A probabilistic network is a graphical model that encodes probabilistic relationships between variables of interest. Such a model records qualitative influences between variables in addition to the numerical parameters of the probability distribution. As such it provides an ideal form for combining prior knowledge, which might be limited solely to experience of the influences between some of the variables of interest, and data. In this paper, we first show how data can be used to revise initial estimates of the parameters of a model. We then progress to showing how the structure of the model can be revised as data is obtained. Techniques for learning with incomplete data are also covered.

This paper uses Bayesian network models for that investigation. Bayesian networks, or directed acyclic graph (DAG) models have proved very useful in representing both causal and statistical hypotheses. The nodes of the graph represent vertices, directed edges represent direct influences, and the topology of the graph encodes statistical constraints. We will consider features of such models that can be determined from data under assumptions that are related to those routinely applied in experimental situations:

We present a hybrid constraint-based/Bayesian algorithm for learning causal networks in the presence of sparse data. The algorithm searches the space of equivalence classes of models (essential graphs) using a heuristic based on conventional constraintbased techniques. Each essential graph is then converted into a directed acyclic graph and scored using a Bayesian scoring metric. Two variants

We develop simple methods for constructing likelihoods and parameter priors for learning about the parameters and structure of a Bayesian network. In particular, we introduce several assumptions that permit the construction of likelihoods and parameter priors for a large number of Bayesian-network structures from a small set of assessments. The most notable assumption is that of likelihood equivalence, which says that data can not help to discriminate network structures that encode the same assertions of conditional independence. We describe the constructions that follow from these assumptions, and also present a method for directly computing the marginal likelihood of a random sample with no missing observations. Also, we show how these assumptions lead to a general framework for characterizing parameter priors of multivariate distributions. Keywords: Bayesian network, learning, likelihood equivalence, Dirichlet, normal-Wishart. 1 Introduction A Bayesian network is a graphical repres...

Abstract not found

Abstract. Extracting sentiments from unstructured text has emerged as an important problem in many disciplines. An accurate method would enable us, for example, to mine on-line opinions from the Internet and learn customers ’ preferences for economic or marketing research, or for leveraging a strategic advantage. In this paper, we propose a two-stage Bayesian algorithm that is able to capture the dependencies among words, and, at the same time, finds a vocabulary that is efficient for the purpose of extracting sentiments. Experimental results on the Movie Reviews data set show that our algorithm is able to select a parsimonious feature set with substantially fewer predictor variables than in the full data set and leads to better predictions about sentiment orientations than several state-of-the-art machine learning methods. Our findings suggest that sentiments are captured by conditional dependence relations

Abstract not found

this paper we will describe how to extend search algorithms developed for non-latent variable DAG models to the case of DAG models with latent variables. We will introduce two generalizations of DAGs, called mixed ancestor graphs (or MAGs) and partial ancestor graphs (or PAGs), and briefly describe how they can be used to search for latent variable DAG models, to classify, and to predict the effects of interventions in causal systems.

A tremendous debt of gratitude is due to the many people who assisted or contributed in different ways to this research. The Robotic Antarctic Meteorites Search project was a team effort, involving many people, all of who deserve credit for the project’s success. Firstly, thanks go to my thesis committee: Martial Hebert, Red Whittaker, Andrew Moore, Reid Simmons and Peter Cheeseman, for their time and effort. Martial Hebert, my academic advisor, has carefully nurtured this thesis and steadily kept me on track over the last few years. In spite of a busy schedule, he has always selflessly managed to make time to help make this research happen, write references, and generally smooth the way. All this in spite of not being directly involved with, or receiving direct credit for the successes of the Robotic Antarctic Meteorite Search project. Dimitrious Apostolopoulos, as project manager for the Robotic Antarctic Meteorite Search project tirelessly and consistently backed my research. Without his leadership and tenacity, the project would not have been completed and this research would not have been possible. William “Red ” Whittaker, as director of the Field Robotics Center, has supported this project and my research throughout. His tenacity and experience at handling funding agencies ensured essential financial support, and his counsel has been invaluable both in the preparation of this document and in furthering this

We propose a new scoring function for learning Bayesian networks from data using score search algorithms. This is based on the concept of mutual information and exploits some well-known properties of this measure in a novel way. Essentially, a statistical independence test based on the chi-square distribution, associated with the mutual information measure, together with a property of additive decomposition of this measure, are combined in order to measure the degree of interaction between each variable and its parent variables in the network. The result is a non-Bayesian scoring function called MIT (mutual information tests) which belongs to the family of scores based on information theory. The MIT score also represents a penalization of the Kullback-Leibler divergence between the joint probability distributions associated with a candidate network and with the available data set. Detailed results of a complete experimental evaluation of the proposed scoring function and its comparison with the well-known K2, BDeu and BIC/MDL scores are also presented.