ID 原文 译文
9244 实验结果表明,该方法学习效率更高,同时能取得较好的本体映射结果。 The experimental results show that the learning efficiency is higher, and can obtain good ontology mapping results.
9245 当无线传感器网络(wireless sensor network,WSN)采用概率覆盖模型时,难以采用几何方法进行网络覆盖率的优化。 When the wireless sensor network (wireless sensor network, WSN) using probability coverage model, it is difficult to using geometry optimization for network coverage.
9246 针对这一问题,通过提出一种改进粒子群优化(particle swarm optimization,PSO)算法,有效提高了WSN网络的覆盖率。 ‭In order to solve this problem, through an improved particle swarm optimization (particle swarm optimization, PSO) algorithm, and effectively improve the network coverage of WSN.
9247 首先对粒子越界处理的方法进行推了广,提高了其适用范围;‭ ‭Crossing the line of the particles is first processing method to push the wide, improved its applicable scope;
9248 其次,针对PSO算法容易陷入局部最优解的问题,通过对粒子探索能力进行增强,提出了一种探索能力增强型PSO(explorative capability enhancement PSO,ECE-PSO)算法,有效改善了种群陷入局部最优解的缺点。 ‭Secondly, in view of the PSO algorithm is easy to fall into local optimal solution of the problem, through the particles to enhance its capability of exploration, this paper proposes a exploring ability enhanced PSO (explorative capability enhancement PSO, ECE - PSO) algorithm, and effectively improve the population into the shortcoming of local optimum solution. ‭
9249 基于概率覆盖模型的WSN覆盖优化的仿真验证表明,ECE-PSO算法显著提高了解的质量,有效改善了算法收敛于局部最优解的缺点,且ECEPSO算法具有较强的稳定性。 WSN coverage optimization based on probability coverage model simulation showed that the ECE - PSO algorithm significantly improve the quality of knowledge, effectively improves the algorithm converges to the shortcoming of local optimum solution, and ECEPSO algorithm has stronger stability.
9250 引入专家知识已成为小数据集条件下贝叶斯网络建模的主流方法,然而,专家知识是否正确直接决定了算法的结果和性能。 Introduction of expert knowledge has become a small data set under the condition of the mainstream of the bayesian network modeling method, however, the expert knowledge is correct directly determines the result and performance of the algorithm.
9251 因此,在考虑专家知识正确性的基础上,本文对贝叶斯网络结构学习问题展开研究。 Therefore, on the basis of considering the accuracy of expert knowledge, in this paper, the bayesian network structure learning problems.
9252 首先,建立一种基于连接概率分布的结构约束模型来表示专家知识,进而结合该约束模型对贝叶斯信息准则(Bayesian information criterions,BIC)评分进行改进; First, to establish a connection probability distribution based on the structural constraint model to represent the expert knowledge, and then combined with the constraint model of Bayesian information criterion (Bayesian information criterions, BIC) score was improved;
9253 最后,利用K2算法学习贝叶斯网络结构。 Finally, using the K2 algorithm for learning bayesian network structure.