We present methods for inferring the cost of interrupting users based on multiple streams of events including information generated by interactions with computing devices, visual and acoustical analyses, and data drawn from online calendars. Following a review of prior work on techniques for deliberating about the cost of interruption associated with notifications, we introduce methods for learning models from data that can be used to compute the expected cost of interruption for a user. We describe the Interruption Workbench, a set of event-capture and modeling tools. Finally, we review experiments that characterize the accuracy of the models for predicting interruption cost and discuss research directions.

The intuition that different text classifiers behave in qualitatively different ways has long motivated attempts to build a better metaclassifier via some combination of classifiers. We introduce a probabilistic method for combining classifiers that considers the context-sensitive reliabilities of contributing classifiers. The method harnesses reliability indicators—variables that provide signals about the performance of classifiers in different situations. We provide background, present procedures for building metaclassifiers that take into consideration both reliability indicators and classifier outputs, and review a set of comparative studies undertaken to evaluate the methodology.

We present a multi-camera system based on Bayesian modality fusion to track multiple people in an indoor environment. Bayesian networks are used to combine multiple modalities for matching subjects between consecutive image frames and between multiple camera views. Unlike other occlusion reasoning methods, we use multiple cameras in order to obtain continuous visual information of people in either or both cameras so that they can be tracked through interactions. Results demonstrate that the system can maintain people’s identities by using multiple cameras cooperatively. 1.

Abstract. Robust visual tracking has become an important topic in the field of computer vision. The integration of cues such as color, edge strength and motion has proved to be a promising approach to robust visual tracking in situations where no single cue is suitable. In this paper, an algorithm is presented which integrates multiple cues in a probabilistic manner. Specifically the likelihood of each cue is calculated and weighted before Bayes ’ rule is applied to obtain the resultant posterior. This posterior is generally not well represented analytically, and is therefore represented as a set of weighted particles, which is updated at each frame by a particle filter. This paper demonstrates how the combination of multiple cue integration and particle filtering results in a robust tracking method. We also demonstrate how each cue’s weight can be adapted on-line during the tracking procedure. 1

Tracking interacting human body parts from a single two-dimensional view is difficult due to occlusion, ambiguity and spatio-temporal discontinuities. We present a Bayesian network method for this task. The method is not reliant upon spatio-temporal continuity, but exploits it when present. Our inferencebased tracking model is compared with a CONDENSATION model augmented with a probabilistic exclusion mechanism. We show that the Bayesian network has the advantages of fully modelling the state space, explicitly representing domain knowledge, and handling complex interactions between variables in a globally consistent and computationally effective manner. 1

Video analysis of human dynamics is an important area of research devoted to detecting people and understanding their dynamic physical behavior in a complex environment that can be used for biometric applications. This paper provides a detailed survey of the various studies in areas related to the tracking of people and body parts such as face, hands, fingers, legs, etc., and modeling behavior using motion analysis. 1.

Most existing methods for text categorization employ induction algorithms that use the words appearing in the training documents as features. While they perform well in many categorization tasks, these methods are inherently limited when faced with more complicated tasks where external knowledge is essential. Recently, there have been efforts to augment these basic features with external knowledge, including semi-supervised learning and transfer learning. In this work, we present a new framework for automatic acquisition of world knowledge and methods for incorporating it into the text categorization process. Our approach enhances machine learning algorithms with features generated from domain-specific and common-sense knowledge. This knowledge is represented by ontologies that contain hundreds of thousands of concepts, further enriched through controlled Web crawling. Prior to text categorization, a feature generator analyzes the documents and maps them onto appropriate ontology concepts that augment the bag of words used in simple supervised learning. Feature generation is accomplished through contextual analysis of document text, thus implicitly performing word sense disambiguation. Coupled with the ability to generalize concepts using the ontology, this approach addresses two significant problems in natural language processing—synonymy and polysemy. Categorizing documents with the aid of knowledge-based features leverages information that cannot be deduced from the training documents alone. We applied our methodology using the Open Directory Project, the largest existing Web directory built by over 70,000 human editors. Experimental results over a range of datasets confirm improved performance compared to the bag of words document representation.

Abstract. This paper describes a unified approach, based on Gaussian Processes, for achieving sensor fusion under the problematic conditions of missing channels and noisy labels. Under the proposed approach, Gaussian Processes generate separate class labels corresponding to each individual modality. The final classification is based upon a hidden random variable, which probabilistically combines the sensors. Given both labeled and test data, the inference on unknown variables, parameters and class labels for the test data is performed using the variational bound and Expectation Propagation. We apply this method to the challenge of classifying a student’s interest level using observations from the face and postures, together with information from the task the students are performing. Classification with the proposed new approach achieves accuracy of over 83%, significantly outperforming the classification using individual modalities and other common classifier combination schemes. 1

Machine-learning researchers face the omnipresent challenge of developing predictive models that converge rapidly in accuracy with increases in the quantity of scarce labeled training data. We introduce Layered Abstraction-Based Ensemble Learning (LABEL), a method that shows promise in improving generalization performance by exploiting additional labeled data drawn from related discrimination tasks within a corpus and from other corpora. LA-BEL first maps the original feature space, targeted at predicting membership in a specific topic, to a new feature space aimed at modeling the reliability of an ensemble of text classifiers. The resulting abstracted representation is invariant across each of the binary discrimination tasks, allowing the data to be pooled. We then construct a context-sensitive combination rule for each task using the pooled data. Thus, we are able to more accurately model domain structure which would not have been possible using only the limited labeled data from each task separately. Using several corpora for an empirical evaluation of topic classification accuracy of text documents, we demonstrate that LABEL can increase the generalization performance across a set of related tasks.

The solutions to many vision problems involve integrating measurements from multiple sources. Most existing methods rely on a hidden assumption, i.e., these measurements are consistent. In reality, unfortunately, this may not hold. The fact that naively fusing inconsistent measurements amounts to failing these methods indicates that this is not a trivial problem. This paper presents a novel approach to handling it. A new theorem is proven that gives two algebraic criteria to examine the consistency and inconsistency. In addition, a more general criterion is presented. Based on the theoretical analysis, a new information integration method is proposed and leads to encouraging results when applied to the task of visual tracking. 1