediction system 10 includes a storage unit 11 and a prediction unit 12. The comfort prediction system 10 may further include a validity evaluation unit 13. The comfort prediction system 10 may further include a correction unit 14. The comfort prediction system 10 may further include a selection unit 15.

[0032] [Storage Unit] The storage unit 11 records the data sent from the display device 5 to the comfort prediction system 10 in the comfort history data 20. For example, the storage unit 11 records the user's evaluation regarding the comfort of the clothing 7 in the comfort history data 20.

[0033] The storage unit 11 stores a database. The database is provided for each registered person. For example, a database is provided for each member of a family. In the database, one comfort history data 20 is associated with each of a plurality of clothing 7. That is, the comfort history data 20 is configured for each clothing 7.

[0034] Referring to FIG. 4, the comfort history data 20 will be described for one piece of clothing 7 selected from a plurality of clothing 7.

[0035] The comfort history data 20 includes an identification number corresponding to the clothing 7. The identification number is a number individually set for the clothing 7. The comfort history data 20 includes an image of the clothing 7. The image of the clothing 7 may be an image of only the clothing 7 or an image when worn. The comfort history data 20 may include a plurality of images taken from different directions for one piece of clothing 7.

[0036] The comfort history data 20 includes information about the clothing 7. The information about the clothing 7 may include information about the type of the clothing 7, information about the use of the clothing 7, information about the suitable season for wearing, information about the elapsed time since the purchase of the clothing 7, information about the user's preference for the clothing 7, etc.

[0037] The comfort history data 20 includes owner information indicating the owner of the clothing 7. The owner information is information used when the comfort prediction system 10 extracts the user's clothing 7 when the user is identified.

[0038] The comfort history data 20 includes one or more data sets 21 for each date and time. The data set 21 includes the wearing date when the clothing 7 was worn, the weight on the wearing date, and comfort information 30 indicating the comfort 31 of the clothing 7 when worn. The data set 21 may include the date when the clothing 7 was purchased.

[0039] The weight is the value measured by the electronic weighing scale 3. The user steps on the electronic weighing scale 3 when entering the fitting room 1. When the electronic weighing scale 3 measures the user's weight, it sends the measured weight as the measured weight to the comfort prediction system 10. The comfort prediction system 10 stores the measured weight in the storage unit 11 together with the date and time. Also, the comfort prediction system 10 associates the user identified by the imaging device 2 at the time of entry with the measured weight. In this way, the measured weight is recorded in the comfort history data 20 of the user.

[0040] The comfort information 30 is input by the user via the display device 5. When the user tries on the clothing 7 displayed on the display device 5, in the state where the clothing 7 is displayed, the user inputs the comfort level 31 of the clothing 7 via the touch panel of the display device 5. The display device 5 sends the input information to the comfort prediction system 10 as the comfort information 30. The comfort prediction system 10 stores the comfort information 30 about the clothing 7 displayed on the display device 5 in the storage unit 11 together with the date and time. Also, the comfort prediction system 10 associates the user identified by the photographing device 2 at the time of entry, the clothing 7, and the comfort information 30 of the clothing 7. Then, the comfort prediction system 10 updates the comfort history data 20 with this information.

[0041] The comfort information 30 indicates the user's comfort level 31 regarding the clothing 7. The comfort information 30 includes information corresponding to each body position. For example, when the clothing 7 is flare pants, the comfort information 30 includes first comfort information corresponding to the waist, second comfort information corresponding to the thighs, and third comfort information corresponding to the buttocks.

[0042] In FIG. 4, a comfortable fit 31 with the correct size is indicated by "C". A tight fit 31 is indicated by "T". A loose fit 31 is indicated by "L".

[0043] The data set 21 may further include a comfort prediction value 40 and prediction accuracy information 45. The prediction accuracy information 45 has a match value and a non-match value. The match value indicates that the comfort prediction value 40 and the comfort level 31 match. In the present embodiment, the match between the comfort prediction value 40 and the comfort level 31 indicates that all three pieces of information, namely the first comfort information, the second comfort information, and the third comfort information, match. The non-match value indicates that the comfort prediction value 40 and the comfort level 31 do not match. In the present embodiment, the non-match between the comfort prediction value 40 and the comfort level 31 indicates that at least one of the three pieces of information, namely the first comfort information, the second comfort information, and the third comfort information, does not match. In FIGS. 4 and 5, the match value is indicated by a circle, and the non-match value is indicated by a cross.

[0044] The dataset 21 may include reference information 34. The reference information 34 is information used as the basis for predicting the comfort prediction value 40 when deriving the comfort prediction value 40. Specifically, the reference information 34 includes the date of the past dataset 21 used as the basis for predicting the comfort prediction value 40 (in FIG. 5, the reference date) and the weight of the past dataset 21 (in FIG. 5, the reference weight).

[0045] The dataset 21 may include a wearing and going-out history 33 indicating that the user went out wearing the clothing 7.

[0046] Furthermore, the dataset 21 may include period information 35 (in FIG. 4, "within a predetermined period from the current day") indicating whether it is within a predetermined period in the past from the day when the clothing 7 is selected (in FIG. 4, the current day). The period information 35 is updated each time the comfort prediction system 10 is activated. In the example of FIG. 4, data within two weeks is given the information "within 2WEEK".

[0047] FIG. 5 shows data extracted as being within a predetermined period from the current day based on the period information 35 from among a plurality of the user's clothing 7. In this example, assuming the current day is June 10, 2023, data of the clothing 7 within two weeks in the past from that day is extracted. According to the period information 35, data within a predetermined period in the past from the current day can be quickly extracted easily.

[0048] [Display device] With reference to FIGS. 6 to 8, the operation of the display device 5 will be described. When the prediction unit 12 sets the comfort prediction value 40 for all of the user's clothing 7, the display device 5 displays recommended clothing 7 that the user can wear comfortably. Specifically, the display device 5 displays an image of the clothing 7 of the dataset 21 selected by the selection unit 15 based on the comfort prediction value 40.

[0049] Figure 6 shows the screen of the display device 5 showing an image of the recommended clothing 7. On the screen, a "Previous" button for returning to the previous image, a "Next" button for proceeding to the next image, and a "Try On" button for selecting the clothing 7 for wearing are shown. When the user touches the "Try On" button, the first identification information of the clothing 7 is sent to the comfort prediction system 10. At this time, the comfort prediction system 10 issues a command to notify the electronic hanger 8 associated with the first identification information. The electronic hanger 8 of the clothing 7 housed in the electronic closet 4 notifies by sound or light. As a result, the user can easily take out the desired clothing 7.

[0050] As shown in Figure 7, when the user touches the "Try On" button, along with the notification of the electronic hanger 8, a column for the comfort prediction value predicted by the comfort prediction system 10 appears on the display unit. In this example, the comfort prediction values are shown for the waist, thigh, and buttocks respectively.

