dprep-package: Data Preprocessing for supervised classification

Description Details Author(s) References

Description

Functions for normalization, treatment of missing values,discretization, outlier detection, feature selection, and visualization

Details

Dprep has been developed by Professor Edgar Acuna and his students at the CASTLE research Group of the University of Puerto Rico-Mayaguez. This is a library of R functions for normalization, handling of missing values, discretization, outlier detection, feature selection, and data visualization classification. Most of the methods handle datasets with numerical and categorical attributes.

Normalization methods: Score Normalization, Min-Max Normalization, Decimal scale, Sigmoidal Normalization, Softmax normalization.

Missing values methods: Imputation by mean, median and mode (categorical features), K-nn Imputation (categorical and numerical data).

Discretization Methods: Equal width bins, Equal Frequency bins, Holte's One R, chiMerge, Entropy Discretization with MDL stopping rule.

Feature Selection Methods: ReliefF, LVF, Finco, Sequential Forward Selection, Sequential Floating Forward Selection.

Outlier Detection Methods: Mahaout, Robout, Bay's algorthm, LOF.

Crossvalidation estimation error: LDA, Naive Bayes, Logistic, Knn, Rpart, Neural Networks.

Author(s)

Maintainer: Edgar Acuna <[email protected]>

References

See website academic.uprm.edu/eacuna/dprep.html


dprep documentation built on May 29, 2017, 11:01 a.m.