Skip to main content
. 2021 May 26;10:156. doi: 10.1186/s13643-021-01700-x

Table 1.

Illustrated workflow steps to process a simple collection of abstracts for machine learning classification

graphic file with name 13643_2021_1700_Tab1_HTML.jpg

DFM  document-feature matrix. IDDM insulin-dependent diabetes mellitus. *The features in this example are words. **The DFM contains frequency counts of the features. Highlighted columns in the DFM denote relatively high-frequency features that are selected and retained for further analysis