| Crates.io | cigar-lodhi-rs |
| lib.rs | cigar-lodhi-rs |
| version | 0.1.0 |
| created_at | 2025-11-22 12:03:55.43274+00 |
| updated_at | 2025-11-22 12:03:55.43274+00 |
| description | Lodhi subsequence kernel on CIGAR strings |
| homepage | |
| repository | https://github.com/rickbeeloo/cigar-lodhi-rs |
| max_upload_size | |
| id | 1945253 |
| size | 16,590 |
A 2002 paper, Lodhi et. al. 2002 introduced "Text Classification using String Kernels". The idea is that instead of treating strings as flat sequences of symbols, we can represent them as weighted sets of all subsequences of length k, where longer consecutive matches contribute more.
todo: finish