Evolutionary Learning Techniques Research Articles

Group theoretic particle swarm optimization for multi-level threshold lung cancer image segmentation.

Image segmentation is an important step during the processing of medical images. For example, for the computer aid diagnostic systems for lung cancer image analysis, the segmented regions of tumors would help doctors in early diagnosis to determine timely and appropriate treatment possibilities and thereby improve the survival rate of the patients. However, general clinical routines of manual segmentation for large number of medical images are very difficult and time consuming, which is the challenge we aim to tackle using our proposed method. A novel image segmentation method with evolutionary learning technique named Group Theoretic Particle Swarm Optimization is proposed. It can tackle multi-level thresholding optimization problem during the segmentation process and rebuild the search paradigm according to the solid mathematical foundation of symmetric group from four designable aspects, which are particle encoding, solution landscape, neighborhood movement and swarm topology, respectively. The Kapur's entropy of multi-level thresholds is assessed as the objective function. In contrast to those conventional metaheuristics methods for lung cancer image segmentation, this newly presented method generates the best performance result among them. Experimental results show that its Kapur's entropy has the value of 9.07, which is 16% higher than the worst case. Computational time is acceptable at the cost of 173.730 seconds, average level of evaluation metrics [Kappa, Precision, Recall, F1-measure, intersection over union (IoU) and receiver operating characteristic (ROC)] is over 90%, and search process of multi-level threshold combination would finally converge in the later phase of iterations after 700. The ablation study indicates that all components are significant to the contributions of our proposed method. Group Theoretic Particle Swarm Optimization for multi-level threshold segmentation is an efficient way to split a medical image into distinct regions and extract tumor tissues regions from the background. It maintains the balanced relationship between diversification and intensification during the search process and helps clinicians to make the diagnosis more accurately. Our proposed method processes potential medical value and clinical meanings.

A prediction model for specific energy required by point attack picks based on a hybrid evolutionary machine learning technique

A prediction model for specific energy required by point attack picks based on a hybrid evolutionary machine learning technique

Lung cancer prediction using multi-gene genetic programming by selecting automatic features from amino acid sequences

Lung cancer is one of the leading causes of cancer related deaths. Early diagnosis of lung cancer using automatic feature selection from large number of features is a challenging task. Conventionally, cancer diagnosis approaches use physical features that appear in later stages, while harmful effects have already been occurred due to abnormal somatic mutations. In order to extract useful novel patterns to efficiently predict cancer at early stages, we analyzed lung cancer related mutated genes that reveal useful information in protein amino acid sequences. For this, we developed a new evolutionary learning technique with biologically inspired multi-gene genetic programming algorithm using discriminant information of protein amino acids. The proposed model efficiently selects 23 discriminant features out of 1500 features. Then it combines the selected features and related primitive functions optimally for prediction of lung cancer. Hence, an efficient predictive model is constructed that helps in understanding the complex heterogeneous nature of lung cancer. The proposed system achieved area under ROC curve and accuracy values of 98.79% and 95.67%, respectively outperforming related lung cancer prediction approaches.

Application of ANFIS hybrids to predict coefficients of curvature and uniformity of treated unsaturated lateritic soil for sustainable earthworks

Unsaturated lateritic soils are complex soils to work with due to moisture effects. So, the determination of its properties requires lots of time, labor and equipment. For this reason, the application of evolutionary learning techniques has been adopted to overcome these complexities. Lateritic soil under unsaturated condition classified as poorly graded and A-7–6 group was subjected to treatment by using hybrid cement and nanostructured quarry fines in a stabilization method. The clay activity, clay content and frictional angle were determined through multiple experiments at different proportions of the additives. 121 datasets were collected through the multiple testing of treated specimens and 70% and 30% of the datasets were used in the model training and testing, respectively to predict the coefficients of curvature and uniformity (Cc and Cu) of the unsaturated lateritic soil. Fist, the multi-linear regression (MLR) model showed that the selected input parameters correlated well with the output parameters. The model performance evaluation and validation selected indicators; R2, RMSE and MAE showed that ANFIS with 0.9999, 0.0021 and 0.0015 respectively, for the training and 0.9994, 0.0077 and 0.0059 respectively outclassed all its hybrid techniques and MLR in both training and testing. However, ANFIS-PSO with performance indicators 0.9996, 0.0062 and 0.0050 respectively (training) and 0.9989, 0.0095 and 0.0073 respectively (testing); followed by ANFIS-GA; 0.9991, 0.0094, and 0.0065 respectively (training) and 0.0089, 0.0099, and 0.0079 (testing) outclassed the other learning techniques for the Cc prediction model while ANFIS-GA; 0.9949, 0.1000, and 0.0798 respectively (training) and 0.9954, 0.0983, and 0.0807 respectively, followed by ANFIS-PSO; 0.9893, 0.1347, and 0.1011 respectively (training) and 0.9951, 0.1127, and 0.0924 respectively outclassed the other techniques for the Cu prediction model. Finally, ANFIS and its evolutionary hybrid techniques have shown their usefulness and flexibility in predicting stabilized unsaturated soil properties for sustainable earthwork design, construction and foundation performance monitoring.

