arXiv subjects

W. F. Smyth

Publications and source records attributed to W. F. Smyth.

16 recordsLinked to original sources

arXiv2025

On the Complexity of Finding Approximate LCS of Multiple Strings

Finding an Approximate Longest Common Substring (ALCS) within a given set $S=\{s_1,s_2,\ldots,s_m\}$ of $m \ge 2$ strings is a key problem in computational biology, such as identifying related mutations across multiple genetic sequences. We study several variants of ALCS problems that, given integers $k$ and $t \le m$, seek the longest string $u$ -- or the longest substring $u$ of any string in $S$ -- that lies within distance $k$ of at least one substring in $t$ distinct strings from $S$. While the general problems are NP-hard, we present efficient algorithms for restricted cases under Hamming and edit distances using the $LCP_k$ and $k$-errata tree data structures. Our methods achieve run times of $\mathcal{O}(N^2)$, $\mathcal{O}(k\ell N^2)$, and $\mathcal{O}(mN\log^k \ell)$, where $\ell$ is the length of the longest string and $N$ is the sum of the lengths of all the strings in $S$. We also establish conditional lower bounds under the Strong Exponential Time Hypothesis and extend our study to indeterminate strings.

cs.DS↗

arXiv2024

V-Words, Lyndon Words and Galois Words

We say that a family $\mathcal{W}$ of strings over $Σ^+$ forms a Unique Maximal Factorization Family (UMFF) if and only if every $w \in \mathcal{W}$ has a unique maximal factorization. Further, an UMFF $\mathcal{W}$ is called a circ-UMFF whenever it contains exactly one rotation of every primitive string $x \in Σ^+$. $V$-order is a non-lexicographical total ordering on strings that determines a circ-UMFF. In this paper we propose a generalization of circ-UMFF called the substring circ-UMFF and extend combinatorial research on $V$-order by investigating connections to Lyndon words. Then we extend these concepts to any total order. Applications of this research arise in efficient text indexing, compression, and search problems.

cs.DS↗

arXiv2024

Practical KMP/BM Style Pattern-Matching on Indeterminate Strings

In this paper we describe two simple, fast, space-efficient algorithms for finding all matches of an indeterminate pattern $p = p[1..m]$ in an indeterminate string $x = x[1..n]$, where both $p$ and $x$ are defined on a "small" ordered alphabet $Σ$ $-$ say, $σ= |Σ| \le 9$. Both algorithms depend on a preprocessing phase that replaces $Σ$ by an integer alphabet $Σ_I$ of size $σ_I = σ$ which (reversibly, in time linear in string length) maps both $x$ and $p$ into equivalent regular strings $y$ and $q$, respectively, on $Σ_I$, whose maximum (indeterminate) letter can be expressed in a 32-bit word (for $σ\le 4$, thus for DNA sequences, an 8-bit representation suffices). We first describe an efficient version KMP Indet of the venerable Knuth-Morris-Pratt algorithm to find all occurrences of $q$ in $y$ (that is, of $p$ in $x$), but, whenever necessary, using the prefix array, rather than the border array, to control shifts of the transformed pattern $q$ along the transformed string $y$. We go on to describe a similar efficient version BM Indet of the Boyer- Moore algorithm that turns out to execute significantly faster than KMP Indet over a wide range of test cases. A noteworthy feature is that both algorithms require very little additional space: $Θ(m)$ words. We conjecture that a similar approach may yield practical and efficient indeterminate equivalents to other well-known pattern-matching algorithms, in particular the several variants of Boyer-Moore.

cs.DS↗

arXiv2023

String Covering: A Survey

The study of strings is an important combinatorial field that precedes the digital computer. Strings can be very long, trillions of letters, so it is important to find compact representations. Here we first survey various forms of one potential compaction methodology, the cover of a given string x, initially proposed in a simple form in 1990, but increasingly of interest as more sophisticated variants have been discovered. We then consider covering by a seed; that is, a cover of a superstring of x. We conclude with many proposals for research directions that could make significant contributions to string processing in future.

cs.DS↗

arXiv2020

A New Approach to Regular & Indeterminate Strings

In this paper we propose a new, more appropriate definition of regular and indeterminate strings. A regular string is one that is "isomorphic" to a string whose entries all consist of a single letter, but which nevertheless may itself include entries containing multiple letters. A string that is not regular is said to be indeterminate. We begin by proposing a new model for the representation of strings, regular or indeterminate, then go on to describe a linear time algorithm to determine whether or not a string $x = x[1..n]$ is regular and, if so, to replace it by a lexicographically least (lex-least) string $y$ whose entries are all single letters. Furthermore, we connect the regularity of a string to the transitive closure problem on a graph, which in our special case can be efficiently solved. We then introduce the idea of a feasible palindrome array MP of a string, and prove that every feasible MP corresponds to some (regular or indeterminate) string. We describe an algorithm that constructs a string $x$ corresponding to given feasible MP, while ensuring that whenever possible $x$ is regular and if so, then lex-least. A final section outlines new research directions suggested by this changed perspective on regular and indeterminate strings.

