KR100880295B1 - User taste estimation device, user profile estimation device, and robot - Google Patents

Abstract

The user's preferences and profiles are estimated with good accuracy. The user preference estimating apparatus 100 is based on the receiving means 2 which receives the information about the article from the tag attached to the article, and the information about the article which the receiving means 2 has received. Article name determination means 3 for acquiring the article name, and symbol determination means 4 for estimating the user's preferences based on the information about the article received by the reception means 2. The preference determination means 4 acquires preference information corresponding to the article name from the database storing the article name and the symbol information corresponding to the article name based on the article name acquired by the article name determination means 3, and the user's preference. Determine.

User preferences, user profile, robot, home robot, hobby judgment

Description

USER TASTE ESTIMATION DEVICE, USER PROFILE ESTIMATION DEVICE, AND ROBOT}

The present invention relates to a user preference estimation apparatus, a user preference estimation method, a user profile estimation apparatus, a user profile estimation method, and a robot.

Conventionally, as a method of estimating a user's hobby in the system, as disclosed in Patent Document 1, for example, the hobby has been estimated from the result of the user's web access.

This technique is used by the system to grasp the user's hobbies, to act more appropriately for the user, to improve the user's satisfaction or to present the user with appropriate information.

For example, in a robot used in a general home or the like, if a user's situation such as a user's hobbies or preferences can be grasped, the robot can speak or act accordingly. As a result, a more pleasant robot can be realized, or an appropriate advertisement can be presented according to the user's hobbies and preferences.

Patent Document 1: Japanese Patent Application Laid-Open No. 2000-339322

<Start of invention>

However, the related art has a problem that since information available for estimating the hobby is limited, it cannot be estimated with good accuracy. As described in Patent Literature 1, in the method of estimating the hobby from the results of the Web access, the hobby is estimated only from the record in which the user accesses the Web with a personal computer or a mobile phone. However, since the time and purpose of actually accessing the web by the user are limited, it is difficult to obtain an accurate estimation result even when the user's hobby is estimated from such limited information.

An object of the present invention is to provide a technique for estimating a user's preferences and profile with good accuracy.

According to the present invention, there is provided a receiving unit for receiving information about an article from a tag attached to the article, and a symbol determination unit for estimating a user's preference based on the information about the article received by the receiving unit. A user preference estimation apparatus is provided.

The user preference estimation apparatus of the present invention may further include an article name acquisition unit for acquiring an article name of the article based on the information about the article received by the reception unit, and the preference determination unit further includes an article name acquired by the article name acquisition unit. On the basis of this, the symbol information corresponding to the article name can be obtained from the database storing the article name and the preference information corresponding to the article name.

The user preference estimating apparatus of the present invention is based on the article name acquiring unit for acquiring the article name of the article based on the information about the article received by the receiver, and the article name and the article based on the article name acquired by the article name acquisition unit. An article category acquiring unit may further include an article category acquiring unit for acquiring a category corresponding to the article name from a database storing a category corresponding to the article name, wherein the preference determining unit is based on the category based on the category acquired by the article category acquiring unit. From the database storing the preference information corresponding to the category, the preference information corresponding to the category can be obtained.

The user preference estimation apparatus of the present invention may further include a threshold storage section for storing a threshold for estimating that the preference is the user's preferences, and the preference determination section includes a magnitude of the sum of the preference information acquired from the database and the threshold value. Based on this, the user's preference can be estimated.

In the user preference estimation apparatus of the present invention, the preference determination unit can also obtain a weighting coefficient defined for each article together with the preference information from the database, and the preference determination unit takes the weighting coefficient into consideration. The cumulative total is calculated, and the preference of the user can be estimated based on the cumulative total and the magnitude of the threshold.

The user preference estimation apparatus of the present invention may further include an owner estimation section for estimating the owner of the article for each article, and estimates the owner's preference for each owner estimated by the owner estimation section. have.

The user preference estimation apparatus of the present invention may further include a presentation processing unit for presenting information to the user, and an input reception unit for accepting input of information from the user, wherein the owner estimating unit inquires an owner of the presentation processing unit. The owner of the article can be estimated by presenting the information to be answered and receiving a reply to the information to be inquired through the input reception unit.

The user preference estimation apparatus of the present invention may further include a clustering processing unit for clustering a plurality of the articles on which the reception unit has received information about the article based on a predetermined criterion, wherein the preference determination unit includes the clustering processing. For each additionally divided cluster, the preference of the article owner included in the cluster can be estimated.

The user preference estimation apparatus of the present invention may further include a position information acquisition unit for acquiring position information of a plurality of the articles from which the reception unit reads the information identifying the article, wherein the clustering processing unit acquires the position information. Based on the positional information of the article acquired additionally, the article can be clustered according to the relative position between the articles.

In the user preference estimation apparatus of the present invention, the information about the article can be information identifying the article.

The user preference estimation apparatus of the present invention may further include a presentation processing unit for presenting information to the user, and an input accepting unit for accepting input of information from the user, wherein the preference determination unit is configured to present a candidate of a determination result as the presentation processing unit. The user's preference can be determined by accepting the selection of the correct determination result from among the candidates of the determination result by the input reception unit.

The user preference estimation apparatus of the present invention may further include a moving unit for autonomous movement.

According to the present invention, there is provided a robot characterized by autonomous movement, comprising the user preference estimating apparatus according to any one of the above, a moving unit for moving the user preference estimation apparatus, and an operation control unit for controlling the moving unit. do.

According to the present invention, there is provided a robot comprising a user preference estimating apparatus according to any one of the above, and an operation control unit which operates on a user based on the preference of the user estimated by the user preference estimation apparatus. do.

According to the present invention, there is provided an information acquisition unit for acquiring information about an article from a tag attached to the article, and a profile estimating section for estimating a user's profile based on the information about the article read by the information acquisition unit. A user profile estimating apparatus is provided.

The user profile estimating apparatus of the present invention may further include an owner estimating unit for estimating the owner of the article for each article, wherein the profile estimating unit estimates the owner's profile for each owner estimated by the owner estimating unit. A user profile estimating apparatus characterized by the above-mentioned.

The user profile estimating apparatus of the present invention may further include a position information acquisition unit for acquiring the position information of the article from which the information acquisition unit reads information on the article, and the owner estimation unit acquires the position information acquisition unit. Based on the positional information of the article, the owner of the article can be estimated according to the relative position between the articles.

The user profile estimating apparatus of the present invention may further include a history storage unit for storing the time when the information acquisition unit acquires the information on the article in correspondence with the information representing the article. A profile of the user can be estimated based on the information indicating the article stored additionally and the time associated with it.

According to the present invention, there is provided a robot characterized by autonomous movement, comprising the user profile estimating apparatus described in any one of the above, a moving unit for moving the user profile estimating apparatus, and an operation control unit for controlling the moving unit. do.

According to the present invention, there is provided a robot comprising a user profile estimating apparatus according to any one of the above, and an operation control unit which operates for a user based on the profile of the user estimated by the user profile estimating apparatus. do.

According to the present invention, a user's preference is estimated based on a step of receiving information on an article from a tag attached to the article and on the information on the article received in the step of receiving information on the article. There is provided a user preference estimation method comprising a step.

The user preference estimation method of the present invention may further include a step of acquiring the article name of the article based on the information about the article received in the step of receiving the information about the article. The estimating step may include obtaining symbol information corresponding to the article name from a database storing the article name and the symbol information corresponding to the article name based on the article name acquired in the step of acquiring the article name. Can be.

According to the present invention, a profile of a user is estimated based on a step of acquiring information about the article from a tag attached to the article and the information about the article obtained in the step of acquiring the information about the article. There is provided a user profile estimation method comprising the step of.

According to the present invention, a computer is provided with a receiving means for receiving information about an article from a tag attached to the article, and a sign determining means for estimating a user's preference based on the information about the article received by the receiving means. There is provided a user preference estimation program characterized by functioning as a function.

According to the present invention, a user included in a computer included in a robot having a moving means, based on the information on the article received by the receiving means for receiving information about the article from the tag attached to the article, and the user. A robot control program is provided, which functions as a symbol determining means for estimating a symbol of a symbol and an operation control means for controlling the moving means to autonomously move the robot.

According to the present invention, since the user's preferences and profiles are estimated based on the possessions of the user, the user preferences and profiles can be estimated with good accuracy. In addition, since a plurality of articles are classified based on information such as the owner of the article, the spatial location of the article, and the hobby is estimated for each group, even if there are possessions of a plurality of users, The profile can be estimated with good precision.

In addition, according to the present invention, since the user reads tag information pre-attached to an article existing around the user, such as an article in the user's house, and estimates the user's hobby based on the information, the user has a hobby or the like. Does not need to be entered directly into the system. Therefore, the burden on a user can be reduced.

The above objects and other objects, features, and advantages will become more apparent from preferred embodiments described below and accompanying drawings.

BRIEF DESCRIPTION OF THE DRAWINGS Fig. 1 is a block diagram showing the configuration of a user preference estimation device in an embodiment of the present invention.

2 is a diagram illustrating an example of an internal configuration of a preference information database.

3 is a diagram showing an example of an internal configuration of a read result storage unit;

4 is a flowchart showing a processing procedure of the user preference estimation apparatus shown in FIG. 1;

5 is a flowchart showing the processing procedure of the user preference estimation apparatus in more detail.

6 is a diagram illustrating another example of an internal configuration of a preference information database.

