Identifying Incident Causal Factors to Improve Aviation Transportation Safety: Proposing a Deep Learning Approach

Table 4

Important statistics about the utilized ASRS data set.

Multilabel statistics	Value
Number of utilized factors	6
Number of valid samples	172,990
Factor cardinality	1.47
Factor density	0.245
Number of distinct label sets	28
Most frequent label set	{Human factor, aircraft}

After cleaning and preprocessing, we use the six most frequent labels from 172,990 reports. On average, every report has 1.47 labels (label density of 0.245).