feature construction
Recently Published Documents
TOTAL DOCUMENTS
H-INDEX
Genetic Programming based Feature Manipulation for Skin Cancer Image Classification
<p>Skin image classification involves the development of computational methods for solving problems such as cancer detection in lesion images, and their use for biomedical research and clinical care. Such methods aim at extracting relevant information or knowledge from skin images that can significantly assist in the early detection of disease. Skin images are enormous, and come with various artifacts that hinder effective feature extraction leading to inaccurate classification. Feature selection and feature construction can significantly reduce the amount of data while improving classification performance by selecting prominent features and constructing high-level features. Existing approaches mostly rely on expert intervention and follow multiple stages for pre-processing, feature extraction, and classification, which decreases the reliability, and increases the computational complexity. Since good generalization accuracy is not always the primary objective, clinicians are also interested in analyzing specific features such as pigment network, streaks, and blobs responsible for developing the disease; interpretable methods are favored. In Evolutionary Computation, Genetic Programming (GP) can automatically evolve an interpretable model and address the curse of dimensionality (through feature selection and construction). GP has been successfully applied to many areas, but its potential for feature selection, feature construction, and classification in skin images has not been thoroughly investigated. The overall goal of this thesis is to develop a new GP approach to skin image classification by utilizing GP to evolve programs that are capable of automatically selecting prominent image features, constructing new high level features, interpreting useful image features which can help dermatologist to diagnose a type of cancer, and are robust to processing skin images captured from specialized instruments and standard cameras. This thesis focuses on utilizing a wide range of texture, color, frequency-based, local, and global image properties at the terminal nodes of GP to classify skin cancer images from multiple modalities effectively. This thesis develops new two-stage GP methods using embedded and wrapper feature selection and construction approaches to automatically generating a feature vector of selected and constructed features for classification. The results show that wrapper approach outperforms the embedded approach, the existing baseline GP and other machine learning methods, but the embedded approach is faster than the wrapper approach. This thesis develops a multi-tree GP based embedded feature selection approach for melanoma detection using domain specific and domain independent features. It explores suitable crossover and mutation operators to evolve GP classifiers effectively and further extends this approach using a weighted fitness function. The results show that these multi-tree approaches outperformed single tree GP and other classification methods. They identify that a specific feature extraction method extracts most suitable features for particular images taken from a specific optical instrument. This thesis develops the first GP method utilizing frequency-based wavelet features, where the wrapper based feature selection and construction methods automatically evolve useful constructed features to improve the classification performance. The results show the evidence of successful feature construction by significantly outperforming existing GP approaches, state-of-the-art CNN, and other classification methods. This thesis develops a GP approach to multiple feature construction for ensemble learning in classification. The results show that the ensemble method outperformed existing GP approaches, state-of-the-art skin image classification, and commonly used ensemble methods. Further analysis of the evolved constructed features identified important image features that can potentially help the dermatologist identify further medical procedures in real-world situations.</p>
<p>Skin image classification involves the development of computational methods for solving problems such as cancer detection in lesion images, and their use for biomedical research and clinical care. Such methods aim at extracting relevant information or knowledge from skin images that can significantly assist in the early detection of disease. Skin images are enormous, and come with various artifacts that hinder effective feature extraction leading to inaccurate classification. Feature selection and feature construction can significantly reduce the amount of data while improving classification performance by selecting prominent features and constructing high-level features. Existing approaches mostly rely on expert intervention and follow multiple stages for pre-processing, feature extraction, and classification, which decreases the reliability, and increases the computational complexity. Since good generalization accuracy is not always the primary objective, clinicians are also interested in analyzing specific features such as pigment network, streaks, and blobs responsible for developing the disease; interpretable methods are favored. In Evolutionary Computation, Genetic Programming (GP) can automatically evolve an interpretable model and address the curse of dimensionality (through feature selection and construction). GP has been successfully applied to many areas, but its potential for feature selection, feature construction, and classification in skin images has not been thoroughly investigated. The overall goal of this thesis is to develop a new GP approach to skin image classification by utilizing GP to evolve programs that are capable of automatically selecting prominent image features, constructing new high level features, interpreting useful image features which can help dermatologist to diagnose a type of cancer, and are robust to processing skin images captured from specialized instruments and standard cameras. This thesis focuses on utilizing a wide range of texture, color, frequency-based, local, and global