Optimizing BEVA with Two-Level Indexes

dc.contributor.advisor1Moura, Edleno Silva de Moura
dc.contributor.advisor1Latteshttp://lattes.cnpq.br/4737852130924504por
dc.contributor.referee1Silva, Altigran Soares da
dc.contributor.referee1Latteshttp://lattes.cnpq.br/3405503472010994por
dc.contributor.referee2Rosa, Thierson Couto
dc.contributor.referee2Latteshttp://lattes.cnpq.br/4414718560764818por
dc.creatorFerreira, Van Den Berg da Gama
dc.creator.Latteshttp://lattes.cnpq.br/9433225118102294por
dc.date.issued2020-07-10
dc.description.abstractQuery autocompletion is an important component of modern search systems that suggests possible queries at each user keystroke to complete the query based on the prefix already typed in the search box. One of the most adopted and successful data structures for query autocompletion is the TRIE which is used to index the possible query suggestions. The TRIE is traversed based on the search prefix typed by the user in order to select suggestions that match the prefix. The use of TRIEs requires a large amount of extra memory for processing queries, which may increase the cost for processing queries and may limit the number of query suggestions indexed. In this work we propose optimized alternative implementations of BEVA algorithm, currently the state-of-the-art in the literature for autocompletion, in order to achieve a reduction in its memory consumption while keeping it efficient in query processing times. First, we propose a novel strategy to build the TRIE, named level-at-a-time (laat), and compare its performance to the way TRIEs are usually built, the key-ata-time (kaat). In the kaat strategy the index is built in depth-ward direction and in the laat strategy the index is built in breadth-ward direction. We implemented the proposed ideas and experimented them with several datasets, where we show that laat strategy allows a significant speedup in query processing of BEVA, being up to four times faster, with improvements achieved specially in queries with high number of errors, which are the most expensive ones. Second, we study the use of two-level indexing and prefix processing approaches for query autocompletion also in BEVA method. Two-level approaches combine the use of indexes with sequential search in order to reduce memory requirements in search systems. In our study, we insert in the TRIE only part of each query inserted, and the leaf nodes reference to a set of queries where a sequential search is performed. We experimented two alternative ways of selecting the portion of each key that remains indexed in the first level and compare their performance. The two-level approach has shown to significantly reduce the memory requirements for storing the index with just a small variation in query processing timeseng
dc.description.resumoIndisponívelpor
dc.embargo.liftdatePeríodo de Embargo: 22/07/2020 - 09/01/2023por
dc.formatapplication/pdf*
dc.identifier.citationFERREIRA, Van Den Berg. Optimizing BEVA with Two-Level Indexes. 2020. 89 f. Dissertação (Mestrado em Informática) - Universidade Federal do Amazonas, Manaus, 2020.por
dc.identifier.urihttps://tede.ufam.edu.br/handle/tede/7840
dc.languageporpor
dc.publisherUniversidade Federal do Amazonaspor
dc.publisher.countryBrasilpor
dc.publisher.departmentInstituto de Computaçãopor
dc.publisher.initialsUFAMpor
dc.publisher.programPrograma de Pós-graduação em Informáticapor
dc.rightsAcesso Abertopor
dc.rights.urihttp://creativecommons.org/licenses/by-nc-nd/4.0/
dc.subject.cnpqCIÊNCIAS EXATAS E DA TERRA: CIÊNCIA DA COMPUTAÇÃOpor
dc.subject.userQuery processingeng
dc.subject.userError-toleranteng
dc.subject.userAutocompletioneng
dc.subject.userTwo leveleng
dc.subject.userTrieeng
dc.subject.userProgramação de sistemas (Computação)por
dc.subject.userProgramas de computador - Precisãopor
dc.thumbnail.urlhttps://tede.ufam.edu.br//retrieve/39514/Reprodu%c3%a7%c3%a3o%20N%c3%a3o%20Autorizada.pdf.jpg*
dc.thumbnail.urlhttps://tede.ufam.edu.br/retrieve/61940/Dissertacao_VanDenBerg_PPGI.pdf.jpg*
dc.titleOptimizing BEVA with Two-Level Indexespor
dc.title.alternativeOtimizando o BEVA com índices de dois níveispor
dc.typeDissertaçãopor

Arquivos

Pacote original

Agora exibindo 1 - 1 de 1
Carregando...
Imagem de Miniatura
Nome:
Dissertacao_VanDenBerg_PPGI.pdf
Tamanho:
3.1 MB
Formato:
Adobe Portable Document Format

Licença do pacote

Agora exibindo 1 - 1 de 1
Carregando...
Imagem de Miniatura
Nome:
license.txt
Tamanho:
2.32 KB
Formato:
Item-specific license agreed upon to submission
Descrição: