作者：
Petrou, MUniv Surrey
Sch Elect Engn Informat Technol & Math Guildford GU2 5XH Surrey England
learning in the context of a patternrecognition system is defined as the process that allows it to cope with real and ambiguous data. The various ways by which artificial decision systems operate are discussed in con...
详细信息
ISBN:
(纸本)3540665994
learning in the context of a patternrecognition system is defined as the process that allows it to cope with real and ambiguous data. The various ways by which artificial decision systems operate are discussed in conjunction with their learning aspects.
We describe the guidelines of a system for monitoring environmental risk situations. The system is based on datamining techniques and in particular classification trees working on the data base collected by the Itali...
详细信息
ISBN:
(纸本)3540665994
We describe the guidelines of a system for monitoring environmental risk situations. The system is based on datamining techniques and in particular classification trees working on the data base collected by the Italian National Hydro-geological Net. The gear of our application is to achieve a better discrimination among cases then that obtained by the system which is presently in use. The decision trees are evaluated and selected via a metric that takes a weighted account of the errors of different kinds.
This paper presents a method for concept formation of a personal learning apprentice (PLA) system that attempts to capture users' internal conceptual structure by observing interactions between user and system. Cu...
详细信息
ISBN:
(纸本)3540665994
This paper presents a method for concept formation of a personal learning apprentice (PLA) system that attempts to capture users' internal conceptual structure by observing interactions between user and system. Current hot topics on techniques of datamining may potentially contribute to the above purpose, but different from the conventional approaches of datamining, we have to consider more about the aspects in which hom the mined knowledge should be used by the human in the consequent processes, not only about what knowledge should be extracted. In this paper we propose such a process-oriented datamining method based upon an idea of soft systems methodologies proposed by P.B. Checkland in 1980's, and we propose an algorithm for its implementation using evolutional computing.
In tackling datamining and patternrecognition tasks, finding a compact but effective set of features is often a crucial step in the whole problem solving process. In this paper we present an empirical study on featu...
详细信息
ISBN:
(纸本)9780769527307
In tackling datamining and patternrecognition tasks, finding a compact but effective set of features is often a crucial step in the whole problem solving process. In this paper we present an empirical study on feature selection for classical instrument recognition, using machinelearning techniques to select and evaluate features extracted from a number of different feature schemes in terms of their classification performance. It is revealed that there is significant redundancy in existing feature schemes commonly used in practice. Our results suggest that further feature analysis research is necessary for optimising feature selection for the instrument recognition problem.
作者：
Jahn, HDLR
Deutsch Zentrum Luft & Raumfahrt EV Inst Weltraumsensor & Planetenerkundung D-12489 Berlin Germany
A parallel-sequential unsupervised learning method for image smoothing is presented which can be implemented with a Multi Layer Neural Network. In contrast to older work of the author which has used 4-connectivity of ...
详细信息
ISBN:
(纸本)3540665994
A parallel-sequential unsupervised learning method for image smoothing is presented which can be implemented with a Multi Layer Neural Network. In contrast to older work of the author which has used 4-connectivity of processing elements (neurons) leading to a very big number of recursions now each neuron of network lever t+1 is connected with (2M+1)*(2M+1) neurons of layer t guaranteeing a significant reduction of network layers with the same good smoothing results.
In the field of patternrecognition, multiple classifier systems based on the combination of the outputs of a set of different classifiers have been proposed as a method for the development of high performance classif...
详细信息
ISBN:
(纸本)3540665994
In the field of patternrecognition, multiple classifier systems based on the combination of the outputs of a set of different classifiers have been proposed as a method for the development of high performance classification systems. Previous work clearly showed that multiple classifier systems are effective only if the classifiers forming them make independent errors. This achievement pointed out the fundamental need for methods aimed to design ensembles of "independent" classifiers. However, the most of the recent work focused on the development of combination methods. In this paper, an approach to the automatic design of multiple classifier systems based on unsupervised learning is proposed. Given an initial set of classifiers, such approach is aimed to identify the largest subset of "independent" classifiers. A proof of the optimality of the proposed approach is given. Reported results on the classification of remote sensing images show that this approach allows one to design effective multiple classifier systems.