FIG. 7 is a diagram showing an example of an internal configuration of a read result storage unit when the preference information database is the configuration shown in FIG. 6; FIG.

8 is a diagram illustrating an example of an internal configuration of a threshold storage unit.

9 is a diagram illustrating a configuration of a user preference estimation device in an embodiment of the present invention.

10 is a diagram showing an example of an internal configuration of a preference information database according to the embodiment of the present invention.

FIG. 11 is a flowchart showing a processing procedure of the user preference estimation apparatus shown in FIG. 9;

12 is a diagram illustrating another example of an internal configuration of a preference information database.

Fig. 13 is a block diagram showing the structure of a user preference estimation device in the embodiment of the present invention.

FIG. 14 is a flowchart showing a processing procedure of the user preference estimation apparatus shown in FIG.

Fig. 15 is a block diagram showing another example of the configuration of a user preference estimation device in the embodiment of the present invention.

FIG. 16 is a flowchart showing a processing procedure of the user preference estimation apparatus shown in FIG. 15;

17 is a diagram illustrating a configuration of a user preference estimation device in an embodiment of the present invention.

18 is a flowchart showing a processing procedure of the user preference estimation apparatus shown in FIG. 17;

19 is a block diagram showing another example of the configuration of a user preference estimation apparatus in the embodiment of the present invention.

20 is a flowchart showing a processing procedure of the user preference estimation apparatus shown in FIG. 19. FIG.

21 is a diagram illustrating a configuration of a user preference estimation device in an embodiment of the present invention.

Fig. 22 is a flowchart showing an example of the processing procedure of the user preference estimation device in the embodiment of the present invention.

Fig. 23 is a diagram showing an example of an internal configuration of a classification result storage unit in the embodiment of the present invention.

24 is a diagram showing an example of an internal configuration of a read result storage unit in the embodiment of the present invention;

25 is a diagram illustrating a configuration of a user preference estimation device in an embodiment of the present invention.

Fig. 26 is a flowchart showing an example of a processing procedure of a user preference estimation device in the embodiment of the present invention.

Fig. 27 is an external configuration diagram showing an example of a robot in the embodiment of the present invention.

28 is a block diagram illustrating an example of an electrical configuration of a robot.

FIG. 29 is a block diagram showing details of the controller of the robot shown in FIG. 28; FIG.

30 is an external configuration diagram showing an example of a user profile estimating apparatus according to the embodiment of the present invention.

<Best Mode for Carrying Out the Invention>

EMBODIMENT OF THE INVENTION Next, embodiment which concerns on this invention is described in detail, referring an accompanying drawing. In the following embodiment, the same code | symbol is attached | subjected to the same component, and description is abbreviate | omitted suitably.

(1st embodiment)

1 is a block diagram showing the configuration of a user preference estimation device in the present embodiment.

In this embodiment, the IC tag 1 is attached to the article arrange | positioned indoors. The IC tag 1 can be a device that generates waves that mediate information such as electromagnetic waves or sound waves, such as RFID (Radio Frequency Identification) tags, ultrasonic tags, or IR (Infra-Red) tags. The IC tag 1 stores at least information for uniquely specifying the article on which the IC tag 1 is attached, and transmits the information by electromagnetic waves (including radio waves and light) or sound waves.

The user preference estimation apparatus 100 is constituted by the operation means 10 and the preference information database 5. The operation means 10 includes the reception means 2 (receiver), the article name determination means 3 (article name acquisition unit), the symbol determination means 4 (the symbol determination unit), the threshold storage unit 15, and the read result. The storage unit 19 is included. The operation means 10 assumes the function of being operated by the CPU of a computer or the like.

The receiving means 2 includes an antenna for receiving information from the IC tag 1, an electric circuit for driving the same, a signal analysis, and the like.

The article name database 6 holds a table in which tag information stored in the IC tag 1 and article names are associated with each other. The article name determination means 3 searches the article name database 6 based on the tag information acquired by the receiving means 2 from the IC tag 1, and acquires the article name corresponding to the tag information.

Although not shown here, in the present embodiment, the user preference estimating apparatus 100 has a communication control function of transmitting and receiving data with other terminals via a network such as the Internet. The article name database 6 is stored in, for example, a Web server connected to the Internet. The article name determination means 3 acquires a article name from the article name database 6 via a network.

The article name determination means 3 notifies the preference determination means 4 of the article name acquired from the article name database 6. The preference information database 5 holds information capable of determining a preference based on at least the article name.

2 is a diagram illustrating an example of an internal configuration of the preference information database 5.

The preference information database 5 includes a hobby column and an article name column in which a product name of an article serving as a reference for determining that the user has the hobby is stored. Here, in the hobby column, "fishing", "classic music listening", "skiing", and "overseas travel" are stored, for example. For example, in the article name field corresponding to "fishing" in the hobby column, "A fishing rod a", "A fishing rod b", and "B fishing company" initiation "are stored.

Returning to FIG. 1, the preference determination means 4 determines the preference of the user by checking the preference information database 5. For example, when a user owns "A fishing rod a", "A fishing rod b", etc., the preference determination means 4 can determine that a user's hobby is "fishing."

The threshold storage unit 15 stores a threshold value that the taste determining unit 4 refers to when determining the user's hobby. The preference determination means 4 reads the hobby corresponding to the article name from the preference information database 5 based on the article name acquired by the article name determination means 3. Each time the taste determination means 4 reads the hobby from the preference information database 5, the taste determination means 4 stores the result in the read result storage unit 19.

3 is a diagram illustrating an example of an internal configuration of the read result storage unit 19.

The reading result storage unit 19 includes a hobby column and a reading count column for storing the number of times the taste determining means 4 read the hobby from the preference information database 5. Each time the taste determination means 4 reads the hobby from the preference information database 5, the preference determination means 4 increments the number of read count columns corresponding to the hobby of the read result storage unit 19.

Returning to Fig. 1, the preference determination means 4 refers to the read result storage unit 19 and the threshold storage unit 15 after the end of the read processing. If it is more than the threshold stored in the threshold memory 15, it is determined that the hobby is the hobby of the user. If the number of times of reading is lower than the threshold for all the hobbies, the preference determination means 4 judges that the reliability of determination is low, and can not output the determination result. The preference determination means 4 outputs the determination result. For example, the preference determination means 4 may output the determination result to other components included in the user preference estimation apparatus 100, or may be output to an external device of the user preference estimation apparatus 100. .

FIG. 4 is a flowchart showing a processing procedure of the user preference estimation apparatus 100 shown in FIG. 1.

First, the reception means 2 reads tag information from the surrounding IC tag 1 (step S1). At this time, the receiving means 2 may read tag information of one IC tag 1 or may read tag information of a plurality of IC tags 1. Subsequently, the article name determination means 3 searches the article name database 6 based on the tag information read by the reception means 2, and acquires the corresponding article name (step S2). Thereafter, the article name determination means 3 notifies the preference determination means 4 of the acquired article name.

In step S2, for example, when the information written in the IC tag 1 is a number attached to a product, the article name determination means 3 uses the number as a search key and supplies the article name database 6 via the network. Can be accessed to obtain a specific product name corresponding to the number.

Next, the preference determination means 4 searches the preference information database 5, and determines the preference of a user (step S3).

5 is a flowchart showing the processing procedure of the user preference estimation apparatus 100 in more detail. A description with reference to FIGS. 1 to 3 is as follows.

First, the reception means 2 acquires tag information from the IC tag 1 (step S100). Subsequently, the article name determination means 3 acquires the corresponding article name from the article name database 6 based on the tag information acquired by the reception means 2 (step S102). Subsequently, the preference determination means 4 reads the corresponding hobby from the preference information database 5 on the basis of the article name acquired by the article name determination means 3, and reads the hobby into the reading result storage unit 19. It is stored in association with the number of readings (step S104).

Subsequently, the control unit (not shown) of the operation means 10 determines whether or not the processing has ended for all the articles (step S106). The determination of whether or not the processing has been completed for all the articles can be made based on whether or not the tag information of all the articles within the predetermined range where the tag information can be acquired by the receiving means 2 is obtained, for example. If the process has not ended (NO in step S106), the process returns to step S100 to repeat the same process for the next article.

On the other hand, when the process is finished in step S106 (example of step S106), the preference determination means 4 refers to the read result storage unit 19 and the threshold value storage unit 15 to read the read result storage unit. For each hobby stored in (19), it is determined whether the number of readings is equal to or greater than the threshold stored in the threshold storage unit 15 (step S108). If there is a hobby more than the threshold value (YES in step S108), the hobby more than the threshold value is determined as the hobby of the user and the determination result is output (step S110). At this time, when there are a lot of hobby more than a threshold value, the preference determination means 4 may output all of them, and may preferentially output from the hobby with many reading times.

On the other hand, in step S108, when there is no hobby greater than or equal to the threshold (NO in step S108), the preference determination means 4 determines that the reliability of the determination is low, and outputs that there is no corresponding hobby (S112).

Hereinafter, specific examples will be described. For example, it is assumed that the threshold value = 15 stores the threshold value = 2. In addition, it is assumed that the article name determination means 3 has acquired "A fishing rod a" and "A fishing rod b" as the product name. In this case, when the preference determination means 4 searches the preference information database 5 using the article name as a retrieval key, the number of times of "hobby" = "fishing" is read twice. Each time the preference determination means 4 reads "hobbies" = "fishing" from the preference information database 5, the preference determination means 4 memorize | stores the effect in the reading result storage part 19. As shown in FIG. After the end of the readout process, the preference determination means 4 refers to the threshold value storage section 15 and the read result storage section 19. Here, since the number of times of reading "hobby" = "fishing" is more than a threshold value, the preference determination means 4 determines that this user's hobby is "fishing". The preference determination means 4 determines that the user's hobby is "fishing", and outputs the effect.

