Efficient and Flexible Data Preprocessing Tools

Efficiently and flexibly preprocess data using a set of data filtering, deletion, and interpolation tools. These data preprocessing methods are developed based on the principles of completeness, accuracy, threshold method, and linear interpolation and through the setting of constraint conditions, time completion & recovery, and fast & efficient calculation and grouping. Key preprocessing steps include deletions of variables and observations, outlier removal, and missing values (NA) interpolation, which are dependent on the incomplete and dispersed degrees of raw data. They clean data more accurately, keep more samples, and add no outliers after interpolation, compared with ordinary methods. Auto-identification of consecutive NA via run-length based grouping is used in observation deletion, outlier removal, and NA interpolation; thus, new outliers are not generated in interpolation. Conditional extremum is proposed to realize point-by-point weighed outlier removal that saves non-outliers from being removed. Plus, time series interpolation with values to refer to within short periods further ensures reliable interpolation. These methods are based on and improved from the reference: Liang, C.-S., Wu, H., Li, H.-Y., Zhang, Q., Li, Z. & He, K.-B. (2020) .


News

Reference manual

It appears you don't have a PDF plugin for this browser. You can click here to download the reference manual.

install.packages("dataprep")

0.1.4 by Chun-Sheng Liang, a month ago


Browse source code at https://github.com/cran/dataprep


Authors: Chun-Sheng Liang <[email protected]> , Hao Wu , Hai-Yan Li , Qiang Zhang , Zhanqing Li , Ke-Bin He , Lanzhou University , Tsinghua University


Documentation:   PDF Manual  


GPL (>= 2) license


Imports ggplot2, scales, foreach, doParallel, dplyr, reshape2, data.table, zoo

Suggests knitr, rmarkdown


See at CRAN