Evolutionary learning algorithm for reliable facility location under disruption

Facility location represents an important supply chain problem aiming at minimizing facility establishment and transportation cost to meet customer demands. Many facility location problem (FLP) instances can be modelled as p-median problems (PMP) and uncapacitated facility location (UFL) problems. While, most solution approaches assume totally reliable deployed facilities, facilities often experience disruptions and their failure often leads to a notably higher cost. Therefore, determination of facility locations and fortification of a subset of them within a limited budget are crucial to supply chain organizations to provide cost effective services in presence of probable disruptions. We propose an evolutionary learning technique to near-optimally solve two research problems: Reliable p-Median Problem and Reliable Uncapacitated Facility Location Problem considering heterogeneous facility failure probabilities, one layer of backup and limited facility fortification budget. The technique is illustrated using a case study and its performance is evaluated via benchmark results. We also provide an analysis on the effects on facility location by prioritizing customer demands and adopting geographic distance calculation. The approach allows fast generation of cost-effective and complete solution using reasonable computing power. Moreover, the underlying technique is customizable offering a trade-off between solution quality and computation time.

An immersive learning model using evolutionary learning

In this article, we have proposed an educational model using virtual reality on a mobile platform by personalizing the simulated environments as per user actions. We have also proposed an evolutionary learning algorithm based on which the user learning path is designed and the corresponding simulated learning environment is modified. The main objective of this study is to create a personalized learning path for each student as per their calibre and make the learning immersive and retainable using virtual reality. Our proposed model emulates the innate natural learning process in humans and uses that to customize the virtual simulations of the lessons by applying the evolutionary learning technique. A quasi-experimental study is conducted by taking different case studies to establish the effectiveness of our learning model. The results show that our learning model is immersive and gives long term retention while enhancing creativity through reinforced customization of the simulations.

An improved shuffled frog leaping algorithm based evolutionary framework for currency exchange rate prediction

Forecasting purchasing power of one currency with respect to another currency is always an interesting topic in the field of financial time series prediction. Despite the existence of several traditional and computational models for currency exchange rate forecasting, there is always a need for developing simpler and more efficient model, which will produce better prediction capability. In this paper, an evolutionary framework is proposed by using an improved shuffled frog leaping (ISFL) algorithm with a computationally efficient functional link artificial neural network (CEFLANN) for prediction of currency exchange rate. The model is validated by observing the monthly prediction measures obtained for three currency exchange data sets such as USD/CAD, USD/CHF, and USD/JPY accumulated within same period of time. The model performance is also compared with two other evolutionary learning techniques such as Shuffled frog leaping algorithm and Particle Swarm optimization algorithm. Practical analysis of results suggest that, the proposed model developed using the ISFL algorithm with CEFLANN network is a promising predictor model for currency exchange rate prediction compared to other models included in the study.

Using rule-based machine learning for candidate disease gene prioritization and sample classification of cancer gene expression data.

Microarray data analysis has been shown to provide an effective tool for studying cancer and genetic diseases. Although classical machine learning techniques have successfully been applied to find informative genes and to predict class labels for new samples, common restrictions of microarray analysis such as small sample sizes, a large attribute space and high noise levels still limit its scientific and clinical applications. Increasing the interpretability of prediction models while retaining a high accuracy would help to exploit the information content in microarray data more effectively. For this purpose, we evaluate our rule-based evolutionary machine learning systems, BioHEL and GAssist, on three public microarray cancer datasets, obtaining simple rule-based models for sample classification. A comparison with other benchmark microarray sample classifiers based on three diverse feature selection algorithms suggests that these evolutionary learning techniques can compete with state-of-the-art methods like support vector machines. The obtained models reach accuracies above 90% in two-level external cross-validation, with the added value of facilitating interpretation by using only combinations of simple if-then-else rules. As a further benefit, a literature mining analysis reveals that prioritizations of informative genes extracted from BioHEL’s classification rule sets can outperform gene rankings obtained from a conventional ensemble feature selection in terms of the pointwise mutual information between relevant disease terms and the standardized names of top-ranked genes.

