Japanese / English

Detail of Publication

Text Language Japanese
Authors Masakazu Iwamura, Tomoyuki Mutoh, Koichi Kise
Title Duplicate Registration of Data in Approximate Nearest Neighbor
Journal IEICE Technical Report
Location 神奈川県横浜市
Reviewed or not Not reviewed
Presentation type Oral
Month & Year June 2011
Abstract We propose a new method based on duplicate registration of data in approximate nearest neighbor problem. In the experiment, the proposed method achieved the same accuracy as a conventional method, LSH, with 18\% of computational cost and 90\% of memory usage needed for LSH. In addition, we consider the phenomenon based on the criterion $\rho$ of efficiency of LSH appeared in [1].
Back to list