Adapts COM and RLM statistical features from visual computing to string similarity, outperforming distance-based and other measures in synthetic tests (P<0.001 in 3/4 cases) and achieving best results on a plagiarism dataset.
Title resolution pending
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.LG 1years
2026 1verdicts
UNVERDICTED 1representative citing papers
citing papers explorer
-
Proposal and study of statistical features for string similarity computation and classification
Adapts COM and RLM statistical features from visual computing to string similarity, outperforming distance-based and other measures in synthetic tests (P<0.001 in 3/4 cases) and achieving best results on a plagiarism dataset.