Destination prediction apparatus and destination prediction method
View Patent ↗To provide a destination prediction apparatus which can predict a destination of a user by using a carried item carried by the user, the destination prediction apparatus according to the present invention includes: a position information detection unit which detects a present position of a mobile terminal; a carried item detection unit which detects the carried item of the user; a history accumulation unit which accumulates a history of date and time information detected by a clock unit, position information detected by the position information detection unit, and information of the carried item of the user detected by the carried item detection unit; a carry frequency judgment unit which calculates a frequency at which the carried item is carried; a causal relationship extraction unit which calculates an infrequently carried item and a related point; a destination prediction unit which predicts the destination using the position information, the carried item information, and a position causal relationship extracted by the causal relationship extraction unit; and the clock unit which detects the date and time information.
1. A destination prediction apparatus which predicts a destination of a user, said apparatus comprising:
a carried item detection unit operable to detect a carried item which the user carries;
an accumulation unit operable to associate at least the carried item detected by said carried item detection unit with position information, and accumulate the carried item and the position information as a history;
a causal relationship extraction unit operable to extract, from the history, a causal relationship between at least the carried item and the position information;
a destination prediction unit operable to predict the destination of the user based on the causal relationship and the carried item; and
a carry frequency judgment unit operable to perform judgment that distinguishes whether the carried item is a frequently carried item which is carried with a frequency greater than or equal to a predetermined percentage, or an infrequently carried item which is carried with a frequency lower than the predetermined percentage,
wherein said causal relationship extraction unit is operable to extract a causal relationship between the infrequently carried item and the frequently carried item.
2. The destination prediction apparatus according to claim 1 ,
wherein said carry frequency judgment unit is operable to judge the carried item with a percentage lower than the predetermined percentage to be the infrequently carried item, and the carried item with a percentage greater than or equal to the predetermined percentage to be the frequently carried item, using, as the predetermined percentage, a percentage at which the user carries the carried item in each of a plurality of predetermined periods.
3. The destination prediction apparatus according to claim 1 ,
wherein said frequency carry judgment unit is operable to judge the carried item with a percentage lower than the predetermined percentage to be the infrequently carried item, and the carried item with a percentage greater than or equal to the predetermined percentage to be the frequently carried item, using, as the predetermined percentage, a percentage at which the user carries the carried item in each of a plurality of areas obtained by dividing a travel route of the user on a map.
4. The destination prediction apparatus according to claim 1 ,
wherein said frequency carry judgment unit is operable to judge the carried item with a percentage lower than the predetermined percentage to be the infrequently carried item, and the carried item with a percentage greater than or equal to the predetermined percentage to be the frequently carried item, using, as the predetermined percentage, a percentage at which the carried item is present in each of a plurality of classes that are obtained by dividing a travel route into the classes.
5. The destination prediction apparatus according to claim 1 , further comprising
a target area prediction unit operable to predict a target area which is the destination of the user,
wherein said causal relationship extraction unit is operable to extract a causal relationship between the infrequently carried item and the target area.
6. The destination prediction apparatus according to claim 5 ,
wherein said carry frequency judgment unit is further operable to judge the carried item with a percentage lower than the predetermined percentage to be the infrequently carried item, and the carried item with a percentage greater than or equal to the predetermined percentage to be the frequently carried item, using, as the predetermined percentage, a percentage at which the carried item is carried when a movement from a departure area to the target area predicted by said target area prediction unit is calculated as one movement.
7. A destination prediction apparatus which predicts a destination of a user, said apparatus comprising:
a carried item detection unit operable to detect a carried item which the user carries;
an accumulation unit operable to associate at least the carried item detected by said carried item detection unit with position information, and accumulate the carried item and the position information as a history;
a causal relationship extraction unit operable to extract, from the history, a causal relationship between at least the carried item and the position information;
a destination prediction unit operable to predict the destination of the user based on the causal relationship and the carried item;
a non-carried item isolation unit operable to isolate a non-carried item which is not carried by the user based on the casual relationship and the carried item detected by said carried item detection unit;
a non-carried item position detection unit operable to detect a position in which the non-carried item is present; and
a non-carried item position relationship judgment unit operable to judge a position relationship between the non-carried item and the user based on the position of the non-carried item detected by said non-carried item position detection unit, a position of the user and a movement direction of the user,
wherein said destination prediction unit is operable to predict, as the destination of the user, a destination corresponding to the non-carried item, in the case where said non-carried item position relationship judgment unit judges that the user is approaching the non-carried item.
8. The destination prediction apparatus according to claim 7 ,
wherein said destination prediction unit is further operable to predict the destination of the user excluding the destination corresponding to the non-carried item.
9. A destination prediction method for predicting a destination of a user, said method comprising:
detecting a carried item which the user carries;
associating at least the carried item detected by said carried item detection unit with position information, and accumulating the carried item and the position information as a history;
judging whether the carried item is a frequently carried item which is carried with a frequency greater than or equal to a predetermined percentage, or an infrequently carried item which is carried with a frequency lower than the predetermined percentage;
extracting, from the history, a causal relationship between at least the carried item and the predetermined position information; and
predicting the destination of the user based on the causal relationship and the carried item,
wherein said extracting comprises extracting a causal relationship between the infrequently carried item and the frequently carried item.
10. A program used in a destination prediction method for predicting a destination of a user, said program being embodied on a computer-readable medium and causing a computer to execute a method comprising the steps of:
detecting a carried item which the user carries;
associating at least the carried item detected by said carried item detection unit with position information, and accumulating the carried item and the position information as a history;
extracting, from the history, a causal relationship between at least the carried item and the position information;
judging whether the carried item is a frequently carried item which is carried with a frequency greater than or equal to a predetermined percentage, or an infrequently carried item which is carried with a frequency lower than the predetermined percentage; and
predicting the destination of the user based on the causal relationship and the carried item,
wherein said extracting comprises extracting a causal relationship between the infrequently carried item and the frequently carried item.
11. A destination prediction method for predicting a destination of a user, said method comprising:
detecting a carried item which the user carries;
associating at least the carried item detected by said carried item detection unit with position information, and accumulating the carried item and the position information as a history;
extracting, from the history, a causal relationship between at least the carried item and the position information;
predicting the destination of the user based on the causal relationship and the carried item;
isolating a non-carried item which is not carried by the user based on the casual relationship and the carried item;
detecting a position in which the non-carried item is present; and
judging a position relationship between the non-carried item and the user based on the detected position of the non-carried item, a position of the user and a movement direction of the user,
wherein said predicting comprises predicting, as the destination of the user, a destination corresponding to the non-carried item, in the case where said judging judges that the user is approaching the non-carried item.
12. A program used in a destination prediction method for predicting a destination of a user, said program being embodied on a computer-readable medium and causing a computer to execute a method comprising the steps of:
detecting a carried item which the user carries;
associating at least the carried item detected by said carried item detection unit with position information, and accumulating the carried item and the position information as a history;
extracting, from the history, a causal relationship between at least the carried item and the position information;
predicting the destination of the user based on the causal relationship and the carried item;
isolating a non-carried item which is not carried by the user based on the casual relationship and the carried item;
detecting a position in which the non-carried item is present; and
judging a position relationship between the non-carried item and the user based on the detected position of the non-carried item, a position of the user and a movement direction of the user,
wherein said predicting comprises predicting, as the destination of the user, a destination corresponding to the non-carried item, in the case where said judging judges that the user is approaching the non-carried item.