6 is a diagram illustrating another example of the internal configuration of the preference information database 5.

Here, the preference information database 5 includes a hobby column and an article data column. The article data column includes an article name column and a weight column. Here, weights are predetermined for each article. For example, in the item name field corresponding to "fishing" in the hobby column, "A fishing rod a", "A fishing rod b", and "B company's book" introductory fishing "" are stored, and each weight is "15". , "15" and "10" are set.

FIG. 7: is a figure which shows an example of the internal structure of the reading result storage part 19 when the preference information database 5 is a structure shown in FIG.

Here, the read result storage unit 19 includes a hobby column and a weighted sum column. The preference determination means 4 reads the weight together with the hobby from the preference information database 5, and adds the weight read to the weight total column corresponding to the hobby of the read result storage unit 19 each time. .

1, in this case, the threshold value storage part 15 also stores the threshold value which considered the weight. After the end of the reading process, the preference determination means 4 refers to the reading result storage unit 19 and the threshold storage unit 15, and the weighted sum is assigned to the threshold storage unit 15 for either hobby. If it is more than the stored threshold, it is determined that the hobby is the hobby of the user. Also in this case, the preference determination means 4 judges that the reliability of the determination is low when all the hobbies are lower than the threshold value, and can not output the determination result.

In the threshold storage unit 15, the weight can be set to be low for articles that are likely to be owned by many people regardless of the hobby. For example, when an article is a "book", if it is a best-selling book, it may be owned by a person who is not very interested in the field of the book. Therefore, the weight of the threshold storage part 15 can be set suitably also according to the sales of each article, etc. In addition, the weight of the threshold memory 15 can be updated as appropriate. In this way, the weight information is stored in the preference information database 5 in association with the article name, and the weight is also taken into consideration when determining the user's hobby, whereby the user's hobby can be determined with better accuracy.

8 is a diagram illustrating an example of an internal configuration of the threshold storage unit 15.

The threshold storage unit 15 may store one threshold value for all the hobbies. However, as shown in FIG. 8, the threshold storage unit 15 may also store different threshold values for each hobby. Here, for example, the threshold of the hobby "fishing" is "2", the threshold of the hobby "classic music appreciation" is "10", the threshold of the hobby "ski" is "2", the threshold of the hobby "overseas travel" is " 10 ”. For example, regardless of the hobby of the user, when a lot of people are likely to own the goods related to the hobby, the threshold for the hobby can be set higher. This also makes it possible to determine the hobby of the user with better accuracy.

In addition, in this embodiment, in order to prevent the same article being observed multiple times at the time of addition, the IC tag 1 contains the ID unique to the article, and the ID is added only once even when observed multiple times. It is also possible not to. For example, the user preference estimation apparatus 100 may include a storage unit (not shown) that stores tag information of the IC tag 1 read by the receiving means 2. When reading the tag information from the arbitrary IC tag 1, the receiving means 2 refers to the storage part, and judges whether the tag information of the IC tag 1 is already read and processed, If the tag information is not read, the subsequent processing can be performed. Thereby, overlapping of the observation result of the same article can be prevented.

In another example, the user preference estimation apparatus 100 may be configured to include the article name database 6 in the own apparatus. In this case, the article name database 6 can hold information associated with the tag information and the article name (product name). The article name determination means 3 can acquire a product name from the article name database 6 using tag information as a key.

According to the user preference estimation apparatus 100 according to the present embodiment, the user's preference information can be obtained without the explicit input of the user. In recent years, at the logistics stage, such as when the article is manufactured or sold, a tag is attached to the article and a process of managing the article using the tag is performed. According to the present invention, the tag information used in the distribution step can be utilized to estimate the preference of the user even after the sale.

In addition, in this embodiment, the user preference estimation apparatus 100 reads tag information from the article in the predetermined range which the user preference estimation apparatus 100 reads tag information, and based on the information, Estimate the symbol. Therefore, the user's preference can be estimated based on the article actually owned by the user at the time of estimation, and the preference can be estimated accurately even when the user's preference changes.

(2nd embodiment)

9 is a diagram showing the configuration of the user preference estimation apparatus 100 in the present embodiment.

In the present embodiment, the operation means 10 further includes an article category determination means 7 (article category acquisition unit) in addition to the components described with reference to FIG. 1 in the first embodiment. In addition, the internal structure of the preference information database 5 differs from the preference information database 5 in 1st Embodiment.

The article category database 8 has a database structure which stores the article name and the category corresponding to the article name in association. The article category database 8 is stored in, for example, a Web server connected to the Internet. The article category determination means 7 acquires a corresponding category from the article category database 8 via a network.

In the present embodiment, the article name determination means 3 notifies the article category determination means 7 of the article name obtained by searching the article name database 6. The article category determination means 7 retrieves the category corresponding to the article name as a search result by searching the article category database 8 using the article name notified from the article name determination means 3 as a search key.

The preference determination means 4 receives a category from the article category determination means 7, performs a search of the preference information database 5 using this category, and determines a user's preference.

10 is a diagram illustrating an example of an internal configuration of the preference information database 5 according to the present embodiment.

The preference information database 5 includes a hobby column and an article category name column of an article serving as a reference for determining that the user has the hobby. Here, in the hobby column, "fishing", "classic music listening", "skiing", and "overseas travel" are stored, for example. For example, "fishing rod", "fishing book" and "reel" are stored in the article category name field corresponding to "fishing" in the hobby column.

FIG. 11 is a flowchart showing a processing procedure of the user preference estimation apparatus 100 shown in FIG. 9.

First, the reception means 2 reads tag information from the surrounding IC tag 1 (step S11). At this time, the reception means 2 may read tag information of one IC tag 1, and may also read tag information of a plurality of IC tags 1. Subsequently, the article name determination means 3 searches the article name database 6 based on the tag information read by the reception means 2, and acquires the corresponding article name (step S12). Thereafter, the article name determination means 3 notifies the article category determination means 7 of the acquired article name.

Subsequently, the article category determination means 7 searches the article category database 8 based on the article name acquired from the article name determination means 3, and specifies the category to which the article belongs (step S13). The article category determination means 7 can access the article category database 8 via a network, and acquire the category name corresponding to a article name. For example, the article category database 8 can store the article name "personal computer of the company number XXXX of A company" in association with a category name "personal computer." In this case, for example, when the article name obtained by the article name determination means 3 is "PC of company No. XXXX", the article category determination means 7 retrieves the article category database 8 to obtain the article category. Becomes a "personal computer".

Subsequently, the preference determination means 4 searches the preference information database 5 to determine the preference of the user (step S14). The process of determining a user's preference can be performed similarly to the steps S104 to S112 described with reference to FIG. 5 in the first embodiment.

In other words, the preference determination means 4 reads the corresponding hobby from the preference information database 5 based on the article category acquired by the article category determination means 7, and reads the corresponding hobby into the reading result storage unit 19. Memorizes the number of readings in association with. Subsequently, the control unit (not shown) of the operation means 10 determines whether or not the reading process has ended for all articles. If the process has not ended, the process returns to Step S11 to repeat the same process for the next article. On the other hand, when the process is finished, the preference determination means 4 refers to the read result storage unit 19 and the threshold value storage unit 15 and, for each hobby stored in the read result storage unit 19. It is determined whether or not the number of readings is equal to or greater than the threshold stored in the threshold storage unit 15. If there is a hobby above the threshold, the hobby above the threshold is determined as the hobby of the user, and the determination result is output. At this time, when there are a plurality of hobby more than a threshold value, the preference determination means 4 may output all of them, and may preferentially output from the hobby with many reading times.

Hereinafter, specific examples will be described. For example, it is assumed that the threshold value = 15 stores the threshold value = 2. In addition, it is assumed that the article category determining means 7 obtained "fishing rod" twice as a category name. In this case, when the preference determination means 4 searches the preference information database 5 using a category name as a search key, the number of times of reading "hobby" = "fishing" is made twice. Each time the preference determination means 4 reads "hobbies" = "fishing" from the preference information database 5, the preference determination means 4 memorize | stores the effect in the reading result storage part 19. As shown in FIG. After the end of the reading process, the preference determination means 4 refers to the threshold value storage section 15 and the read result storage section 19. Here, since the number of times of reading "hobby" = "fishing" is more than a threshold value, the preference determination means 4 determines that the hobby of a user is "fishing", and outputs the effect.

12 is a diagram illustrating another example of the internal configuration of the preference information database 5.

Here, the preference information database 5 includes a hobby column and an article data column. The article data column includes an article category name column and a weight column. Here, weights are predetermined for each article category name. At this time, the read result storage unit 19 has a structure similar to that described with reference to FIG. 7 in the first embodiment.

The preference determination means 4 reads the weight together with the hobby from the preference information database 5 every time the article category determination means 7 acquires the article category name, The weight read out is added to the total weight column corresponding to the hobby. After the end of the reading process, the preference determination means 4 refers to the reading result storage unit 19 and the threshold storage unit 15, and the weighted sum is added to the threshold storage unit 15 for either hobby. If it is more than the stored threshold value, it determines with the hobby of a user, and outputs the effect. Also in this case, if the sum of weights for all the hobbies is lower than the threshold, the preference determination means 4 determines that the reliability of the determination is low, and can not output the determination result.

