Skip to main content
. Author manuscript; available in PMC: 2018 Jan 1.
Published in final edited form as: Pac Symp Biocomput. 2018;23:204–215.

Table 3.

Average of user responses to 4 most important usability questions for various user groups

Question ALL users (13) FEATURE and RF NON-experts (4) FEATURE experts and RF NON-experts (3) FEATURE NON-experts and RF experts (2) FEATURE and RF experts (4)
Estimate your increase in confidence of RF classification of FEATURE data after using RFEX summaries 2.7 (SD 2.2) 2.5 2 0.5 4.5
Estimate your increase in understanding why and how RF works on FEATURE data (e.g. RF Explainability) using RFEX one page summaries 3.3 (SD 1.7) 3.25 3.7 1 4.25
I believe RFEX approach will be useful for other applications of RF 4.4 (SD 0.5) 4.5 4.3 4 4.5
I believe RFEX approach (or its modifications) will be useful for other machine learning methods 4.0 (SD 0.8) 4.5 3.3 3.5 4.25