The Design of the Context Quantizer on the basis of Amazing Measure

DOI

10.2991/meici-15.2015.176How to use a DOI?

Keywords

Context modeling; Context quantization; Amazing measure; Clustering algorithm

Abstract

When using the clustering algorithm, the implementation of the context quantization not only expands the application range of the quantizer, but also obtains better coding performance. However, those clustering algorithms depend on the choice of the similarity measure. In the the previous works, the increment of the description length was suggested but the result is that it cannot fully meets the similarity measure of various attributes, resulting in the performance of the clustering result deviation. In this paper, a new similarity measure which holds better mathematical description is given. The increment of the amazing measure, which denotes the similarity measure, two count vectors are discussed in this paper and its corresponding properties are also explained. The experimental results indicate that when using the proposed similarity measure, both the stability of the context quantizer and the corresponding coding results can be optimized at the same time.

Copyright

© 2015, the Authors. Published by Atlantis Press.

Open Access

This is an open access article distributed under the CC BY-NC license (http://creativecommons.org/licenses/by-nc/4.0/).

Download article (PDF)

TY  - CONF
AU  - Yiping Zhang
AU  - Min Chen
PY  - 2015/06
DA  - 2015/06
TI  - The Design of the Context Quantizer on the basis of Amazing Measure
BT  - Proceedings of the 2015 International Conference on Management, Education, Information and Control
PB  - Atlantis Press
SP  - 1013
EP  - 1016
SN  - 1951-6851
UR  - https://doi.org/10.2991/meici-15.2015.176
DO  - 10.2991/meici-15.2015.176
ID  - Zhang2015/06
ER  -