大规模语料中频繁模式增量发现算法

提出一种适用于大规模语料的频繁模式增量发现算法。统计局部区域提取的字符串频度,对局部相对低频字符串进行剪枝。利用多模式串匹配算法,统计剪枝后局部相对高频字符串在整个语料中的频度,得到频度大于阈值的频繁模式。实验结果表明,该算法具有较低的空间复杂度和时间复杂度,内存消耗为基于后缀数组的频繁模式发现算法的20%左右。     

数据流优化匹配在数字海洋中的应用

提出一种利用图形处理单元(GPU)协同CPU的实时数据流分析匹配机制,将海洋实时数据流与国家标准数据进行比对,分析水质是否异常.该方法将存储到分布式基站上的数据分类规整并阶段化存储,接受数据流匹配请求后,在实时数据流匹配分析系统中利用GPU进行匹配分析处理,并自动生成分析结果.实验结果表明,该方法可以缩短海洋监测的检测...     

一种基于子串识别的多模式串匹配算法

多模式串匹配算法是网络内容过滤系统的核心技术。巨大的存储空间开销是制约多模式匹配串算法应用的瓶颈之一。提出一种基于子串识别的多模式匹配算法—HashBOM,该算法利用位哈希表存储模式串的子串信息以大幅度减少存储空间,利用递归哈希函数计算字符串的哈希值以实现快速匹配。理论分析表明,该算法的空间复杂度为O(rm~2),优于基于子串识别的匹配算法BOM的空间复杂度O(mr|∑|log_2mr);该算法搜索匹配过程的平均时间复杂度为O(nlog|∑|)mr/m,与BOM算法相同(其中m为最短模式串的长度,r为模式串的个数,n为待匹配文本的长度,|∑|为字母表的大小)。在随机数据集和真实数据集上的实验表明,该算法的存储空间远远低于BOM算法,而匹配速度与BOM算法相当,非常适合在线实时匹配的应用环境。     

基于二维网状结构的应用层组播协议研究

为了改善网状优先的应用层组播协议可扩展性差等缺点,提出一种新的应用组播协议,协议采用特殊的二维层次结构。在低层,将距离近的结点划分为同一个域,域内成员组织成车轮状,域成员只和邻居结点交换信息,减少了维护通信量。在高层,将距离较远的域首组织成网状结构,上下层通过不同的路由算法并行地传输数据,使得数据可以尽快到达远处的结点,提高了传输效率。仿真结果表明,该组播协议具有较小的控制开销、链路压力和端到端延迟。     

An artificial negotiating agent modeling approach embedding dynamic offer generating and cognitive layer

In this paper, a dynamic offer generating unit and cognitive layer are suggested for artificial agents based negotiation systems. For this purpose, first, adaptive time and behavior dependent tactics are developed taking advantages from time continuity and dynamics aspects (features) integrated in their modeling. Then, a negotiation strategy (bilateral over single issue) based on these two tactics is suggested. Second, a cognitive negotiation model for a negotiator agent is developed using Win-Lose and Win-Win orientations which will be formed based on personality factors. Afterwards, an experimental validation is conducted for testing applicability of time dependent tactics, the effect of offering time, and the effect of cognitive orientations (Win-Lose and Win-Win) on final negotiation outcomes. The results prove the applicability of the suggested time and behavior dependent tactics as well as the proposed cognitive negotiation model.     

Novel e-nose for the discrimination of volatile organic biomarkers with an array of carbon nanotubes (CNT) conductive polymer nanocomposites (CPC) sensors

We report the original design of a new type of electronic nose (e-nose) consisting of only five sensors made of hierarchically structured conductive polymer nanocomposites (CPC). Each sensor benefits from both the exceptional electrical properties of carbon nanotubes (CNT) used to build the conductive architecture and the spray layer by layer (sLbL) assembly technique, which provides the transducers with a highly specific 3D surface structure. Excellent sensitivity and selectivity were obtained by optimizing the amount of CNT with five different polymer matrices: poly(caprolactone) (PCL), poly(lactic acid) (PLA), poly(carbonate) (PC), poly(methyl methacrylate) (PMMA) and a biobased polyester (BPR). The ability of the resulting e-nose to detect nine organic solvent vapours (isopropanol, tetrahydrofuran, dichloromethane, n-heptane, cyclohexane, methanol, ethanol, water and toluene), as well as biomarkers for lung cancer detection in breath analysis, has been demonstrated. Principal component analysis (PCA) proved to be an excellent pattern recognition tool to separate vapour clusters.     

Screen-printed carbon electrode for choline based on MnO2 nanoparticles and choline oxidase/polyelectrolyte layers

This paper presents the amperometric biosensor that determines choline and cholinesterase activity using a screen printed graphite electrode. In order to detect H₂O₂ we have blanket modified the electrode material with manganese dioxide nanoparticles layer. Using layer-by-layer technique on the developed hydrogen peroxide sensitive electrode surface choline oxidase was incorporated into the interpolyelectrolyte nanofilm. Its ability to serve as a detector of choline in bulk analysis and cholinesterase assay was investigated. We examined the interferences from red-ox species and heavy metals in the blood and in the environmental sample matrixes. The sensor exhibited a linear increase of the amperometric signal at the concentration of choline ranging from 1.3 × 10⁻⁷ to 1.0 × 10⁻⁴ M, with a detection limit (evaluated as 3σ) of 130 nM and a sensitivity of 103 mA M⁻¹ cm⁻² under optimized potential applied (480 mV vs. Ag/AgCl). The biosensor retained its activity for more than 10 consecutive measurements and kept 75% of initial activity for three weeks of storage at 4 °C. The R.S.D. was determined as 1.9% for a choline concentration of 10⁻⁴ M (n = 10) with a typical response time of about 10 s. The developed choline biosensor was applied for butyrylcholinesterase assay showing a detection limit of 5 pM (3σ). We used the biosensor to develop the cholinesterase inhibitor assay. Detection limit for chlorpyrifos was estimated as 50 pM.     

基于语义相似度的Web服务匹配研究

现有的语义Web服务匹配算法没有考虑到本体概念间的多元关系,导致概念的语义不能被完整地反映出来,从而影响了算法的匹配性能。利用本体概念间的多元关系定义了一种语义距离,并通过该语义距离给出了概念间的语义相似度计算方法,在此基础上提出基于语义相似度的Web服务匹配算法。该算法通过本体概念间的语义相似度来反映Web服务的匹配程度。最后,通过对比实验验证了该算法的可行性和有效性。     

室性早搏的多模板匹配自适应识别算法

为实现室性早搏的自动判别,提出一种基于多模板匹配的自适应识别算法。采用滤波提高数据信噪比,利用样本数据通过特定的自学习机制建立模板库,设计模板的压缩算法实现待测心搏与模板心搏宽度的对齐,采用模板队列匹配和模板库匹配相结合的匹配策略,设计自适应更新规则实时优化队列与库的结构,利用相关系数检测室早。使用MIT-BIH数据库进行测试,达到99.39%的灵敏度和98.79%的特异度。