cs.DS↗

arXiv2018

More properties of the Fibonacci word on an infinite alphabet

Recently the Fibonacci word $W$ on an infinite alphabet was introduced by [Zhang et al., Electronic J. Combinatorics 24-2 (2017) #P2.52] as a fixed point of the morphism $ϕ: (2i) \mapsto (2i)(2i+ 1),\ (2i+ 1) \mapsto (2i+ 2)$ over all $i \in \mathbb{N}$. In this paper we investigate the occurrence of squares, palindromes, and Lyndon factors in this infinite word.

math.CO↗

arXiv2018

Palindromes in starlike trees

In this note, we obtain an upper bound on the maximum number of distinct non-empty palindromes in starlike trees. This bound implies, in particular, that there are at most $4n$ distinct non-empty palindromes in a starlike tree with three branches each of length $n$. For such starlike trees labelled with a binary alphabet, we sharpen the upper bound to $4n-1$ and conjecture that the actual maximum is $4n-2$. It is intriguing that this simple conjecture seems difficult to prove, in contrast to the straightforward proof of the bound.

math.CO↗

arXiv2017

On the Parikh-de-Bruijn grid

We introduce the Parikh-de-Bruijn grid, a graph whose vertices are fixed-order Parikh vectors, and whose edges are given by a simple shift operation. This graph gives structural insight into the nature of sets of Parikh vectors as well as that of the Parikh set of a given string. We show its utility by proving some results on Parikh-de-Bruijn strings, the abelian analog of de-Bruijn sequences.

cs.DM↗

arXiv2017

Lyndon Array Construction during Burrows-Wheeler Inversion

In this paper we present an algorithm to compute the Lyndon array of a string $T$ of length $n$ as a byproduct of the inversion of the Burrows-Wheeler transform of $T$. Our algorithm runs in linear time using only a stack in addition to the data structures used for Burrows-Wheeler inversion. We compare our algorithm with two other linear-time algorithms for Lyndon array construction and show that computing the Burrows-Wheeler transform and then constructing the Lyndon array is competitive compared to the known approaches. We also propose a new balanced parenthesis representation for the Lyndon array that uses $2n+o(n)$ bits of space and supports constant time access. This representation can be built in linear time using $O(n)$ words of space, or in $O(n\log n/\log\log n)$ time using asymptotically the same space as $T$.

cs.DS↗

arXiv2017

Counting Lyndon factors

In this paper, we determine the maximum number of distinct Lyndon factors that a word of length $n$ can contain. We also derive formulas for the expected total number of Lyndon factors in a word of length $n$ on an alphabet of size $σ$, as well as the expected number of distinct Lyndon factors in such a word. The minimum number of distinct Lyndon factors in a word of length $n$ is $1$ and the minimum total number is $n$, with both bounds being achieved by $x^n$ where $x$ is a letter. A more interesting question to ask is what is the minimum number of distinct Lyndon factors in a Lyndon word of length $n$? In this direction, it is known (Saari, 2014) that an optimal lower bound for the number of distinct Lyndon factors in a Lyndon word of length $n$ is $\lceil\log_ϕ(n) + 1\rceil$, where $ϕ$ denotes the golden ratio $(1 + \sqrt{5})/2$. Moreover, this lower bound is attained by the so-called finite "Fibonacci Lyndon words", which are precisely the Lyndon factors of the well-known "infinite Fibonacci word" -- a special example of a "infinite Sturmian word". Saari (2014) conjectured that if $w$ is Lyndon word of length $n$, $n\ne 6$, containing the least number of distinct Lyndon factors over all Lyndon words of the same length, then $w$ is a Christoffel word (i.e., a Lyndon factor of an infinite Sturmian word). We give a counterexample to this conjecture. Furthermore, we generalise Saari's result on the number of distinct Lyndon factors of a Fibonacci Lyndon word by determining the number of distinct Lyndon factors of a given Christoffel word. We end with two open problems.

math.CO↗

arXiv2016

Algorithms to Compute the Lyndon Array

We first describe three algorithms for computing the Lyndon array that have been suggested in the literature, but for which no structured exposition has been given. Two of these algorithms execute in quadratic time in the worst case, the third achieves linear time, but at the expense of prior computation of both the suffix array and the inverse suffix array of x. We then go on to describe two variants of a new algorithm that avoids prior computation of global data structures and executes in worst-case n log n time. Experimental evidence suggests that all but one of these five algorithms require only linear execution time in practice, with the two new algorithms faster by a small factor. We conjecture that there exists a fast and worst-case linear-time algorithm to compute the Lyndon array that is also elementary (making no use of global data structures such as the suffix array).

cs.DS↗

arXiv2015

String Comparison in $V$-Order: New Lexicographic Properties & On-line Applications

$V$-order is a global order on strings related to Unique Maximal Factorization Families (UMFFs), which are themselves generalizations of Lyndon words. $V$-order has recently been proposed as an alternative to lexicographical order in the computation of suffix arrays and in the suffix-sorting induced by the Burrows-Wheeler transform. Efficient $V$-ordering of strings thus becomes a matter of considerable interest. In this paper we present new and surprising results on $V$-order in strings, then go on to explore the algorithmic consequences.

