Results 1 to 6 of about 6 (6)

Automatic search for fragments containing biographical information in a natural language text

Труды Института системного программирования РАН, 2019
The search and classification of text documents are used in many practical applications. These are the key tasks of information retrieval. Methods of text searching and classifying are used in search engines, electronic libraries and catalogs, systems ...
A. V. Glazkova
doaj   +1 more source

Vector search using method of clustering using ensemble of oblivious trees

Научно-технический вестник информационных технологий, механики и оптики
Information retrieval using machine learning algorithms is based on transforming the original multimodal documents into vector representations. These vectors are then indexed, and the search is performed within this index.
N. A. Tomilov, V. P. Turov, A. A. Babayants, A. V. Platonov   +3 more
doaj   +1 more source

A method of storing vector data in compressed form using clustering

Научно-технический вестник информационных технологий, механики и оптики
The development of the machine learning algorithms for information search in recent years made it possible to represent text and multimodal documents in the form of vectors.
N. A. Tomilov, V. P. Turov, A. A. Babayants, A. V. Platonov   +3 more
doaj   +1 more source

Vector embeddings compression using clustering with the ensemble of oblivious decision trees and separate centroids storage

Научно-технический вестник информационных технологий, механики и оптики
The modern approach to search textual and multimodal data in large collections involves the transformation of the documents into vector embeddings. To store these embeddings efficiently different approaches could be used, such as quantization, which ...
N. A. Tomilov
doaj   +1 more source

K-sparse encoder for efficient information retrieval

Научно-технический вестник информационных технологий, механики и оптики
Modern industrial search engines typically employ a two-stage pipeline: fast candidate retrieval followed by reranking. This approach inevitably leads to the loss of some relevant documents due to the simplicity of algorithms used in the first stage ...
V. Yu. Dobrynin
doaj   +1 more source

Efficient sparse retrieval through embedding-based inverted index construction

Научно-технический вестник информационных технологий, механики и оптики
Modern search engines use a two-stage architecture for efficient and high-quality search over large volumes of data. In the first stage, simple and fast algorithms like BM25 are applied, while in the second stage, more precise but resourceintensive ...
V. Yu. Dobrynin, R. K. Abramovich, A. V. Platonov   +2 more
doaj