Abstract
In the predictive modeling tasks, a clear distinction is often made between learning problems that are supervised or unsupervised, the first involving only labeled data (training patterns with known category labels) while the latter involving only unlabeled data. There is a growing interest in a hybrid setting, called semi-supervised learning, in semi-supervised classification, the labels of only a small portion of the training data set are available. The unlabeled data, instead of being discarded, are also used in the learning process. Motivated by a breast cancer application, in this work we address a new learning task, in-between classification and semi-supervised classification. Each example is described using two different feature sets, not necessarily both observed for a given example. If a single view is observed, then the class is only due to that feature set, if both views are present the observed class label is the maximum of the two values corresponding to the individual views. We propose new learning methodologies adapted to this learning paradigm and experimentally compare them with baseline methods from the conventional supervised and unsupervised settings. The experimental results verify the usefulness of the proposed approaches. © 2011 IEEE.

Abstract

Abstract

Abstract
Breast Cancer Conservative Treatment (BCCT) is considered nowadays to be the most widespread form of locor-regional breast cancer treatment. However, aesthetic results are heterogeneous and difficult to evaluate in a standardized way. The limited reproducibility of subjective aesthetic evaluation in BCCT motivated the research towards objective methods. A recent computer system (BCCT. core) was developed to objectively and automatically evaluate the aesthetic result of BCCT. The system is centered on a support vector machine (SVM) classifier with a radial basis function (RBF) used to predict the overall cosmetic result from features computed on a digital photograph of the patient. However, this classifier is not ideal for the interpretation of the factors being used in the prediction. Therefore, an often suggested improvement is the interpretability of the model being used to assess the overall aesthetic result. In the current work we investigate the accuracy of different interpretable methods against the model currently deployed in the BCCT. core software. We compare the performance of decision trees and linear classifiers with the RBF SVM currently in BCCT. core. In the experimental study, these interpretable models shown a similar accuracy to the currently used RBF SVM, suggesting that the later can be replaced without sacrificing the performance of the BCCT.core.

Abstract

Abstract
Feature selection is a topic of growing interest mainly due to the increasing amount of information, being an essential task in many machine learning problems with high dimensional data. The selection of a subset of relevant features help to reduce the complexity of the problem and the building of robust learning models. This work presents an adaptation of a recent quadratic programming feature selection technique that identifies in one-fold the redundancy and relevance on data. Our approach introduces a non-probabilistic measure to capture the relevance based on Minimum Spanning Trees. Three different real datasets were used to assess the performance of the adaptation. The results are encouraging and reflect the utility of feature selection algorithms.