Prototypebased classifiers and their applications in the lifesciences
Prototype-based classifiers and their applications in the life-sciences
requirement:
moves prototypes towards / away from sample with prefactors
Classification of adrenal tumors
log-transformed steroid excretion in ACA/ACC rescaled using healthy control group values
projection on 2 nd eigenvector projection on first eigenvector of Λ
matrix square root is not unique
given transformation: is possible if the rows of are in the null-space of → identical mapping of examples, different for is singular if features are correlated, dependent possible to extend by prototypes
training process yields determine with eigenvectors and eigenvalues
regularized mapping after/during training retains original features flexible K may include prototypes pre-processing of data (PCA-like) mapped feature space fixed K prototypes yet unknown
medium low alcohol content high
over-fitting effect
regularization - enhances generalization - smoothens relevance profile/matrix - removes ‘false relevances’ - improves interpretability of Λ
Early diagnosis of Rheumatoid Arthritis
synovium tissue section m. RNA extraction real-time PCR
true positive rate initialization of relevances as prior knowledge false positive rate
PF 4 (platelet factor 4) PPBP (pro-platelet basic protein) = CXCL 4 chemokine (C-X-C motif) ligand 4 = CXCL 7 chemokine (C-X-C motif) ligand 7 cytokines associated with platelets (historically), also produced by other cell types
true positive rate initialization of relevances false positive rate
- Slides: 46