In addition, in this embodiment, in order to prevent observing the same article plural times at the time of addition, the IC tag 1 includes the ID unique to the article, and the ID is added only once even when plural times are observed. It is also possible to avoid.

In another example, the user preference estimation apparatus 100 may be configured to include an article name database 6 and an article category database 8 in its own apparatus. In this case, the article name database 6 can store the information which matched tag information and article name (product name). The article category database 8 can also hold information associated with the article name and the category name.

In addition, the article name database 6 and the article category database 8 may be formed integrally. That is, one database may be configured to hold information associated with tag information, article name, and category name.

Also in this embodiment, the effect similar to the user preference estimation apparatus 100 in 1st Embodiment is acquired. In addition, since the preference information database 5 of this embodiment can arrange a plurality of article names into a category name, it can be set as the structure which stored a hobby in correspondence with it, and can prepare the preference information database 5 easily. . In addition, the preference information database 5 can be updated easily.

(Third embodiment)

13 is a block diagram showing the configuration of the user preference estimation apparatus 100 in the present embodiment.

In the present embodiment, the user preference estimating apparatus 100 estimates the owner for each article, and determines the preference for each owner. Thereby, even when a plurality of articles each owned by a plurality of persons are arranged indoors, preference can be estimated for each owner.

The operation means 10 further includes an owner judging means 9 (owner estimating section) in addition to the components described with reference to FIG. 1 in the first embodiment. The owner determining means 9 estimates the owner for each article. The owner determination means 9 can estimate the owner of each article as follows, for example.

For example, the owner judging means 9 may include a storage unit (not shown) in which the user memorizes the owner with respect to each article in advance. Specifically, for example, the storage unit can store information (tag information) for identifying the article stored in the IC tag 1 attached to the article and information for identifying the user in association with each other. The owner determination means 9 can read the information identifying the user from the storage unit on the basis of the tag information read by the reception means 2 from the IC tag 1, and can estimate the owner of the article based thereon. have.

Further, in another example, the owner judging means 9 may determine the owner by acquiring an individual purchase record from information such as a credit card or electronic payment and specifying the purchaser. Specifically, for example, the owner judging means 9 can access a database storing a purchase record and estimate the owner. The database storing the purchase record can be configured to perform access authentication, and the owner determination means 9 can hold a password or ID for performing access authentication. The owner determination means 9 can access a database storing a purchase record, perform access authentication, and obtain necessary information from the database.

In another example, the user preference estimating apparatus 100 can detect a person and an article by processing sensor information such as a camera, and can determine a person to be frequently used as the owner. That is, the user preference estimating apparatus 100 can record the use history of the article by a person, and can estimate the owner based on the record. The usage history of the article detects the presence of an arbitrary person, for example, by image recognition by a camera or the like, and when the IC tag 1 attached to the article is moved at that time, the person has moved the article. It can be obtained by judging that it is used and storing the person and the article in correspondence.

FIG. 14 is a flowchart showing the processing procedure of the user preference estimation apparatus 100 shown in FIG. A description with reference to FIG. 13 is also given below.

Step S21 and step S22 are the same as step S1 and step S2 described with reference to FIG. 4 in the first embodiment, respectively, and thus description thereof is omitted.

In this embodiment, the owner determination means 9 determines the owner of each article. Here, the owner is set to p [n] (p [0] to P [N]) (step S23).

In addition, the owner determination means 9 can store the result of the owner determination in a database (not shown) in association with the article name of the article. After the determination of the owner of the observed article is completed, the preference determination means 4 first sets n = O (step S24). Subsequently, the preference determination means 4 searches the database for the article name of the article whose owner is p [n] by referring to the database of the owner determination means 9, and determines the hobby of the owner p [n] ( Step S25). Determination of the hobby for each owner can be performed similarly to step S3 of FIG. 4 demonstrated by 1st Embodiment, for example.

Next, let n = n + 1 (step S26). After that, it is determined whether n is larger than N (step S27), and when n is not N (NO in step S27), the process returns to step S25 and the same process is repeated. On the other hand, in step S27, when n> N (the example of step S27), a process is complete | finished. By the above processing, the hobby of owners p [0] to P [N] can be determined.

15 is a block diagram showing another example of the configuration of the user preference estimation apparatus 100 according to the present embodiment.

Here, the user preference estimation apparatus 100 further includes article category determination means 7 as described with reference to FIG. 9 in the second embodiment. Other components are the same as the configuration shown in FIG.

FIG. 16 is a flowchart showing a processing procedure of the user preference estimation apparatus 100 shown in FIG. 15.

Since step S31 to step S33 are the same as step S11 to step S13 demonstrated with reference to FIG. 11 in 2nd Embodiment, respectively, description is abbreviate | omitted. In addition, step S34 to step S38 are the same as step S23 to step S27 demonstrated with reference to FIG. 14 in this embodiment, respectively, and abbreviate | omit description suitably. The owner determination means 9 can store the result of determining the owner of each article in step S34 in a database (not shown) in association with the category to which the article belongs. In step S36, the preference determination means 4 refers to the database of the owner determination means 9, searches from the database for the category to which the article whose owner is p [n] belongs, and determines the hobby of the owner p [n]. do.

Also in this embodiment, the effect similar to 1st Embodiment and 2nd Embodiment is acquired. In addition, according to the user preference estimation apparatus 100 according to the present embodiment, even when articles of a plurality of owners are arranged within a predetermined range in which the user preference estimation apparatus 100 reads tag information, the owner is considered. Thus, the preference can be estimated for each owner.

(4th embodiment)

17 is a block diagram showing the configuration of the user preference estimation apparatus 100 according to the present embodiment.

In the present embodiment, the operation means 10 clusters a plurality of articles on which the reception means 2 has read tag information based on a predetermined criterion, and determines the preference for each cluster. As a result, the user's preference can be estimated with better accuracy.

The operation means 10 further includes a clustering means 11 (clustering processing section) and a clustering result storage section 21 in addition to the components described with reference to FIG. 1 in the first embodiment. The clustering means 11 clusters the plurality of articles whose tag information has been read by the receiving means 2 based on predetermined criteria. The clustering result storage unit 21 stores the clustering result by the clustering means 11.

In this embodiment, the clustering means 11 can acquire the positional information of the article which the receiving means 2 read the tag information. The clustering means 11 can classify the thing in which the position in which an article is placed, and the relative positional relationship between articles is close to the same cluster based on the positional information of the article.

The clustering means 11 can acquire the positional information of each article by various methods, and is not specifically limited, For example, the clustering means 11 can acquire the positional information of each article as follows.

For example, the clustering means 11 can be set as the structure which acquires the magnetic position of the user preference estimation apparatus 100. FIG. When the reading range of the tag information from the IC tag 1 by the receiving means 2 is not very wide, the user preference estimating apparatus 100 moves to an article placed in a predetermined range by being moved or autonomously moved by the user. Tag information is acquired from the attached IC tag 1. In this case, the position information of the article can be obtained by acquiring the magnetic position of the user preference estimation apparatus 100 when the reception means 2 acquires the tag information.

In another example, the clustering means 11 is based on the radio wave strength of the IC tag 1 and the receiving means 2 and the like when the receiving means 2 acquires the tag information from the IC tag 1. The positional information of the article can be obtained.

FIG. 18 is a flowchart showing the processing procedure of the user preference estimation apparatus 100 shown in FIG. 17. A description with reference to FIG. 17 is also given below.

Step S41 and step S42 are the same as step S1 and step S2 described with reference to FIG. 4 in the first embodiment, respectively, and thus description thereof is omitted.

The clustering means 11 clusters the plurality of articles observed in step S41 and step S42. Here, the name of each cluster is set to c [n] (c [O] to c [N]) (step S43). The clustering means 11 stores the clustering result in the clustering result storage unit 21 in association with the article name of the article.

After clustering of the observed article is finished, the preference determination means 4 first sets n = O (step S44). Subsequently, the preference determination means 4 reads the article name of the article whose cluster name is c [n] from the clustering result storage section 21 with reference to the clustering result storage section 21. Subsequently, the preference determination means 4 reads the hobby corresponding to the article name from the preference information database 5 and stores it in the read result storage unit 19 in correspondence with the number of times of reading the article name. When the processing is finished for all articles having the cluster name c [n], the number of reads stored in the read result storage unit 19 and the threshold stored in the threshold storage unit 15 are compared to determine the cluster name c [n. The hobby is determined based on the article belonging to the cluster (step S45).

Next, n = n + 1 (step S46). After that, it is determined whether n is larger than N (step S47), and when n is not N (NO in step S47), the process returns to step S45 and the same process is repeated. On the other hand, in step S47, when n> N (the example of step S47), a process is complete | finished. Through the above processing, the taste of the cluster whose cluster names are c [O] to c [N] can be determined.

19 is a block diagram showing another example of the configuration of the user preference estimation apparatus 100 according to the present embodiment.

Here, the user preference estimation apparatus 100 further includes article category determination means 7 as described with reference to FIG. 9 in the second embodiment. The other components are the same as the configuration shown in FIG.

FIG. 20 is a flowchart showing a processing procedure of the user preference estimation apparatus 100 shown in FIG. 19.

