论文部分内容阅读
贝叶斯网络是以概率理论为基础的不确定知识表示模型,联合树算法是一种应用广泛的贝叶斯网络推理算法。提出了基于邻接点优先的联合树算法,从图模型和计算效率两个方面对联合树算法(JT)和基于图的邻接点优先的联合树(AD-JT)算法进行推理时间的比较,实验表明:基于图的邻接点优先的联合树算法能够有效地处理大规模数据,极大地减少了消耗时间,计算效率有显著改进。
Bayesian network is an indefinite knowledge representation model based on probability theory. Joint tree algorithm is a widely used Bayesian network reasoning algorithm. A joint tree algorithm based on the contiguous point prioritization is proposed to compare the reasoning time between the joint tree algorithm (JT) and the graph-based adjacency-first joint tree algorithm (AD-JT) from the aspects of graph model and computational efficiency. It is shown that the joint tree algorithm based on graph-based adjacency-point prioritization can effectively deal with large-scale data, greatly reducing the elapsed time and improving computational efficiency.