[0051] As shown in Figure 8, when the user touches the "Try On" button again, an evaluation column appears on the display unit. Buttons for evaluating comfort are displayed in the evaluation column. After trying on, the user can evaluate the clothing 7. In this example, the comfort can be evaluated for the waist, thigh, and buttocks respectively. The display device 5 sends the user's evaluation to the comfort prediction system 10. The storage unit 11 of the comfort prediction system 10 records the user's evaluation of the clothing 7 in the comfort history data 20 of that clothing 7. In this way, each time the clothing 7 is worn, the comfort history data 20 is updated.

[0052] Figure 8 shows the screen when the evaluation of the clothing 7 has been input. As shown in Figure 8, when the evaluation of the clothing 7 is completed, a "Decide" button and a "Select Another" button are displayed. The "Decide" button is for deciding to wear the clothing 7. The "Select Another" button is for displaying the next recommended clothing 7.

[0053] [Prediction Unit] The prediction unit 12 predicts the comfort of all of the user's clothes 7 or the clothes 7 within a predetermined range of the user. The clothes 7 within the predetermined range are specified by the user. An example of the predetermined range is classified by season.

[0054] The prediction unit 12 predicts a comfort prediction value 40 for the user for each of a plurality of clothes 7.

[0055] The prediction unit 12 predicts the comfort for each of the clothes 7 based on the measured weight of the user at the time of predicting comfort and the comfort history data 20. In the present embodiment, at the time of predicting comfort, it is the same as when choosing the clothes 7. The prediction date including the time of predicting comfort is the current day.

[0056] The prediction unit 12 predicts the comfort for the user of the clothes 7 (hereinafter referred to as "target clothes 7") displayed on the display unit of the display device 5 based on the comfort history data 20 for the clothes 7. The user is the person who measured the weight.

[0057] The prediction unit 12 extracts a data set 21 including a weight equal to or close to the measured weight from the comfort history data 20 by a comparison process. The extracted data set 21 is data for the basis of predicting the comfort prediction value 40. The prediction unit 12 sets the comfort 31 of the extracted data set 21 as the comfort prediction value 40 for the clothes 7 for the user. The data set 21 for the basis of prediction is recorded in the comfort history data 20 together with the measured weight as data for the current day as reference information.

[0058] In the above-described comparison process, the prediction unit 12 compares the measured weight with the weight of the data set 21 for all of the data sets 21 in the comfort history data 20.

[0059] With reference to FIGS. 9 to 16, several examples will be described regarding the derivation of the comfort prediction value 40 for a predetermined piece of clothing 7. In FIGS. 9 to 16, the black circles are past data. The white circles are current-day data. Here, the current day indicates the day when the comfort prediction system 10 is predicting the comfort of the clothes 7.

[0060] The following (A1) and (A2) are examples when there is data equal to the measured weight in the wearing comfort history data 20 for the target clothing. (A3) to (A7) are examples when there is no data equal to the measured weight in the wearing comfort history data 20.

[0061] (A1) As shown in FIG. 9, when the prediction unit 12 detects only one data set 21 including a weight equal to the measured weight by the comparison process, the wearing comfort 31 of the data set 21 detected by the comparison process is set as the wearing comfort prediction value 40 of the clothing 7 for the user.

[0062] (A2) As shown in FIG. 10, when the prediction unit 12 detects a plurality of data sets 21 including a weight equal to the measured weight by the comparison process, the wearing comfort 31 with the highest wearing discomfort among the wearing comforts 31 of the plurality of data sets 21 is set as the wearing comfort prediction value 40 of the clothing 7 for the user.

[0063] In the present embodiment, the wearing discomfort is defined as follows. In the order of loose wearing comfort, comfortable wearing comfort, and tight wearing comfort, the latter has a higher wearing discomfort. That is, the tight wearing comfort has the highest wearing discomfort.

[0064] (A3) As shown in FIG. 11, when the prediction unit 12 determines that there is no data set 21 including a weight equal to the measured weight by the comparison process, the wearing comfort 31 of the data set 21 including the weight closest to the measured weight is set as the wearing comfort prediction value 40 of the clothing 7 for the user.

[0065] (A4) As shown in FIG. 11, when the following X1 and X2 are satisfied, the wearing comfort 31 of the data set 21 including the weight closest to the measured weight is set as the wearing comfort prediction value 40 of the clothing 7 for the user. X1 is the case where "it is determined by comparison processing that there is no dataset 21 including a weight equal to the measured weight". X2 is the case where "for all of the datasets 21, it is determined that the weight of the dataset 21 is greater than the measured weight, or for all of the datasets 21, it is determined that the weight of the dataset 21 is less than the measured weight". In the example of FIG. 11, the prediction unit 12 determines that for all of the datasets 21, the weight of the dataset 21 is less than the measured weight.

[0066] (A5) As shown in FIG. 12, when the prediction unit 12 satisfies the following Y1, Y2, and Y3, the comfort level 31 of YA or the comfort level 31 of YB is set as the comfort level prediction value 40 of the clothing 7 for the user.

[0067] Y1 is the case where "it is determined that there is no dataset 21 including a weight equal to the measured weight". Y2 is the case where "it is determined by comparison processing that there are a dataset 21 including a weight greater than the measured weight and a dataset 21 including a weight less than the measured weight within the comfort level history data 20". Y3 is the case where "the comfort level 31 of the dataset 21 including a weight greater than the measured weight and closest to the measured weight is equal to the comfort level 31 of the dataset 21 including a weight less than the measured weight and closest to the measured weight".

[0068] The comfort level 31 of YA is the comfort level 31 of the dataset 21 including a weight greater than the measured weight and closest to the measured weight. The comfort level 31 of YB is the comfort level 31 of the dataset 21 including a weight less than the measured weight and closest to the measured weight.

[0069] (A6) As shown in FIG. 13, when the prediction unit 12 satisfies the following Z1, Z2, and Z3, the comfort level 31 of the dataset 21 including the weight closest to the measured weight is set as the comfort level prediction value 40 of the clothing 7 for the user.

[0070] Z1 is the case where "it is determined that there is no dataset 21 including a weight equal to the measured weight". Z2 is the case where "it is determined by comparison processing that there are a dataset 21 including a weight greater than the measured weight and a dataset 21 including a weight less than the measured weight in the comfort history data 20". Z3 is the case where "the comfort 31 of the dataset 21 including a weight greater than the measured weight and closest to the measured weight is different from the comfort 31 of the dataset 21 including a weight less than the measured weight and closest to the measured weight".

[0071] (A7) As shown in FIG. 14, when the prediction unit 12 satisfies the following R1, R2, and R3, the comfort 31 with the highest discomfort level among the comfort 31 of the first dataset and the comfort 31 of the second dataset is set as the comfort prediction value 40 of the clothing 7 for the user.

[0072] R1 is the case where "the prediction unit 12 determines that there is no dataset 21 including a weight equal to the measured weight". R2 is the case where "it is determined by comparison processing that there are a dataset 21 including a weight greater than the measured weight and a dataset 21 including a weight less than the measured weight in the comfort history data 20".

