The bootstrap aggregating procedure at the core of ensemble tree classifiers reduces, in most cases, the variance of such models while offering good generalization capabilities. The average predictive performance of those ensembles is known to improve up to a certain point while increasing the ensemble size. The present work studies this convergence in contrast to the stability of the class pre...

We propose and study a new technique for aggregating an ensemble of bootstrapped classifiers. In this method we seek a linear combination of the base-classifiers such that the weights are optimized to reduce variance. Minimum variance combinations are computed using quadratic programming. This optimization technique is borrowed from Mathematical Finance where it is called Markowitz Mean-Varianc...

This paper develops an adaptive ensemble Genetic Algorithm (GA) model for bankruptcy classification of firms cited in the SEC’s Accounting and Auditing Enforcement Release (AAER). Our research contributes to the bankruptcy literature in several ways. First of all, it fills a gap in bankruptcy classification by developing a domain specific model for AAER firms. Secondly, by using financial and n...

Topological spatial data can be useful for the classification and analysis of biomedical data. Neural networks have been used previously to make diagnostic classifications of corneal disease using summary statistics as network inputs. This approach neglects global shape features (used by clinicians when they make their diagnosis) and produces results that are difficult to interpret clinically. ...

This paper presents a machine learning technique for bird species identification at large scale. It automatically identifies about a thousand different species in a large number of audio recordings and provides the basis for the winning solution to the LifeCLEF 2015 Bird Identification Task. To process the very large amounts of audio data and to achieve similar good results compared to previous...

We present a list of “best possible” estimates of low-degree p-mode frequencies, from 8640 days of observations made by the Birmingham Solar-Oscillations Network (BiSON). This is the longest stretch of helioseismic data ever used for this purpose, giving exquisite precision in the estimated frequencies. Every effort has been made in the analysis to ensure that the frequency estimates are also a...

a r t i c l e i n f o a b s t r a c t In this article we address the issue of generating diversified translation systems from a single Statistical Machine Translation (SMT) engine for system combination. Unlike traditional approaches, we do not resort to multiple structurally different SMT systems, but instead directly learn a strong SMT system from a single translation engine in a principled w...

This paper presents ensemble approaches in single-layered complex-valued neural network (CVNN) to solve real-valued classification problems. Each component CVNN of an ensemble uses a recently proposed activation function for its complex-valued neurons (CVNs). A gradient-descent based learning algorithm was used to train the component CVNNs. We applied two ensemble methods, negative correlation ...

Decorrelated and CELS are two ensembles that modify the learning procedure to increase the diversity among the networks of the ensemble. Although they provide good performance according to previous comparatives, they are not as well known as other alternatives, such as Bagging and Boosting, which modify the learning set in order to obtain classifiers with high performance. In this paper, two di...

Ensembling neural classifiers can significantly improve the generalization ability of classification systems. In this paper, GASEN, a genetic algorithm based selective ensemble method that has been shown to be excellent in ensembling neural regressors, is applied to neural classifiers. Experiments on four large data sets show that this method can generate ensembles of neural classifiers with st...