Research Article

Augmenting Multi-Instance Multilabel Learning with Sparse Bayesian Models for Skin Biopsy Image Analysis

Table 4

Annotation result evaluated by accuracy.

Term

78.2% 76.1% 70.6% 75.9% 68.3%
80.3% 75.9% 76.1% 74.5% 73.8%
77.7% 79.5% 77.8% 76.2% 68.5%
81.3% 81.2% 80.5% 82.4% 81.2%
69.3% 66.5% 67.9% 70.1% 67.4%
76.3% 75.0% 71.7% 74.2% 72.3%
77.8% 77.4% 76.5% 75.8% 75.9%
85.1% 85.2% 84.6% 83.8% 80.9%
87.3% 86.8% 81.4% 83.0% 78.2%
75.9% 75.4% 74.5% 73.8% 72.0%
69.9% 71.5% 68.9% 70.7% 69.6%
78.0% 76.1% 73.2% 75.8% 73.2%
79.2% 80.1% 77.2% 78.8% 72.5%
80.6% 81.2 77.2% 81.9% 73.5%
87.9% 86.4% 82.6% 83.1% 80.2%