[0073] R3 is the case where the comfort 31 of the first dataset defined below is different from the comfort 31 of the second dataset defined below, and the difference between the weight W1 of the first dataset and the measured weight is equal to the difference between the weight W2 of the second dataset and the measured weight. Here, the first dataset is the dataset 21 including a weight greater than the measured weight and closest to the measured weight. The second dataset is the dataset 21 including a weight less than the measured weight and closest to the measured weight.

[0074] [Validity Evaluation Unit] The validity evaluation unit 13 evaluates the validity of the comfort prediction value 40 for predicting comfort. The validity is evaluated based on whether the predicted comfort value 40 matches the actual comfort 31 in the most recent predetermined period. For all the data sets 21 in the most recent predetermined period, if the predicted comfort value 40 when worn outside matches the actual comfort 31, it is evaluated that the current predicted comfort value 40 is also valid.

[0075] On the other hand, for all the data sets 21 in the most recent predetermined period, if there is a data set 21 where the predicted comfort value 40 when worn outside does not match the actual comfort 31, since there is a high possibility that the prediction is off, the current predicted comfort value 40 is corrected. Hereinafter, the validity evaluation unit 13 will be specifically described.

[0076] When the prediction unit 12 evaluates the validity of the predicted comfort value 40 of the clothing 7, the validity evaluation unit 13 performs the following. The validity evaluation unit 13 determines whether there is information indicating that the prediction accuracy information 45 is a mismatch value from the past comfort history data 20 within a predetermined period before the prediction date (i.e., the current day), including the time of predicting comfort, for one or more other pieces of clothing 7 different from the target clothing 7 (hereinafter, the reference clothing for validity evaluation). The clothing 7 targeted by the validity evaluation unit 13 for the information extraction process is clothing 7 of the same category as the clothing 7 targeted for validity evaluation.

[0077] Based on the determination of whether there is information indicating that the prediction accuracy information 45 is a mismatch value, the validity evaluation unit 13 determines the validity of the predicted comfort value 40 as follows.

[0078] When the prediction unit 12 predicts the predicted comfort value 40 of the clothing 7, the validity evaluation unit 13 determines that the predicted comfort value 40 is valid if there is no mismatch information within the past comfort history data 20 within a predetermined period from the prediction date (i.e., the current day) for one or more other pieces of clothing 7 (i.e., the reference clothing for validity evaluation). The mismatch information indicates the data set 21 where the prediction accuracy information 45 is a mismatch value. The mismatch information is extracted from the comfort history data 20 based on the prediction accuracy information 45. And in this case, the validity evaluation unit 13 determines the confirmed value 42 of the predicted comfort value 40.

[0079] When the prediction unit 12 predicts the comfort prediction value 40 of the clothing 7, if there is inconsistent information within the past comfort history data 20 within a predetermined period from the prediction date (i.e., the current day) for one or more other pieces of clothing 7 (i.e., the reference clothing for validity evaluation), the validity evaluation unit 13 determines that the comfort prediction value 40 is not valid. When the comfort prediction value 40 is not valid, the correction unit 14 corrects the comfort prediction value 40.

[0080] [Correction Unit] With reference to FIGS. 15 and 16, the operation of the correction unit 14 and the method for correcting the comfort prediction value 40 will be described. FIG. 15 shows the comfort history data 20 of the clothing 7 and the measured weight of the user on the current day when one piece of the user's clothing 7 is the object of prediction. FIG. 15 further shows the corrected comfort prediction value 40 corrected by the correction unit 14.

[0081] FIG. 16 shows the comfort history data 20 of the clothing 7 that satisfies the following condition Q. Condition Q is that the clothing 7 is of the same category as the clothing 7 that is the object of prediction, is the clothing 7 worn within a predetermined period in the past from the current day, and the comfort prediction value 40 is different from the comfort 31 when actually worn. In short, FIG. 16 shows information regarding the clothing 7 including the inconsistent information in the case where there is the above-mentioned inconsistent information.

[0082] The correction unit 14 corrects the comfort prediction value 40. When the validity evaluation unit 13 determines that the comfort prediction value 40 is not valid, the correction unit 14 corrects the comfort prediction value 40. Hereinafter, specific examples will be described.

[0083] When the validity evaluation unit 13 determines that the comfort prediction value 40 is not valid, the correction unit 14 corrects the comfort prediction value 40 based on the comparison between the first weight deviation degree dw1 and the second weight deviation degree dw2.

[0084] The first weight deviation degree dw1 is the difference between the weight included in the dataset 21 of another piece of clothing 7, which is used for predicting the comfort of the other piece of clothing 7 in the dataset 21 where the prediction accuracy information 45 is the inconsistent value, and the weight of the user on the day when the comfort of the other piece of clothing 7 was predicted, for other pieces of clothing 7. In short, the first weight deviation degree dw1 is the difference between the weight (hereinafter referred to as weight A) in the dataset 21 of the past day, which is the basis for the prediction on the day when the prediction deviated, and the measured weight (hereinafter referred to as weight B) on the day when the prediction deviated, for the piece of clothing 7 for which the prediction deviated among the most recently worn pieces of clothing 7. Specifically, the first weight deviation degree dw1 is weight B - weight A.

[0085] The second weight deviation degree dw2 is the difference between the weight (hereinafter referred to as weight C) included in the dataset 21 of the piece of clothing 7, which is used for predicting the comfort of the piece of clothing 7, and the weight of the user (hereinafter referred to as weight D) on the day when the comfort of the piece of clothing 7 is predicted, for the target piece of clothing. Specifically, the day for prediction is the day when the piece of clothing 7 is selected. In short, the second weight deviation degree dw2 indicates the difference between the weight in the dataset 21 of the past day, which is the basis for the prediction, and the measured weight on the day, in the prediction on the day. Specifically, the second weight deviation degree dw2 is weight D - weight C.

[0086] When the second weight deviation degree dw2 is greater than the first weight deviation degree dw1, the correction unit 14 corrects the comfort prediction value 40 towards the difficult - to - wear side. When the second weight deviation degree dw2 is less than or equal to the first weight deviation degree dw1, the correction unit 14 does not correct the comfort prediction value 40.

[0087] [Selection Unit] The selection unit 15 selects the piece of clothing 7 with high comfort 31 from a plurality of pieces of clothing 7. The selection unit 15 selects one or more recommended pieces of clothing 7 based on the comfort prediction values 40 of each of the plurality of pieces of clothing 7. The selection unit 15 evaluates the height of the comfort 31 according to the value of C. In one example, when the piece of clothing 7 is flare pants, the selection unit 15 selects the piece of clothing 7 for which the comfort 31 of the waist, thigh, and buttocks is all set to "C" and the comfort prediction value 40 is set.

[0088] [Prediction Process] Referring to FIG. 17, the prediction process executed by the prediction unit 12 will be described. When the prediction unit 12 acquires the measured weight from the electronic weighing scale 3, for the user's clothing 7 identified by the imaging device 2, the prediction unit 12 executes a prediction process for all of the user's clothing 7. The prediction unit 12 predicts the comfort level of the user wearing the clothing 7 on the day when the user selects the clothing 7 for all of the clothing 7.