Since step S51 to step S53 are the same as that of step S11 to step S13 which were demonstrated with reference to FIG. 11 in 2nd Embodiment, respectively, description is abbreviate | omitted. In addition, since step S53-step S58 are the same as that of step S43-step S47 which were demonstrated with reference to FIG. 18 in this embodiment, respectively, description is abbreviate | omitted suitably. The clustering means 11 stores the result of clustering the plurality of articles in step S54 in the clustering result storage unit 21 in association with the category to which the articles belong. In step S56, the preference determination means 4 refers to the clustering result storage unit 21 and reads from the result storage unit 19 the category of the article whose cluster name is c [n]. Subsequently, the preference determination means 4 reads the hobby corresponding to a category from the preference information database 5, and stores it in the read result storage unit 19 in correspondence with the number of times of reading the category. When the processing is finished for all articles having the cluster name c [n], the number of reads stored in the read result storage unit 19 and the threshold stored in the threshold storage unit 15 are compared, and the cluster name is c [n. The hobby is determined based on the article belonging to the cluster.

Also in this embodiment, the effect similar to 1st Embodiment and 2nd Embodiment is acquired. In addition, according to the user preference estimating apparatus 100 of the present embodiment, since preferences can be estimated for each cluster, errors can be less likely to occur, and preferences can be estimated with better accuracy.

(5th embodiment)

21 is a diagram illustrating the configuration of the user preference estimation apparatus 100 in the present embodiment.

In the present embodiment, as described in the third embodiment, the user preference estimation apparatus 100 estimates the owner for each article and determines the preference for each owner. Thereby, even when a plurality of articles each owned by a plurality of persons are arranged indoors, preference can be estimated for each owner.

The user preference estimating apparatus 100 includes an information acquisition unit 102, an article information acquisition unit 108, a preference determination unit 114, a classification processing unit 120, an input reception unit 130, a presentation processing unit 132, and a preference. The information database 5, the threshold storage unit 15, the read result storage unit 19 and the classification result storage unit 122 are included. In the present embodiment, the information acquisition unit 102, the article information acquisition unit 108, the preference determination unit 114, the classification processing unit 120, the input reception unit 130, and the presentation processing unit 132 are first to fifth. It is realized by the operating means described in the fourth embodiment. In addition, each component of the user preference estimation apparatus 100 is controlled by the control part which is not shown in figure.

Hereinafter, the relationship between each component in this embodiment and each component in 1st-3rd embodiment is demonstrated. The information acquisition unit 102 has the same function as the reception means 2. The article information acquisition unit 108 has either or both of the functions of the article name determination means 3 and the article category determination means 7. The article information database 150 is a structure including any one or both of the article name database 6 and the article category database 8. The preference determination unit 114 has the same function as the preference determination means 4.

In the present embodiment, the input accepting unit 130 receives an instruction or the like from the user. The input reception unit 130 can be realized by, for example, a combination of a voice input means and a voice recognition means, a graphical user interface (GUI) or a button. The presentation processing unit 132 performs a process of presenting information to the user. The presentation processing unit 132 performs display processing, audio output, or the like on a display or a speaker.

The classification processing unit 120 classifies the article on which the information acquisition unit 102 reads the tag information based on a predetermined criterion. In this embodiment, the classification processing part 120 has a function similar to the owner determination means 9 of 3rd embodiment. The classification result storage unit 122 stores the results classified by the classification processing unit 120.

For example, each time the classification processing unit 120 reads the tag information, the classification processing unit 120 estimates the owner of the article and associates the tag information of the article with the owner of the article. Can remember. Further, whenever the article information acquisition unit 108 acquires article information from the article information database 150, the classification processing section 120 estimates the owner of the article, and gives tag information of the article to the owner of the article. The information can be stored in the classification processing unit 120 in association.

In this embodiment, the classification processing part 120 can estimate an owner for every article by the method similar to what was demonstrated about the owner determination means 9 in 3rd Embodiment, or other various methods.

For example, the classification processing unit 120 can estimate the owner of the article based on the positional information of the article. In this case, it can be performed as follows, for example. When the user preference estimating apparatus 100 acquires the information of the some goods arrange | positioned in the arbitrary house, the user preference estimation apparatus 100 stores the memory which memorize | stores the map information which matched the position information of each room with the owner of each room. It may include wealth. For example, when the information acquisition unit 102 reads the tag information from the IC tag 1 attached to the article, the classification processing unit 120 acquires the position information of the article and based on the map information, for example. The owner of the item can be estimated.

In the user preference estimating apparatus 100, when the owner of the article is estimated, the classification processing unit 120 inquires of the owner through the presentation processing unit 132 and asks the owner from the user through the input reception unit 130. You can enter a reply. Thereby, the classification processing part 120 can estimate the owner of an article. In this case, the classification processing unit 120 may inquire the user for each article. However, when a response from the user is obtained for any article, the classification processing unit 120 may assume that the owners are the same for the articles in the surroundings. . As a result, the burden on the user can be reduced.

As another example, a tag different from the IC tag 1 that holds information identifying the owner may be attached to the article. In this case, when the information acquisition unit 102 reads the tag information from the IC tag 1, the information acquisition unit 102 also reads the tag information from other tags. The classification processing part 120 can estimate the owner of each article based on the information which identifies the owner which the information acquisition part 102 read. Also in this case, all the articles may be tagged, but only a few articles are attached, and the sorting processing unit 120 can assume that the owners are the same for the articles around them. For example, a bookshelf in any room may be tagged with information that identifies its owner, and for books placed on the bookshelf, it may be assumed that the owner and owner of the bookshelf are the same. have.

22 is a flowchart showing an example of a processing procedure of the user preference estimation apparatus 100 according to the present embodiment.

First, the information acquisition part 102 acquires tag information from the IC tag 1 (step S200). Subsequently, the classification processing unit 120 determines whether to classify the article (step S202). In this embodiment, the classification of the article is to determine the owner. At what stage the sorting processing unit 120 classifies articles can be set in advance. For example, when the information acquisition unit 102 is set to classify articles every time the tag information is acquired from the IC tag 1, the classification processing unit 120 determines that the classification is performed in step S202. (YES in step S202), the information acquisition unit 102 performs classification processing of the article in which the tag information has been acquired (step S204). The classification processing unit 120 stores the classification result in the classification result storage unit 122.

If it is determined after step S204 or not to be classified in step S202 (NO in step S202), the flow proceeds to step S206 and the article information acquisition unit 108 based on the tag information acquired by the information acquisition unit 102. The corresponding article information is obtained from the article information database 150 (step S206). Here, the article information can be a product name or an article category name.

Subsequently, the classification processing unit 120 determines whether to classify the article (step S208). For example, when the article information acquisition unit 108 is set to classify articles each time the article information is acquired from the article information database 150, the classification processing unit 120 performs classification in step S208. It judges that it is (YES in step S208), and the article information acquisition part 108 performs the classification process of the article which acquired the article information (step S210). The classification processing unit 120 stores the classification result in the classification result storage unit 122.

If it is determined after step S210 or not to be classified in step S208 (NO in step S208), the flow proceeds to step S212, and the preference determination unit 114 based on the article information acquired by the article information acquisition unit 108. The corresponding hobby is read from the preference information database 5. In the present embodiment, the article information acquisition unit 108 also notifies the preference determination unit 114 of the tag information acquired from the information acquisition unit 102 together with the article information acquired from the article information database 150. do. When the preference determination unit 114 receives the notification from the article information acquisition unit 108, the preference determination unit 114 reads the hobby corresponding to the article information with reference to the preference information database 5. At the same time, the preference determination unit 114 refers to the classification result storage unit 122 and reads out information indicating the owner of the article using the tag information as a key. The taste determining unit 114 stores the number of readings for each owner in the reading result storage unit 19 in association with the hobby read out from the preference information database 5 (step S212).

Subsequently, the control unit (not shown) of the user preference estimation apparatus 100 determines whether or not the processing has ended for all articles (step S214). If the process has not ended (NO in step S214), the process returns to step S200 and the same process is repeated for the next article. On the other hand, when the process ends in step S214 (YES in step S214), the preference determination unit 114 refers to the read result storage unit 19 and the threshold storage unit 15 to read out for each owner. For each hobby stored in the result storage unit 19, it is determined whether the number of readings is equal to or greater than the threshold stored in the threshold storage unit 15 (step S216). If there is a hobby above the threshold, the hobby above the threshold is determined as the hobby of the owner, and the determination result is output (step S218). On the other hand, when there is no hobby more than a threshold in step S216, in step S218, the preference determination part 114 determines that the determination reliability is low, and can output that it does not have a corresponding hobby.

FIG. 23 is a diagram showing an example of an internal configuration of the classification result storage unit 122 in the present embodiment. The classification result storage unit 122 includes a tag information field and an owner information field. Here, tag information "0001" and "0003" are associated with owner information "p [0]". The tag information "0002" is associated with the owner information "p [1]". That is, articles of tag information "0001" and "0003" are classified into groups identified by owner information "p [0]". Articles of tag information "0002" are classified into groups identified by owner information "p [1]".

24 is a diagram showing an example of an internal configuration of the read result storage unit 19 according to the present embodiment.

Here, about the owner information "p [0]", "fishing" and "overseas travel" are memorize | stored as a hobby, and the number of readings is 2 times and 6 times, respectively. Similarly, about the owner information "p [1]", "ski" is memorize | stored as a hobby, and is read once. The preference determination unit 114 determines the hobby in comparison with the threshold stored in the threshold storage unit 15 for each classification.

