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 多��������樽存息族贈����損��
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