Institutional Repository of Vilnius University
Sensitivity analysis of online string-matching algorithms /
Abstract
dc:descriptionString-matching is a well known and studied problem in computer science that allows us to find occurrences of a pattern inside a text. As datasets grow larger every year, it becomes more and more important to know which algorithms work better in specific situations, as there is not a single algorithm that will be well suited for every scenario. The aim of this paper is to analyse a new algorithm proposed by Faro et al, which takes the approach of subsampling the text and patterns before calculating the distance of the elements in the subsample. Sensitivity analysis is used to allow us to test the robustness of the proposed algorithm. An extensive experimental comparison of the parameters, together with a comparison with classical algorithms, was presented. A dataset was generated from the performance measurements applied during the experiments and in addition, some suggestions about case scenarios of when to use the newly proposed algorithm.
Degree
thesis:*- Grantor dc:publisher
- Institutional Repository of Vilnius University
- Year dc:date
- 2021
Author and committee
dc:creator, dc:contributor.*- Author dc:creator
-
- Rivera Montano, Alvaro,
Subjects
dc:subject × 1Rights
dc:rights- Statement dc:rights
-
- info:eu-repo/semantics/openAccess
- Language dc:language
- eng
Identifiers
dc:identifier.*- Repository record dc:identifier
- https://repository.vu.lt/VU:ELABAETD193202412&prefLang=en_US