ID	原文	译文
4643	所提构造方法得到的序列具有良好的自相关性和平衡性，扩大了现有理想平衡四进制序列的存在范围，为实际应用提供了更多性能优良的序列。	Through these constructions, all the almost quaternary sequences constructed are balanced and optimal. These constructed se-quences extend the existence range of the balanced optimal quaternary sequences and provide more optimal sequences for practical applications.
4644	针对无线传感器网络分簇路由协议所筛选簇头节点的位置分布不均衡及转发节点的数据传输路径不合理会加剧节点能量消耗、缩短网络生存周期的问题，提出一种基于改进粒子群优化算法的分簇路由协议。	Aiming at the problem that the location distribution of cluster head nodes filtered by wireless sensor networkclustering routing protocol was unbalanced and the data transmission path of forwarding nodes was unreasonable, whichwould increase the energy consumption of nodes and shorten the network life cycle, a clustering routing protocol basedon improved particle swarm optimization algorithm was proposed.
4645	在簇头选举过程中，通过定义节点的能量因子和位置均衡因子建立新的适应度函数，评估和选择更优的候选簇头节点；通过优化的自适应学习因子调整候选簇头节点的位置更新速度，扩大局部搜索并加快全局搜索的收敛速度。	In the process of cluster head election, a new fitness function was established by defining the energy factor and position equalization factor of the node, the better candidate cluster head node was evaluated and selected, the position update speed of the candidate cluster head nodes was adjustedby the optimized update learning factor, the local search and speeded up the convergence of the global search was ex-panded.
4646	根据转发节点与基站的距离确定采用单跳还是多跳传输方式，设计一种基于最小生成树的多跳方法，为转发节点数据传输选择最优的多跳路径。	According to the distance between the forwarding node and the base station, the single-hop or multi-hop trans-mission mode was adopted, and a multi-hop method was designed based on the minimum spanning tree to select an op-timal multi-hop path for the data transmission of the forwarding node.
4647	仿真测试结果表明，基于改进粒子群算法的分簇路由协议能够选举能量与位置更均衡的簇头节点和转发节点，缩短了网络的通信距离，	Simulation results show that the clustering routing protocol based on improved particle swarm optimization algorithm can elect cluster head nodes and forwarding nodeswith more balanced energy and location, which shortened the communication distance of the network.
4648	节点的能耗更低且更均衡，有效延长了网络生存周期。	The energy con-sumption of nodes is lower and more balanced, effectively extending the network life cycle.
4649	针对机器学习算法在应用中存在的问题，构建基于智能启发算法的机器学习模型优化体系。	Aiming at the problems existing in the application of machine learning algorithm, an optimization system of the machine learning model based on the heuristic algorithm was constructed.
4650	首先，介绍已有智能启发算法类型及其建模过程。	Firstly, the existing types of heuristic algo-rithms and the modeling process of heuristic algorithms were introduced.
4651	然后，从智能启发算法在机器学习算法中的应用，包括神经网络等参数结构优化、特征优化、集成约简、原型优化、加权投票集成和核函数学习等方面说明智能启发算法的优势。	Then, the advantages of the heuristic algorithmwere illustrated from its applications in machine learning, including the parameter and structure optimization of neuralnetwork and other machine learning algorithms, feature optimization, ensemble pruning, prototype optimization,weighted voting ensemble and kernel function learning.
4652	最后，结合实际需求展望智能启发算法及在机器学习领域的发展方向。	Finally, the heuristic algorithms and their development directionsin the field of machine learning were given according to the actual needs.