[0089] The prediction unit 12 sequentially executes a comfort level prediction process S1, a validity evaluation process S2, and a comfort level prediction value output process S3.

[0090] In the comfort level prediction process S1, for the clothing 7, a provisional value 41 of the comfort level prediction value 40 is obtained. In the validity evaluation process S2, the validity of the provisional value 41 of the comfort level prediction value 40 is evaluated. Further, in the validity evaluation process S2, when the comfort level prediction value 40 is valid, the provisional value 41 of the comfort level prediction value 40 is determined as the confirmed value 42 of the comfort level prediction value 40. And in the validity evaluation process S2, when the comfort level prediction value 40 is not valid, the provisional value 41 of the comfort level prediction value 40 is corrected, and the corrected comfort level prediction value 40 is determined as the confirmed value 42 of the comfort level prediction value 40.

[0091] In the comfort level prediction value output process S3, the comfort level prediction value 40 of the clothing 7 is output to the storage unit 11 together with the first identification information of the clothing 7. Based on receiving the comfort level prediction value 40 of the clothing 7, the storage unit 11 updates the comfort level history data 20 of the corresponding clothing 7. Hereinafter, the comfort level prediction process S1 and the validity evaluation process S2 will be described in detail.

[0092] Referring to FIGS. 18 and 19, the comfort level prediction process S1 executed by the prediction unit 12 will be described. As described above, the prediction unit 12 executes the comfort level prediction process S1 for all of the user's clothing 7. Hereinafter, the clothing 7 related to the comfort level prediction process S1 will be referred to as the target clothing 7.

[0093] In step S10, when the prediction unit 12 acquires the measured weight, it determines whether there is a data set 21 having a weight equal to the measured weight in the comfort level history data 20 of the target clothing 7.

[0094] When there is a dataset 21 with a weight equal to the measured weight, in step S11, the prediction unit 12 executes "selection processing based on the number of corresponding datasets" (see FIG. 19). By the "selection processing based on the number of corresponding datasets", the corresponding dataset 21 is selected. Subsequently, in step S12, the prediction unit 12 sets the comfort 31 of the corresponding dataset 21 as the provisional value 41 of the comfort prediction value 40. When there is no dataset 21 with a weight equal to the measured weight, the prediction unit 12 executes steps S13 and subsequent steps.

[0095] In step S13, the prediction unit 12 determines whether there are both a dataset 21 larger than the measured weight and a dataset 21 smaller than the measured weight in the comfort history data 20 of the target clothing 7. If there are not both a dataset 21 larger than the measured weight and a dataset 21 smaller than the measured weight, that is, if there is only a dataset 21 larger than the measured weight or only a dataset 21 smaller than the measured weight, the prediction unit 12 executes steps S14 to S16.

[0096] In step S14, the prediction unit 12 extracts the dataset 21 closest to the measured weight from the comfort history data 20 of the target clothing 7. Subsequently, in step S15, the prediction unit 12 executes "selection processing based on the number of corresponding datasets" (see FIG. 19). By the "selection processing based on the number of corresponding datasets", the corresponding dataset 21 is selected. Subsequently, in step S16, the prediction unit 12 sets the comfort 31 of the corresponding dataset 21 as the provisional value 41 of the comfort prediction value 40.

[0097] If, as a result of the determination in step S13, there are a dataset 21 larger than the measured weight and a dataset 21 smaller than the measured weight, the prediction unit 12 executes steps S17 to S19.

[0098] In step S17, the prediction unit 12 extracts the comfort level 31 of the dataset 21 that is greater than the measured weight and closest to the measured weight (hereinafter referred to as the former comfort level 31), and the comfort level 31 of the dataset 21 that is less than the measured weight and closest to the measured weight (hereinafter referred to as the latter comfort level 31). Then, the prediction unit 12 determines whether the former comfort level 31 is equal to the latter comfort level 31. When the prediction unit 12 determines that the former comfort level 31 is equal to the latter comfort level 31, in step S18, the comfort level 31 of the corresponding dataset 21 is set as the provisional value 41 of the comfort prediction value 40. That is, the prediction unit 12 sets the comfort level 31 of the former or latter dataset 21 as the provisional value 41 of the comfort prediction value 40. When the prediction unit 12 determines that the former comfort level 31 is not equal to the latter comfort level 31, the comfort level 31 with the highest discomfort level among the former and the latter is set as the provisional value 41 of the comfort prediction value 40.

[0099] Referring to FIG. 19, the selection process based on the number of corresponding datasets will be described. In step S20, the prediction unit 12 determines whether there are a plurality of corresponding data. In step S20, when the prediction unit 12 determines that the number of the corresponding datasets 21 is not a plurality, that is, when it is determined that the number of the corresponding datasets 21 is one, in step S21, the comfort level 31 of the corresponding dataset 21 is set as the comfort prediction value 40.

[0100] In step S20, when the prediction unit 12 determines that the number of the corresponding datasets 21 is a plurality, step S22 is executed. In step S22, the prediction unit 12 determines whether the comfort levels 31 of the plurality of datasets 21 are all the same by comparing the comfort levels 31 of the plurality of datasets 21. When the prediction unit 12 determines that the comfort levels 31 of the plurality of datasets 21 are all the same, step S21 is executed. Specifically, the comfort level 31 of any one of the plurality of datasets 21 is set as the comfort prediction value 40.

[0101] In step S22, when the prediction unit 12 determines that the comfort levels 31 of the plurality of data sets 21 are not the same, step S23 is executed. In step S23, the prediction unit 12 sets the comfort level 31 with the highest wearing difficulty among the comfort levels 31 of the plurality of data sets 21 as the comfort level prediction value 40.

[0102] Referring to FIG. 20, the validity evaluation process S2 executed by the prediction unit 12 and the correction unit 14 will be described. As described above, the validity evaluation process S2 is executed after the comfort level prediction process S1. In the comfort level prediction process S1, a provisional value 41 of the comfort level prediction value 40 is set for the target clothing 7. In the validity evaluation process S2, the validity of the provisional value 41 of the comfort level prediction value 40 for the target clothing 7 is evaluated.

[0103] In step S30, the prediction unit 12 determines whether the comfort level prediction matches the comfort level 31 in all of the data sets 21 having a wearing-outgoing history 33 in the past two weeks among the clothing 7 of the same category as the target clothing 7. In this determination, if the comfort level prediction matches the comfort level 31 in all of the data sets 21 having a wearing-outgoing history 33 in the past two weeks, the prediction unit 12, in step S31, determines that the provisional value 41 of the comfort level prediction value 40 is valid and determines the provisional value 41 of the comfort level prediction value 40 as the confirmed value 42 of the comfort level prediction value 40.

[0104] In the determination of step S30, if there is at least one data set 21 in which the comfort level prediction does not match the comfort level 31 for all of the data sets 21 having a wearing-outgoing history 33 in the past two weeks, the next step S32 is executed.

[0105] In step S32, the prediction unit 12 calculates the first weight deviation degree dw1 in the inconsistent data set 21 for the data set 21 in which the comfort level prediction does not match the comfort level 31. In addition, the prediction unit 12 calculates the second weight deviation degree dw2 in the prediction for the current day. The first weight deviation degree dw1 and the second weight deviation degree dw2 are as defined above.

