{"responseHeader":{"status":0,"QTime":7,"params":{"q":"{!q.op=AND}id:\"102814\"","hl":"true","hl.simple.post":"","hl.fragsize":"5000","fq":"!embargo_tdt:[NOW TO *]","hl.fl":"ocr_t","hl.method":"unified","wt":"json","hl.simple.pre":""}},"response":{"numFound":1,"start":0,"docs":[{"file_name_t":"Soller-Automated_Detection.pdf","thumb_s":"/e8/f3/e8f3f77e7360aeb65751b0f6de18962b49584cdc.jpg","oldid_t":"compsci 10927","setname_s":"ir_computersa","restricted_i":0,"format_t":"application/pdf","modified_tdt":"2016-05-26T00:00:00Z","file_s":"/a8/3c/a83cf5ba830f695a14be8e6fb57d9655d9b2073e.pdf","title_t":"Page 57","ocr_t":"42 As a result of the standardization, the covariance matrix of this transformed data set equals its correlation matrix. Principle component analysis, as described earlier in this chapter, extracted the principle component factors. The sorted factor loadings and eigenvalue distributions after varimax rotation are listed in Appendices G and H. Factor analysis extracted two factors from DRS scale elements using an eigenvalue cutoff of 1. 7. The selection of this cutoff resulted from qualitative judgement. The factor loadings were not simple. In the MMSE scale, four factors had eigenvalues greater than 1. 7. However, the loadings on the fourth factor did not add anything to the conceptual model. Therefore, the subsequent models extracted and utilized three MMSE factors. For the analysis by linear discriminant classifiers and logistic regression, the scale observations are weighted by the factor weights to produce two DRS factors and three MMSE factors. The DRS factors are independent of each other, but not of the MMSE factors . Each set of factors had an associated clinical observation of delirium. 8.6 Linear Discriminant Classifiers 8.6.1 Known Parameter Formulation Discriminant functions [31] [94] are used to generate a statistically optimal decision given evidence, the clinical observations and laboratory results analyzed in this dissertation. Given observations partitioned into multivariate normal groups, linear discriminant classifiers find the linear combination of the observations that produces the greatest difference between the two groups relative to the variance between the two groups. Conceptually, the mean determines the center of each group's statistical distribution, and the covariance matrix determines the shape of the distribution. The loci of points with constant probability density form hyperellipsoids (31], where the quadratic form, (8.13) is constant. The principal axis of each hyperellipsoid is given by the eigenvectors","id":102814,"created_tdt":"2016-05-26T00:00:00Z","parent_i":102961,"_version_":1642982670125236224}]},"highlighting":{"102814":{"ocr_t":[]}}}