Also in this embodiment, the effect similar to 1st Embodiment and 2nd Embodiment is acquired. In addition, according to the user preference estimation apparatus 100 according to the present embodiment, even when articles of a plurality of owners are arranged within a predetermined range in which the user preference estimation apparatus 100 reads tag information, the owner is considered. Thus, the preference can be estimated for each owner. In addition, if necessary, the user can inquire about necessary information or input an answer, and user preference can be estimated with good accuracy.

(6th Embodiment)

25 is a diagram illustrating the configuration of the user preference estimation apparatus 100 in the present embodiment.

In the present embodiment, after obtaining tag information attached to a plurality of articles, the user preference estimating apparatus 100 clusters those articles, and determines the preferences of the user based on articles belonging to the cluster for each cluster. The user preference estimating apparatus 100 may estimate the owner for each cluster after clustering a plurality of articles. As a result, errors can be less likely to occur, and the symbol can be estimated with better accuracy.

In the present embodiment, the user preference estimation apparatus 100 further includes an acquisition information storage unit 124 in addition to the configuration described with reference to FIG. 21 in the fifth embodiment. The acquisition information storage unit 124 stores information about an article for which the information acquisition unit 102 has acquired tag information. In the present embodiment, the acquisition information storage unit 124 stores the positional information of the article, for example. The classification processing unit 120 refers to the acquisition information storage unit 124 to classify a plurality of articles.

FIG. 26 is a flowchart showing a processing procedure of the user preference estimation apparatus 100 according to the present embodiment.

First, the information acquisition part 102 acquires tag information from the IC tag 1 (step S230). At this time, the classification processing part 120 acquires the positional information of each article. The classification processing unit 120 stores the positional information of each article in the acquisition information storage unit 124 in association with the tag information. Subsequently, the article information acquisition unit 108 obtains the corresponding article information from the article information database 150 based on the tag information acquired by the information acquisition unit 102 (step S232). Subsequently, the preference determination unit 114 reads the corresponding hobby from the preference information database 5 based on the article information acquired by the article information acquisition unit 108. Here, the article information acquisition unit 108 also notifies the preference determination unit 114 of the tag information acquired from the information acquisition unit 102 together with the article information acquired from the article information database 150. The preference determination unit 114 stores the hobby read out from the preference information database 5 in the read result storage unit 19 in association with the tag information (step S234).

Subsequently, the control unit (not shown) of the user preference estimation apparatus 100 determines whether or not the processing has been completed for all articles (step S236). If the process has not ended (NO in step S236), the process returns to step S230 to repeat the same process for the next article. On the other hand, when the process ends in step S236 (YES in step S236), the classification processing unit 120 refers to the acquisition information storage unit 124 and classifies, that is, clusters, the plurality of articles based on the positional information. (Step S238). Here, the classification processing unit 120 is similar to the owner determination means 9 described in the third embodiment, the clustering means 11 described in the fourth embodiment, and the classification processing unit 120 described in the fifth embodiment. A plurality of articles can be classified.

The preference determination unit 114 refers to the acquisition information storage unit 124, the read result storage unit 19, and the threshold value storage unit 15, and stores the angles stored in the read result storage unit 19 for each cluster. For the hobby, it is determined whether the number of readings is equal to or higher than the threshold stored in the threshold storage unit 15 (step S240). If there is a hobby greater than or equal to the threshold, the hobby greater than or equal to the threshold is determined to be a hobby in the cluster, and the determination result is output (step S242). On the other hand, if there is no hobby greater than or equal to the threshold in step S240, in step S242, the preference determination unit 114 determines that the reliability of the determination is low and can output that there is no corresponding hobby.

Moreover, also in this embodiment, the classification processing part 120 can inquire a user about the owner of an article through the presentation processing part 132, and can receive a reply from a user via the input reception part 130. As shown in FIG. Thereby, the classification processing part 120 can acquire the information about the owner of each cluster.

In the present embodiment, as described in the fifth embodiment, the classification processing unit 120 estimates the owner of the article every time the information acquisition unit 102 acquires the tag information, and the position information and the owner of the article. May be stored in the acquisition information storage unit 124 in association with the tag information. The classification processing unit 120 may classify articles according to the owners by referring to the acquisition information storage unit 124 after the processing is completed for all articles. At this time, the classification processing unit 120 can change the estimated owner in accordance with the positional information of the article. In this way, the owner can be estimated with better precision. The classification processing unit 120 stores the changed result in the classification result storage unit 122.

In another example, when the classification processing unit 120 cannot acquire the position information of the article, the classification processing unit 120 classifies the article according to the article information acquired by the article information acquisition unit 108. You may. For example, the article information acquisition unit 108 can acquire information regarding the age to be used for each article based on the tag information acquired by the information acquisition unit 102. In this case, the classification processing unit 120 can classify articles according to the age of use.

Also in this embodiment, the effect similar to 1st Embodiment and 2nd Embodiment is acquired. In addition, according to the user preference estimating apparatus 100 of the present embodiment, since preferences can be estimated for each cluster, errors can be less likely to occur, and preferences can be estimated with better accuracy.

(Seventh embodiment)

In the present embodiment, the user preference estimating apparatus is incorporated in a robot capable of autonomous running. The robot in this embodiment has a function of communicating with a user.

By using such a robot, the following effects can be further acquired in addition to the effect similar to the user preference estimation apparatus 100 in 1st-6th embodiment.

Since the robot moves autonomously, the robot can automatically read the IC tag 1 attached to the article in the room. In addition, since the robot has a function of communicating with the user, it is possible to receive an instruction from the user or acquire necessary information in the conversation with the user. Thereby, the estimation function of the user preference estimation apparatus 100 can be improved. Moreover, the robot can communicate with a user using the estimation result of the user preference estimation apparatus 100, and the robot and user can be smoothly communicated.

27 is an external configuration diagram showing an example of a robot according to the present embodiment.

The robot 200 is comprised by connecting the fuselage part 201 and the head 202, for example. Wheels 203A and wheels 203B are mounted on the lower left and right sides of the trunk portion 201, respectively, and these wheels can independently rotate back and forth.

The head 202 may rotate in a range determined about a vertical axis mounted vertically to the trunk portion 201 and a horizontal axis installed at an angle of 90 degrees with respect to the vertical axis. The vertical axis is provided so as to pass through the center of the head 202, and the horizontal axis is horizontally installed in the left and right direction while passing through the center of the head 202 with the fuselage 201 and the head 202 facing to the front. . That is, the head 202 can rotate within the determined range in two degrees of freedom, left and right and up and down.

The speaker 212 and the microphone 213 are provided on the surface of the body part 201. In addition, a CCD camera 221A and a CCD camera 221B are provided on the surface of the head 202.

In addition, the robot 200 is not limited to the structure which has the trunk | drum 201 and the head 202, as shown in FIG. 27, and can be made in any shape.

28 is a block diagram illustrating an example of an electrical configuration of the robot 200.

The body portion 201 includes a controller 210 for controlling the entire robot, a battery 211 serving as a power source of the robot, a speaker 212, a microphone 213, an actuator 214A for moving two wheels, and The actuator 214B and the like are housed.

The microphone 213 picks up surrounding voices and transmits the obtained voice signals to the controller 210. In addition, although only one microphone 213 is shown here, a plurality of microphones 213 may be installed in the robot 200. Thereby, the process which removes a noise etc. can be performed, for example.

The controller 210 incorporates a CPU 210A or a memory 210B, and executes various processes by executing a control program stored in the memory 210B in the CPU 210A.

The head 202 houses a CCD camera 221A, a CCD camera 221B, an actuator 222A, an actuator 222B, and the like for rotating the head 202.

The CCD camera 221A and the CCD camera 221B image the surrounding situation, and transmit the obtained image signal to the controller 210. The actuator 222A and the actuator 222B rotate the head 202 of the robot 200 up, down, left and right.

The controller 210 appropriately reads information from the memory 210B based on an audio signal or an image signal obtained from the microphone 213, the CCD camera 221A, and the CCD camera 221B, and reads the information from the surrounding situation or the user. By interpreting the command of the robot 200 to determine the behavior of the robot 200, or the like.

When the controller 210 determines the behavior of the robot, based on the determination result, the controller 210 controls the actuator 214A, the actuator 214B, the actuator 222A, and the actuator 222B to move the head 202 up, down, left, and right. Rotate, or move or rotate the robot 200.

When the controller 210 determines the utterance of the robot, the controller 210 generates a synthesized sound based on the determination result and outputs it from the speaker 212.

FIG. 29 is a block diagram showing in detail the controller 210 of the robot 200 shown in FIG. 28. Here, the robot 200 can be set as the structure which has the function similar to the user preference estimation apparatus 100 demonstrated in 1st-6th embodiment.

The controller 210 includes the information acquisition unit 102, the article information acquisition unit 108, the classification processing unit 120, the preference determination unit 114, the input reception unit 130, the presentation processing unit 132, and the control unit 250. , The mechanical control unit 254, the preference information database 5, the threshold storage unit 15, the read result storage unit 19, the classification result storage unit 122, and the robot motion database 252. Although not shown here, the robot 200 may be configured to have an acquisition information storage unit 124 as described with reference to FIG. 25 in the sixth embodiment.

