Classifying an item to one of a plurality of groups
View Patent ↗A method of classifying an item to one of a plurality of groups includes providing a plurality of predictors associated with the item. For each predictor, the item is assigned to one of the groups. An assignment number is determined for each group. The item is classified to one of the groups based on the assignment number for each group.
1. A method of classifying an item to one of a plurality of groups, comprising: providing a plurality of predictors associated with the item; for each predictor, assigning the item to one of the groups based on predictor distributions for known classified items; determining an assignment number for each group, wherein the assignment number for each group represents a number of times that the item was assigned to that group based on the plurality of predictors; and classifying the item to one of the groups based on the assignment number for each group.
2. The method of claim 1 , wherein classifying the item to one of the groups based on the assignment number for each group comprises:
classifying the item to the group with a largest assignment number.
3. The method of claim 1 , and further comprising: providing a plurality of known classified items, wherein each of the known classified items is known to be classified to one of the groups; providing a plurality of predictors associated with each known classified item; and determining the predictor distributions for the known classified items based on values of the predictors.
4. The method of claim 3 , wherein the assigning of the item to one of the groups for each predictor is performed using a one-sample t-test.
5. The method of claim 3 , wherein the plurality of predictor distributions include N distributions for each of the predictors, wherein N represents a number of the groups in the plurality of groups.
6. The method of claim 3 , wherein a number of the predictors associated with each known classified item is substantially larger than a number of the known classified items.
7. The method of claim 6 , wherein the number of predictors exceeds the number of known classified items by at least a factor of 100.
8. The method of claim 1 , wherein the predictors represent biomagnetism data, and wherein the item is a human subject.
9. The method of claim 8 , wherein the biomagnetism data is generated with a magnetoencephalography (MEG) device.
10. The method of claim 1 , wherein the assigning, determining, and classifying are performed by at least one processor.
11. A system for automatically classifying an item to one of a plurality of groups, comprising:
a measurement device configured to generate a plurality of predictors associated with the item; and
a controller configured to perform, for each predictor, an assignment of the item to one of the groups based on predictor distributions for known classified items, and classify the item to one of the groups based on a number of the assignments for each group.
12. The system of claim 11 , wherein the controller is configured to classify the item to the group with a largest number of the assignments.
13. The system of claim 11 , wherein the measurement device is configured to generate a set of predictors for each of the known classified items, and wherein the predictor distributions for the known classified items are determined based on values of the predictors for the known classified items, and wherein the predictor distributions are weighted.
14. The system of claim 13 , wherein the controller is configured to perform, for each predictor, the assignment of the item to one of the groups based on the predictor distributions for the known classified items using a one-sample t-test.
15. The system of claim 14 , wherein a number of the predictors for each known classified item exceeds a number of the known classified items by at least a factor of 100.
16. The system of claim 11 , wherein the predictor distributions include N distributions for each of the predictors, wherein N represents a number of the groups in the plurality of groups.
17. The system of claim 11 , wherein the predictors represent biomagnetism data, and wherein the item is a human subject.
18. The system of claim 17 , wherein the measurement device comprises a magnetoencephalography (MEG) device.
19. A system for automatically classifying an item to one of a plurality of groups, comprising: a measurement device configured to generate a plurality of predictors associated with the item; and a controller configured to perform, for each predictor, an assignment of the item to one of the groups based on predictor distributions for known classified items, determine an assignment number for each group, and classify the item to one of the groups based on the assignment number for each group, wherein the assignment number for each group represents a number of times that the item was assigned to that group based on the plurality of predictors.