cs.DS↗

arXiv2015

Enhanced Covers of Regular & Indeterminate Strings using Prefix Tables

A \itbf{cover} of a string $x = x[1..n]$ is a proper substring $u$ of $x$ such that $x$ can be constructed from possibly overlapping instances of $u$. A recent paper \cite{FIKPPST13} relaxes this definition --- an \itbf{enhanced cover} $u$ of $x$ is a border of $x$ (that is, a proper prefix that is also a suffix) that covers a {\it maximum} number of positions in $x$ (not necessarily all) --- and proposes efficient algorithms for the computation of enhanced covers. These algorithms depend on the prior computation of the \itbf{border array} $β[1..n]$, where $β[i]$ is the length of the longest border of $x[1..i]$, $1 \le i \le n$. In this paper, we first show how to compute enhanced covers using instead the \itbf{prefix table}: an array $π[1..n]$ such that $π[i]$ is the length of the longest substring of $x$ beginning at position $i$ that matches a prefix of $x$. Unlike the border array, the prefix table is robust: its properties hold also for \itbf{indeterminate strings} --- that is, strings defined on {\it subsets} of the alphabet $Σ$ rather than individual elements of $Σ$. Thus, our algorithms, in addition to being faster in practice and more space-efficient than those of \cite{FIKPPST13}, allow us to easily extend the computation of enhanced covers to indeterminate strings. Both for regular and indeterminate strings, our algorithms execute in expected linear time. Along the way we establish an important theoretical result: that the expected maximum length of any border of any prefix of a regular string $x$ is approximately 1.64 for binary alphabets, less for larger ones.

cs.DS↗

arXiv2015

Computing Covers Using Prefix Tables

An \emph{indeterminate string} $x = x[1..n]$ on an alphabet $Σ$ is a sequence of nonempty subsets of $Σ$; $x$ is said to be \emph{regular} if every subset is of size one. A proper substring $u$ of regular $x$ is said to be a \emph{cover} of $x$ iff for every $i \in 1..n$, an occurrence of $u$ in $x$ includes $x[i]$. The \emph{cover array} $γ= γ[1..n]$ of $x$ is an integer array such that $γ[i]$ is the longest cover of $x[1..i]$. Fifteen years ago a complex, though nevertheless linear-time, algorithm was proposed to compute the cover array of regular $x$ based on prior computation of the border array of $x$. In this paper we first describe a linear-time algorithm to compute the cover array of regular string $x$ based on the prefix table of $x$. We then extend this result to indeterminate strings.

cs.DS↗

arXiv2015

Inferring an Indeterminate String from a Prefix Graph

An \itbf{indeterminate string} (or, more simply, just a \itbf{string}) $\s{x} = \s{x}[1..n]$ on an alphabet $Σ$ is a sequence of nonempty subsets of $Σ$. We say that $\s{x}[i_1]$ and $\s{x}[i_2]$ \itbf{match} (written $\s{x}[i_1] \match \s{x}[i_2]$) if and only if $\s{x}[i_1] \cap \s{x}[i_2] \ne \emptyset$. A \itbf{feasible array} is an array $\s{y} = \s{y}[1..n]$ of integers such that $\s{y}[1] = n$ and for every $i \in 2..n$, $\s{y}[i] \in 0..n\- i\+ 1$. A \itbf{prefix table} of a string $\s{x}$ is an array $\sπ = \sπ[1..n]$ of integers such that, for every $i \in 1..n$, $\sπ[i] = j$ if and only if $\s{x}[i..i\+ j\- 1]$ is the longest substring at position $i$ of \s{x} that matches a prefix of \s{x}. It is known from \cite{CRSW13} that every feasible array is a prefix table of some indetermintate string. A \itbf{prefix graph} $\mathcal{P} = \mathcal{P}_{\s{y}}$ is a labelled simple graph whose structure is determined by a feasible array \s{y}. In this paper we show, given a feasible array \s{y}, how to use $\mathcal{P}_{\s{y}}$ to construct a lexicographically least indeterminate string on a minimum alphabet whose prefix table $\sπ = \s{y}$.

cs.DS↗

arXiv2014

Indeterminate Strings, Prefix Arrays & Undirected Graphs

An integer array y = y[1..n] is said to be feasible if and only if y[1] = n and, for every i \in 2..n, i \le i+y[i] \le n+1. A string is said to be indeterminate if and only if at least one of its elements is a subset of cardinality greater than one of a given alphabet Sigma; otherwise it is said to be regular. A feasible array y is said to be regular if and only if it is the prefix array of some regular string. We show using a graph model that every feasible array of integers is a prefix array of some (indeterminate or regular) string, and for regular strings corresponding to y, we use the model to provide a lower bound on the alphabet size. We show further that there is a 1-1 correspondence between labelled simple graphs and indeterminate strings, and we show how to determine the minimum alphabet size |Sigma| of an indeterminate string x based on its associated graph Gx. Thus, in this sense, indeterminate strings are a more natural object of combinatorial interest than the strings on elements of Sigma that have traditionally been studied.

cs.DM↗