Home

About

Advanced Search

Browse by Discipline

Scientific Societies

E-print Alerts

Add E-prints

E-print Network
FAQHELPSITE MAPCONTACT US


  Advanced Search  

 
Time Series Classification by Imprecise Hidden Markov Models
 

Summary: Time Series Classification
by Imprecise Hidden Markov Models
Alessandro ANTONUCCI a,1
and Rocco DE ROSA b
,
a
IDSIA, Galleria 2, Via Cantonale, CH-6928 Manno-Lugano (Switzerland)
alessandro@idsia.ch
b
Computer Science Dept., University of Milan, Via Comelico 39/41, 20135 Milan, Italy
rocco.derosa@studenti.unimi.it
Abstract. Hidden Markov models (HMMs) are powerful tools for modelling the
generative and observational processes behind time series. For short sequences, the
small amount of data can make unreliable the estimates returned by the EM algo-
rithm, which is generally used to learn HMMs. To gain robustness in these cases, an
imprecise version of the EM algorithm, achieving an interval-valued quantification
of the model parameters can be considered instead. The bounds of the likelihood
assigned to a particular sequence with respect to these intervals can be efficiently
computed. Overall, this provides a time series classification algorithm. To classify
a new sequence, the bounds of the likelihood associated to the HMMs learned from

  

Source: Antonucci, Alessandro - Istituto Dalle Molle di Studi sull'Intelligenza Artificiale (IDSIA)

 

Collections: Computer Technologies and Information Sciences