izpis_h1_title_alt

Vzporedni algoritmi za problem Levenshteinove razdalje in najdaljše skupno podzaporedje
ID BAJIĆ, LUKA (Author), ID Mihelič, Jurij (Mentor) More about this mentor... This link opens in a new window

.pdfPDF - Presentation file, Download (441,70 KB)
MD5: D1FCC9B6252CB1CE39ED574EC5A28144

Abstract
Algoritmi za iskanje podobnosti med nizi so pomemben gradnik raznovrstnih aplikacij, ki so dandanes v vsakodnevni uporabi, od črkovalnikov do orodij za vodenje izvedb izvorne kode (npr. Git). Diplomsko delo opiše različne pristope optimizacije dveh tovrstnih algoritmov: Levenshteinove razdalje in najdaljšega skupnega podzaporedja, z uporabo koncepta večnitnosti in dveh temeljnih pristopov vzporejanja -- diagonalni in naprej-nazaj. Glavni cilj je časovna pohitritev algoritmov, ki jo ustrezno izmerimo in primerjamo nove implementacije z obstoječimi. Tako za diagonalni pristop kot za pristop naprej-nazaj razvijemo tudi prostorsko optimizirano metodo, ki je smiselna za uporabo pri velikih količinah vhodnih podatkov.

Language:Slovenian
Keywords:vzporedni algoritmi, večnitnost, Levenshteinova razdalja, najdaljše skupno podzaporedje, dinamično programiranje
Work type:Bachelor thesis/paper
Typology:2.11 - Undergraduate Thesis
Organization:FRI - Faculty of Computer and Information Science
Year:2024
PID:20.500.12556/RUL-160507 This link opens in a new window
COBISS.SI-ID:208457219 This link opens in a new window
Publication date in RUL:29.08.2024
Views:183
Downloads:39
Metadata:XML DC-XML DC-RDF
:
Copy citation
Share:Bookmark and Share

Secondary language

Language:English
Title:Parallel algorithms for the Levenshtein distance and the longest common subsequence problems
Abstract:
Algorithms for finding similarity between strings are a building block of a large variety of modern applications, from spell checkers to version control tools (e.g. Git). This BSc thesis describes various approaches for optimizing two such algorithms: Levenshtein distance and Longest Common Subsequence, utilizing the concept of multithreading and two fundamental parallelization approaches -- diagonal and forward-backward. Main objective is improved execution time of said algorithms, which we measure and compare our results with existing implementations. For both diagonal and forward-backward approaches we develop an additional space optimization, which is useful for extremely large amounts of input data.

Keywords:parallel algorithms, multithreading, Levenshtein distance, longest common subsequence, dynamic programming

Similar documents

Similar works from RUL:
Similar works from other Slovenian collections:

Back