Mixture modelling of class-conditional densities is a standard patternrecognition technique. Although most research on mixture models has concentrated on mixtures for continuous data, emerging patternrecognition app...
详细信息
Mixture modelling of class-conditional densities is a standard patternrecognition technique. Although most research on mixture models has concentrated on mixtures for continuous data, emerging patternrecognition applications demand extending research efforts to other data types. This paper focuses on the application of mixtures of multivariate Bernoulli distributions to binary data. More concretely, a text classification task aimed at improving language modelling for machine translation is considered. (C) 2002 patternrecognition Society. Published by Elsevier Science Ltd. All rights reserved.
Process mining, and in particular process discovery, provides useful tools for extracting process models from event-based data. Nevertheless, certain types of processes are too complex and unstructured to be able to b...
详细信息
ISBN:
(纸本)9783030116408;9783030116415
Process mining, and in particular process discovery, provides useful tools for extracting process models from event-based data. Nevertheless, certain types of processes are too complex and unstructured to be able to be represented with a start-to-end process model. For such cases, instead of extracting a model from a complete event log, it is interesting to zoom in on some parts of the data and explore behavioral patterns on a local level. Recently, local process model mining has been introduced, which is a technique in-between sequential patternmining and process discovery. Other process mining methods can also used for mining local patterns, if combined with certain data preprocessing. In this paper, we explore discovery of local patterns in the data representing learning processes. We exploit real-life event logs from JMermaid, a Smart learning Environment for teaching Information System modeling with built-in feedback functionality. We focus on a specific instance of feedback provided in JMermaid, which is a reminder to simulate the model, and locally explore how students react to this feedback. Additionally, we discuss how to tailor local process model mining to a certain case, in order to avoid the computationally expensive task of discovering all available patterns, by combining it with other techniques for dealing with unstructured data, such as trace clustering and window-based data preprocessing.
Many application domains make use of specific datastructures such as sequences and graphs to represent knowledge. These datastructures are ill-fitted to the standard representations used in machinelearning and data...
详细信息
Many application domains make use of specific datastructures such as sequences and graphs to represent knowledge. These datastructures are ill-fitted to the standard representations used in machinelearning and data-mining algorithms: propositional representations are not expressive enough, and first order ones are not efficient enough. In order to efficiently represent and reason on these datastructures, and the complex patterns that are related to them, we use domain-specific logics. We show these logics can be built by the composition of logical components that model elementary datastructures. The standard strategies of top-down and bottom-up search are ill-suited to some of these logics, and lack flexibility. We therefore introduce a dichotomic search strategy, that is analogous to a dichotomic search in an ordered array. We prove this provides more flexibility in the search, while retaining completeness and non-redundancy. We present a novel algorithm for learning using domain specific logics and dichotomic search, and analyse its complexity. We also describe two applications which illustrates the search for motifs in sequences;where these motifs have arbitrary length and length-constrained gaps. In the first application sequences represent the trains of the East-West challenge;in the second application they represent the secondary structure of Yeast proteins for the discrimination of their biological functions.
A fast support vector machine (SVM) training algorithm is proposed under SVM's decomposition framework by effectively integrating kernel caching, digest and shrinking policies and stopping conditions. Kernel cachi...
详细信息
A fast support vector machine (SVM) training algorithm is proposed under SVM's decomposition framework by effectively integrating kernel caching, digest and shrinking policies and stopping conditions. Kernel caching plays a key role in reducing the number of kernel evaluations by maximal reusage of cached kernel elements. Extensive experiments have been conducted on a large handwritten digit database MNIst to show that the proposed algorithm is much faster than Keerthi et al.'s improved SMO, about nine times. Combined with principal component analysis, the total training for ten one-against-the-rest classifiers on MNIst took less than an hour. Moreover, the proposed fast algorithm speeds up SVM training without sacrificing the generalization performance. The 0.6% error rate on MNIst test set has been achieved. The promising scalability of the proposed scheme paves a new way to solve more large-scale learning problems in other domains such as datamining.
暂无评论