Statistical applications in fields such as bioinformatics, information retrieval, speech processing, image processing and communications often involve large-scale models in which thousands or millions of random variables are linked in complex ways. Graphical models provide a general methodology for approaching these problems, and indeed many of the models developed by researchers in these applied fields are instances of the general graphical model formalism. We review some of the basic ideas underlying graphical models, including the algorithmic ideas that allow graphical models to be deployed in large-scale data analysis problems. We also present examples of graphical models in bioinformatics, error-control coding and language processing. Key words and phrases: Probabilistic graphical models, junction tree algorithm, sum-product algorithm, Markov chain Monte Carlo, variational inference, bioinformatics, error-control coding.

This paper introduces a hierarchical Markov model that can learn and infer a user’s daily movements through the community. The model uses multiple levels of abstraction in order to bridge the gap between raw GPS sensor measurements and high level information such as a user’s mode of transportation or her goal. We apply Rao-Blackwellised particle filters for efficient inference both at the low level and at the higher levels of the hierarchy. Significant locations such as goals or locations where the user frequently changes mode of transportation are learned from GPS data logs without requiring any manual labeling. We show how to detect abnormal behaviors (e.g. taking a wrong bus) by concurrently tracking his activities with a trained and a prior model. Experiments show that our model is able to accurately predict the goals of a person and to recognize situations in which the user performs unknown activities.

We present a method of learning a Bayesian model of a traveler moving through an urban environment. This technique is novel in that it simultaneously learns a unified model of the traveler's current mode of transportation as well as his most likely route, in an unsupervised manner. The model is implemented using particle filters and learned using Expectation-Maximization. The training data is drawn from a GPS sensor stream that was collected by the authors over a period of three months. We demonstrate that by adding more external knowledge about bus routes and bus stops, accuracy is improved.

In sequence modeling, we often wish to represent complex interaction between labels, such as when performing multiple, cascaded labeling tasks on the same sequence, or when longrange dependencies exist. We present dynamic conditional random fields (DCRFs), a generalization of linear-chain conditional random fields (CRFs) in which each time slice contains a set of state variables and edges---a distributed state representation as in dynamic Bayesian networks (DBNs)---and parameters are tied across slices. Since exact

Most probabilistic inference algorithms are specified and processed on a propositional level. In the last decade, many proposals for algorithms accepting first-order specifications have been presented, but in the inference stage they still operate on a mostly propositional representation level. [Poole, 2003] presented a method to perform inference directly on the first-order level, but this method is limited to special cases. In this paper we present the first exact inference algorithm that operates directly on a first-order level, and that can be applied to any first-order model (specified in a language that generalizes undirected graphical models). Our experiments show superior performance in comparison with propositional exact inference. 1

Learning patterns of human behavior from sensor data is extremely important for high-level activity inference. We show how to extract a person’s activities and significant places from traces of GPS data. Our system uses hierarchically structured conditional random fields to generate a consistent model of a person’s activities and places. In contrast to existing techniques, our approach takes high-level context into account in order to detect the significant places of a person. Our experiments show significant improvements over existing techniques. Furthermore, they indicate that our system is able to robustly estimate a person’s activities using a model that is trained from data collected by other persons. 1

Many Web sites, especially those that dynamically generate HTML pages to display the results of a user’s query, present information in the form of list or tables. Current tools that allow applications to programmatically extract this information rely heavily on user input, often in the form of labeled extracted records. The sheer size and rate of growth of the Web make any solution that relies primarily on user input is infeasible in the long term. Fortunately, many Web sites contain much explicit and implicit structure, both in layout and content, that we can exploit for the purpose of information extraction. This paper describes an approach to automatic extraction and segmentation of records from Web tables. Automatic methods do not require any user input, but rely solely on the layout and content of the Web source. Our approach relies on the common structure of many Web sites, which present information as a list or a table, with a link in each entry leading to a detail page containing additional information about that item. We describe two algorithms that use redundancies in the content of table and detail pages to aid in information extraction. The first algorithm encodes additional information provided by detail pages as constraints and finds the segmentation by solving a constraint satisfaction problem. The second algorithm uses probabilistic inference to find the record segmentation. We show how each approach can exploit the web site structure in a general, domain-independent manner, and we demonstrate the effectiveness of each algorithm on a set of twelve Web sites. 1.

Abstract. We present an automated transportation routing system, called “Opportunity Knocks, ” whose goal is to improve the efficiency, safety and independence of individuals with mild cognitive disabilities. Our system is implemented on a combination of a Bluetooth sensor beacon that broadcasts GPS data, a GPRS-enabled cell-phone, and remote activity inference software. The system uses a novel inference engine that does not require users to explicitly provide information about the start or ending points of their journeys; instead this information is learned from users ’ past behavior. Futhermore, we demonstrate how route errors can be detected and how the system helps to correct the errors with real-time transit information. In addition we present a novel solution to the problem of labeling positions with place names. 1

We present a probabilistic generarive model for timing deviations in expressive music performance. The structure of the proposed model is equivalent to a switching state space model. The switch variables correspond to discrete note locations as in a musical score. The continuous hidden variables denote the tempo. We formulate two well known music recognition problems, namely tempo tracking and automatic transcription (rhythm quantization) as filtering and maximum a posteriori (MAP) state estimation tasks. Ex- act computation of posterior features such as the MAP state is intractable in this model class, so we introduce Monte Carlo methods for integration and optimization. We compare Markov Chain Monte Carlo (MCMC) methods (such as Gibbs sampling, simulated annealing and iterative improvement) and sequential Monte Carlo methods (particle filters). Our simulation results suggest better results with sequential methods. The methods can be applied in both online and batch scenarios such as tempo tracking and transcription and are thus potentially useful in a number of music applications such as adaptive automatic accompaniment, score typesetting and music information retrieval.

Context-aware applications are applications that implicitly take their context of use into account by adapting to changes in a user's activities and environments. No one has more intimate knowledge about these activities and environments than end-users themselves. Currently there is no support for end-users to build context-aware applications for these dynamic settings. To address this issue, we present a CAPpella, a programming by demonstration Context-Aware Prototyping environment intended for end-users. Users "program " their desired context-aware behavior (situation and associated action) in situ, without writing any code, by demonstrating it to a CAPpella and by annotating the relevant portions of the demonstration. Using a meeting and medicine-taking scenario, we illustrate how a user can demonstrate different behaviors to a CAPpella. We describe a CAPpella's underlying system to explain how it supports users in building behaviors and present a study of 14 endusers to illustrate its feasibility and usability.