Modification of the Wu-Mendel approach for linguistic summarization

This paper presents versatile modification of the Wu-Mendel approach for linguistic summarisation and concentrates on eliminating the following main drawbacks of the approach: significant user/expert involvement, significant and rapidly increasing computational cost, some flaws in quality measures formulae. The techniques and formulae proposed to eliminate the aforesaid drawbacks are put through the experimental verification process involving several real-world datasets. The results of the experimental verification demonstrate the increased efficiency and effectiveness of the modified Wu-Mendel approach for linguistic summarisation.     

Multiphase topology optimization with a single variable using the phase-field design method

In this study, a multimaterial topology optimization method using a single variable is proposed by combining the solid isotropic material with penalization method and the reaction-diffusion equation. Unlike ordinary multimaterial optimization, which requires several variables depending on the number of material types, this method intends to represent various materials as one variable. The proposed method combines two special functions in the sensitivity analysis of the objective function to converge the design variable into prespecified density values defined for each of the multimaterials. The composition constraint based on a normal distribution function is also introduced to estimate the distribution of each target density value in a single variable. It enables density exchange between multiple materials by increasing or decreasing the amount of a specific material. The proposed method is applied to structural and electromagnetic problems to verify its effectiveness, and its usefulness is also confirmed from the viewpoint of cost and computation time.     

Optimal interval type-2 fuzzy fractional order super twisting algorithm: A second order sliding mode controller for fully-actuated and under-actuated nonlinear systems

In this paper, a novel interval type-2 fuzzy fractional order super twisting algorithm (IT2FFOSTA) which is essentially a second order sliding mode controller is presented. The proposed IT2FFOSTA enhances fractional order super twisting algorithm (FOSTA) by taking advantage of an interval type-2 fuzzy fractional order sliding surface (IT2FFOSS) for some classes of fully-actuated and under-actuated nonlinear systems in presence of uncertainty. The FOSTA significantly reduces the amount of chattering and the IT2FFOSS results in decreasing the tracking error, control effort, and chattering level. In order to control under-actuated systems, a hierarchical sliding surface is employed. The multi-tracker optimization algorithm is utilized to adjust the controller’s parameters; this leads to an optimal performance for the IT2FFOSTA. To examine the performance of the IT2FFOSTA, some simulation and experimental tests on three examples of different classes of fully-actuated and under-actuated systems, including ball and plate, inverted pendulum, and ball and beam systems are carried out. The simulation and experimental results demonstrate the superiority of the IT2FFOSTA in reducing the amount of chattering, tracking error, and control effort compared to those of the other control methods.     

基于(火用)效率目标的环冷机余热回收系统操作参数优化

摘要：针对烧结环冷机余热回收利用率不高的难题，采用分析法建立了评价某钢铁厂烧结环冷机余热回收系统运行效率的效率模型。基于多孔介质模型、局部非热平衡方程、真实气体SRK方程建立环冷机内气固两相换热模型。通过CFD仿真模拟，探究料层高度、循环风机输入烟气温度、烧结矿底部入口风速三项可控环冷机运行工艺参数对系统效率的影响规律。结果表明，料层厚度在1～1.5 m区间每增加0.1 m，效率增加0.8%～1.1%；循环风温在100～140℃之间每增加10℃，效率增加1.4%～1.5%；烧结矿底部入口风速在0.9～1.9 m/s之间每增加0.1 m/s，效率降低0.18%～0.24%。在此基础上，基于工业运行数据建立效率正交试验优化模型，提高了该余热回收系统3.42%的效率。     

采用信息增益率的混合入侵检测模型设计

针对现有混合入侵检测模型仅定性选取特征而导致检测精度较低的问题，同时为了充分结合误用检测模型和异常检测模型的优势，提出一种采用信息增益率的混合入侵检测模型.首先，利用信息增益率定量地选择特征子集，最大程度地保留样本信息；其次，采用余弦时变粒子群算法确定支持向量机参数构建误用检测模型，使其更好地平衡粒子在全局和局部的搜索能力，然后，选取灰狼算法确定单类支持向量机参数构建异常检测模型，以此来提高对最优参数的搜索效率和精细程度，综合提高混合入侵检测模型对攻击的检测效果；最后，通过两种数据集进行仿真实验，验证了所提混合入侵检测模型具有较好的检测性能.     