image properties at the terminal nodes of GP to classify skin cancer images from multiple modalities effectively. This thesis develops new two-stage GP methods using embedded and wrapper feature selection and construction approaches to automatically generating a feature vector of selected and constructed features for classification. The results show that wrapper approach outperforms the embedded approach, the existing baseline GP and other machine learning methods, but the embedded approach is faster than the wrapper approach. This thesis develops a multi-tree GP based embedded feature selection approach for melanoma detection using domain specific and domain independent features. It explores suitable crossover and mutation operators to evolve GP classifiers effectively and further extends this approach using a weighted fitness function. The results show that these multi-tree approaches outperformed single tree GP and other classification methods. They identify that a specific feature extraction method extracts most suitable features for particular images taken from a specific optical instrument. This thesis develops the first GP method utilizing frequency-based wavelet features, where the wrapper based feature selection and construction methods automatically evolve useful constructed features to improve the classification performance. The results show the evidence of successful feature construction by significantly outperforming existing GP approaches, state-of-the-art CNN, and other classification methods. This thesis develops a GP approach to multiple feature construction for ensemble learning in classification. The results show that the ensemble method outperformed existing GP approaches, state-of-the-art skin image classification, and commonly used ensemble methods. Further analysis of the evolved constructed features identified important image features that can potentially help the dermatologist identify further medical procedures in real-world situations.</p>
<p>Classification is a major task in machine learning and data mining. Many real-world datasets suffer from the unavoidable issue of missing values. Classification with incomplete data has to be carefully handled because inadequate treatment of missing values will cause large classification errors. Existing most researchers working on classification with incomplete data focused on improving the effectiveness, but did not adequately address the issue of the efficiency of applying the classifiers to classify unseen instances, which is much more important than the act of creating classifiers. A common approach to classification with incomplete data is to use imputation methods to replace missing values with plausible values before building classifiers and classifying unseen instances. This approach provides complete data which can be then used by any classification algorithm, but sophisticated imputation methods are usually computationally intensive, especially for the application process of classification. Another approach to classification with incomplete data is to build a classifier that can directly work with missing values. This approach does not require time for estimating missing values, but it often generates inaccurate and complex classifiers when faced with numerous missing values. A recent approach to classification with incomplete data which also avoids estimating missing values is to build a set of classifiers which then is used to select applicable classifiers for classifying unseen instances. However, this approach is also often inaccurate and takes a long time to find applicable classifiers when faced with numerous missing values. The overall goal of the thesis is to simultaneously improve the effectiveness and efficiency of classification with incomplete data by using evolutionary machine learning techniques for feature selection, clustering, ensemble learning, feature construction and constructing classifiers. The thesis develops approaches for improving imputation for classification with incomplete data by integrating clustering and feature selection with imputation. The approaches improve both the effectiveness and the efficiency of using imputation for classification with incomplete data. The thesis develops wrapper-based feature selection methods to improve input space for classification algorithms that are able to work directly with incomplete data. The methods not only improve the classification accuracy, but also reduce the complexity of classifiers able to work directly with incomplete data. The thesis develops a feature construction method to improve input space for classification algorithms with incomplete data by proposing interval genetic programming-genetic programming with a set of interval functions. The method improves the classification accuracy and reduces the complexity of classifiers. The thesis develops an ensemble approach to classification with incomplete data by integrating imputation, feature selection, and ensemble learning. The results show that the approach is more accurate, and faster than previous common methods for classification with incomplete data. The thesis develops interval genetic programming to directly evolve classifiers for incomplete data. The results show that classifiers generated by interval genetic programming can be more effective and efficient than classifiers generated the combination of imputation and traditional genetic programming. Interval genetic programming is also more effective than common classification algorithms able to work directly with incomplete data. In summary, the thesis develops a range of approaches for simultaneously improving the effectiveness and efficiency of classification with incomplete data by using a range of evolutionary machine learning techniques.</p>