[0106] Subsequently, in step S33, the correction unit 14 corrects the provisional value 41 of the comfort prediction value 40 based on the first weight deviation degree dw1 and the second weight deviation degree dw2.

[0107] Specifically, when the second weight deviation degree dw2 is greater than the first weight deviation degree dw1 (hereinafter referred to as condition A), the correction unit 14 corrects the provisional value 41 of the comfort prediction value 40 so that the wearing difficulty increases. Note that even if the first weight deviation degree dw1 and the second weight deviation degree dw2 are positive values or negative values, the correction is executed when condition A is satisfied. When the second weight deviation degree dw2 is smaller than or equal to the first weight deviation degree dw1, the provisional value 41 of the comfort prediction value 40 is not corrected.

[0108] Subsequently, in step S34, the prediction unit 12 determines the comfort prediction value 40 that has completed the process of step S33 as the confirmed value 42 of the comfort prediction value 40.

[0109] [Operation of this Embodiment] The operation of this embodiment will be described. In the comfort prediction system 10, the prediction unit 12 compares the measured weight with the weight in the data set 21. Then, the prediction unit 12 extracts the data set 21 including the weight equal to or close to the measured weight from the comfort history data 20. The prediction unit 12 sets the comfort 31 of the extracted data set 21 as the comfort prediction value 40 of the clothing 7 for the user. In this way, in the comfort prediction system 10, the data set 21 in which the weight and the comfort 31 are linked is used as a reference for prediction. Therefore, the comfort 31 of the clothing 7 on the current day can be accurately derived.

[0110] [Effect of this Embodiment] The effect of this embodiment will be described. (1) In the comfort prediction system 10, the prediction unit 12 extracts, by comparison processing, a data set 21 that includes a weight equal to or close to the measured weight from the comfort history data 20. Then, the prediction unit 12 sets the comfort 31 of the extracted data set 21 as the comfort prediction value 40 of the clothing 7 for the user. According to this configuration, since the comfort prediction value 40 is set based on the comfort 31 when the weight is equal to or close to the measured weight, appropriate comfort can be provided.

[0111] (2) When the prediction unit 12 detects only one data set 21 that includes a weight equal to the measured weight by comparison processing, the prediction unit 12 sets the comfort 31 of the data set 21 detected by the comparison processing as the comfort prediction value 40 of the clothing 7 for the user. According to this configuration, appropriate comfort can be provided.

[0112] (3) When the prediction unit 12 detects a plurality of data sets 21 that include a weight equal to the measured weight by comparison processing, the prediction unit 12 sets the comfort 31 with the highest wearing discomfort among the comforts 31 of the plurality of data sets 21 as the comfort prediction value 40 of the clothing 7 for the user. According to this configuration, appropriate comfort can be provided.

[0113] (4) When the prediction unit 12 determines by comparison processing that there is no data set 21 that includes a weight equal to the measured weight, the prediction unit 12 sets the comfort 31 of the data set 21 that includes the weight closest to the measured weight as the comfort prediction value 40 of the clothing 7 for the user. According to this configuration, appropriate comfort can be provided.

[0114] (5) When the prediction unit 12 satisfies the following X1 and X2 by comparison processing, the prediction unit 12 sets the comfort 31 of the data set 21 that includes the weight closest to the measured weight as the comfort prediction value 40 of the clothing 7 for the user. According to this configuration, appropriate comfort can be provided.

[0115] X1 is the case where it is determined that there is no dataset 21 including a weight equal to the measured weight. X2 is the case where it is determined that for all of the datasets 21, the weight of the dataset 21 is greater than the measured weight, or it is determined that for all of the datasets 21, the weight of the dataset 21 is less than the measured weight.

[0116] (6) When the prediction unit 12 satisfies the following Y1, Y2, and Y3, the comfort level 31 of YA or the comfort level 31 of VB is set as the comfort level prediction value 40 of the clothing 7 for the user. According to this configuration, an appropriate comfort level can be provided.

[0117] Y1 is the case where it is determined that there is no dataset 21 including a weight equal to the measured weight. Y2 is the case where it is determined by the comparison process that there are a dataset 21 including a weight greater than the measured weight and a dataset 21 including a weight less than the measured weight in the comfort level history data 20. Y3 is the case where the comfort level 31 of the dataset 21 including a weight greater than the measured weight and closest to the measured weight is equal to the comfort level 31 of the dataset 21 including a weight less than the measured weight and closest to the measured weight. YA is the comfort level 31 of the dataset 21 including a weight greater than the measured weight and closest to the measured weight. YB is the comfort level 31 of the dataset 21 including a weight less than the measured weight and closest to the measured weight.

[0118] (7) When the prediction unit 12 satisfies the following Z1, Z2, and Z3, the comfort level 31 of the dataset 21 including the weight closest to the measured weight is set as the comfort level prediction value 40 of the clothing 7 for the user. According to this configuration, an appropriate comfort level can be provided.

[0119] Z1 is the case where it is determined that there is no dataset 21 including a weight equal to the measured weight. Z2 is the case where it is determined by comparison processing that there are a dataset 21 including a weight greater than the measured weight and a dataset 21 including a weight less than the measured weight in the comfort history data 20. Z3 is the case where the comfort 31 of the dataset 21 including a weight greater than the measured weight and closest to the measured weight is different from the comfort 31 of the dataset 21 including a weight less than the measured weight and closest to the measured weight.

[0120] (8) When the prediction unit 12 satisfies R1, R2, and R3, the comfort 31 with the highest wearing discomfort among the comfort 31 of the first dataset and the comfort 31 of the second dataset is set as the comfort prediction value 40 of the clothing 7 for the user. According to this configuration, an appropriate comfort can be provided.

[0121] R1 is the case where it is determined that there is no dataset 21 including a weight equal to the measured weight. R2 is the case where it is determined by comparison processing that there are a dataset 21 including a weight greater than the measured weight and a dataset 21 including a weight less than the measured weight in the comfort history data 20. R3 is the case where the comfort 31 of the first dataset is different from the comfort 31 of the second dataset, and the difference between the weight W1 of the first dataset and the measured weight is equal to the difference between the weight W2 of the second dataset and the measured weight. Here, the first dataset is the dataset 21 including a weight greater than the measured weight and closest to the measured weight. The second dataset is the dataset 21 including a weight less than the measured weight and closest to the measured weight.

[0122] (9) In the comfort prediction system 10, the dataset 21 includes a comfort prediction value 40 and prediction accuracy information 45. The prediction accuracy information 45 has, as information, a matching value indicating that the comfort prediction value 40 matches the comfort 31 and a non-matching value indicating that the comfort prediction value 40 does not match the comfort 31. According to this configuration, the validity of the comfort prediction value 40 can be evaluated.

