Glass-Classification (Q6036843)

Context\NThis is a Glass Identification Data Set from UCI. It contains 10 attributes including id. The response is glass type(discrete 7 values)\NContent\NAttribute Information:\N\NId number: 1 to 214 (removed from CSV file)\NRI: refractive index \NNa: Sodium (unit measurement: weight percent in corresponding oxide, as are attributes 4-10) \NMg: Magnesium \NAl: Aluminum \NSi: Silicon \NK: Potassium \NCa: Calcium \NBa: Barium \NFe: Iron \NType of glass: (class attribute) \N-- 1 buildingwindowsfloatprocessed \N-- 2 buildingwindowsnonfloatprocessed \N-- 3 vehiclewindowsfloatprocessed \N-- 4 vehiclewindowsnonfloatprocessed (none in this database) \N-- 5 containers \N-- 6 tableware \N-- 7 headlamps\N\NAcknowledgements\Nhttps://archive.ics.uci.edu/ml/datasets/Glass+Identification\NSource:\NCreator: \NB. German \NCentral Research Establishment \NHome Office Forensic Science Service \NAldermaston, Reading, Berkshire RG7 4PN \NDonor: \NVina Spiehler, Ph.D., DABFT \NDiagnostic Products Corporation \N(213) 776-0180 (ext 3014)\NInspiration\NData exploration of this dataset reveals two important characteristics :\N1) The variables are highly corelated with each other including the response variables:\NSo which kind of ML algorithm is most suitable for this dataset Random Forest , KNN or other? Also since dataset is too small is there any chance of applying PCA or it should be completely avoided?\N2) Highly Skewed Data:\NIs scaling sufficient or are there any other techniques which should be applied to normalize data? Like BOX-COX Power transformation?