Table 4 - uploaded by Jeannie Fitzgerald
Content may be subject to copyright.
Source publication
![Table 3 : Data Sets [23]](profile/Jeannie-Fitzgerald/publication/264670826/figure/tbl2/AS:613508001583137@1523283114622/Data-Sets-23_Q320.jpg)
In Machine Learning classification tasks, the class imbalance problem is an important one which has
received a lot of attention in the last few years. In binary classification, class imbalance occurs when there are
significantly fewer examples of one class than the other. A variety of strategies have been applied to the problem
with varying degrees...
Context in source publication
Similar publications
Class imbalance, as a phenomenon of asymmetry, has an adverse effect on the performance of most machine learning and overlap is another important factor that affects the classification performance of machine learning algorithms. This paper deals with the two factors simultaneously, addressing the class overlap under imbalanced distribution. In this...
Citations
... For comparison purposes we choose simple classification accuracy as a performance metric. It has been empirically established in the GP literature that simple classification accuracy is not a reliable measure of classification on unbalanced datasets (Bhowan et al., 2012), and that other measures such as average accuracy or Matthews Correlation Co-efficient might be more appropriate especially if combined with a sampling approach (Fitzgerald and Ryan, 2013). However, in this preliminary investigation, the classes are balanced which allows us to consider simple classification accuracy as a reasonable measure, particularly as we want to be able to observe differences in performance across the various levels of learning. ...
This paper introduces a novel evolutionary approach which can be applied to supervised, semi-supervised and unsupervised learning tasks. The method, Grammatical Evolution Machine Learning (GEML) adapts machine learning concepts from decision tree learning and clustering methods, and integrates these into a Grammatical Evolution framework. With minor adaptations to the objective function the system can be trivially modified to work with the conceptually different paradigms of supervised, semi-supervised and unsupervised learning. The framework generates human readable solutions which explain the mechanics behind the classification decisions , offering a significant advantage over existing paradigms for unsupervised and semi-supervised learning. GEML is studied on a range of multi-class classification problems and is shown to be competitive with several state of the art multi-class classification algorithms.
... For the volumes studied, of the 75 usable positive cases, 3 of difficulty of the classification problem. The imbalance in the data was mitigated in all cases by using Proportional Individualised Random Sampling, as described in [8] Based on this master dataset, we consider several setups representing different configurations of breasts, segments and views (see Table 1). The following terminology is used to describe the composition of instances for a given setup: BXSY VZ, where X is the number of breasts, Y the number of segments and Z the number of views for a given instance. ...
We present an automated, end-to-end approach for Stage~1 breast cancer detection. The first phase of our proposed work-flow takes individual digital mammograms as input and outputs several smaller sub-images from which the background has been removed. Next, we extract a set of features which capture textural information from the segmented images.
In the final phase, the most salient of these features are fed into a Multi-Objective Genetic Programming system which then evolves classifiers capable of identifying those segments which may have suspicious areas that require further investigation.
A key aspect of this work is the examination of several new experimental configurations which focus on textural asymmetry between breasts. The best evolved classifier using such a configuration can deliver results of 100% accuracy on true positives and a false positive per image rating of just 0.33, which is better than the current state of the art.
... Each of these three distributions exhibit very significant class imbalance which, in and of itself, increases the level of difficulty of the classification problem. The imbalance in the data was handled in all cases by using Proportional Individualised Random Sampling (Fitzgerald and Ryan 2013), as described in Sect. 10.4.2 ...
... Also, the use of over-sampling generally results in increased computational cost because of the increased size of the dataset. In this study, we have employed a proportional sampling approach (Fitzgerald and Ryan 2013) which eliminates or mitigates these disadvantages. ...
