Year: 2023
Type: dissertation
Abstract: The main objective of this thesis is to show the extraction of sections relevant to a query from ranked documents in the form of a snippet (summary). To achieve this, we adapt the local sequence alignment method, commonly used in bioinformatics to identify similarities between gene sequences. By applying this method in information retrieval, we can spot the relevant sections within the ranked documents. To address the shortcomings of this approach in terms of data processing, including time and computational power, we reduce them to a negligible level by applying the method to only relevant documents. The quality of the snippets is assessed by employing various metrics and they are compared with Google snippets. In addition, the benefits of the local sequence alignment method are emphasized by adapting this method to query searches in YouTube videos.
Author: Yılmaz, Mert Anıl
Institution: Middle East Technical University
Language: English
Subfield: Information Systems
Field: Computer Science
Domain: Physical Sciences
Open Access status: green
OpenAlex ID: W7110617918
