敏感图片自动检测系统的研究

设计并实现了一个敏感图片自动识别系统。该系统首先利用肤色模型从敏感图片中检测出肤色区域,再从图片肤色区域中,提取大量经验特征表示图像内容,然后采用Haar特征算法设计一个检测敏感部位对象的分类器,最后通过敏感部位的几何特征逐级识别正常图片与敏感图片。该检测系统具有良好的系统性能。     

选择分类器进行入侵检测

使用集成分类器的方法进行入侵检测,但差的个体分类器往往会对集成性能造成不良影响。因此,使用信息增益法评价各分类器性能,并剔除表现不好的若干个分类器。一方面,降低了分类器空间维数。另一方面,提高了集成效果。在公用的入侵检测数据集上的实验结果表明,本文方法具有较好的集成效果,优于单分类器性能。     

在癌症分类中基于分层抽样的神经网络集成算法

在基因表达谱数据的分析中,针对有效合理地选择特征基因集的问题,本文将分层抽样技术引入特征基因选择,提高特征基因集的分类能力。以神经网络作为分量分类器,神经网络集成进行分类预测。并在结肠癌数据集上进行实验,实验结果表明该方法能有效地降低特征基因集选择的复杂性,提高对于未知样本的分类预测效果。     

Multi-strategy ensemble particle swarm optimization for dynamic optimization

Optimization in dynamic environments is important in real-world applications, which requires the optimization algorithms to be able to find and track the changing optimum efficiently over time. Among various algorithms for dynamic optimization, particle swarm optimization algorithms (PSOs) are attracting more and more attentions in recent years, due to their ability of keeping good balance between convergence and diversity maintenance. To tackle the challenges of dynamic optimization, several strategies have been proposed to enhance the performance of PSO, and have gained success on various dynamic optimization problems. But there still exist some issues in dynamic optimization which need to be studied carefully, i.e. the robustness of the algorithm to problems of various dynamic features. In this paper, a new multi-strategy ensemble particle swarm optimization (MEPSO) for dynamic optimization is proposed. In MEPSO, all particles are divided into two parts, denoted as part I and part II, respectively. Two new strategies, Gaussian local search and differential mutation, are introduced into these two parts, respectively. Experimental analyses reveal that the mechanisms used in part I can enhance the convergence ability of the algorithm, while mechanisms used in part II can extend the searching area of the particle population to avoid being trapped into the local optimum, and can enhance the ability of catching up with the changing optimum in dynamic environments. The whole algorithm has few parameters that need to be tuned, and all of them are not sensitive to problems. We compared MEPSO with other PSOs, including MQSO, PHPSO and Standard PSO with re-initialization, on moving peaks Benchmark and dynamic Rastrigin function. The experimental results show that MEPSO has pretty good performance on almost all testing problems adopted in this paper, and outperforms other algorithms when the dynamic environment is unimodal and changes severely, or has a great number of local optima as dynamic Rastrigin function does.     

Supervised classification of share price trends

Share price trends can be recognized by using data clustering methods. However, the accuracy of these methods may be rather low. This paper presents a novel supervised classification scheme for the recognition and prediction of share price trends. We first produce a smooth time series using zero-phase filtering and singular spectrum analysis from the original share price data. We train pattern classifiers using the classification results of both original and filtered time series and then use these classifiers to predict the future share price trends. Experiment results obtained from both synthetic data and real share prices show that the proposed method is effective and outperforms the well-known K-means clustering algorithm.     

基于集成的年龄估计方法

近十年来, 由于广泛的应用前景, 关于人脸识别的研究得到了广泛的关注. 但目前有一种影响人脸识别技术的因素尚未被研究者所重视, 那就是年龄变化. 而在适用于年龄变化的人脸识别技术中有一个重要的问题, 即年龄估计. 本文基于典型相关分析和代价敏感学习提出了两种年龄估计算法, 并在此基础上利用集成技术来提高年龄估计的准确性. 最终实验结果验证了本文方法的有效性.     

Early bankruptcy prediction using ENPC

Bankruptcy prediction has long time been an active research field in finance. One of the main approaches to this issue is dealing with it as a classification problem. Among the range of instruments available, we focus our attention on the Evolutionary Nearest Neighbor Classifier (ENPC). In this work we assess the performance of the ENPC comparing it to six alternatives. The results suggest that this algorithm might be considered a good choice.     

AptaCDSS-E: A classifier ensemble-based clinical decision support system for cardiovascular disease level prediction

Conventional clinical decision support systems are generally based on a single classifier or a simple combination of these models, showing moderate performance. In this paper, we propose a classifier ensemble-based method for supporting the diagnosis of cardiovascular disease (CVD) based on aptamer chips. This AptaCDSS-E system overcomes conventional performance limitations by utilizing ensembles of different classifiers. Recent surveys show that CVD is one of the leading causes of death and that significant life savings can be achieved if precise diagnosis can be made. For CVD diagnosis, our system combines a set of four different classifiers with ensembles. Support vector machines and neural networks are adopted as base classifiers. Decision trees and Bayesian networks are also adopted to augment the system. Four aptamer-based biochip data sets including CVD data containing 66 samples were used to train and test the system. Three other supplementary data sets are used to alleviate data insufficiency. We investigated the effectiveness of the ensemble-based system with several different aggregation approaches by comparing the results with single classifier-based models. The prediction performance of the AptaCDSS-E system was assessed with a cross-validation test. The experimental results show that our system achieves high diagnosis accuracy (>94%) and comparably small prediction difference intervals (<6%), proving its usefulness in the clinical decision process of disease diagnosis. Additionally, 10 possible biomarkers are found for further investigation.     

一种基于Vague-Sigmoid核的支持向量机研究

Sigmoid核最初起源于神经网络,目前在支持向量机中也得到了广泛应用,但由于核矩阵的非半正定性,其应用受到一些限制.研究表明Sigmoid核可以用简单的模糊三角隶属函数来近似替代,使得其学习效率能进一步提高.本文首先分析模糊支持向量机的特性,将模糊理论用于支持向量机的核中,并在此基础上提出了基于Vague-Sigmoid核函数的支持向量分类器.该方法充分结合了Vague集的自身优势,用基于Vague集的相似度量来代替了常规中的样本间的点积计算方法.将文中提出的方法应用于标准数据集中,并与传统的Sigmoid核方法、Fuzzy -Sigmoid核方法进行了实验分析, 实验表明文中提出的方法在不损失精度的情况下,能较好的提高算法的执行效率,取得了较好的实验结果;同时也表明在支持向量机中能利用Vague-S igmoid核取代替传统的Sigmoid核,从而减少对Sigmoid核的限制.     

RSM组合方法用于高光谱遥感数据的分类研究

RSM(Random Subspace Method)是一种新型的组合方法,利用随机挑选的特征子集生成每个基模型(Base Model),因此对于高维数据,RSM具有很大的应用潜力.本文将RSM方法应用于高光谱数据分类中,验证了RSM方法的有效性,并采用遗传算法(Genetic Algorithm)选择部分基模型构成组合来进一步提高分类精度.本文的工作为进一步在高光谱领域开展组合算法研究起到了很好的示范作用.