Modern personalised medicine uses high dimensional genomic data to perform customised diagnostic/prognostic. In addition, physicians record several medical parameters to evaluate some clinical status. In this thesis we are interested in jointly using those different but complementary kinds of variables to perform classification tasks. Our main goal is to provide interpretability to predictive m...

Data classification plays important role in the field of data mining. The increasing rate of data diversity and size decrease the performance and efficiency of classifier. The decreasing performance of classifier compromised with unvoted data of classifier. Now the merging of two or more classifier for better prediction and voting of data are used, such techniques are called Ensemble classifier...

This paper describes the development of a predictive model for corporate insolvency risk in Australia. The model building methodology is empirical with out-ofsample future year test sets. The regression method used is logistic regression after pre-processing by quantisation of interval (or numeric) attributes. We show that logistic regression matches the performance of ensemble methods, such as...

Methods that use an !1-norm to encourage model sparsity are now widely applied across many disciplines. However, aggregating such sparse models across fits to resampled data remains an open problem. Because resampling approaches have been shown to be of great utility in reducing model variance and improving variable selection, a method able to generate a single sparse solution from multiple fit...

With the rapid development of Web 2.0 applications, social media have increasingly become a major factor influencing the purchase decisions of customers. Massive user behavioral, i.e., longitudinal individual behavioral and engagement behavioral, data generated on social media sites post challenges to integrate diverse heterogeneous data to improve prediction performance in customer response mo...

Protein trafficking or protein sorting is the mechanism by which a cell transports proteins to the appropriate position in the cell or outside of it. This targeting is based on the information contained in the protein. Many methods predict the subcellular location of proteins in eukaryotes from the sequence information. However, most of these methods use a flat structure to perform prediction. ...

Ensemble methods for supervised machine learning have become popular due to their ability to accurately predict class labels with groups of simple, lightweight “base learners.” While ensembles offer computationally efficient models that have good predictive capability, they tend to be large and offer little insight into the patterns or structure in a dataset. In this study, we extend an ensembl...

Comparing different novel feature sets and classifiers for speech processing based fatigue detection is the primary aim of this study. Thus, we conducted a within-subject partial sleep deprivation design (20.00–04.00 h, N1⁄477 participants) and recorded 372 speech samples of sustained vowel phonation. The self-report on the Karolinska Sleepiness Scale (KSS) and an observer report on the KSS, th...

Finding relevant subspaces in very highdimensional data is a challenging task not only for microarray data. The selection of features must be stable, but on the other hand learning performance is to be increased. Ensemble methods have succeeded in the increase of stability and classification accuracy, but their runtime prevents them from scaling up to real-world applications. We propose two met...

The naive Bayes classifier has proved to be very effective on many real data applications. Its performance usually benefits from an accurate estimation of univariate conditional probabilities and from variable selection. However, although variable selection is a desirable feature, it is prone to overfitting. In this paper, we introduce a Bayesian regularization technique to select the most prob...