Improving the scalability of rule-based evolutionary learning

Evolutionary learning techniques are comparable in accuracy with other learning methods such as Bayesian Learning, SVM, etc. These techniques often produce more interpretable knowledge than, e.g. SVM; however, efficiency is a significant drawback. This paper presents a new representation motivated by our observations that Bioinformatics and Systems Biology often give rise to very large-scale datasets that are noisy, ambiguous and usually described by a large number of attributes. The crucial observation is that, in the most successful rules obtained for such datasets, only a few key attributes (from the large number of available ones) are expressed in a rule, hence automatically discovering these few key attributes and only keeping track of them contributes to a substantial speed up by avoiding useless match operations with irrelevant attributes. Thus, in effective terms this procedure is performing a fine-grained feature selection at a rule-wise level, as the key attributes may be different for each learned rule. The representation we propose has been tested within the BioHEL machine learning system, and the experiments performed show that not only the representation has competent learning performance, but that it also manages to reduce considerably the system run-time. That is, the proposed representation is up to 2–3 times faster than state-of-the-art evolutionary learning representations designed specifically for efficiency purposes.

KEEL: a software tool to assess evolutionary algorithms for data mining problems

This paper introduces a software tool named KEEL which is a software tool to assess evolutionary algorithms for Data Mining problems of various kinds including as regression, classification, unsupervised learning, etc. It includes evolutionary learning algorithms based on different approaches: Pittsburgh, Michigan and IRL, as well as the integration of evolutionary learning techniques with different pre-processing techniques, allowing it to perform a complete analysis of any learning model in comparison to existing software tools. Moreover, KEEL has been designed with a double goal: research and educational.

A study of the contribution made by evolutionary learning on dynamic load-balancing problems in distributed computing systems

A computer simulation model that investigated the contribution made by evolutionary learning techniques on load-balancing problems was proposed. Three parameters for controlling the load-balancing activity of a node were used. The system was tested in different distributed systems, including different processing and communication speeds as well as network structures. Our experimental results showed that the system demonstrated an effective learning capability in balancing load among different processing nodes. It also showed that each of these three parameters played an important role in contributing to load-balancing, and that the system performance increased upon increasing the number of parameter changes simultaneously. The contribution made by evolutionary learning was significant as the variety of node processing speeds increased.

E-Net: Evolutionary neural network synthesis

E-Net is a new distributed evolutionary learning system that evolves neural-network-based pattern recognition systems (PRSs) with limited human interaction. This system orchestrates a multiplicity of evolutionary and classical learning techniques to synthesize feature detectors, select sets of cooperative features, and assemble classifiers. Feature detectors are represented as feed-forward neural networks and recognition systems are defined using a collection of networks. E-Net evolves network topologies and trains weights to form accurate recognition systems using a computationally efficient process that gradually extends primitive network topologies to form increasingly discriminating structures. The evolutionary search process effectively explores the space of candidate topologies by manipulating populations of feature detectors and recognition systems using variation operators such as crossover and mutation. The majority of evolutionary learning techniques have been designed to perform parameter optimization. E-Net is designed to perform both synthesis and optimization. Consequently, many novel concepts and techniques are introduced in this research that expedite the gradual synthesis of structure, such as the new multitiered selection process used in E-Net's evolutionary algorithm that avoids premature convergence to complex topological structures.

AN EVOLUTION-BASED APPROACH TO PROGRAM UNDERSTANDING USING CULTURAL ALGORITHMS

Program understanding plays a very important role in the software engineering field as an essential part of the widely accepted reuse process. Unfortunately, the lack of general tools for this purpose often prevents users from effectively retrieving software modules for reuse. The primary difficulty is in determining which of the already-available modules performs the desired function. This paper explores the task of extracting functional knowledge from a program object (i.e., coded module) using an evolutionary-learning technique, cultural algorithms, to automatically learn classification rules for a reuse library. The modules are stored in a reuse library in the PM system using a faceted classification scheme. Therefore, the goal is to learn a set of rules to characterize the semantics of code modules based on their syntactic structure. The most difficult part of the process is to identify those parts of the program that provide specific evidence of the presence of a concept. Here, evolutionary-learning techniques are employed to interactively identify lines in a module that are "believed" to provide evidence for the presence of a facet value (concept) in the code by a user. A prototype is used to learn concepts relating to the "list" and "stack" programming concepts.