<p>Classification is a major task in machine learning and data mining. Many real-world datasets suffer from the unavoidable issue of missing values. Classification with incomplete data has to be carefully handled because inadequate treatment of missing values will cause large classification errors. Existing most researchers working on classification with incomplete data focused on improving the effectiveness, but did not adequately address the issue of the efficiency of applying the classifiers to classify unseen instances, which is much more important than the act of creating classifiers. A common approach to classification with incomplete data is to use imputation methods to replace missing values with plausible values before building classifiers and classifying unseen instances. This approach provides complete data which can be then used by any classification algorithm, but sophisticated imputation methods are usually computationally intensive, especially for the application process of classification. Another approach to classification with incomplete data is to build a classifier that can directly work with missing values. This approach does not require time for estimating missing values, but it often generates inaccurate and complex classifiers when faced with numerous missing values. A recent approach to classification with incomplete data which also avoids estimating missing values is to build a set of classifiers which then is used to select applicable classifiers for classifying unseen instances. However, this approach is also often inaccurate and takes a long time to find applicable classifiers when faced with numerous missing values. The overall goal of the thesis is to simultaneously improve the effectiveness and efficiency of classification with incomplete data by using evolutionary machine learning techniques for feature selection, clustering, ensemble learning, feature construction and constructing classifiers. The thesis develops approaches for improving imputation for classification with incomplete data by integrating clustering and feature selection with imputation. The approaches improve both the effectiveness and the efficiency of using imputation for classification with incomplete data. The thesis develops wrapper-based feature selection methods to improve input space for classification algorithms that are able to work directly with incomplete data. The methods not only improve the classification accuracy, but also reduce the complexity of classifiers able to work directly with incomplete data. The thesis develops a feature construction method to improve input space for classification algorithms with incomplete data by proposing interval genetic programming-genetic programming with a set of interval functions. The method improves the classification accuracy and reduces the complexity of classifiers. The thesis develops an ensemble approach to classification with incomplete data by integrating imputation, feature selection, and ensemble learning. The results show that the approach is more accurate, and faster than previous common methods for classification with incomplete data. The thesis develops interval genetic programming to directly evolve classifiers for incomplete data. The results show that classifiers generated by interval genetic programming can be more effective and efficient than classifiers generated the combination of imputation and traditional genetic programming. Interval genetic programming is also more effective than common classification algorithms able to work directly with incomplete data. In summary, the thesis develops a range of approaches for simultaneously improving the effectiveness and efficiency of classification with incomplete data by using a range of evolutionary machine learning techniques.</p>
<p>More and more high-dimensional data appears in machine learning, especially in classification tasks. With thousands of features, these datasets bring challenges to learning algorithms not only because of the curse of dimensionality but also the existence of many irrelevant and redundant features. Therefore, feature selection and feature construction (or feature manipulation in short) are essential techniques in preprocessing these datasets. While feature selection aims to select relevant features, feature construction constructs high-level features from the original ones to better represent the target concept. Both methods can decrease the dimensionality and improve the performance of learning algorithms in terms of classification accuracy and computation time. Although feature manipulation has been studied for decades, the task on high-dimensional data is still challenging due to the huge search space. Existing methods usually face the problem of stagnation in local optima and/or require high computation time. Evolutionary computation techniques are well-known for their global search. Particle swarm optimisation (PSO) and genetic programming (GP) have shown promise in feature selection and feature construction, respectively. However, the use of these techniques to high-dimensional data usually requires high memory and computation time. The overall goal of this thesis is to investigate new approaches to using PSO for feature selection and GP for feature construction on high-dimensional classification problems. This thesis focuses on incorporating a variety of strategies into the evolutionary process and developing new PSO and GP representations to improve the effectiveness and efficiency of PSO and GP for feature manipulation on high-dimensional data. This thesis proposes a new PSO based feature selection approach to high-dimensional data by incorporating a new local search to balance global and local search of PSO. A hybrid of wrapper and filter evaluation method which can be sped up in the local search is proposed to help PSO achieve better performance, scalability and robustness on high-dimensional data. The results show that the proposed method significantly outperforms the compared methods in 80% of the cases with an increase up to 16% average accuracy while reduces the number of features from one to two orders of magnitude. This thesis develops the first PSO based feature selection via discretisation method that performs both multivariate discretisation and feature selection in a single stage to achieve better solutions than applying these techniques separately in two stages. Two new PSO representations are proposed to evolve cut-points for multiple features simultaneously. The results show that the proposed method selects less than 4.6% of the features in all cases to improve the classification performance from 5% to 23% in most cases. This thesis proposes the