No abstract

No abstract

P300 is an event-related potential that is elicited by an oddball paradigm. In several neuropsychiatric diseases, differences in latencies and amplitude compared to healthy subjects have been reported. Because of its clinical significance, several investigations have tried to elucidate the intracranial origins of the P300 component. In the present study we could demonstrate a network of P300 generators. Investigated were 15 healthy subjects with an acoustical oddball paradigm within a fMRI block design, which enabled us to exclude attention or acoustical processing effects. The inferior and middle frontal, superior temporal, lower parietal cortex, the insula and the anterior cingulum were significantly activated symmetrical in both hemispheres.

As databases for real-world problems increase in size, there is a need in many situations to select and keep relevant training data for efficient storage and processing reasons. Support vector machines (SVMs) reportedly exhibit certain desirable properties in selecting and preserving useful training data as support vectors.This paper attempts to quantify the extent to which SVM training behaves like a model independent example selection procedure.Using several common machine-learning training databases, we compare the prediction results obtained by different classifiers, trained with data selected by SVMs and by two other example selection methods (IB2 and random sampling). Some interesting observations are made with explanations.

No abstract