The robot motion database 252 stores utterance data and motion data of the robot in a specific situation. The robot behavior database 252 stores, for example, the correspondence table of the speech of the corresponding answer and the speech data for each word input by the user. For example, when the input word from the user is "returned", the speech "come over" is stored as the response of the robot 200.

The input reception unit 130 recognizes the user's motion or ambient state based on the audio signal, the image signal, etc. transmitted from the microphone 213, the CCD camera 221A, the CCD camera 221B, and controls the recognition result. Notify 250. In addition, as described in the fifth and sixth embodiments, the input reception unit 130 can also be implemented by a GUI (Graphical User Interface) or a button.

The mechanical control unit 254 generates a control signal for driving the actuator 214A, the actuator 214B, the actuator 222A, and the actuator 222B based on the action command sent out from the control unit 250. It sends out to actuator 214A, 214B, 222A, and 222B. The actuators 214A, 214B, 222A, and 222B are driven in accordance with a control signal.

The controller 250 determines the behavior of the robot 200 based on the recognition result of the input receiver 130 and the utterance motion information stored in the robot behavior database 252. In the present embodiment, the presentation processing unit 132 includes a voice synthesis unit that generates synthesized sound, and controls the output of the synthesized sound synthesized by the voice synthesis unit.

Also in this embodiment, the user's preference is estimated by the preference determination unit 114 in the same order as described in the first to sixth embodiments. In the present embodiment, since the robot 200 has a function of communicating with the user, the robot 200 can inquire a user or obtain a reply from the user in the step of estimating a user's preference. Thereby, the user's preference can be estimated efficiently with good accuracy.

In addition, when the preference determination part 114 determines a user's preference, the robot 200 operates according to a determination result. For example, when the preference determination unit 114 determines that the user's hobby is "classic music listening", the control unit 250 refers to the robot activating database 252, and the user's hobby is "classic music listening". In this case, the motion of the associated robot 200 is read. For example, when the user's hobby is "listening to classical music," the robot music database (252), "the classic music makes you feel at ease. Can you hear me? ” In this case, the control unit 250 instructs the mechanical control unit 254, and the mechanical control unit 254, from the speaker 212, says, "Classic music is comfortable. Can you hear me? ” By realizing such a function in, for example, a home robot, the robot can ignite according to the hobby of the user, and the user can feel intimacy and cleverness by the robot.

According to the robot 200 in this embodiment, since the robot 200 moves autonomously, for example, indoors, IC tag information attached to the article arrange | positioned in arbitrary rooms can be read sequentially. As a result, the robot 200 can automatically acquire the article information of the room, and the user's preference can be determined.

In addition, since the robot 200 has a function of talking with the user, the robot 200 can receive an instruction from the user or obtain necessary information from the conversation with the user. Thereby, user convenience can be aimed at.

Moreover, the robot can communicate with a user using the estimation result of the user preference estimation apparatus 100, and the robot and user can be smoothly communicated.

(8th Embodiment)

30 is a block diagram showing an example of the configuration of the user profile estimating apparatus according to the present embodiment. The user profile estimating apparatus 300 estimates the user's profile based on the possession of the user.

The user profile estimating apparatus 300 includes an information acquisition unit 102, an article information acquisition unit 108, a classification processing unit 120, an input reception unit 130, a presentation processing unit 132, a profile estimation unit 314, and a classification. The result storage unit 122 and the history database 320 are included.

The profile estimating unit 314 stores the article information acquired by the article information acquisition unit 108 in the history database 320 in association with the date and time at which the article information was obtained. In addition, also in this embodiment, the classification processing part 120 can classify a some goods and store the classification result in the classification result storage part 122 similarly to what was demonstrated in 5th and 6th embodiment. The profile estimating unit 314 can estimate the profile for each group of articles classified by the classification processing unit 120 with reference to the classification result storage unit 122.

The profile estimator 314 may refer to the history database 320 when estimating the profile. For example, when an article is stored in the history database 320 in correspondence with the date and time at which the article information was acquired for a certain period of time, the user immediately selects a new article according to the change of the article information. It can be determined whether it is a type that is easy to purchase or discard, or that the same item is used for a long time. The profile estimating unit 314 can estimate the personality of the user, such as "easy to get tired" or "prefer things", according to the determination result.

In addition, by storing the product information in the history database 320 in correspondence with the date and time, it is possible to determine when any item is in the user's hands. For that purpose, a list of belongings of the user may be prepared according to the acquisition date and time. In this way, when the user wants to purchase a new product, it can be used to determine which product should be preferentially purchased.

In addition, in this embodiment, even when the goods of a plurality of people are arrange | positioned in the indoor of a subject, the property of each person can be distinguished and managed, For example, the belongings of a family can be grasped | ascertained, respectively. In this way, for example, in the case where someone in a family of plural purchases a coat of one person, it can be used for determining which coat should be preferentially purchased.

In addition, also in this embodiment, the user profile estimating apparatus 300 can be set as the structure containing the preference information database 5 similarly to the user preference estimation apparatus 100 demonstrated in 1st-7th embodiment. . In this case, the profile estimating unit 314 can determine the user's preferences, similar to the preference determination unit 114 of the user preference estimation apparatus 100. In addition, the user profile estimating apparatus 300 in this embodiment can also be incorporated in the robot 200.

As described above, according to the user profile estimating apparatus 300 according to the present embodiment, the user's profile such as personality of the user and belongings of the user can be estimated widely in addition to preferences and hobbies.

Each component of the user preference estimating apparatus, the user profile estimating apparatus, and the controller of the robot described in the above embodiments stores a program for realizing the components of the drawing loaded in the CPU, memory, and memory of an arbitrary computer, and the programs. It is realized by any combination of hardware and software centering on a storage unit such as a hard disk and an interface for network connection. And it is understood by those skilled in the art that various modifications can be made in the realization method and apparatus. Each figure described in the embodiment shows a block of functional units, not a configuration of hardware units.

As mentioned above, although embodiment of this invention was described with reference to drawings, these are illustrations of this invention, Various structures of that excepting the above are also employable.

For example, the symbol determination unit 114 estimates the user's preferences, and then presents the plurality of preference candidates to the user, thereby determining the user's preferences by selecting the actual preferences. In this manner, the user's preference can be accurately determined while reducing the burden on the user's preference. In addition, when the user's preference estimated by the user preference estimation apparatus 100 differs from the preference of the actual user, the difference can be fed back and reflected in the determination processing of the user preference from the next time.

For example, in the first embodiment, in step S108 shown in FIG. 5, it is determined whether the number of times of reading of any one hobby is equal to or greater than the threshold value, and the hobby is output according to the determination result (step S110). Outputting the absence of the corresponding hobby (step S112). However, in another example, a predetermined number of hobbies can be outputted and presented to the user, in the order of the number of readings, regardless of whether or not it is above the threshold. At this time, a message such as "Please select your hobby" is presented at the same time, and the user selects one hobby. If the hobby selected by the user is determined to be equal to or greater than the threshold in step S108, the estimation result and the actual hobby of the user coincide. In this case, the processing can be continued as it is. In addition, when the user preference estimating apparatus 100 is built in the robot 200, it is possible to have a conversation with the user according to the hobby selected by the user.

On the other hand, when the hobby selected by the user is judged not to be greater than or equal to the threshold in step S108, or even when the hobby determined to be greater than or equal to the threshold in step S108, when the user does not actually select the hobby, the estimation result and the actual user Hobbies do not match. In this case, it is also considered that the threshold set in the threshold storage 15 is not appropriate. It is also conceivable that the weighting coefficient is not appropriate when the weighting coefficient is set in the preference information database 5. The user preference estimating apparatus 100 may automatically reset the threshold of the threshold storage unit 15 and the weighting coefficient of the preference information database 5, and present to the user that the estimation accuracy is not good. It may also prompt the reset of.

In addition, even in the hobby judged to be more than the threshold value in step S108, when the user does not actually select it as a hobby, there is a possibility that the thing which is not owned by the user is arranged in the target room or the owner's estimation is wrong. In this case, for example, in the seventh embodiment, the robot 200 may make an inquiry such as "Are not the fishing rod yours?" To obtain an answer from the user. About this inquiry, user said, "It is not mine. When A answered, etc., remember that the owner is A, not the user, to the identification information indicating the fishing rod. In this way, the next time the user's hobby is estimated, the owner of the fishing rod can be judged correctly, and the user's hobby can be estimated with good accuracy.

In addition, when performing the process of estimating the hobby of a user, the user preference estimation apparatus 100 can store the result in correspondence with the date and time when the process was performed. In this way, for example, when the user's hobby is estimated next time, a change in the hobby of the user can be detected. As a result, when the user's hobby changes, for example, "I used to enjoy fishing, but now I enjoy listening to classical music," etc. can be uttered, and communication with a user can be expanded. In addition, a plurality of changes in the hobby of the user can be stored as a history. For example, if a change in the user's hobbies takes place in a short period of time, I can grasp that the user is easy to get tired of, and I'm already tired of fishing. I might get tired of listening to classical music soon, so it's a good idea to buy new CDs as well. ”

Moreover, also in 1st-4th embodiment, the user preference estimation apparatus 100 can be set as the structure which further contains the moving means which is a mobile unit, such as a wheel, a leg, and a crawler. The control unit (not shown) of the user preference estimation apparatus 100 controls the moving means. With such a configuration, a large number of tag information can be obtained. In addition, when classifying a plurality of articles, not only the position of the article can be known more broadly, but also the current position of the apparatus can be known by using a moving means, the article position can be estimated with better accuracy. Clustering is done with better precision.

