Graphical models for probabilistic reasoning are now in widespread use. Many approaches have been developed such as Bayesian network. A newly developed approach named as dynamic uncertain causality graph (DUCG) is initially presented in a previous paper, in which only the inference algorithm in terms of individual events and probabilities is addressed. In this paper, we first explain the statistic basis of DUCG. Then, we extend the algorithm to the form of matrices of events and probabilities. It is revealed that the representation of DUCG can be incomplete and the exact probabilistic inference may still be made. A real application of DUCG for fault diagnoses of a generator system of a nuclear power plant is demonstrated, which involves > 600 variables. Most inferences take < 1 s with a laptop computer. The causal logic between inference result and observations is graphically displayed to users so that they know not only the result, but also why the result obtained.

With the advancement of agricultural modernization, many suppliers of agricultural means of production have delivery delay problems and have created environmental pollution and other issues, which affect the coordination and overall efficiency of the agricultural supply chain. Focusing on the green suppliers, this paper puts forward a series of evaluation indexes and considers the influence of environmental performance for performing uncertainty event reasoning based on a Bayesian network – establishing a complete selection and evaluation system for retail enterprises and downstream customers. In addition, an improved genetic algorithm is combined with the Bayesian approach to quantify the evaluation indicators, which solves the problems of the traditional methods of information occlusion and an unreasonable selection scheme, and provides an intelligent and efficient selection of green suppliers.

Brain activity is a dynamic combination of different sensory responses and thus brain activity/state is continuously changing over time. However, the brain's dynamical functional states recognition at fast time-scales in task fMRI data have been rarely explored. In this work, we propose a novel 5-layer deep sparse recurrent neural network (DSRNN) model to accurately recognize the brain states across the whole scan session. Specifically, the DSRNN model includes an input layer, one fully-connected layer, two recurrent layers and a softmax output layer. The proposed framework

No abstract

With the popularization of the online to offline (O2O) e-commerce on fresh food products, how to control the quality is becoming increasingly important. To adequately address this problem, this paper presents a fuzzy Bayesian network model for effectively controlling the quality in O2O ecommerce. Reasoning about uncertain events and incomplete data through an intelligent simulation with Bayesian networks provides a convenient and fast method of evaluation and analysis for e-commerce platforms to quickly select fresh food suppliers. Such a model is capable of appropriately modelling the uncertainty inherent in the fresh food product distribution process. It focuses on the identification of the critical factors that affect the food product quality along the supply chain. This leads to the development of a complete selection and evaluation system for the quality in O2O e-commerce. A simulation study is conducted that shows the proposed model is applicable for effectively controlling the quality in O2O e-commerce. Ultimately, the unloading level, warehouse inspection and warehouse monitoring are determined as the entry points for quality control, with corresponding degrees of influence of 44%, 37%, and 34%. The main points to protect the quality of food are introduced, which provides a theoretical basis for solving fresh food safety problems for business platforms.