[0123] (10) The comfort prediction system 10 further includes a validity evaluation unit 13. When the prediction unit 12 predicts the comfort prediction value 40 of the clothing 7, the validity evaluation unit 13 determines the validity of the comfort prediction value 40 based on whether there is information indicating that the prediction accuracy information 45 is a mismatch value in the past comfort history data 20 within a predetermined period before the prediction date including the prediction time for one or more other pieces of clothing 7 different from the clothing 7. According to this configuration, the validity of the comfort prediction value 40 can be evaluated.

[0124] (11) When the prediction unit 12 predicts the comfort prediction value 40 of the clothing 7, if there is no information indicating that the prediction accuracy information 45 is a mismatch value in the past comfort history data 20 within a predetermined period from the prediction date for one or more other pieces of clothing 7, the validity evaluation unit 13 determines that the comfort prediction value 40 is valid and determines the comfort prediction value 40 as the confirmed value 42 of the comfort prediction value 40. According to this configuration, a high-accuracy comfort prediction value 40 can be provided.

[0125] (12) When the prediction unit 12 predicts the comfort prediction value 40 of the clothing 7, if there is information indicating that the prediction accuracy information 45 is a mismatch value in the past comfort history data 20 within a predetermined period from the prediction date for one or more other pieces of clothing 7, the validity evaluation unit 13 determines that the comfort prediction value 40 is not valid. According to this configuration, it is possible to deal with an invalid comfort prediction value 40.

[0126] (13) The comfort prediction system 10 further includes a correction unit 14 for correcting the comfort prediction value 40. When the validity evaluation unit 13 determines that the comfort prediction value 40 is not valid, the correction unit 14 corrects the comfort prediction value 40 based on the comparison between the first body weight deviation degree dw1 and the second body weight deviation degree dw2. According to this configuration, when it is determined that the comfort prediction value 40 is not valid, the comfort prediction value 40 is corrected, so a high-accuracy comfort prediction value 40 can be provided.

[0127] (14) When the second body weight deviation degree dw2 is greater than the first body weight deviation degree dw1, the correction unit 14 corrects the comfort prediction value 40 toward the difficult-to-wear side. According to this configuration, a high-accuracy comfort prediction value 40 can be provided.

[0128] (15) The comfort prediction system 10 further includes a selection unit 15 that selects a highly comfortable piece of clothing 7 from a plurality of pieces of clothing 7. The selection unit 15 selects one or more recommended pieces of clothing 7 based on the comfort prediction value 40 of each of the plurality of pieces of clothing 7. According to this configuration, one or more pieces of clothing 7 can be provided as highly comfortable pieces of clothing 7.

[0129] <Modification Example> The above embodiment is an exemplification of the forms that the comfort prediction system 10 can take, and is not intended to limit the form. The comfort prediction system 10 can take a form different from the form exemplified in the above embodiment. Examples thereof are forms in which part of the configuration of the embodiment is replaced, changed, or omitted, or forms in which a new configuration is added to the embodiment. Modification examples of the embodiment are shown below.

[0130] · This system technology may be configured as the technology of the following method. The comfort prediction method is a method for predicting the comfort of a piece of clothing 7. The comfort prediction method includes a preparation step and a prediction step. In the preparation step, comfort history data 20 of the piece of clothing 7 is prepared for the user. The comfort history data 20 includes one or more data sets 21 for each date and time. The data set 21 includes the wearing date of the piece of clothing 7, the weight on the wearing date, and comfort information 30 indicating the comfort of the piece of clothing 7 at the time of wearing.

[0131] In the prediction step, the comfort is predicted based on the measured weight of the user at the time of comfort prediction and the comfort history data 20. In the prediction step, by a comparison process of comparing the measured weight with the weight of the data set 21, a data set 21 including a weight equal to or close to the measured weight is extracted from the comfort history data 20. Then, the comfort 31 of the extracted data set 21 is set as the comfort prediction value 40 of the piece of clothing 7 for the user.

[0132] · This system technology may be configured as the technology of the following program. The comfort prediction program is a program for predicting the comfort of clothing 7. The comfort prediction program includes a preparation step and a prediction step. In the preparation step, the computer is made to prepare the comfort history data 20 of the user for clothing 7. The comfort history data 20 includes one or more data sets 21 for each date and time. The data set 21 includes the wearing date of wearing clothing 7, the weight on the wearing date, and the comfort information 30 indicating the comfort of clothing 7 at the time of wearing. In the prediction step, the computer is made to predict the comfort based on the measured weight of the user at the time of comfort prediction and the comfort history data 20. In the prediction step, by a comparison process of comparing the measured weight with the weight of the data set 21, a data set 21 including a weight equal to or close to the measured weight is extracted from the comfort history data 20. And in the prediction step, the comfort 31 of the extracted data set 21 is made the comfort prediction value 40 of clothing 7 for the user.

[0133] This specification discloses the following technology. [Appendix 1] The technology according to Appendix 1 is a comfort prediction system for predicting the comfort of clothing. The comfort prediction system includes a storage unit that stores the comfort history data of the user for the clothing, and a prediction unit that predicts the comfort based on the measured weight of the user at the time of comfort prediction and the comfort history data. The comfort history data includes one or more data sets for each date and time. The data set includes the wearing date of wearing the clothing, the weight on the wearing date, and the comfort information indicating the comfort of the clothing at the time of wearing. The prediction unit extracts the data set including a weight equal to or close to the measured weight from the comfort history data by a comparison process of comparing the measured weight with the weight of the data set, and makes the comfort of the extracted data set the comfort prediction value of the clothing for the user.

[0134] [Appendix 2] In the comfort prediction system described in Appendix 1, when the prediction unit detects only one data set including a weight equal to the measured weight by the comparison process, the comfort of the data set detected by the comparison process is used as the comfort prediction value of the clothing for the user.

[0135] [Appendix 3] In the comfort prediction system described in Appendix 2, when the prediction unit detects a plurality of data sets including a weight equal to the measured weight by the comparison process, the comfort with the highest wearing discomfort among the comforts of the plurality of data sets is used as the comfort prediction value of the clothing for the user.

[0136] [Appendix 4] In the comfort prediction system described in Appendix 1, when the prediction unit determines by the comparison process that there is no data set including a weight equal to the measured weight, the comfort of the data set including the weight closest to the measured weight is used as the comfort prediction value of the clothing for the user.

[0137] [Appendix 5] In the comfort prediction system described in Appendix 1, when the prediction unit determines by the comparison process that there is no data set including a weight equal to the measured weight, and determines that the weight of the data set is greater than the measured weight for all of the data sets, or determines that the weight of the data set is less than the measured weight for all of the data sets, the comfort of the data set including the weight closest to the measured weight is used as the comfort prediction value of the clothing for the user.

[0138] [Appendix 6] In the comfort prediction system according to Supplementary Note 1, when the prediction unit determines that there is no data set including the weight equal to the measured weight, and when, by the comparison process, it is determined that there are data sets including weights greater than the measured weight and data sets including weights less than the measured weight in the comfort history data, and when the comfort of the data set including the weight greater than the measured weight and closest to the measured weight is equal to the comfort of the data set including the weight less than the measured weight and closest to the measured weight, the comfort of the data set including the weight greater than the measured weight and closest to the measured weight or the comfort of the data set including the weight less than the measured weight and closest to the measured weight is used as the comfort prediction value of the clothing for the user.

