Research Article

Effective and Fast Near Duplicate Detection via Signature-Based Compression Metrics

Table 1

Sensitivity of SoptSigs to various settings.

Settings Chain min_idf max_idf min_spotsigs

Setting 1 1 1 0.2 0.85 4 0.3 0.83
Setting 2 2 2 0.2 0.85 4 0.3 0.76
Setting 3 2 3 0.2 0.85 4 0.3 0.73
Setting 4 1 1 0.4 1.0 4 0.3 0.76
Setting 5 1 1 0.5 1.0 4 0.3 0.55
Setting 6 1 1 0.2 0.85 20 0.3 0.78
Setting 7 2 3 0.5 1.0 20 0.3 0.59