Institutional Repository of Vilnius University
HNSW grafų algoritmas apytikslio artimiausio kaimyno paieškai /
Abstract
dc:descriptionThe bachelor thesis focused on the HNSW algorithm, aiming to understand its functioning. A notable aspect of the thesis was the exploration of parallels between the HNSW algorithm and skip-lists in one-dimensional data. The structural similarities between these two data structures were analyzed, focusing on their common characteristics and case, when HNSW corresponds to the skp-list structure. Furthermore, the thesis delved into the behavior of the HNSW algorithm when applied to clustered data. It aimed to identify patterns and assess the algorithm's performance in scenarios where data points exhibit clustering tendencies. This analysis provided valuable insights into the algorithm's behavior and shed light on its strengths and limitations. Overall, the bachelor thesis successfully investigated the workings of the HNSW algorithm, drew parallels between HNSW and skip-lists, and explored the algorithm's performance in clustered data. The findings contribute to a deeper understanding of the HNSW algorithm and its potential applications in approximate nearest neighbor search tasks.
Degree
thesis:*- Grantor dc:publisher
- Institutional Repository of Vilnius University
- Year dc:date
- 2023
Author and committee
dc:creator, dc:contributor.*- Author dc:creator
-
- Pliška, Brygida Ivona,
Rights
dc:rights- Statement dc:rights
-
- info:eu-repo/semantics/openAccess
- Language dc:language
- lit
Identifiers
dc:identifier.*- Repository record dc:identifier
- https://repository.vu.lt/VU:ELABAETD192844026&prefLang=en_US