Results 1  10
of
10
A Comparison of Methods for Learning of Highly NonSeparable Problems
"... Abstract. Learning in cases that are almost linearly separable is easy, but for highly nonseparable problems all standard machine learning methods fail. Many strategies to build adaptive systems are based on the “divideandconquer ” principle. Constructive neural network architectures with novel t ..."
Abstract

Cited by 2 (2 self)
 Add to MetaCart
(Show Context)
Abstract. Learning in cases that are almost linearly separable is easy, but for highly nonseparable problems all standard machine learning methods fail. Many strategies to build adaptive systems are based on the “divideandconquer ” principle. Constructive neural network architectures with novel training methods allow to overcome some drawbacks of standard backpropagation MLP networks. They are able to handle complex multidimensional problems in reasonable time, creating models with small number of neurons. In this paper a comparison of our new constructive c3sep algorithm based on kseparability idea with several sequential constructive learning methods is reported. Tests have been performed on parity function, 3 artificial Monks problems, and a few benchmark problems. Simple and accurate solutions have been discovered using c3sep algorithm even in highly nonseparable cases. 1
A Machine Learning Algorithm to Estimate Minimal Cut and Path Sets from a Monte Carlo Simulation
 Proceedings Probabilistic Safety Assessment and Management PSAM7/ESREL'04
, 2004
"... In this paper a novel approach based on a machine learning algorithm (Hamming Clustering) is proposed to estimate the minimal cut and path sets, using the samples generated by a Monte Carlo simulation and any Evaluation Function. Two examples show the potential of the proposed approach. 1 ..."
Abstract

Cited by 1 (0 self)
 Add to MetaCart
(Show Context)
In this paper a novel approach based on a machine learning algorithm (Hamming Clustering) is proposed to estimate the minimal cut and path sets, using the samples generated by a Monte Carlo simulation and any Evaluation Function. Two examples show the potential of the proposed approach. 1
Network Reliability Assessment through Empirical Models using a Machine Learning Approach
, 2006
"... The Machine Learning (ML) paradigm offers an interesting approach for assessing various aspects related to the reliability of any system that can be represented as a network. The main idea is to employ a specific ML technique, trained on a restricted subset of data, to produce an estimate of the Str ..."
Abstract
 Add to MetaCart
The Machine Learning (ML) paradigm offers an interesting approach for assessing various aspects related to the reliability of any system that can be represented as a network. The main idea is to employ a specific ML technique, trained on a restricted subset of data, to produce an estimate of the Structure Function. In this chapter, three ML techniques (Support Vector Machines, Decision Trees and Shadow Clustering) are presented in detail and their behavior is carefully examined through different applications involving: reliability evaluation, reconstruction of approximate reliability expressions and determination of cut and path sets.
Estimating female labor force participation through statistical and machine learning methods: A comparison
, 2005
"... ..."
minigene, Hamming clustering,
"... The 1785 nucleotides of the coding region of the estrogen receptor = (ER=) are dispersed over a region of more than 300.000 nucleotides in the primary transcript. Splicing of this precursor RNA frequently leads to variants lacking one or more exons that have been associated to breast cancer progres ..."
Abstract
 Add to MetaCart
The 1785 nucleotides of the coding region of the estrogen receptor = (ER=) are dispersed over a region of more than 300.000 nucleotides in the primary transcript. Splicing of this precursor RNA frequently leads to variants lacking one or more exons that have been associated to breast cancer progression. The most frequent splice variant lacks exon 4 and is expressed in the human mammary carcinoma cell line MCF7 at a level similar to that of the fulllength messenger. The in silico analysis of ER = splice sites by Hamming clustering, a self learning method trained on more than 28.000 experimentally proved splice sites, reveals high relevance for the 5 ' and 3' splice sites of exon 4. The splicing analysis of transfected minigene constructs containing drastically shortened introns excludes that weak splice sites, intron or exon lengths or splice enhancers are responsible for exon skipping. Exon 6 is never skipped in MCF7 cells but is spliced out from minegene derived primary transcripts if inserted between exons 3 and 5 instead of exon 4. As a consequence, it appears that a particular splice site affinity of exon 3 donor (5 ' splice site) and exon 5 acceptor sites (3 ' splice site) is responsible for skipping of the exon in between.
