Presentation 2005/12/14
An Efficient Segment Searching and its Experimental Evaluation
Tomoko OKAMOTO, Masahide SUGIYAMA,
PDF Download Page PDF download Page Link
Abstract(in Japanese) (See Japanese page)
Abstract(in English) The problem of searching query signal from stored signal is formulated as a vector search problem where a series of feature vectors extracted from signal is converted to an output probability function. As an efficient searching algorithm we have proposed DSCC (Determination using Segmentation-Clustering-Centers) method and have shown its effectiveness. DSCC is composed three methods; M1 (Determination using Segmentation-Centers), M2 (Determination using Clustering-Centers) and M3 (Distance Pruning). When a current vector can not be determined using these three methods, AS (Active Search) is applied. This report describes the difference between M2 and M3 methods. Furthermore, the improvement of M1 and its effectiveness is described.
Keyword(in Japanese) (See Japanese page)
Keyword(in English) Segment Searching / Clustering / Output Probability / VQ
Paper # NLC2005-64,SP2005-97
Date of Issue

Conference Information
Committee NLC
Conference Date 2005/12/14(1days)
Place (in Japanese) (See Japanese page)
Place (in English)
Topics (in Japanese) (See Japanese page)
Topics (in English)
Chair
Vice Chair
Secretary
Assistant

Paper Information
Registration To Natural Language Understanding and Models of Communication (NLC)
Language JPN
Title (in Japanese) (See Japanese page)
Sub Title (in Japanese) (See Japanese page)
Title (in English) An Efficient Segment Searching and its Experimental Evaluation
Sub Title (in English)
Keyword(1) Segment Searching
Keyword(2) Clustering
Keyword(3) Output Probability
Keyword(4) VQ
1st Author's Name Tomoko OKAMOTO
1st Author's Affiliation Graduate School of Computer Science and Engineering, The Univ. of Aizu()
2nd Author's Name Masahide SUGIYAMA
2nd Author's Affiliation Graduate School of Computer Science and Engineering, The Univ. of Aizu
Date 2005/12/14
Paper # NLC2005-64,SP2005-97
Volume (vol) vol.105
Number (no) 493
Page pp.pp.-
#Pages 6
Date of Issue