Hidden Markov Models for Lithological Well Log Classification
نویسنده
چکیده
The aim of this study is to explore the usefulness of hidden Markov models for lithological classification of geophysical measurements in boreholes. Standard statistical methods like linear and quadratic discriminant or cluster analysis treat the individual measurement points of well logs as independent samples of the underlying distribution. With hidden Markov models (HMM) the knowledge of statistical dependency between successive data points in a well log can be incorporated in the classification process. The sequence of lithological classes is modelled as a discrete, finite, homogeneous, first order Markov chain of transitions between states. This state sequence can be unobserved or is generally hidden from the observer. The only necessary information about the hidden state sequence is an observed sequence of state-dependent outputs. This output information is used to estimate the hidden state sequence which most probably has caused the given output sequence. The performance of HMM is evaluated on simulated data and on a real data set from eight wells near Berlin (Germany) with final depths between 60 and 140 metres. The results are compared with the outcome of the quadratic discriminant analysis.
منابع مشابه
Introducing Busy Customer Portfolio Using Hidden Markov Model
Due to the effective role of Markov models in customer relationship management (CRM), there is a lack of comprehensive literature review which contains all related literatures. In this paper the focus is on academic databases to find all the articles that had been published in 2011 and earlier. One hundred articles were identified and reviewed to find direct relevance for applying Markov models...
متن کاملSpeaker Independent Speech Recognition Using Hidden Markov Models for Persian Isolated Words
متن کامل
Hidden Markov Models for Spectral Similarity of Songs
Hidden Markov Models (HMM) are compared to Gaussian Mixture Models (GMM) for describing spectral similarity of songs. Contrary to previous work we make a direct comparison based on the log-likelihood of songs given an HMM or GMM. Whereas the direct comparison of log-likelihoods clearly favors HMMs, this advantage in terms of modeling power does not allow for any gain in genre classification acc...
متن کاملSpeaker Independent Speech Recognition Using Hidden Markov Models for Persian Isolated Words
متن کامل
Describing user's search behaviour with Hidden Markov Models
This paper introduces Hidden Markov Models (HMM) for modelling user’s search behaviour in a digital library search scenario. Three potential applications of these models have been identified in the literature: System evaluation, simulation of user activity and user guidance. Models have been generated from 36 eye tracking log files, where smaller models up to five hidden states provided best pr...
متن کامل