The Parameterized Suffix Tray

TL;DR

This paper introduces the parameterized suffix tray, an efficient indexing structure for parameterized string matching that uses linear space and supports fast pattern queries.

Contribution

It presents the parameterized suffix tray, a novel data structure for p-matching, with optimal space and query time complexities, and details its construction from the parameterized suffix tree.

Findings

The parameterized suffix tray uses O(n) space.

Supports pattern matching in O(m + log(σ+π) + occ) time.

Constructible in O(n) time from the parameterized suffix tree.

Abstract

Let $\Sigma$ and $\Pi$ be disjoint alphabets, respectively called the static alphabet and the parameterized alphabet. Two strings $x$ and $y$ over $\Sigma \cup \Pi$ of equal length are said to parameterized match (p-match) if there exists a renaming bijection $f$ on $\Sigma$ and $\Pi$ which is identity on $\Sigma$ and maps the characters of $x$ to those of $y$ so that the two strings become identical. The indexing version of the problem of finding p-matching occurrences of a given pattern in the text is a well-studied topic in string matching. In this paper, we present a state-of-the-art indexing structure for p-matching called the parameterized suffix tray of an input text $T$, denoted by $\mathsf{PSTray}(T)$. We show that $\mathsf{PSTray}(T)$ occupies $O(n)$ space and supports pattern matching queries in $O(m + \log (\sigma+\pi) + \mathit{occ})$ time, where $n$ is the length of $T$, $m$ is the length of a query pattern $P$, $\pi$ is the number of distinct symbols of $|\Pi|$ in $T$, $\sigma$ is the number of distinct symbols of $|\Sigma|$ in $T$ and $\mathit{occ}$ is the number of p-matching occurrences of $P$ in $T$. We also present how to build $\mathsf{PSTray}(T)$ in $O(n)$ time from the parameterized suffix tree of $T$.