Problems involving estimation and inference under linear inequality constraints arise often in statistical modeling. In this paper we propose an algorithm to solve the quadratic programming problem of minimizing ψ(θ) = θ′Qθ−2c′θ for positive-definite Q, where θ is constrained to be in a closed polyhedral convex cone C = {θ : Aθ ≥ d}, and the m×n matrix A is not necessarily full row-rank. The th...

This study isolates specific dimensions of service quality expectations and investigates how service-contact-personnel dress, as one of the dimensions, influences the other dimensions in the service context of banking. A conceptual model is proposed based on inference theory. Moderating variables of age and gender are incorporated. A 2 x 2 x 2 full factorial design is utilized with the independ...

Problem statement: Automated subway train-braking system require perfection, efficiency and fast response. In order to cope with this concerns, an appropriate algorithm need to be developed which need to be implemented in hardware for faster response. Approach: In this research, the FPGA realization of fuzzy based subway train braking system has been presented on an Alter FLEX10K device to prov...

Introduction: Epilepsy is a clinical syndrome in which seizures have a tendency to recur. Sodium valproate is the most effective drug in the treatment of all types of generalized seizures. Finding the optimal dosage (the lowest effective dose) of sodium valproate is a real challenge to all neurologists. In this study, a new approach based on Adaptive Neuro-Fuzzy Inference System (ANFIS) was pre...

We tackle the facial landmark localization problem as an inference problem over a Markov Random Field. Efficient inference is implemented using Gibbs sampling with approximated full conditional distributions in a latent variable model. This approximation allows us to improve the runtime performance 1000-fold over classical formulations with no perceptible loss in accuracy. The exceptional robus...

Stevens Institute of Technology Hoboken, NJ, USA {hwang,[email protected]} Abstract. We study the privacy threat by publishing data that contains full functional dependencies (FFDs). We show that the cross-attribute correlations by FFDs can bring potential vulnerability to privacy. Unfortunately, none of the existing anonymization principles can effectively prevent against the FFD-based priv...

Due to the proliferation of data networks, applications and services, distributed system management have to face the complexity, heterogeneity, uncertainty and dynamic changes in the managed domain. Multiagent technology provides an appropriate method to deal with the full distributed system because of its autonomy and cooperation. On the other side, the probability reasoning is needed for the ...