[0139] [Supplementary Note 7] In the comfort prediction system according to Supplementary Note 1, when the prediction unit determines that there is no data set including the weight equal to the measured weight, and when, by the comparison process, it is determined that there are data sets including weights greater than the measured weight and data sets including weights less than the measured weight in the comfort history data, and when the comfort of the data set including the weight greater than the measured weight and closest to the measured weight is different from the comfort of the data set including the weight less than the measured weight and closest to the measured weight, the comfort of the data set including the weight closest to the measured weight is used as the comfort prediction value of the clothing for the user.

[0140] [Supplementary Note 8] In the comfort prediction system according to Supplementary Note 1, when the prediction unit determines that there is no data set including a weight equal to the measured weight, in the case where it is determined by the comparison process that there are a data set including a weight greater than the measured weight and a data set including a weight less than the measured weight in the comfort history data, and a data set including a weight greater than the measured weight and closest to the measured weight is defined as a first data set, and a data set including a weight less than the measured weight and closest to the measured weight is defined as a second data set, if the comfort of the first data set is different from the comfort of the second data set, and the difference between the weight of the first data set and the measured weight is equal to the difference between the weight of the second data set and the measured weight, Of the comfort of the first data set and the comfort of the second data set, the comfort with the highest wearing difficulty is set as the comfort prediction value of the clothing for the user.

[0141] [Supplementary Note 9] In the comfort prediction system according to Supplementary Note 1, the data set further includes the comfort prediction value and prediction accuracy information. The prediction accuracy information has, as information, a matching value indicating that the comfort prediction value matches the comfort, and a non-matching value indicating that the comfort prediction value does not match the comfort.

[0142] [Supplementary Note 10] In the comfort prediction system according to Supplementary Note 9, further provided is a validity evaluation unit that evaluates the validity of the comfort prediction value for predicting the comfort. The storage unit stores a database including the comfort history data of each of a plurality of items of clothing. When the prediction unit predicts the comfort prediction value of the clothing, the validity evaluation unit determines the validity of the comfort prediction value based on whether there is information indicating that the prediction accuracy information is a non-matching value from the past comfort history data within a predetermined period before the prediction date including the time of the comfort prediction for one or more other items of clothing different from the clothing.

[0143] [Supplementary Note 11] In the comfort prediction system described in Supplementary Note 10, when the prediction unit predicts the comfort prediction value of the clothing, for one or more of the other clothing items, if there is no information in the past comfort history data within a predetermined period from the prediction date that the prediction accuracy information is a mismatch value, the validity evaluation unit determines that the comfort prediction value is valid and finalizes the comfort prediction value.

[0144] [Supplementary Note 12] In the comfort prediction system described in Supplementary Note 10, when the prediction unit predicts the comfort prediction value of the clothing, for one or more of the other clothing items, if there is information in the past comfort history data within a predetermined period from the prediction date that the prediction accuracy information is a mismatch value, the validity evaluation unit determines that the comfort prediction value is not valid.

[0145] [Supplementary Note 13] The comfort prediction system described in Supplementary Note 12 further includes a correction unit that corrects the comfort prediction value. When the validity evaluation unit determines that the comfort prediction value is not valid, the correction unit corrects the comfort prediction value based on a comparison between a first weight deviation degree and a second weight deviation degree. The first weight deviation degree is the difference between the weight included in the dataset of the other clothing item used for predicting the comfort of the other clothing item, which includes information that the prediction accuracy information is a mismatch value, and the weight of the user on the day when the comfort of the other clothing item was predicted. The second weight deviation degree is the difference between the weight included in the dataset of the clothing item used for predicting the comfort of the clothing item and the weight of the user on the day when the comfort of the clothing item is predicted.

[0146] [Supplementary Note 14] In the comfort prediction system described in Supplementary Note 13, when the second weight deviation degree is greater than the first weight deviation degree, the correction unit corrects the comfort prediction value toward the side of wearing difficulty.

[0147] [Supplementary Note 15] In the comfort prediction system according to any one of Supplementary Notes 1 to 14, a selection unit is further provided for selecting a piece of clothing with high comfort from a plurality of pieces of clothing. The prediction unit predicts the comfort prediction value of the user for each of the plurality of pieces of clothing. The selection unit selects one or more of the recommended pieces of clothing based on the comfort prediction values of each of the plurality of pieces of clothing.

[0148] [Supplementary Note 16] The comfort prediction method according to Supplementary Note 16 is a method for predicting the comfort of clothing. The comfort prediction method includes a preparation step of preparing the comfort history data of the clothing for the user, and a prediction step of predicting the comfort based on the measured weight of the user at the time of comfort prediction and the comfort history data. The comfort history data includes one or more data sets for each date and time. The data set includes the wearing date of the clothing, the weight on the wearing date, and comfort information indicating the comfort of the clothing at the time of wearing. In the prediction step, by a comparison process of comparing the measured weight with the weight of the data set, a data set including a weight equal to or close to the measured weight is extracted from the comfort history data, and the comfort of the extracted data set is used as the comfort prediction value of the clothing for the user.

[0149] [Supplementary Note 17] The comfort prediction program according to Supplementary Note 17 is a comfort prediction program for predicting the comfort of clothing. The comfort prediction program includes a preparation step of causing a computer to prepare comfort history data of the user with respect to the clothing, and a prediction step of causing the computer to predict comfort based on the measured weight of the user at the time of comfort prediction and the comfort history data. The comfort history data includes one or more data sets for each date and time. The data set includes a wearing date when the clothing was worn, a weight on the wearing date, and comfort information indicating the comfort of the clothing at the time of wearing. In the prediction step, by a comparison process of comparing the measured weight with the weight of the data set, the data set including the weight equal to or close to the measured weight is extracted from the comfort history data, and the comfort of the extracted data set is set as the comfort prediction value of the clothing for the user.

Explanation of Signs

[0150] dw1…First weight deviation degree, dw2…Second weight deviation degree, 1…, 7…Clothing, 10…Comfort prediction system, 11…Storage unit, 12…Prediction unit, 13…Validity evaluation unit, 14…Correction unit, 15…Selection unit, 20…Comfort history data, 21…Data set, 30…Comfort information, 31…Comfort, 40…Comfort prediction value, 45…Prediction accuracy information.

Claims

1. A comfort prediction system for predicting the comfort of clothing, a storage unit that stores the comfort history data of the clothing for the user, a prediction unit that predicts comfort based on the measured weight of the user at the time of comfort prediction and the comfort history data, and the comfort history data includes one or more data sets for each date and time, the data set includes the wearing date when the clothing was worn, the weight on the wearing date, and comfort information indicating the comfort of the clothing at the time of wearing, the prediction unit extracts, by a comparison process of comparing the measured weight with the weight of the data set, the data set including a weight equal to or close to the measured weight from the comfort history data, and sets the comfort of the extracted data set as the comfort prediction value of the clothing for the user, Comfort prediction system.

