Дослідження, статті, розробки | Публікації Лабораторії. Штучний інтелект в освіті. Дистанційне навчанняSearch index building for numeric vectors with O(1) memory consumption
Sukhodolsky A. Search index building for numeric vectors with O(1) memory consumption / Sukhodolsky A., Tytenko S. // Modern Aspects of Software Development: Proceedings of VI International Scientific and Practical Virtual Conference of Software Development Specialists, June, 24 2019. – Kyiv: Igor Sikorsky KPI, 2019. – pp. 54-60.
Introduction
Nowadays, the world of information technologies produces a large amount of information, and therefore we should find new ways for its effective storage and retrieval. For classical data types (text, numbers, dates, etc.) there is a large number of algorithms that allow us to index and search for almost linear time. For a relatively young vector data representation, there is also a set of algorithms, but in most cases it is necessary to find a balance between memory consumption, search speed and quality of the results.
PDFЗверніть увагу на додаткові посилання
Якщо вас цікавить...
Головний розділ
| ||||||||||||||
