Scribd
Upload
78 views41 pages

HMM

- Hidden Markov models (HMMs) are statistical models where the system is assumed to be a Markov process with hidden states. HMMs have a set of states with transition and emission probabilities. - Three central issues in HMMs are model evaluation, decoding the most probable path, and model training. - HMMs are applied in areas like speech recognition, gesture recognition, language modeling, and motion video analysis.

Uploaded by

sachin verma
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
78 views41 pages

HMM

- Hidden Markov models (HMMs) are statistical models where the system is assumed to be a Markov process with hidden states. HMMs have a set of states with transition and emission probabilities. - Three central issues in HMMs are model evaluation, decoding the most probable path, and model training. - HMMs are applied in areas like speech recognition, gesture recognition, language modeling, and motion video analysis.

Uploaded by

sachin verma
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 41

CONTENTS

• Introduction
• Markov Model
• Hidden Markov model (HMM)
• Three central issues of HMM
– Model evaluation
– Most probable path decoding
– Model training
• Application Areas of HMM

• References
Hidden Markov Models

 Hidden Markow Models:

– A hidden Markov model (HMM) is a statistical


model,in which the system being modeled is
assumed to be a Markov process (Memoryless
process: its future and past are independent )
with hidden states.
Hidden Markov Models

 Hidden Markow Models:


– Has a set of states each of which has limited
number of transitions and emissions,
– Each transition between states has an
assisgned probability,
– Each model strarts from start state and ends
in end state,
Hidden Markov Models
Hidden Markov Models

 Markow Models :
 Talk about weather,
 Assume there are three types of weather:

– Sunny,

– Rainy,

– Foggy.
Markov Models

 Weather prediction is about the what would be the weather


tomorrow,
– Based on the observations on the past.
Markov Models

 Weather at day n is qn ∈{ sunny , rainy , foggy }

– qn depends on the known weathers of the past


days (qn-1, qn-2,…)
Markov Models

 We want to find that:

– means given the past weathers what is the


probability of any possible weather of today.
Markov Models

 Markow Models:
 For example:
 if we knew the weather for last three days was:

 the probability that tomorrow would be is:

P(q4 = | q3 = , q2 = , q1 = )
Markov Models

 Markow Models and Assumption (cont.):


– Therefore, make a simplifying assumption Markov
assumption:
 For sequence:

 the weather of tomorrow only depends on today


(first order Markov model)
Markov Models

 Markow Models and Assumption (cont.):


Examples:
 HMM:
Markov Models

 Markow Models and Assumption (cont.):


Examples:
 If
the weather yesterday was rainy and today is foggy
what is the probability that tomorrow it will be sunny?
Markov Models

 Markow Models and Assumption (cont.):


– Examples:
 If
the weather yesterday was rainy and today is foggy
what is the probability that tomorrow it will be sunny?

Markov assumption
Hidden Markov Models

 Hidden Markov Models (HMMs):


– What is HMM:
 Suppose that you are locked in a room for several days,
 you try to predict the weather outside,
 The only piece of evidence you have is whether the
person who comes into the room bringing your daily
meal is carrying an umbrella or not.
Hidden Markov Models

 Hidden Markov Models (HMMs):


– What is HMM (cont.):
 assume probabilities as seen in the table:
Hidden Markov Models

 Hidden Markov Models (HMMs):


– What is HMM (cont.):
 Finding the probability of a certain weather
qn ∈{ sunny , rainy , foggy }

 is based on the observations xi:


Hidden Markov Models

 Hidden Markov Models (HMMs):


– What is HMM (cont.):
 Using Bayes rule:

 For n days:
Hidden Markov Models

 Hidden Markov Models (HMMs):


– Examples:
 Suppose the day you were locked in it was sunny. The
next day, the caretaker carried an umbrella into the
room.
 You would like to know, what the weather was like on
this second day.
20
Discrete Markov Processes
(Markov Chains)
21

Hiddden Markov Models


22

Hidden Markov Models


23

Hidden Markov Models


24

Hidden Markov Model Examples


25

Hidden Markov Models


26

Hidden Markov Models


27

Hidden Markov Models


28
Three Fundamental Problems for
HMMs
29

HMM Evaluation Problem


30

HMM Evaluation Problem


31

HMM Evaluation Problem


32

HMM Evaluation Problem


33

HMM Evaluation Problem


34

HMM Decoding Problem


35

HMM Decoding Problem


36

HMM Decoding Problem


37

HMM Learning Problem


38

HMM Learning Problem


39

HMM Learning Problem


40

HMM Learning Problem


Application Areas of HMM

• On-line handwriting recognition


• Speech recognition
• Gesture recognition
• Language modeling
• Motion video analysis and tracking
• Stock price prediction
and many more….
References

 R.O. Duda, P.E. Hart, and D.G. Stork, Pattern Classification, New York: John Wiley, 2001.
 Selim Aksoy, “Pattern Recognition Course Materials”, Bilkent University, 2011.

You might also like