Content deleted Content added
→LSH algorithm for nearest neighbor search: rm ref to title in heading |
|||
Line 272:
Semantic hashing is a technique that attempts to map input items to addresses such that closer inputs have higher [[semantic similarity]].<ref>{{Cite journal|last1=Salakhutdinov|first1=Ruslan|last2=Hinton|first2=Geoffrey|date=2008|title=Semantic hashing|journal=International Journal of Approximate Reasoning|language=en|volume=50|issue=7|pages=969–978|doi=10.1016/j.ijar.2008.11.006|doi-access=free}}</ref> The hashcodes are found via training of an [[artificial neural network]] or [[graphical model]].{{cn|date=September 2021}}
==
One of the main applications of LSH is to provide a method for efficient approximate [[nearest neighbor search]] algorithms. Consider an LSH family <math>\mathcal F</math>. The algorithm has two main parameters: the width parameter {{mvar|k}} and the number of hash tables {{mvar|L}}.
| |||