Pivot Selection for Narrow Sketches by Optimization Algorithms
| Autor: | Kouichi Hirata, Naoya Higuchi, Yasunobu Imamura, Tetsuji Kuboyama, Vladimir Mic, Takeshi Shinohara |
|---|---|
| Rok vydání: | 2020 |
| Předmět: |
Optimization algorithm
Computer science 020204 information systems Nearest neighbor search 0202 electrical engineering electronic engineering information engineering 020201 artificial intelligence & image processing 02 engineering and technology Selection method 16. Peace & justice Algorithm Selection (genetic algorithm) |
| Zdroj: | Similarity Search and Applications ISBN: 9783030609351 SISAP |
| Popis: | Sketches are compact bit strings that are considered as products of an LSH for high-dimensional data. We use them in filtering for narrowing down solution candidates in similarity search. We propose a pivot selection method for narrow sketches with a length such as 16-bits by optimization algorithms with the accuracy of filtering itself as the objective function. |
| Databáze: | OpenAIRE |
| Externí odkaz: |
|