Skip to main content

Table 1 Illustrated workflow steps to process a simple collection of abstracts for machine learning classification

From: Text mining to support abstract screening for knowledge syntheses: a semi-automated workflow

  1. 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
Back to article page

Systematic Reviews

ISSN: 2046-4053