2. The prediction unit when only one data set including a weight equal to the measured weight is detected by the comparison process, sets the comfort of the data set detected by the comparison process as the comfort prediction value of the clothing for the user, The comfort prediction system according to claim 1.

3. The prediction unit when a plurality of data sets including a weight equal to the measured weight are detected by the comparison process, sets the comfort with the highest wearing difficulty among the comforts of each of the plurality of data sets as the comfort prediction value of the clothing for the user, The comfort prediction system according to claim 2.

4. The prediction unit when it is determined by the comparison process that there is no data set including a weight equal to the measured weight, sets the comfort of the data set including the weight closest to the measured weight as the comfort prediction value of the clothing for the user, The comfort prediction system according to claim 1.

5. The prediction unit when it is determined by the comparison process that there is no data set including a weight equal to the measured weight, and it is determined that the weight of the data set is greater than the measured weight for all of the data sets, or it is determined that the weight of the data set is less than the measured weight for all of the data sets, sets the comfort of the data set including the weight closest to the measured weight as the comfort prediction value of the clothing for the user, The comfort prediction system according to claim 1.

6. The prediction unit When it is determined that there is no such dataset including the weight equal to the measured weight, and when it is determined by the comparison process that there are such datasets including a weight greater than the measured weight and a weight less than the measured weight in the comfort history data, and when the comfort of the dataset including a weight greater than the measured weight and closest to the measured weight is equal to the comfort of the dataset including a weight less than the measured weight and closest to the measured weight, the comfort of the dataset including a weight greater than the measured weight and closest to the measured weight, or the comfort of the dataset including a weight less than the measured weight and closest to the measured weight is set as the predicted value of the comfort of the clothing for the user. The comfort prediction system according to claim 1.

7. The prediction unit when it is determined that there is no such dataset including the weight equal to the measured weight, and when it is determined by the comparison process that there are such datasets including a weight greater than the measured weight and a weight less than the measured weight in the comfort history data, and when the comfort of the dataset including a weight greater than the measured weight and closest to the measured weight is different from the comfort of the dataset including a weight less than the measured weight and closest to the measured weight, the comfort of the dataset including the weight closest to the measured weight is set as the predicted value of the comfort of the clothing for the user. The comfort prediction system according to claim 1.

8. The prediction unit when it is determined that there is no such dataset including the weight equal to the measured weight, and when it is determined by the comparison process that there are such datasets including a weight greater than the measured weight and a weight less than the measured weight in the comfort history data, and when the dataset including a weight greater than the measured weight and closest to the measured weight is defined as the first dataset, and the dataset including a weight less than the measured weight and closest to the measured weight is defined as the second dataset, and the comfort of the first dataset is different from the comfort of the second dataset, and When the difference between the weight of the first dataset and the measured weight is equal to the difference between the weight of the second dataset and the measured weight, Of the comfort levels of the first dataset and the second dataset, the comfort level with the highest discomfort is used as the predicted comfort value of the clothing for the user. The comfort prediction system according to claim 1.

9. The dataset further includes the predicted comfort value and prediction accuracy information. The prediction accuracy information has, as information, a match value indicating that the predicted comfort value matches the comfort level, and a non-match value indicating that the predicted comfort value does not match the comfort level. The comfort prediction system according to claim 1.

10. Furthermore, it includes a validity evaluation unit that evaluates the validity of the predicted comfort value for predicting the comfort level. The storage unit stores a database including the comfort history data of each of a plurality of pieces of clothing. When the prediction unit predicts the predicted comfort value of the clothing, the validity evaluation unit determines whether there is information indicating that the prediction accuracy information is a non-match value in the past comfort history data within a predetermined period before the prediction date, including the prediction time, for one or more other pieces of clothing different from the clothing, and determines the validity of the predicted comfort value based on this. The comfort prediction system according to claim 9.

11. When the prediction unit predicts the predicted comfort value of the clothing, if there is no information indicating that the prediction accuracy information is a non-match value in the past comfort history data within a predetermined period from the prediction date for one or more of the other pieces of clothing, the validity evaluation unit determines that the predicted comfort value is valid and determines the predicted comfort value. The comfort prediction system according to claim 10.

12. When the prediction unit predicts the predicted comfort value of the clothing, if there is information indicating that the prediction accuracy information is a non-match value in the past comfort history data within a predetermined period from the prediction date for one or more of the other pieces of clothing, the validity evaluation unit determines that the predicted comfort value is not valid. The comfort prediction system according to claim 10.

13. It further includes a correction unit that corrects the predicted comfort value. When the validity evaluation unit determines that the predicted comfort value is not valid, the correction unit corrects the predicted comfort value based on a comparison between a first weight deviation degree and a second weight deviation degree. The first weight deviation is the difference between the weight included in the dataset of the other clothing item used for predicting the comfort of the other clothing item, which is related to the dataset in which the prediction accuracy information is a non-matching value, and the weight of the user on the day when the comfort of the other clothing item was predicted. The second weight deviation is the difference between the weight included in the dataset of the clothing item used for predicting the comfort of the clothing item and the weight of the user on the day when the comfort of the clothing item is predicted. The comfort prediction system according to claim 12.

14. When the second weight deviation is greater than the first weight deviation, the correction unit corrects the comfort prediction value towards the difficult-to-wear side. The comfort prediction system according to claim 13.

15. Further comprising a selection unit for selecting clothing items with high comfort from a plurality of clothing items. The prediction unit predicts the comfort prediction value for the user for each of the plurality of clothing items. The selection unit selects one or more of the recommended clothing items based on the comfort prediction values of each of the plurality of clothing items. The comfort prediction system according to any one of claims 1 to 14.

16. A comfort prediction method for predicting the comfort of clothing, comprising: A preparation step of preparing the comfort history data of the clothing for the user; A prediction step of predicting the comfort based on the measured weight of the user at the time of comfort prediction and the comfort history data. The comfort history data includes one or more datasets for each date and time. The dataset includes the wearing date when the clothing was worn, the weight on the wearing date, and comfort information indicating the comfort of the clothing when worn. In the prediction step, by a comparison process of comparing the measured weight with the weight in the dataset, a dataset including a weight equal to or close to the measured weight is extracted from the comfort history data. The comfort of the extracted dataset is used as the comfort prediction value of the clothing for the user. Comfort prediction method.

17. A comfort prediction program for predicting the comfort of clothing, comprising: A preparation step of causing a computer to prepare the comfort history data of the clothing for the user; A prediction step of causing the computer to predict the comfort based on the measured weight of the user at the time of comfort prediction and the comfort history data. The comfort history data includes one or more datasets for each date and time. The dataset includes the wearing date when the clothing was worn, the weight on the wearing date, and comfort information indicating the comfort of the clothing when worn. In the prediction step, by a comparison process of comparing the measured weight with the weight in the dataset, the dataset including the weight equal to or close to the measured weight is extracted from the comfort history data. The comfort of the extracted dataset is used as the predicted value of the comfort of the clothing for the user. Comfort prediction program.

Images