first clustering-based feature construction method to improve the performance of single-tree GP on high-dimensional data. A new feature clustering method is proposed to automatically group similar features into the same group based on a given redundancy level. The results show that compared with standard GP, the new method can select less than half of the features to construct a new high-level feature that achieves significantly better accuracy in most cases. The combination of the single constructed feature and the selected ones achieves the best performance among different feature sets created from a single tree. This thesis develops the first class-dependent multiple feature construction method using multi-tree GP for high-dimensional data. A new GP representation and a new filter fitness function that combines two filter measures are proposed to evaluate the whole set of constructed features more effectively and efficiently. The results show that in 83% of the cases, with less than 10 constructed features, the class-dependent method increases up to 32% average accuracy on using all the original thousands of features and 10% on using those constructed by the class-independent method.</p>
<p>More and more high-dimensional data appears in machine learning, especially in classification tasks. With thousands of features, these datasets bring challenges to learning algorithms not only because of the curse of dimensionality but also the existence of many irrelevant and redundant features. Therefore, feature selection and feature construction (or feature manipulation in short) are essential techniques in preprocessing these datasets. While feature selection aims to select relevant features, feature construction constructs high-level features from the original ones to better represent the target concept. Both methods can decrease the dimensionality and improve the performance of learning algorithms in terms of classification accuracy and computation time. Although feature manipulation has been studied for decades, the task on high-dimensional data is still challenging due to the huge search space. Existing methods usually face the problem of stagnation in local optima and/or require high computation time. Evolutionary computation techniques are well-known for their global search. Particle swarm optimisation (PSO) and genetic programming (GP) have shown promise in feature selection and feature construction, respectively. However, the use of these techniques to high-dimensional data usually requires high memory and computation time. The overall goal of this thesis is to investigate new approaches to using PSO for feature selection and GP for feature construction on high-dimensional classification problems. This thesis focuses on incorporating a variety of strategies into the evolutionary process and developing new PSO and GP representations to improve the effectiveness and efficiency of PSO and GP for feature manipulation on high-dimensional data. This thesis proposes a new PSO based feature selection approach to high-dimensional data by incorporating a new local search to balance global and local search of PSO. A hybrid of wrapper and filter evaluation method which can be sped up in the local search is proposed to help PSO achieve better performance, scalability and robustness on high-dimensional data. The results show that the proposed method significantly outperforms the compared methods in 80% of the cases with an increase up to 16% average accuracy while reduces the number of features from one to two orders of magnitude. This thesis develops the first PSO based feature selection via discretisation method that performs both multivariate discretisation and feature selection in a single stage to achieve better solutions than applying these techniques separately in two stages. Two new PSO representations are proposed to evolve cut-points for multiple features simultaneously. The results show that the proposed method selects less than 4.6% of the features in all cases to improve the classification performance from 5% to 23% in most cases. This thesis proposes the first clustering-based feature construction method to improve the performance of single-tree GP on high-dimensional data. A new feature clustering method is proposed to automatically group similar features into the same group based on a given redundancy level. The results show that compared with standard GP, the new method can select less than half of the features to construct a new high-level feature that achieves significantly better accuracy in most cases. The combination of the single constructed feature and the selected ones achieves the best performance among different feature sets created from a single tree. This thesis develops the first class-dependent multiple feature construction method using multi-tree GP for high-dimensional data. A new GP representation and a new filter fitness function that combines two filter measures are proposed to evaluate the whole set of constructed features more effectively and efficiently. The results show that in 83% of the cases, with less than 10 constructed features, the class-dependent method increases up to 32% average accuracy on using all the original thousands of features and 10% on using those constructed by the class-independent method.</p>
In feature selection, redundancy is one of the major concerns since the removal of redundancy in data is connected with dimensionality reduction. Despite the evidence of such a connection, few works present theoretical studies regarding redundancy. In this work, we analyze the effect of redundant features on the performance of classification models. We can summarize the contribution of this work as follows: (i) develop a theoretical framework to analyze feature construction and selection, (ii) show that certain properly defined features are redundant but make the data linearly separable, and (iii) propose a formal criterion to validate feature construction methods. The results of experiments suggest that a large number of redundant features can reduce the classification error. The results imply that it is not enough to analyze features solely using criteria that measure the amount of information provided by such features.