基于平面不封闭图形切割的路径规划

在平面切割环节中，如何确定更短切割路径以减少加工成本、降低设备损耗和提升切割质量是工业应用和学术研究的重点。目前国内外对平面切割路径的研究主要集中于封闭图形，为此，针对在激光刀模行业中不封闭图形的问题，提出一种基于禁忌搜索的贪婪算法和贪婪准则的局部优化。首先使用构建型的贪婪算法和改进型的禁忌搜索算法相结合的方式对加工过程中的图元路径进行优化排序，随后提出贪婪准则的局部优化系数，削弱贪婪算法的“贪心”思想，解决加工路径的规划和优化问题。实验数据表明，禁忌搜索的贪婪算法和局部优化在对切割路径的规划和空行程优化上有显著效果，空行程减少50%以上，并且其优化性能和图元数量成正比，能有效地解决刀模行业以及其他激光雕刻行业中图形不封闭的切割难题。     

基于PSO-ELM模型的潜油柱塞泵冲次优化方法

为改进油井后期泵效低下、能耗高等缺陷，提出了一种基于PSO-ELM模型的潜油柱塞泵冲次优化方法.采用粒子群（PSO）算法与极限学习机（ELM）相结合的方式来实现动液面软测量建模；根据动液面及潜油柱塞泵工作电流变化，以油井运行经济性最优为目的建立目标函数得到潜油柱塞泵冲次，解决了在油井生产时不能准确调节抽油机冲次问题；最后以目标函数关系建立模糊控制器模型，根据输入参数调整潜油柱塞泵冲次.实验结果表明，建立的软测量模型预测动液面精度高，模糊控制器能够更加合理地调整抽油机冲次，最终达到智能调整冲次大小、提高油井采油率及节能的目的.     

基于模糊谱聚类的不确定蛋白质相互作用网络功能模块挖掘

针对谱聚类融合模糊C-means（FCM）聚类的蛋白质相互作用（PPI）网络功能模块挖掘方法准确率不高、执行效率较低和易受假阳性影响的问题，提出一种基于模糊谱聚类的不确定PPI网络功能模块挖掘（FSC-FM）方法。首先，构建一个不确定PPI网络模型，使用边聚集系数给每一条蛋白质交互作用赋予一个存在概率测度，克服假阳性对实验结果的影响；第二，利用基于边聚集系数流行距离（FEC）策略改进谱聚类中的相似度计算，解决谱聚类算法对尺度参数敏感的问题，进而利用谱聚类算法对不确定PPI网络数据进行预处理，降低数据的维数，提高聚类的准确率；第三，设计基于密度的概率中心选取策略（DPCS）解决模糊C-means算法对初始聚类中心和聚类数目敏感的问题，并对预处理后的PPI数据进行FCM聚类，提高聚类的执行效率以及灵敏度；最后，采用改进的边期望稠密度（EED）对挖掘出的蛋白质功能模块进行过滤。在酵母菌DIP数据集上运行各个算法可知，FSC-FM与基于不确定图模型的检测蛋白质复合物（DCU）算法相比，F-measure值提高了27.92%，执行效率提高了27.92%；与在动态蛋白质相互作用网络中识别复合物的方法（CDUN）、演化算法（EA）、医学基因或蛋白质预测算法（MGPPA）相比也有更高的F-measure值和执行效率。实验结果表明，在不确定PPI网络中，FSC-FM适合用于功能模块的挖掘。     

Title加TextRank抽取关键句的情感分类研究

考虑到不同句子对判断文档情感倾向的重要程度不同，因而区分文档的关键句和细节句将有助于提高情感分类的性能。同时，考虑到Title和上下文信息，提出了一种基于Title和加权TextRank抽取关键句的情感分析方法SKTT，实现了高效的情感分析。根据文档Title的情感权重计算Title贡献度，考虑到标点和语义规则对情感倾向的影响；根据加权TextRank算法思想，在文档正文中构建了一个情感句有向图来提取关键句；计算所有关键句的情感倾向进行情感分类。在4个领域上进行实验，实验结果表明，该SKTT方法性能明显优于Baseline，具有高效性。     

A Dynamic Memory Allocation Optimization Mechanism Based on Spark

Spark is a distributed data processing framework based on memory. Memory allocation is a focus question of Spark research. A good memory allocation scheme can effectively improve the efficiency of task execution and memory resource utilization of the Spark. Aiming at the memory allocation problem in the Spark2.x version, this paper optimizes the memory allocation strategy by analyzing the Spark memory model, the existing cache replacement algorithms and the memory allocation methods, which is on the basis of minimizing the storage area and allocating the execution area according to the demand. It mainly including two parts: cache replacement optimization and memory allocation optimization. Firstly, in the storage area, the cache replacement algorithm is optimized according to the characteristics of RDD Partition, which is combined with PCA dimension. In this section, the four features of RDD Partition are selected. When the RDD cache is replaced, only two most important features are selected by PCA dimension reduction method each time, thereby ensuring the generalization of the cache replacement strategy. Secondly, the memory allocation strategy of the execution area is optimized according to the memory requirement of Task and the memory space of storage area. In this paper, a series of experiments in Spark on Yarn mode are carried out to verify the effectiveness of the optimization algorithm and improve the cluster performance.