FUZZYGRANULAR BASED DATA MINING FOR EFFECTIVE DECISION
, 2006
"... Due to complexity of biomedical problems, adaptive and intelligent knowledge discovery and data mining systems are highly needed to help humans to understand the inherent mechanism of diseases. For biomedical classification problems, typically it is impossible to build a perfect classifier with 100 ..."
Abstract
 Add to MetaCart
(Show Context)
Due to complexity of biomedical problems, adaptive and intelligent knowledge discovery and data mining systems are highly needed to help humans to understand the inherent mechanism of diseases. For biomedical classification problems, typically it is impossible to build a perfect classifier with 100 % prediction accuracy. Hence a more realistic target is to build an effective Decision Support System (DSS). In this dissertation, a novel adaptive Fuzzy Association Rules (FARs) mining algorithm, named FARMDS, is proposed to build such a DSS for binary classification problems in the biomedical domain. Empirical studies show that FARMDS is competitive to stateoftheart classifiers in terms of prediction accuracy. More importantly, FARs can provide strong decision support on disease diagnoses due to their easy interpretability. This dissertation also proposes a fuzzygranular method to select informative and discriminative genes from huge microarray gene expression data. With fuzzy granulation, information loss in the process of gene selection is decreased. As a result, more informative genes for cancer classification are selected and more accurate classifiers can
Empirical Models Based on Machine Learning Techniques for Determining Approximated Reliability Expressions
"... this paper two machine learning algorithms, Decision Trees (DT) and Hamming Clustering (HC), are compared in building approximated Reliability Expression (RE). The main idea is to employ a classification technique, trained on a restricted subset of data, to produce an estimate of the RE, which provi ..."
Abstract
 Add to MetaCart
this paper two machine learning algorithms, Decision Trees (DT) and Hamming Clustering (HC), are compared in building approximated Reliability Expression (RE). The main idea is to employ a classification technique, trained on a restricted subset of data, to produce an estimate of the RE, which provides reasonably accurate values of the reliability. The experiments show that although both methods yield excellent predictions, the HC procedure achieves better results with respect to the DT algorithm
Approximate MultiState Reliability Expressions Using a New Machine Learning Technique
"... The machinelearningbased methodology, previously proposed by the authors for approximating binary reliability expressions, is now extended to develop a new algorithm, based on the procedure of Hamming Clustering, which is capable to deal with multistate systems and any success criterion. The prop ..."
Abstract
 Add to MetaCart
(Show Context)
The machinelearningbased methodology, previously proposed by the authors for approximating binary reliability expressions, is now extended to develop a new algorithm, based on the procedure of Hamming Clustering, which is capable to deal with multistate systems and any success criterion. The proposed technique is presented in details and verified on literature cases: experiment results show that the new algorithm yields excellent predictions.
Rule Generation Methods Based on Logic Synthesis
"... One of the most relevant problems in artificial intelligence is allowing a synthetic device to perform inductive reasoning, i.e. to infer a set of rules consistent with a collection of data pertaining to a given real world problem. A variety of approaches, ..."
Abstract
 Add to MetaCart
(Show Context)
One of the most relevant problems in artificial intelligence is allowing a synthetic device to perform inductive reasoning, i.e. to infer a set of rules consistent with a collection of data pertaining to a given real world problem. A variety of approaches,
A Machine Learning Approach to Estimate Frequency, Duration & Availability Indexes in Complex Networks
"... Frequency, duration and availability are key measures in the evaluation of complex networks. Although efficient techniques have been developed, the calculation of these indexes is, however very difficult in certain type of networks, such as complex capacitylimited networks or in kterminal problems ..."
Abstract
 Add to MetaCart
(Show Context)
Frequency, duration and availability are key measures in the evaluation of complex networks. Although efficient techniques have been developed, the calculation of these indexes is, however very difficult in certain type of networks, such as complex capacitylimited networks or in kterminal problems. In this paper the machine learning algorithm Hamming Clustering (HC), belonging to the family of rule generation methods, is employed to obtain an approximated Availability Expression (AE) for a network, under any success criterion. The AE can be used to evaluate the system availability and then could be transformed, using a set of specific rules, to evaluate system frequency. Two examples related to a complex network are evaluated using the proposed approach. The experiments show that the proposed method, using samples from a Monte Carlo simulation, yield excellent predictions for availability, frequency and duration indexes, with errors less than 1 %. 1.