User grouping apparatus and user grouping method
View Patent ↗An RFID aggregate storage stores a radio-frequency ID aggregate structure. The RFID aggregate structure is an aggregated data structure of a plurality of RFIDs, in which each RFID is associated with a location and time read out from a RFID tag, and RFID related information, hierarchically constructed by classifying RFIDs based on the readout location and further by classifying them based on the readout time. A proximity evaluator evaluates the proximity between the first user's RF ID aggregate and the second user's RFID aggregate. A user classifier classifies the first user and the second user in one group when the proximity is smaller than a predetermined threshold value.
1. A user grouping apparatus comprising:
a storage that stores a wireless ID aggregate structure, which is an aggregated data structure of a plurality of wireless IDs set up for each wireless management apparatus used by a user, in which each wireless ID is associated with a location and time, where and when the corresponding wireless ID was read out from a wireless ID tag by a wireless ID management apparatus, and wireless ID related information which is information associated with the corresponding wireless ID, hierarchically constructed by classifying the plurality of wireless IDs based on the readout location and further by classifying them based on the readout time;
an evaluator, which extracts one or more feature quantities from wireless ID related information of a wireless ID aggregate of a user as a user's preferences, and which evaluates affinity between a first user's preferences as detected from a first wireless ID aggregate and a second user's preferences as detected from a second wireless ID aggregate; and
a classifier, which classifies the first user and the second user in one group when the affinity is stronger than a predetermined threshold value,
wherein the evaluator evaluates the affinity between a selected feature quantity of information related to a plurality of wireless IDs included in the first user's wireless ID aggregate that has been chosen by a user and a similar feature quantity of information related to a plurality of wireless IDs included in the second user's wireless ID aggregate.
2. The user grouping apparatus according to claim 1 , wherein the evaluator, in addition to evaluating the affinity of preferences from respective one or more feature quantities, further evaluates the proximity between the readout locations of a plurality of wireless IDs included in the first user's wireless ID aggregate and the readout locations of a plurality of wireless IDs included in the second user's wireless ID aggregate.
3. The user grouping apparatus according to claim 2 , wherein the evaluator further evaluates the proximity between the readout times of a plurality of wireless IDs included in the first user's wireless ID aggregate and the readout times of a plurality of wireless IDs included in the second user's wireless ID aggregate, and
wherein, if the first user and the second user are classified into one group based on affinity of preferences from respective one or more feature quantities and if the proximity of readout times between the first user and the second user is measured to be farther than a predetermined proximity threshold, the classifier cancels the classification of the first user and the second user into one group.
4. A user grouping method comprising the steps of:
reading out from a storage apparatus a wireless ID aggregate structure, which is an aggregated data structure of a plurality of wireless IDs set up for each wireless management apparatus used by a user, in which each wireless ID is associated with a location and time, where and when the corresponding wireless ID was read out from a wireless ID tag by a wireless ID management apparatus, and wireless ID related information which is information associated with the corresponding wireless ID, hierarchically constructed by classifying the plurality of wireless IDs based on the readout location and further by classifying them based on the readout time;
extracting one or more feature quantities from wireless ID related information of a wireless ID aggregate of a user as a user's preferences;
evaluating affinity between a first user's preferences as detected from a first wireless ID aggregate and a second user's preferences as detected from a second wireless ID aggregate; and
classifying the first user and the second user in one group when the affinity is stronger than a predetermined threshold value,
wherein the step of evaluating includes evaluating the affinity between a selected feature quantity of information related to a plurality of wireless IDs included in the first user's wireless ID aggregate that has been chosen by a user and a similar feature quantity of information related to a plurality of wireless IDs included in the second user's wireless ID aggregate.
5. A user grouping apparatus comprising:
a storage that stores a wireless ID aggregate structure, which is an aggregated data structure of a plurality of wireless IDs set up for each wireless management apparatus used by a user, in which each wireless ID is associated with a location and time, where and when the corresponding wireless ID was read out from a wireless ID tag by a wireless ID management apparatus, and wireless ID related information which is information associated with the corresponding wireless ID, hierarchically constructed by classifying the plurality of wireless IDs based on the readout location and further by classifying them based on the readout time;
an evaluator, which extracts one or more feature quantities from wireless ID related information of a wireless ID aggregate of a user as a user's preferences, and which evaluates affinity between a first user's preferences as detected from a first wireless ID aggregate and a second user's preferences as detected from a second wireless ID aggregate; and
a classifier, which classifies the first user and the second user in one group when the affinity is stronger than a predetermined threshold value,
wherein:
the evaluator, in addition to evaluating the affinity of preferences from respective one or more feature quantities, further evaluates the proximity between the readout locations of a plurality of wireless IDs included in the first user's wireless ID aggregate and the readout locations of a plurality of wireless IDs included in the second user's wireless ID aggregate,
the evaluator further evaluates the proximity between the readout times of a plurality of wireless IDs included in the first user's wireless ID aggregate and the readout times of a plurality of wireless IDs included in the second user's wireless ID aggregate, and
when the first user and the second user are classified into one group based on affinity of preferences from respective one or more feature quantities and if the proximity of readout times between the first user and the second user is measured to be farther than a predetermined proximity threshold, the classifier cancels the classification of the first user and the second user into one group.
6. A user grouping method comprising the steps of:
reading out from a storage apparatus a wireless ID aggregate structure, which is an aggregated data structure of a plurality of wireless IDs set up for each wireless management apparatus used by a user, in which each wireless ID is associated with a location and time, where and when the corresponding wireless ID was read out from a wireless ID tag by a wireless ID management apparatus, and wireless ID related information which is information associated with the corresponding wireless ID, hierarchically constructed by classifying the plurality of wireless IDs based on the readout location and further by classifying them based on the readout time;
extracting one or more feature quantities from wireless ID related information of a wireless ID aggregate of a user as a user's preferences;
evaluating affinity between a first user's preferences as detected from a first wireless ID aggregate and a second user's preferences as detected from a second wireless ID aggregate; and
classifying the first user and the second user in one group when the affinity is stronger than a predetermined threshold value, wherein:
the step of evaluating, in addition to evaluating the affinity of preferences from respective one or more feature quantities, further comprises evaluating the proximity between the readout locations of a plurality of wireless IDs included in the first user's wireless ID aggregate and the readout locations of a plurality of wireless IDs included in the second user's wireless ID aggregate,
the step of evaluating includes evaluating the proximity between the readout times of a plurality of wireless IDs included in the first user's wireless ID aggregate and the readout times of a plurality of wireless IDs included in the second user's wireless ID aggregate, and
when the first user and the second user are classified into one group based on affinity of preferences from respective one or more feature quantities and if the proximity of readout times between the first user and the second user is measured to be farther than a predetermined proximity threshold, the step of classifying further includes cancelling the classification of the first user and the second user into one group.