This chapter describes a general approach for image classification using Genetic Programming (GP) and demonstrates this approach through the application of GP to the task of stage 1 cancer detection in digital mammograms. We detail an automated work-flow that begins with image processing and culminates in the evolution of classification models which identify suspicious segments of mammograms. Early detection of breast cancer is directly correlated with survival of the disease and mammography has been shown to be an effective tool for early detection, which is why many countries have introduced national screening programs. However, this presents challenges, as such programs involve screening a large number of women and thus require more trained radiologists at a time when there is a shortage of these professionals in many countries.Also, as mammograms are difficult to read and radiologists typically only have a few minutes allocated to each image, screening programs tend to be conservative—involving many callbacks which increase both the workload of the radiologists and the stress and worry of patients.Fortunately, the relatively recent increase in the availability of mammograms in digital form means that it is now much more feasible to develop automated systems for analysing mammograms. Such systems, if successful could provide a very valuable second reader function.We present a work-flow that begins by processing digital mammograms to segment them into smaller sub-images and to extract features which describe textural aspects of the breast. The most salient of these features are then used in a GP system which generates classifiers capable of identifying which particular segments may have suspicious areas requiring further investigation. An important objective of this work is to evolve classifiers which detect as many cancers as possible but which are not overly conservative. The classifiers give results of 100 % sensitivity and a false positive per image rating of just 0.33, which is better than prior work. Not only this, but our system can use GP as part of a feedback loop, to both select and help generate further features.
... We choose AUC as the primary measure of performance in considering bias error for these experiments. See [16] for a detailed explanation of why AUC should be preferred to accuracy as a measure of classifier performance, and also [5] as to why overall classification accuracy is a poor and misleading choice, even when the data is even mildly unbalanced. ...
There have been many studies undertaken to
determine the efficacy of parameters and algorithmic
components of Genetic Programming, but historically,
generalization considerations have not been of central
importance in such investigations. Recent contributions have
stressed the importance of generalisation to the future
development of the field. In this paper we investigate aspects of
selection bias as a component of generalisation error, where
selection bias refers to the method used by the learning system to
select one hypothesis over another. Sources of potential bias
include the replacement strategy chosen and the means of
applying selection pressure. We investigate the effects on
generalisation of two replacement strategies, together with
tournament selection with a range of tournament sizes. Our
results suggest that larger tournaments are more prone to
overfitting than smaller ones, and that a small tournament
combined with a generational replacement strategy produces
relatively small solutions and is least likely to over-fit.
... Note that changing the perspective from mammograms to segments changes the distribution of decision classes, as each negative mammogram contributes three negative segments, but also, each positive mammogram contributes two negative segments (typically, only one segment contains cancerous growth). The imbalance in the data was handled in all cases by using Proportional Individualised Random Sampling [17]. ...
We describe a fully automated workflow for performing stage 1 breast cancer detection with GP as its cornerstone. Mammograms are by far the most widely used method for detecting breast cancer in women, and its use in national screening can have a dramatic impact on early detection and survival rates. With the increased availability of digital mammography, it is becoming increasingly more feasible to use automated methods to help with detection.
A stage 1 detector examines mammograms and highlights suspicious areas that require further investigation. A too conservative approach degenerates to marking every mammogram (or segment of) as suspicious, while missing a cancerous area can be disastrous.
Our workflow positions us right at the data collection phase such that we generate textural features ourselves. These are fed through our system, which performs PCA on them before passing the most salient ones to GP to generate classifiers. The classifiers give results of 100% accuracy on true positives and a false positive per image rating of just 1.5, which is better than prior work. Not only this, but our system can use GP as part of a feedback loop, to both select and help generate further features.
In this paper, we propose a hybrid approach to solving multi-class problems which combines evolutionary computation with elements of traditional machine learning. The method, Grammatical Evolution Machine Learning (GEML) adapts machine learning concepts from decision tree learning and clustering methods and integrates these into a Grammatical Evolution framework. We investigate the effectiveness of GEML on several supervised, semi-supervised and unsupervised multi-class problems and demonstrate its competitive performance when compared with several well known machine learning algorithms. The GEML framework evolves human readable solutions which provide an explanation of the logic behind its classification decisions, offering a significant advantage over existing paradigms for unsupervised and semi-supervised learning. In addition we also examine the possibility of improving the performance of the algorithm through the application of several ensemble techniques.
