Urban living in modern large cities has significant adverse effects on health, increasing the risk of several chronic diseases. We focus on the two leading clusters of chronic diseases, heart disease and diabetes, and develop data-driven methods to predict hospitalizations due to these conditions. We base these predictions on the patients' medical history, recent and more distant, as described in their Electronic Health Records (EHRs). We formulate the prediction problem as a binary classification problem and consider a variety of machine learning methods, including kernelized and sparse Support Vector Machines (SVMs), sparse logistic regression, and random forests. To strike a balance between accuracy and interpretability of the prediction, which is important in a medical setting, we propose two novel methods: K -LRT, a likelihood ratio test-based method, and a Joint Clustering and Classification (JCC) method which identifies hidden patient clusters and adapts classifiers to each cluster. We develop theoretical out-of-sample guarantees for the latter method. We validate our algorithms on large data sets from the Boston Medical Center, the largest safety-net hospital system in New England.

We consider the problem of detecting the formation of a set of wireless sensor nodes based on the pairwise measurements of signal strength corresponding to all transmitter/receiver pairs. We assume that formations take values in a discrete set and develop a composite hypothesis testing approach which uses a Generalized Likelihood Test (GLT) as the decision rule. The GLT distinguishes between a set of probability density function (pdf) families constructed using a custom pdf interpolation technique. The GLT is compared with the simple Likelihood Test (LT). We also adapt one prevalent supervised learning approach, Multiple Support Vector Machines (MSVMs), and compare it with our probabilistic methods. Due to the highly variant measurements from the wireless sensor nodes, and these methods' different adaptability to multiple observations, our analysis and experimental results suggest that GLT is more accurate and suitable for formation detection. The formation detection problem has interesting applications in posture detection with Wireless Body Area Networks (WBANs), which is extremely useful in health monitoring and rehabilitation. Another valuable application we explore concerns autonomous robot systems.

Many applications of machine learning involve analysis of sparse high-dimensional data, in which the number of input features is larger than the number of data samples. Standard inductive learning methods may not be sufficient for such data, and this provides motivation for nonstandard learning settings. This paper investigates a new learning methodology called learning through contradictions or Universum support vector machine (U-SVM). U-SVM incorporates a priori knowledge about application data, in the form of additional Universum samples, into the learning process. This paper investigates possible advantages of U-SVM versus standard SVM, and describes the practical conditions necessary for the effectiveness of the U-SVM. These conditions are based on the analysis of the univariate histograms of projections of training samples onto the normal direction vector of (standard) SVM decision boundary. Several empirical comparisons are presented to illustrate the practical utility of the proposed approach.