一种基于粗糙集理论的规则获取算法

获取规则是数据挖掘中的一项重要技术 ,根据粗糙集理论及决策值归纳函数的概念 ,可以把不相容的决策系统转化为相容的决策系统 ,并提出规则参数的合并方法 .在此基础上 ,利用决策矩阵和决策函数 ,提出了一种在决策系统中获取规则的算法 .同传统的算法相比 ,该算法得出的规则集没有信息丢失的现象发生 .最后以例子作了说明 .     

基于邻接网络的频繁项目集发现算法

频繁项目集发现一直都是关联规则研究领域中最繁项目集发现算法,该算法的特别之处在于事先利用有向图进行的一次数据预处理,在预处理过程中将数据库预先存贮为每个结点都有一个域来记录其支持度的项目集邻接网络,从而把复杂的频繁项目集的发现问题转化为简单的图中搜索问题,这就大大提高了频繁项目集发现过程的效率.同时为了有效地解决预处理过程中的项目集支持度计算问题,采用了一种纵向的数据库表示格式.最后对所采用的算法给出实验结果.     

融合视角下创业生态系统阶段特征演化研究：基于深圳上市公司的文本分析

融合创业学、生态隐喻与系统论3个理论视角,建构创业生态系统特征分析框架,并以深圳证券交易所271家本地上市公司的年报作为文本分析样本,在识别深圳创业生态系统演化阶段的同时对特征框架进行验证。研究表明：创业机会集创新程度、生态位重叠程度以及主体间共生程度,是创业生态系统阶段特征演进的判别标准;深圳创业生态系统遵循“生态位低重叠—创造型机会集—线共生”→“生态位高重叠—复制型机会集—面共生”→“生态位低重叠—改进型机会集—体共生”的特征演化路径。     

一种求解双目标flow shop排序问题的进化算法

提出一种求解双目标flow shop排序的递进多目标进化算法.算法采用改进的精英复制策略,在实现精英保留的前提下降低了计算复杂性;通过递进进化模式增加群体多样性,改善了算法收敛性;通过群体进化过程中对非劣解集进行竞争型可变邻域启发式搜索,增强了算法局部搜索性能.采用新算法和参照算法NSGA-II对31个标准双目标flow shop算例进行优化.研究结果表明,新算法在所有算例的求解中均获得了优于NSGA-II的非劣解集,验证了算法的有效性.     

航空票务公司免费接送服务中的车次调度模型及算法

本文从顾客满意度和成本出发，对航空票务公司免费接送顾客的车次分配和调度问题进行研究，首先分析了票务公司免费接送服务的业务流程；然后给出了基于时间的顾客满意度函数，并建立了在一定顾客满意度下成本最小的车次调度模型；最后利用集划分方法求解该模型。集划分包括以下两步：第一步采用广度优先搜索生成全部车次集；第二步提出了一种启发式算法。通过仿真分析比较，验证了启发式算法的有效性。     

一种基于PIM与核方法的模糊聚类新算法

本文针对模糊C均值聚类在大数据量时收敛较慢以及不能对多种数据结构有效聚类的缺点,结合PIM算法与核方法提出了一种新的高效聚类算法———KPIM算法,并从理论上证明了该算法的收敛性.最后利用标准实验数据IRIS数据集测试,结果表明KPIM算法在保证收敛速度的同时,聚类效果更有效.     

基于动态分类器集成选择的不完整数据客户分类方法实证研究

在管理决策的制定中,分类已经成为一种十分重要的方法和技术。由于现实客户数据常常是不完整的,因此,研究不完整数据的客户分类问题具有重要意义。通过分析以往分类过程中对不完整数据的处理方法,提出了一种基于动态分类器集成选择的不完整数据分类方法DCES-ID。分别在UCI客户分类数据集以及某券商客户数据集上进行分类的实验和实证分析。结果表明,与已有的6种分类算法相比,DCES-ID算法具有更高的分类准确性及稳定性,能够更有效地进行客户分类。     

共生营销策略探析

<正> 一 共生营销及其主要形式 共生营销作为一种全新的营销策略,成为企业发展的新支点。所谓共生营销,美国管理学专家阿德勒这样定义:它是通过两个或更多个相互独立的商业组织在资源或项目上的合作,达到增强市场竞争力的一种营销行为,它的主要特征是互利性、互助性、互补性。共生营销主要有以下几种形式:     

一种基于Rough集的客户流失风险分析方法

为了对客户流失风险分析过程中的大量的冗余特征进行约简或压缩,本文利用粗集理论中的特征约简方法来研究客户流失风险分析,提出了一种基于粗集的客户流失风险的分析方法.通过一个客户流失风险的分析实例对该算法进行了检验,实验结果表明,在保证分类质量基本不变的情况下,该算法可以查找出对分类起主要作用的特征,从而达到了特征约简的目的,成功地将粗集理论应用到客户流失风险的分析和预警中,为客户流失风险的分析和预警提供了一条新的研究思路和方法.     

基于PSO和SVM的上市公司财务危机预警模型

提出了一种将经过改进的离散粒子群(PSO)算法和支持向量机(SVM)相结合的算法,以选择最优的指标集并用于财务危机预警。将此算法应用于上市公司的数据,检验模型提前3年的预警效果,最后与常见的主成分分析方法与SVM相结合的模型进行对比,结果证明了PSO-SVM模型的合理性和优越性。     

A novel representation of graph structures in web mining and data analysis

The paper presents a new graph representation, the graph matrix, which combines the adjacency matrix with the linked lists allowing for the fastest possible access to different types of information on a graph. This is increasingly important for a high search performance, for instance, for rapidly extracting information from the link structure in a hub and authority graph of the World-Wide-Web. A very recent application for the proposed data structure arises from categorical data clustering defining proximity and similarity of data through their patterns of co-occurrence.     