<p><b>Feature manipulation refers to the process by which the input space of a machine learning task is altered in order to improve the learning quality and performance. Three major aspects of feature manipulation are feature construction, feature ranking and feature selection. This thesis proposes a new filter-based methodology for feature manipulation in classification problems using genetic programming (GP). The goal is to modify the input representation of classification problems in order to improve classification performance and reduce the complexity of classification models. The thesis regards classification problems as a collection of variables including conditional variables (input features) and decision variables (target class labels). GP is used to discover the relationships between these variables. The types of relationship and the ways in which they are discovered vary with the three aspects of feature manipulation.</b></p> <p>In feature construction, the thesis proposes a GP-based method to construct high-level features in the form of functions of original input features. The functions are evolved by GP using an entropy-based fitness function that maximises the purity of class intervals. Unlike existing algorithms, the proposed GP-based method constructs multiple features and it can effectively perform transformational dimensionality reduction, using only a small number of GP-constructed features while preserving good classification performance.</p> <p>In feature ranking, the thesis proposes two GP-based methods for ranking single features and subsets of features. In single-feature ranking, the proposed method measures the influence of individual features on the classification performance by using GP to evolve a collection of weak classification models, and then measures the contribution of input features to the making of good models. In ranking of subsets of features, a virtual structure for GP trees and a new binary relevance function is proposed to measure the relationship between a subset of features and the target class labels. It is observed that the proposed method can discover complex relationships - such as multi-modal class distributions and multivariate correlations - that cannot be detected by traditional methods. In feature selection, the thesis provides a novel multi-objective GP-based approach to measuring the goodness of subsets of features. The subsets are evaluated based on their cardinality and their relationship to target class labels. The selection is performed by choosing a subset of features from a GP-discovered Pareto front containing suboptimal solutions (subsets). The thesis also proposes a novel method for measuring the redundancy between input features. It is used to select a subset of relevant features that do not exhibit redundancy with respect to each other. It is found that in all three aspects of feature manipulation, the proposed GP-based methodology is effective in discovering relationships between the features of a classification task. In the case of feature construction, the proposed GP-based methods evolve functions of conditional variables that can significantly improve the classification performance and reduce the complexity of the learned classifiers. In the case of feature ranking, the proposed GP-based methods can find complex relationships between conditional variables and decision variables. The resulted ranking shows a strong linear correlation with the actual classification performance. In the case of feature selection, the proposed GP-based method can find a set of sub-optimal subsets of features which provids a trade-off between the number of features and their relevance to the classification task. The proposed redundancy removal method can remove redundant features from a set of features. Both proposed feature selection methods can find an optimal subset of features that yields significantly better classification performance with a much smaller number of features than conventional classification methods.</p>
<p>Mass spectrometry (MS) is currently the most commonly used technology in biochemical research for proteomic analysis. The primary goal of proteomic profiling using mass spectrometry is the classification of samples from different experimental states. To classify the MS samples, the identification of protein or peptides (biomarker detection) that are expressed differently between the classes, is required. However, due to the high dimensionality of the data and the small number of samples, classification of MS data is extremely challenging. Another important aspect of biomarker detection is the verification of the detected biomarker that acts as an intermediate step before passing these biomarkers to the experimental validation stage. Biomarker detection aims at altering the input space of the learning algorithm for improving classification of proteomic or metabolomic data. This task is performed through feature manipulation. Feature manipulation consists of three aspects: feature ranking, feature selection, and feature construction. Genetic programming (GP) is an evolutionary computation algorithm that has the intrinsic capability for the three aspects of feature manipulation. The ability of GP for feature manipulation in proteomic biomarker discovery has not been fully investigated. This thesis, therefore, proposes an embedded methodology for these three aspects of feature manipulation in high dimensional MS data using GP. The thesis also presents a method for biomarker verification, using GP. The thesis investigates the use of GP for both single-objective and multi-objective feature selection and construction. In feature ranking, the thesis proposes a GP-based method for ranking subsets of features by using GP as an ensemble approach. The proposed algorithm uses GP capability to combine the advantages of different feature ranking metrics and evolve a new ranking scheme for the subset of the features selected from the top ranked features. The capability of GP as a classifier is also investigated by this method. The results show that GP can select a smaller number of features and provide a better ranking of the selected features, which can improve the classification performance of five classifiers. In feature construction, this thesis proposes a novel multiple feature construction method, which uses a single GP tree to generate a new set of high-level features from the original set of selected features. The results show that the proposed new algorithm outperforms two feature selection algorithms. In feature selection, the thesis introduces the first GP multi-objective method for biomarker detection, which simultaneously increase the classification accuracy and reduce the number of detected features. The proposed multi-objective method can obtain better subsets of features than the single-objective algorithm and two traditional multi-objective approaches for feature selection. This thesis also develops the first multi-objective multiple feature construction algorithm for MS data. The proposed method aims at both maximising the classification performance and minimizing the cardinality of the constructed new high-level features. The results show that GP can dis- cover the complex relationships between the features and can significantly improve classification performance and reduce the cardinality. For biomarker verification, the thesis proposes the first GP biomarker verification method through measuring the peptide detectability. The method solves the imbalance problem in the data and shows improvement over the benchmark algorithms. Also, the algorithm outperforms a well-known peptide detection method. The thesis also introduces a new GP method for alignment of MS data as a preprocessing stage, which will further help in improving the biomarker detection process.</p>
