基于频繁模式树的普遍化关联规则挖掘

提出了基于频繁模式树的普遍化关联规则挖掘算法 MGAR- FP,充分利用频繁模式树的性质 ,避免大量候选模式的生成和频繁模式匹配 ,提高了挖掘的效率和速度 .实验表明 ,算法是有效的 ,比传统的普遍化关联规则挖掘算法Cum ulate快     

神经网络集成

神经网络集成通过训练多个神经网络并将成结论进行合成，可以显著地提高学习系统的泛化能力。它不仅有助于科学家对机器学习和神经的深入研究，还有助于普通工程技术人员利用神经网络技术来解决真实世界中的问题。因此，它被视为一种广阔应用前景的工程化神经计算技术，已经成为机器学习和神经计算领域的研究热点。该文从实现方法、理论分析和应用成果等三个方面综述了神经网络集成的国际研究现状，并对该领域值得进一步研究的一些问题进行了讨论。     

广义序偶的优化方法

本文在文献［１］的基础上引入了松驰量的概念，解决了在ＣＰＭ网络图中，在Ｎ＞２个平行工序中，预先指定一个工序的前提下，在剩下的Ｎ－１个工序中，选择一个工序与指定工序形成一个序偶的最优化问题，并举例说明了它的应用及该方法的优点。     

An Efficient Method To Estimate Bagging's Generalization Error

Bagging (Breiman, 1994a) is a technique that tries to improve a learning algorithm's performance by using bootstrap replicates of the training set (Efron & Tibshirani, 1993, Efron, 1979). The computational requirements for estimating the resultant generalization error on a test set by means of cross-validation are often prohibitive, for leave-one-out cross-validation one needs to train the underlying algorithm on the order of m times, where m is the size of the training set and is the number of replicates. This paper presents several techniques for estimating the generalization error of a bagged learning algorithm without invoking yet more training of the underlying learning algorithm (beyond that of the bagging itself), as is required by cross-validation-based estimation. These techniques all exploit the bias-variance decomposition (Geman, Bienenstock & Doursat, 1992, Wolpert, 1996). The best of our estimators also exploits stacking (Wolpert, 1992). In a set of experiments reported here, it was found to be more accurate than both the alternative cross-validation-based estimator of the bagged algorithm's error and the cross-validation-based estimator of the underlying algorithm's error. This improvement was particularly pronounced for small test sets. This suggests a novel justification for using bagging—more accurate estimation of the generalization error than is possible without bagging.     

Generalization of a Suffix Tree for RNA Structural Pattern Matching

In molecular biology, it is said that two biological sequences tend to have similar properties if they have similar three-dimensional structures. Hence, it is very important to find not only similar sequences in the string sense, but also structurally similar sequences from databases. In this paper we propose a new data structure that is a generalization of a parameterized suffix tree (p-suffix tree for short) introduced by Baker. We call it the structural suffix tree or s-suffix tree for short. The s-suffix tree can be used for finding structurally related patterns of RNA or single-stranded DNA. Furthermore, we propose an O(n(log|| + log||)) on-line algorithm for constructing it, where n is the sequence length, || is the size of the normal alphabet, and || is that of the alphabet called parameter, which is related to the structure of the sequence. Our algorithm achieves linear time when it is used to analyze RNA and DNA sequences. Furthermore, as an algorithm for constructing the p-suffix tree, it is the first on-line algorithm, though the computing bound of our algorithm is the same as that of Kosarajus best-known algorithm. The results of computational experiments using actual RNA and DNA sequences are also given to demonstrate our algorithms practicality.     

Enhancing the experience of 3D virtual worlds with a cartographic generalization approach

In this work we propose a new approach for fast visualization and exploration of virtual worlds based on the use of cartographic concepts and techniques. Versions of cartographic maps with different levels of details can be created by using a set of operations named cartographic generalization. Cartographic generalization employs twelve operators and domain-specific knowledge, being the contribution of this work their transposition to 3D virtual worlds. The architecture of a system for 3D generalization is proposed and the system is implemented. Differently from traditional cartographic processes, we use artificial intelligence for both selecting the key objects and applying the operators. As a case study, we present the simplification of the historical quarter of Recife (Brazil).     

最小一乘回归神经网络集成方法股市建模研究

提出了一种新的神经网络集成股市建模方法,采用偏最小二乘方法构造神经网络输入矩阵,利用Bagging技术和不同的神经网络学习算法生成集成个体,再用遗传算法选择参与集成的个体,以"误差绝对值和最小"为最优准,建立最小一乘回归神经网络集成模型,通过上证指数开盘价、收盘价进行实例分析,计算结果表明该方法具有较好的学习能力和泛化能力,在股市预测中预测精度高、稳定性好.     

基于综合进展率的GM(1,1)进度控制模型及其应用

首先介绍了宝钢进行进度控制的参数——综合进展率,该数据具有宏观性和直观性.综合进展率可以较准确的从多方面综合反映工程进展.然后以实际综合进展率为原始数据,建立灰色GM(1,1)模型进行进度预测,并进行了模型的通用性检验.最后使用MATLAB语言编程实现.通过实例验证,得出了灰色GM(1,1)模型具有很好的适用性和易操作性的结论,并认为以该模型得到的预测值为依据进行业主的进度控制,易于保证大型冶金项目建设进度目标的实现.     

基于支持向量回归模型的电力系统谐波分析新方法

当前电力系统中的谐波问题日益严重,对谐波的准确检测和分析是抑制谐波畸变的重要依据。将基于改进的SMO算法的支持向量回归模型应用于电力系统谐波的检测,该算法克服了常规算法计算规模大和建模复杂的困难,通过引入一个中间因子,将原来问题的计算规模减半,并利用迭代算法求解中间因子,使得该算法简单可行。对三相桥式整流电路交流侧产生的特征谐波和非特征谐波电流进行了分析,仿真结果通过与FFT算法和ADALINE神经网络的检测分析结果对比,表明该方法无论是在理想情况下还是在考虑了各种影响因素的情况下,都具有很高的检测精度,可以满足电力系统的谐波分析的要求。该方法的不足之处是计算量会随着输入量分辨率的提高而增大。     

ON THE CONNECTION BETWEEN THE COMPLEXITY AND CREDIBILITY OF INFERRED MODELS

Abstract

The connection between the simplicity of scientific theories and the credence attributed to their predictions seems to permeate the practice of scientific discovery. When a scientist succeeds in explaining a set of nobservations using a model Mof complexity c then it is generally believed that the likelihood of finding another explanatory model with similar complexity but leading to opposite predictions decreases with increasing nand decreasing c. This paper derives formal relationships between n, c and the probability of ambiguous predictions by examining three modeling languages under binary classification tasks: perceptrons, Boolean formulae, and Boolean networks. Bounds are also derived for the probability of error associated with the policy of accepting only models of complexity not exceeding c. Human tendency to regard the simpler as the more trustworthy is given a qualified justification.