English

GaKCo: a Fast GApped k-mer string Kernel using COunting

Machine Learning 2017-09-19 v3 Artificial Intelligence Computational Complexity Computation and Language Data Structures and Algorithms

Abstract

String Kernel (SK) techniques, especially those using gapped kk-mers as features (gk), have obtained great success in classifying sequences like DNA, protein, and text. However, the state-of-the-art gk-SK runs extremely slow when we increase the dictionary size (Σ\Sigma) or allow more mismatches (MM). This is because current gk-SK uses a trie-based algorithm to calculate co-occurrence of mismatched substrings resulting in a time cost proportional to O(ΣM)O(\Sigma^{M}). We propose a \textbf{fast} algorithm for calculating \underline{Ga}pped kk-mer \underline{K}ernel using \underline{Co}unting (GaKCo). GaKCo uses associative arrays to calculate the co-occurrence of substrings using cumulative counting. This algorithm is fast, scalable to larger Σ\Sigma and MM, and naturally parallelizable. We provide a rigorous asymptotic analysis that compares GaKCo with the state-of-the-art gk-SK. Theoretically, the time cost of GaKCo is independent of the ΣM\Sigma^{M} term that slows down the trie-based approach. Experimentally, we observe that GaKCo achieves the same accuracy as the state-of-the-art and outperforms its speed by factors of 2, 100, and 4, on classifying sequences of DNA (5 datasets), protein (12 datasets), and character-based English text (2 datasets), respectively. GaKCo is shared as an open source tool at \url{https://github.com/QData/GaKCo-SVM}

Keywords

Cite

@article{arxiv.1704.07468,
  title  = {GaKCo: a Fast GApped k-mer string Kernel using COunting},
  author = {Ritambhara Singh and Arshdeep Sekhon and Kamran Kowsari and Jack Lanchantin and Beilun Wang and Yanjun Qi},
  journal= {arXiv preprint arXiv:1704.07468},
  year   = {2017}
}

Comments

@ECML 2017

R2 v1 2026-06-22T19:26:36.956Z