DataPreparator

DataPreparator is a free software tool which is designed to assist with common tasks of data preparation (or data preprocessing) in data analysis and data mining.

DataPreparator offers features such as character removal, text replacement, date conversion, remove selected attributes, move selected attributes, equal width, equal frequency, equal frequency from grouped data, delete records containing missing values, remove attributes containing missing values, impute missing values, predict missing values from model (dependence tree, Naive Bayes model), include missing value patterns, Z-score metho.

Box-plot method, create binary attributes, replace nominal values by indices, reduce number of labels, decimal, linear, hyperbolic tangent, soft-max, Z-score, other transformations (log(x), 1/x, x2, x3), manual selection, mutual information selection, robust mutual information selection.

Sampling (random, every k-th item, first-k), select records by key, create data sets, create missing values, append, balance, change names, merge, sort, smooth, statistics, table, file, database, visualize, Bar chart, cumulative frequency chart, Box plot (single, conditional), Histogram (single, conditional, normalized, overlaid, histogram matrix), Lag plot, Linear regression plot, Normal-quantile plot, Quantile plot, Quantile-quantile plot, Run sequence plot, Scatter plot, Pareto chart, Stacked chart, Dependency tree, Parallel coordinates, Create data sets from raw data, Create samples from raw data, Shuffle raw data and Configure database drivers.

DataPreparator can be used for data cleaning facilities and include character removal, text replacement, and date conversion. DataPreparator is written in Java and requires Java Runtime Environment (JRE) to be installed on user’s machine and also needs JRE 1.6 or later. DataPreparator runs on Windows, Mac OS/X, and Linux operating systems.