Papers by Keyword: Dynamic Bayesian Network

Abstract: This paper describes a method for inferring user destinations and routes based on logs collected by smartphones. A challenging problem is coping with the uncertainty of smartphone sensor data. In this study, we represent a user transportation model with probabilistic models based on temporal smartphone sensor data, including GPS and accelerometer data. In our model, the travel behavior and spatio-temporal information of users are factors that affect route decisions. We propose hierarchical particle filters to enhance the performance and efficiency by sampling the route model based on hierarchical and semantic relationships.

Abstract: Being important in the field of information security, it is essential to study the major process and the mathematical models of information security risk assessment.This paper proposed amodel based on the discrete dynamic Bayesian network and further studied the major process of the assessment.The experiment which works on the implementing platform shows the results are reasonable and feasible. The model proves that it iseffective to calculate the probability of the risk.

Abstract: Facing asymmetric threats in a network centric environment, modern naval command and control systems confront increasingly demanding challenges in data fusion. It is very important to efficiently and promptly predict the enemy’s or adversary tactical intention from level 2 data fusion. In this paper, a layered intention model is proposed to represent the uncertain elements relating to adversarial intention and their uncertain relations in naval battlefield domain. The main ideal of this paper is to develop a hierarchical Bayesian network based on situation-specific Bayesian network (SSBN) and dynamic Bayesian network (DBN) that can be adapted to cope with the multi-timescales layered intention recognition problem.

Abstract: By now, Attack Graph (AG) is widely applied to the field of network security assessment. In the AG, each vertex has a value that implies the probability of the exploit and each edge represents the relationship between the exploits. In this paper we design an AG model and propose an approach which integrates the AG model with the Dynamic Bayesian Network (DBN). The approach not only strengthens the rationality of uncertain reasoning, but also provides a quantitative assessment of network security status. We evaluated the approach by experiment. The results showed that our model is rather accurate and the performance of it is competitive.

Abstract: With the rapid growth and wide application of electronic commerce, lots of information comes forth to people. However, our experiences and knowledge often do not enough to process the vast amount of information. The problem of obtaining useful information becomes more and more serious. To deal with the problem, the personalized service and recommender system play a more important role in many fields and collaborative filtering is one of the most successful technologies in recommender systems. However, with the tremendous growth in the amount of items and users, most collaborative filtering algorithms suffer from data sparsity which leads to inaccuracy of recommendation. Aiming at the problem of data sparsity, a hybrid personalized recommendation method based on dynamic Bayesian networks is presented. This method uses the dynamic Bayesian network technique to fill the vacant ratings at first, and then employs the user-based collaborative filtering to produce recommendations.

Abstract: When supply channels varied increasingly, key material supply forecasting has become indispensable to effective operations management. Rapid technological changes and an abundance of product configurations mean that the supply for key material is frequently volatile and hard to forecast. The paper describes a key material supply forecasting diagnostics tools based on Dynamic Bayesian Network (DBN). The tool was embodied parametric description of some important factors in key material supply forecasting. Furthermore, we developed this tool to pool supply patterns of little or no supply history data. Finally, we solve this reasoning problem with stochastic simulation.

Abstract: The paper focus on dynamic bayesian network (DBN) inference system and its application on the natural gas transmission and distribution network for the system fault diagnosis which is including time restriction (that is the earliest start time and the latest end time) for task. DBN is a graphic model which can describe the relationships between variable data, and used for reasoning. After we research the basic model of DBN, and the maximum depend algorithm is improved which which can increase the feasibility of natural gas transmission and distribution network system. we can see the new model is more accuracy than the traditional technical fault diagnosis clearly. Also, simulation result can verify the design of DBN is effectively and valuable.

Abstract: Dynamic Bayesian networks can be well dealt with the time-varying multivariable problem. The state model based on Dynamic Bayesian networks can more accurately describe the relationship between the system state and the influencing factors. In this paper, the width of the reasoning is used to simplify the amount of data in the reasoning process. Multi-step state prediction is achieved by extending time-slice. Experiment has shown that the proposed algorithm can achieve better prediction results.