Table 4 Setting for baseline datasets
From: G-bic: generating synthetic benchmarks for biclustering
Properties | Dataset R | Dataset S | DatasetB | Dataset I | Dataset C | Dataset H | |
|---|---|---|---|---|---|---|---|
dataset | Data type | Real-valued | Symbolic | Integer | Integer | Real-valued | Heterogeneous (50% real, 50% symbolic) |
Size | \(1000 \times 100\) | \(1000 \times 100\) | \(1000 \times 100\) | \(1000 \times 100\) | \(1000 \times 100\) | \(1000 \times 100\) | |
Alphabet | [-100,100] | {a,b,c,d,e} | {1,0} | [0,100] | [-100,100] | [-100,100] and {a,b,c,d,e} | |
Background | N(0,30) | Missing | Missing | Uniform | Uniform | Missing | |
bics | Number | 1 | 1 | 1 | 1 | 1 | 1 |
Dimensions | \(50 \times 50\) | \(50 \times 50\) | \(50 \times 50\) | \(50 \times 50\) | \(50 \times 50\) | \(50 \times 50\) | |
Contiguity | No | No | No | No | Yes | No |