Description of the Essex Entry on Data Set V with Precomputed Features John Q. Gan and Louis C.S. Tsui Department of Computer Science University of Essex Colchester CO4 3SQ UK Email: jqgan@essex.ac.uk,csltsu@essex.ac.uk We treat the classification of the precomputed features as a high-dimensional noisy data problem. The focus in developing our methods is on how to combat the overfitting problem. The methods used in preparing this entry are briefly described as follows: 1. Preprocessing The main purpose of preprocessing is to reduce the dimension of the feature space, leading to better generalisation. Techniques used include mutual information as relevance measure, two types of PCA methods, ICA with supervision, and cross-validation for choosing optimal feature subset. 2. Classification: LDA (one against the rest) and neural networks have been investigated for classification. Decision fusion has also been considered with LDA as the dominant classifier. 3. Postprocessing: The purpose of postprocessing is to obtain reliable/robust classification. Techniques used include smoothing window on previous classification outputs, mental task change detection and confirmation. The classification results on the testing sessions of 3 subjects are included in the attached file EssexEntryDataSetV_PF.mat. In the data set description, there are two requirements for the classification output: the first is to provide an output for every input vector, and the second is to provide an output every 0.5 seconds. We are not sure which would be used for final evaluation of all the entries. Therefore, in EssexEntryDataSetV_PF.mat there are two vectors for each subject, named as subjectiTest and subjectiTest8 respectively. The number of estimated class labels in subjectiTest equals the number of input vectors in the testing session of subject i (i=1,2,3), whilst the number of estimated class labels in subjectiTest8 is 1/8 of that of subjectiTest.