Asynchronous Hidden Markov Model Method and System

Abstract

A system, method and program storage device implementing a method for modeling a data generating process, wherein the modeling comprises observing a data sequence comprising irregularly sampled data, obtaining an observation sequence based on the observed data sequence, assigning a time index sequence to the data sequence, obtaining a hidden state sequence of the data sequence, and decoding the data sequence based on a combination of the time index sequence and the hidden state sequence to model the data sequence. The method further comprises assigning a probability distribution over time stamp values of the observation sequence, wherein the decoding comprises using a Hidden Markov Model. The method further comprises using an expectation maximization methodology to learn the Hidden Markov Model.

1 . A system for modeling a data generating process, said system comprising:
 a first processing component configured for observing a data sequence comprising irregularly sampled data;   a second processing component configured for obtaining an observation sequence based on the observed data sequence;   a third processing component configured for assigning a time index sequence to said data sequence;   a fourth processing component configured for obtaining a hidden state sequence of said data sequence; and
 a decoder operable for decoding said data sequence based on a combination of said time index sequence and said hidden state sequence to model said data sequence. 
   
   
   
       2 . The system of  claim 1 , wherein said second processing component being configured for assigning a probability distribution over time stamp values of said observation sequence. 
   
   
       2 . The system of  claim 1 , wherein said decoding comprises using a Hidden Markov Model. 
   
   
       3 . The system of  claim 1 , wherein said modeling is used in applications comprising any of speech recognition applications, jitter cancellation systems, video compression systems, business intelligence knowledge management systems, and bioinformatics. 
   
   
       4 . The system of  claim 1 , wherein said irregularly sampled data comprises missing data. 
   
   
       5 . A system for modeling a data generating process, said system comprising:
 means for observing a data sequence comprising irregularly sampled data;   means for obtaining an observation sequence based on the observed data sequence;   means for assigning a time index sequence to said data sequence;   means for obtaining a hidden state sequence of said data sequence; and   means for decoding said data sequence based on a combination of said time index sequence and said hidden state sequence to model said data sequence.