We develop a novel approach for supervised learning based on adaptively partitioning the feature space into different regions and learning local region-specific classifiers. We formulate an empirical risk minimization problem that incorporates both partitioning and classification in to a single global objective. We show that space partitioning can be equivalently reformulated as a supervised le...

ALTHOUGH HEALTH CARE HAS long been lagging behind other industries in adopting computer technologies, the health care industry is now using information technology in almost every health care sector across the globe. This has resulted in major transformations of the entire industry, aimed at maximizing the quality of medical care while minimizing costs.1 However, storing health care information ...

Multiple myeloma is a genetically complex, invariably fatal, neoplasm of plasma cells. Cytogenetic and other laboratory tests identify markers to classify newly diagnosed multiple myeloma patients into high, intermediate and standard clinical risk categories. The level of risk reflects the aggressiveness of the disease, and thus dictates the intensity of initial treatment. Thus, a riskadapted a...

In the Superset Label Learning (SLL) problem, weak supervision is provided in the form of a superset of labels that contains the true label. If the classifier predicts a label outside of the superset, it commits a superset error. Most existing SLL algorithms learn a multiclass classifier by minimizing the superset error. However, only limited theoretical analysis has been dedicated to this appr...

1942 Emerging Infectious Diseases • www.cdc.gov/eid • Vol. 20, No. 11, November 2014 restricted to plasma for fractionation. Some protection against blood donations from HEV-infected persons may occur because HEV and malaria are coendemic to many countries. Our findings showed a higher HEV seroprevalence among donors with prior malaria or diarrhea deferrals; thus, malariaand diarrhea-related sc...

Cloud computing is a distributed computing model that still faces problems. Many new innovative and emerge technique to take advantage of its features and among the research challenges found in the cloud. In this article, we explore the security threats and Risk Assessment to Cloud Computing, attack mitigation framework and the risk-based dynamic access control for cloud computing. Common secur...

This article describes one risk-adjustment method useful for minimizing threats to internal validity that stem from the impact on the outcomes under investigation from patients' own characteristics and their associated risks. Mortality is the outcome used to illustrate the risk-adjustment approach. A two-step approach to outcomes research is suggested. The first step includes risk-adjusting out...

In this paper, a customer incentive scheme is proposed for retailers to build an effective demand response program over the peak demand period to minimize the financial risk. Firstly, an objective function is formulated based on the market operation and an optimal incentive price is derived from this objective function. Secondly, the incentive price is employed as a part of an incentive scheme ...

In this paper, we consider the net loss of a life insurance company issuing identical equity-linked pure endowment contracts in the case of periodic premiums. Under this construction, financial risks as well as the mortality risk are included. Based on Møller (1998), we particularly investigate the situation where the company applies a time-discretized risk-minimizing hedging strategy, i.e., a ...