Lecture 6 The EM Algorithm , Mixture Models , and Motif
نویسندگان
چکیده
In a previous class, we discussed an algorithm for learning a probabilistic matrix model which describes a fixed-length motif in a set of sequences S : : : S over an alphabet A. This algorithm is one of a class of methods collectively known as expectation maximization, or EM. We will describe the general EM algorithm, then derive the motif-finding algorithm by applying EM to learn a specific probabilistic model of sequences with motifs. Finally, we will describe an EM-based algorithm for learning a partition of sequences into r disjoint classes, each described by its own matrix model.
منابع مشابه
Machine Learning for Data Science (CS 4786) Lecture 16-17: EM Algorithm: Why EM works!, EM for Gaussian Mixture Models and Mixture of Multinomials
متن کامل
Mixture Model based MAP Motif Discovering
In this paper a new maximum a posteriori (MAP) approach based on mixtures of multinomials is proposed for discovering probabilistic motifs in sequences. The main advantage of the proposed methodology is the ability to bypass the problem of overlapping motif occurrences among neighborhood positions in sequences through the use of a Markov Random Field (MRF) as a prior. This model consists of two...
متن کاملA marginal mixture model for discovering motifs in sequences
In this study we present a marginal mixture model for discovering probabilistic motifs in categorical sequences. The proposed method is based on a general framework for developing, extending and marginalizing expressive motif models that encapsulates spatial information. Two alternative schemes for constructing expressive models are described, the extend and the marginal approach. The EM algori...
متن کاملImage Segmentation using Gaussian Mixture Model
Abstract: Stochastic models such as mixture models, graphical models, Markov random fields and hidden Markov models have key role in probabilistic data analysis. In this paper, we used Gaussian mixture model to the pixels of an image. The parameters of the model were estimated by EM-algorithm. In addition pixel labeling corresponded to each pixel of true image was made by Bayes rule. In fact,...
متن کاملIMAGE SEGMENTATION USING GAUSSIAN MIXTURE MODEL
Stochastic models such as mixture models, graphical models, Markov random fields and hidden Markov models have key role in probabilistic data analysis. In this paper, we have learned Gaussian mixture model to the pixels of an image. The parameters of the model have estimated by EM-algorithm. In addition pixel labeling corresponded to each pixel of true image is made by Bayes rule. In fact, ...
متن کامل