In this research work an ensemble of bagging, boosting, rotation forest, decorate and random subspace methods with 5 symbolic sub-classifiers in each one is presented. then a voting methodology is used for the final p...
详细信息
ISBN:
(纸本)9781467393126
In this research work an ensemble of bagging, boosting, rotation forest, decorate and random subspace methods with 5 symbolic sub-classifiers in each one is presented. then a voting methodology is used for the final prediction. In order to decrease training time, before building the ensemble redundant features were removed using a slight filter feature selection method. A comparison with simple bagging, boosting, rotation forest, decorate and random subspace methods ensembles with 25 symbolic sub-classifiers is performed, as well as other well-known combining methods, on standard benchmark datasets. the proposed technique is shown to be more accurate than other related methods in most cases.
Measuring similarity or distance between two data points is fundamental to many machinelearning algorithms such as K-Nearest-Neighbor, Clustering etc. Depending on the nature of the data point, various measurements c...
详细信息
Measuring similarity or distance between two data points is fundamental to many machinelearning algorithms such as K-Nearest-Neighbor, Clustering etc. Depending on the nature of the data point, various measurements can be used. DTW is largely used for mining time series but it is not adopted to large data sets because of its quadratic complexity. Global constraints narrow the search path in the matrix which results in a significant decrease in the number of performed calculations. the distance between examples from the same class is small. Instances from different classes are with large distances. A field called metric learning is introduced to make such criteria. In some time series classification tasks, it is a common case that two time series are out of phase, even they share the same class label. An appropriate constraint of DTW can strongly improve the classification performance. It is to choose the appropriate size of the global constraint. A Tabu search algorithm is used to find the optimal size of the global constraint. Results show the efficiency of the proposed method in terms of the improvement of the classification results and the CPU time.
the proceedings contain 61 papers. the topics discussed include improved comprehensibility and reliability of explanations via restricted halfspace discretization;selection of subsets of ordered features in machine le...
ISBN:
(纸本)3642030696
the proceedings contain 61 papers. the topics discussed include improved comprehensibility and reliability of explanations via restricted halfspace discretization;selection of subsets of ordered features in machinelearning;combination of vector quantization and visualization;discretization of target attributes for subgroup discovery;preserving privacy in time series data classification by discretization;sequential EM for unsupervised adaptive Gaussian mixture model based classifier;optimal double-kernel combination for classification;a linear classification method in a very high dimensional space using distributed representation;PMCRI: a parallel modular classification rule induction framework;dynamic score combination: a supervised and unsupervised score combination method;and ODDboost.
In this paper, we study the computer recognition of emotions involved in facial expressions. We propose a recognition system based on a support vector machine (SVM) system as a classifier for detecting of spontaneous ...
详细信息
In this paper we present a novel approach to the problem of understanding, monitoring, and controlling the machining process of composites materials. the approach is called Logical Analysis of data (LAD). It is based ...
详细信息
Metadata provide a high-level description of digital library resources and represent the key to enable the discovery and selection of suitable resources. However the growth in size and diversity of digital collections...
详细信息
Missing data methods, within the datamining context, are limited in computational complexity due to large data amounts. Amongst the computationally simple yet effective imputation methods are the hot deck procedures....
详细信息
Proceedings oftheSixthinternationalconference on Intelligent System and Knowledge Engineering presents selected papers from the conference ISKE 2011, held December 15-17 in Shanghai, China. this proceedings doesnt on...
详细信息
ISBN:
(纸本)9783642256578
Proceedings oftheSixthinternationalconference on Intelligent System and Knowledge Engineering presents selected papers from the conference ISKE 2011, held December 15-17 in Shanghai, China. this proceedings doesnt only examine original research and approaches in the broad areas of intelligent systems and knowledge engineering, but also present new methodologies and practices in intelligent computing paradigms. the book introduces the current scientific and technical advances in the fields of artificial intelligence, machinelearning, patternrecognition, datamining, information retrieval, knowledge-based systems, knowledge representation and reasoning, multi-agent systems, natural-language processing, etc. Furthermore, new computing methodologies are presented, including cloud computing, service computing and pervasive computing with traditional intelligent methods. the proceedings will be beneficial for both researchers and practitioners who want to utilize intelligent methods in their specific research fields. Dr. Yinglin Wang is a professor at the Department of Computer Science and Engineering, Shanghai Jiao Tong University, China; Dr. Tianrui Li is a professor at the School of Information Science and Technology, Southwest Jiaotong University, China.
Unsupervised classification or clustering is an important data analysis technique demanded in various fields including machinelearning, datamining, patternrecognition, image analysis and bioinformatics. Recently a ...
详细信息
data Warehousing and datamining are two mature disciplines in their own right. Yet, they have developed largely separate from each other, despite the fact that techniques developed for patternrecognition such as Clu...
详细信息
ISBN:
(纸本)9788988678312
data Warehousing and datamining are two mature disciplines in their own right. Yet, they have developed largely separate from each other, despite the fact that techniques developed for patternrecognition such as Clustering and Visualization in the datamining discipline have much to offer in the design of data Warehouses. this is somewhat surprising, given that the two disciplines have broadly the same set of objectives, although the techniques that they employ are admittedly quite different from each other. this may be due to the lack of a suitable methodology for integrating methods such as clustering and pattern visualization into data warehousing design. In this research, we propose such a methodology and report on its application to two case studies involving real world data taken from the UCI machinelearning repository. We demonstrate how data clustering and visualization methods, working in conjunction with each other can be used to gain new insights and build more meaningful dimensions which may not be obvious to human data warehouse designers.
暂无评论