Locality Preserving Discriminating Projections for cancer classification

Cancer classification of gene expression data helps determine appropriate treatment and prognosis. Its accurate prediction to the type or size of tumors relies on adopting powerful and efficient classification models such that patients can be provided with better treatment. As a graph-based method for linear dimensionality reduction, Locality Preserving Projections(LPP) searches for an embedding space in which the similarity among the local neighborhoods is preserved. However LPP doesn´t take the label information into consideration which is crucial for classification tasks. In order to gain better classification, in this study, a feature dimensionality reduction method termed the Locality Preserving Discriminating Projections(LPDP) is proposed. LPDP allows both locality and class label information to be incorporated which improves the performance of classification. Experimental results using public gene expression data show the superior performance of the method.