A monothetic clustering method
Pattern Recognition Letters
Self-Organizing Maps
Planar Hidden Markov Modeling: From Speech to Optical Character Recognition
Advances in Neural Information Processing Systems 5, [NIPS Conference]
Information Extraction with HMM Structures Learned by Stochastic Optimization
Proceedings of the Seventeenth National Conference on Artificial Intelligence and Twelfth Conference on Innovative Applications of Artificial Intelligence
A Hybrid Self-Organizing Model for Sequence Analysis
ICTAI '08 Proceedings of the 2008 20th IEEE International Conference on Tools with Artificial Intelligence - Volume 02
The spherical hidden markov self organizing map for learning time series data
ICANN'12 Proceedings of the 22nd international conference on Artificial Neural Networks and Machine Learning - Volume Part I
Hi-index | 0.00 |
We propose in this paper a novel approach which makes self-organizing maps (SOM) and the Hidden Markov Models (HMMs) cooperate. Our approach (SOS-HMM: Self Organizing Structure of HMM) allows to learn the Hidden Markov Models topology. The main contribution for the proposed approach is to automatically extract the structure of a hidden Markov model without any prior knowledge of the application domain. This model can be represented as a graph of macro-states, where each state represents a micro model. Experimental results illustrate the advantages of the proposed approach compared to a fixed structure approach.