In the above embodiment, after reading tag information from the IC tag 1 attached to the article arrange | positioned in the range which can be read by the receiving means 2 (or the information acquisition part 102), the symbol determination means 4 ( Or the form which performs the process which determines the user's preference by the preference determination part 114 was described. However, for example, the receiving means 2 (or the information acquisition unit 102) sequentially acquires the tag information from the IC tag 1 and, from the preference information database 5, for any hobby. When the number of readings is equal to or greater than the threshold stored in the threshold storage unit 15, the preference determination unit 114 may determine that the hobby is the hobby of the user and output the result.

In addition, although the case where the receiving means 2 or the information acquisition part 102 acquired the tag information wirelessly from the IC tag 1 was demonstrated as an example in the above embodiment, the user preference estimation apparatus 100 It is also possible to have a configuration in which tag information is directly read from the IC tag 1 instead of wirelessly. That is, in the first to fourth embodiments, the user preference estimation apparatus 100 can be configured to include an information acquisition unit that reads tag information directly from the IC tag 1 instead of the reception means 2. have. In the fifth to seventh embodiments, the information acquisition unit 102 of the user preference estimation apparatus 100 can be configured to read tag information directly from the IC tag 1.

In addition, this invention also includes the following forms.

(1) user state estimation, comprising: receiving means for receiving information about the article from a tag attached to the article, and determining means for estimating a user's hobby from the information of the article received by the receiving means Device.

(2) receiving means for receiving information about the article from a tag attached to the article, article name determining means for obtaining a product name of the article based on the information about the article received by the receiving means, and the article name And a determination means for determining a preference by searching a database storing an article name and preference information corresponding to the article name based on the article name obtained from the determination means.

(3) receiving means for receiving information about the article from a tag attached to the article, article name determination means for acquiring the article name of the article based on the information about the article received at the receiving means, and the article name An article category determination means for determining a category corresponding to the article name by searching a database storing an article name and a category corresponding to the article name based on the article name obtained from the determination means, and the category obtained from the article category determination means And determining means for determining a preference by searching a database storing a category and preference information corresponding to the category.

(4) In the user state estimating apparatus according to (2) or (3), the determination means determines the preference as having a predetermined threshold value, and the number of matches in the database exceeds the threshold value. A user state estimation device.

(5) The user state estimating apparatus according to (2), wherein the determining means determines that the hobby whose sum of the weights defined for each article exceeds the threshold is the hobby of the person.

(6) The user state estimating apparatus according to (3), wherein the determining means determines that the hobby whose sum of the weights defined for each category exceeds the threshold is the hobby of the person.

(7) The user state estimating apparatus according to any one of (1) to (6), wherein the determination means determines the hobby for each owner of each article.

(8) The user state estimating apparatus according to any one of (1) to (7), comprising owner estimating means for estimating who is the owner for each article, wherein the determination means determines the hobby for each owner of the article. A user state estimating apparatus characterized in that it is carried out.

(9) The user state estimating apparatus according to any one of (1) to (8), comprising clustering means for clustering information about the article received by the receiving means based on an arbitrary criterion, and determining The means for estimating a hobby for each cluster is a user state estimating apparatus characterized by the above-mentioned.

(10) The user state estimating apparatus according to (9), wherein the clustering means clusters the information about the article received by the receiving means according to its relative positional relationship, and determines the hobby for each cluster. A user state estimation device.

(11) The user state estimating apparatus according to any one of (1) to (10), wherein the user state is detected from a change in time of the user's hobby from a time change of information on the article received by the receiving means. Estimation device.

(12) The user state estimating apparatus according to any one of (1) to (11), further comprising moving means for moving by itself.

(13) a first step of receiving information about the article from a tag attached to the article, a second step of acquiring the article name of the article based on the information about the article received by the first step; And a third step of determining a preference by searching a database storing an article name and preference information corresponding to the article name on the basis of the article name obtained in the second step.

Claims (27)

delete delete delete delete delete delete A receiver for receiving information about the article from a tag attached to the article; A preference determination unit for estimating a preference of a user based on the information about the article received by the reception unit; A presentation processing unit for presenting information to the user, Input reception unit that accepts input of information from the user Including, The preference determination unit further includes an owner estimating unit for estimating the owner of the article for each article by presenting information for inquiring the owner to the presentation processing unit and receiving a response to the inquiring information through the input reception unit. And an owner preference estimating device for each owner estimated by the owner estimating unit. A receiver for receiving information about the article from a tag attached to the article; An article name acquisition unit for acquiring an article name of the article based on the information about the article received by the receiver; A user preference determination unit for estimating a user's preference based on the article name acquired by the article name acquisition unit; A clustering processing unit for clustering a plurality of the article received the information about the article on the basis of a predetermined criterion Including, And the preference determination unit estimates a preference of the owner of the article included in the cluster for each cluster divided by the clustering processing unit. The method of claim 8, A positional information acquisition unit for acquiring positional information of a plurality of the articles in which the reception unit reads information for identifying the article; More, And the clustering processing unit clusters the articles according to the relative positions between the articles based on the positional information of the articles acquired by the positional information obtaining unit. The method according to claim 7 or 8, And the information on the article is information for identifying the article. The method according to claim 7 or 8, A presentation processing unit for presenting information to the user, Input reception unit that accepts input of information from the user More, The preference determination unit judges a preference of the user by presenting a candidate of a determination result by the presentation processing unit and accepting selection of a correct determination result among candidates of the determination result by the input accepting unit. A user preference estimation device. The method according to claim 7 or 8, A user preference estimation apparatus further comprising a moving unit for autonomous movement. The user preference estimation device of claim 7 or 8, A moving unit which moves the user preference estimating apparatus; And a motion control unit for controlling the moving unit, wherein the robot moves autonomously. The user preference estimation device of claim 7 or 8, And an operation controller for operating the user based on the preference of the user estimated by the user preference estimating apparatus. delete An information acquisition unit for acquiring information about the article from a tag attached to the article; A profile estimating unit estimating a profile of the user based on the information about the article read by the information acquiring unit; For each article, including an owner estimating section for estimating the owner of the article, And the profile estimating unit estimates a profile of the owner for each owner estimated by the owner estimating unit. The method of claim 16, A location information acquisition section for acquiring the location information of the article from which the information acquisition section reads information about the article; And the owner estimating unit estimates the owner of the article according to the relative position between the articles, based on the positional information of the article obtained by the position information obtaining unit. The method according to claim 16 or 17, A history storage section for storing the time at which the information acquisition section acquires the information on the article in association with the information representing the article; And the profile estimating unit estimates the profile of the user based on the information indicating the article stored in the history storage unit and the time associated with it. The user profile estimating apparatus of claim 16 or 17, A moving unit which moves the user profile estimating apparatus; And a motion control unit for controlling the moving unit, wherein the robot moves autonomously. The user profile estimating apparatus of claim 16 or 17, An operation control unit that operates on the user based on the user profile estimated by the user profile estimating apparatus; Robot comprising a. delete delete Obtaining information about the article from a tag attached to the article; Estimating a profile of the user based on the information about the article obtained in the step of obtaining information about the article; Estimating the owner of the article for each article Including, And estimating the profile of the owner for each owner estimated in the estimating step. Computer, Receiving means for receiving information about the article from a tag attached to the article, Preference determination means for estimating a preference of a user based on the information about the article received by the reception means; Presentation processing means for presenting information to a user, Input receiving means for receiving input of information from a user Function as The preference determination means presents the information for inquiring the owner to the presentation processing means, and receives an answer to the information to be inquired through the input reception means, thereby estimating the owner of the article for each article. And a user preference estimation program recorded for each owner estimated by said owner estimation means. A computer included in a robot having a moving means, Receiving means for receiving information about the article from a tag attached to the article, Preference determination means for estimating a preference of a user based on the information about the article received by the reception means; Presentation processing means for presenting information to a user, Input receiving means for receiving an input of information from a user, Motion control means for controlling the movement means Function as The preference determination means presents the information for inquiring the owner to the presentation processing means, and receives an answer to the information to be inquired through the input reception means, thereby estimating the owner of the article for each article. Further comprising, for each owner estimated by the owner estimating means, estimating a preference of the owner, A computer-readable recording medium having recorded thereon a robot control program, wherein the robot autonomously moves. Computer, Receiving means for receiving information about the article from a tag attached to the article, Article name acquiring means for acquiring an article name of the article based on the information about the article received by the receiving means; User preference determination means for estimating a preference of a user based on the article name acquired by the article name acquisition means; Clustering processing means for clustering a plurality of the articles on which the receiving means has received the information about the article based on a predetermined criterion; Function as And the preference determining means estimates the preferences of the owners of the articles included in the cluster, for each cluster divided by the clustering processing means. A computer included in a robot having a moving means, Receiving means for receiving information about the article from a tag attached to the article, Article name acquiring means for acquiring an article name of the article based on the information about the article received by the receiving means; User preference determination means for estimating a preference of a user based on the article name acquired by the article name acquisition means; Clustering processing means for clustering the plurality of the articles on which the receiving means has received the information about the article based on a predetermined criterion; Operation control means for controlling the moving means, Function as The preference determining means estimates the preference of the owner of the article included in the cluster for each cluster divided by the clustering processing means. A computer-readable recording medium having recorded thereon a robot control program, wherein the robot autonomously moves.

Images