A Two-Phase Exact Algorithm for MAX-SAT and Weighted MAX-SAT Problems

We describe a two-phase algorithm for MAX-SAT and weighted MAX-SAT problems. In the first phase, we use the GSAT heuristic to find a good solution to the problem. In the second phase, we use an enumeration procedure based on the Davis-Putnam-Loveland algorithm, to find a provably optimal solution. The first heuristic stage improves the performance of the algorithm by obtaining an upper bound on the minimum number of unsatisfied clauses that can be used in pruning branches of the search tree.We compare our algorithm with an integer programming branch-and-cut algorithm. Our implementation of the two-phase algorithm is faster than the integer programming approach on many problems. However, the integer programming approach is more effective than the two-phase algorithm on some classes of problems, including MAX-2-SAT problems.     

无人仓系统储位分配问题的优化模型与算法

考虑到无人仓系统补货阶段货架上只有部分空余储位的特点，研究了补货商品储位分配问题的优化模型与算法。以同一货架上存放的商品之间关联度之和最大化为目标建立了混合整数规划模型；结合贪婪算法和邻域搜索算法设计了求解模型的两阶段方法。第一阶段利用贪婪算法求初始可行解；第二阶段利用邻域搜索算法对初始可行解进行优化。利用一个具体算例验证了邻域搜索算法的优化效果，结果显示，通过邻域搜索算法对初始可行解的优化，可以使目标函数值至少提升27%。进一步利用多个小规模算例分析了两阶段算法的近似比和求解速度，验证了算法的快速有效性。本文的研究结果不仅解决了货架初始状态非空情况下的储位分配问题，同样适合解决货架初始状态为空的情况，因此更加符合实际场景，可以作为无人仓管理信息系统的核心模型和算法。     

An online algorithm for continuous slab caster scheduling

Lee et al. (Lee, K., Chang, S.Y., and Hong, Y., 2004. Continuous slab caster scheduling and interval graphs. Production Planning & Control, 13 (5), 495–501) have introduced a slab caster scheduling problem and developed an optimal algorithm. Their algorithm is efficient but an offline algorithm that we need the information on all the customer orders a priori to implement. In this article, we propose an online algorithm that we can implement without knowledge of the orders yet to arrive. We show that the offline version of our new algorithm also provides an optimal solution and the online version has the worst case performance ratio of 3. We also give a short proof on the correctness of Lee et al.'s algorithm.     

投资项目集合选择问题的非线性规划模型与解法研究

基于项目集合选择问题的定义,给出了项目集合选择问题求解的一般步骤。依据投资方案组合选择问题的非线性特性,构建了投资项目集合选择问题的非线性规划模型,在此模型的基础上提出了基于外点法求解此类问题的改进贪婪搜索算法。研究了采用surrogate松弛模型确定初始点和运用改进的贪婪算法搜索最优解的具体实现方法,给出了实现算法的具体步骤。     

The hierarchical model for load balancing on two machines

Following previous work, we consider the hierarchical load balancing model on two machines of possibly different speeds. We first focus on maximizing the minimum machine load and show that no competitive algorithm exists for this problem. We overcome this barrier in two ways, both related to previously known models. The first one is fractional assignment, where each job can be arbitrarily split between the machines. The second one is a semi-online model where the sum of jobs is known in advance. We design algorithms of best possible competitive ratios for both these cases. Furthermore, we show that the combination of the two models leads to the existence of an optimal algorithm (i.e., an algorithm of competitive ratio 1). This algorithm is clearly optimal for the makespan minimization problem as well. For the latter problem, we consider the fractional assignment model and design an algorithm of best possible competitive ratio for it. This work was submitted as the M.Sc. thesis of the first author.     

Multi-neighborhood based path relinking for two-sided assembly line balancing problem

This paper presents a multi-neighborhood based path relinking algorithm (MN-PR) for solving the two-sided assembly line balancing problem. By incorporating an effective local search into a path relinking framework, the proposed MN-PR algorithm integrates a number of distinguishing features, such as a multi-neighborhood based local search procedure, a dedicated path relinking operator to generate new solutions and a strategy to fix an infeasible solution generated by the path relinking procedure to a feasible one. Our proposed MN-PR algorithm is tested on a set of totally 45 public instances widely used in the literature. Comparisons with other reference algorithms show the efficacy of the proposed algorithm in terms of the solution quality. Particularly, the proposed MN-PR algorithm is able to improve the best upper bounds for one instance with 65 tasks and 326 cycle time. This paper also presents an analysis to show the significance of the main components of the proposed algorithm.     

A pairwise interchange algorithm for parallel machine scheduling

This paper considers the problem of non-preemptive scheduling n tasks on m identical parallel processors to minimize makespan for simultaneous arrivals. Based on a pairwise interchange method, an efficient algorithm ispresented which is able to give a near-optimal schedule in a short time through suitable pairwise interchange between tasks, after an initial solution is constructed. The behaviour of the algorithm is discussed. Testing results prove its high performance in comparison with available simple heuristic procedures. Finally, the algorithm is generalized for the problems of non-identical processors and non-simultaneous arrivals.