Colony Optimization Based Feature Selection Research Articles

Extreme Learning Machine for Thyroid Nodule Classification with Graph Cluster Ant Colony Optimization Based Feature Selection

Thyroid nodule is defined as an endocrine malignancy that occurs in humans due to abnormal growth of cells. Recently, an increasing level of thyroid incidence has been identified worldwide. Thus, it is necessary to detect the nodules at an early stage. Ultrasonography is an important tool that is utilized for the detection as well as differentiation of malignant thyroid nodules from benign nodules. The nodules in ultrasound appear in different heterogenic forms, which are difficult to differentiate by the physicians. Further, large number of features available in US characteristics increases the computation time as well as complexity of classification. In this paper, Graph-Clustering Ant Colony Optimization based Extreme Learning Machine approach is proposed to achieve efficient diagnosis of thyroid nodules. It will enhance thyroid nodule classification by selecting only the optimal features and further using it for improving the function of classifier. The main goal of this technique is to differentiate the malignant nodules from the benign nodules. The performance of both feature selection and classification are evaluated through parameters such as accuracy, AUC, sensitivity and specificity. From the experimental results, it is revealed that the proposed method is significantly better than the existing methods. Thus, it is considered to be an effective tool for diagnosing the thyroid nodules with less complexity and reduced computation time.

Improved Ant Colony on Feature Selection and Weighted Ensemble to Neural Network Based Multimodal Disease Risk Prediction (WENN-MDRP) Classifier for Disease Prediction Over Big Data

As the big data is growing in biomedical and healthcare communities, so are precise analyses of medical data aids, premature disease identification, patient care as well as community services. On the other hand, the accuracy of the analysis decreases, if the medical data quality is imperfect. As a result, the choice of features from the dataset turns out to be an extremely significant task. Feature selection has exposed its efficiency in numerous applications by means of constructing modest and more comprehensive models, enlightening learning performance and preparing clean and clear data. The proposed method analyzes the difficulties of feature selection for big data analytics. Improved Ant Colony Optimization based Feature Selection (IACO) algorithm is presented for resolving this issue. The reconstruction of missing data before the incomplete data available was performed with help of latent factor mode. Therefore, it was not easy to choose the best features from the structured and unstructured data. the unheard technique which is called Weighted Ensemble Based Neural Network for multimodal disease risk prediction(WENN-MDRP) algorithm is implemented in order to provide the best features selection among structured as well as unstructured data. The research method provides improved prediction accuracy when matched with conventional techniques. In the MATLAB environment, the presented classifiers are implemented. The outcomes are computed in regard to recall, precision, accuracy, f-measure and error rate.

Computer aided diagnosis of pulmonary hamartoma from CT scan images using ant colony optimization based feature selection

BackgroundComputer-aided diagnosis (CAD) systems for the detection of lung disorders play an important role in clinical decision making. CAD systems provide a second opinion to the physician in interpreting computed tomography (CT) images. In this work, a CAD system to diagnose pulmonary hamartoma nodules from chest CT images is proposed. MethodsSegmentation of lung parenchyma from CT images is carried out using Otsu’s thresholding method. Nodules are considered to be the region of interests (ROIs) in this work. Texture, shape and run length based features are extracted from the ROIs. Cosine similarity measure (CSM) and rough dependency measure (RDM) are used independently as filter evaluation functions with ant colony optimization (ACO) to select two subsets of features. The selected subsets are used to train two classifiers namely support vector machine (SVM) and Naive Bayes (NB) classifiers using 10-fold cross validation. All the four trained classifiers are tested and the performance measures are estimated. ResultsCT slices of patients affected with pulmonary cancer and hamartoma are used for experimentation. From the lung parenchymal tissues of 300 CT slices, 390 nodules are extracted. The feature selection algorithms, ACO-CSM and ACO-RDM are run for different feature subset sizes. The selected features are used to train SVM and NB classifiers. From the results obtained, it is inferred that SVM classifier with the feature subsets chosen by ACO-RDM feature selection approach yielded a maximum classification accuracy of 94.36% with 38 features. ConclusionFrom the results, it can be clearly inferred that selecting relevant features to train the classifier has a definite impact on the performance of the classifier.

A discrete bacterial algorithm for feature selection in classification of microarray gene expression cancer data

When mining in high dimensional data, the curse of dimensionality is one of the major difficulty to overcome. In this paper, a weighted feature selection strategy is developed and embedded in bacterial based algorithms to reduce the feature dimension in classification. The proposed weighted feature selection strategy distinguishes the features by their classification performances as well as the occurrence frequency in population according to the two matrices. The objectives of minimizing the number of features, maximizing the performance, and minimizing the computational cost are all considered. Regarding the drawback of bacterial based algorithms, Bacterial Colony Optimization based feature selection algorithm is proposed to decrease the computational complexity as well as improve the search ability even in discrete optimization problems. To test the effectiveness of the proposed feature selection method, four bacterial based methods with the weighted strategy embedded have been compared with four classical feature selection methods and three well-known population based algorithms using 15 cancer micro-array datasets with different numbers of features and classes. The results show that the weighted feature selection strategies embedded have improved the feature selection capability of bacterial algorithms. The new proposed mechanisms embedded in Bacterial Colony Optimization method can overcome the limitation of the traditional bacterial based algorithms using premature termination to decrease the computational time, and provide comparable or in most cases better solutions than other feature selection methods considered in the comparison.

Ant Colony Optimization Based Feature Selection for Opinion Mining Classification

Ant Colony Optimization Based Feature Selection for Opinion Mining Classification

Optimal Feature Subset Selection using Ant Colony Optimization

Background/Objectives: Data mining is the process of extracting large volumes of raw data from hidden knowledge. The health care industry requires the use of data mining techniques as it generates huge and complex volumes of data. The applications of data mining techniques to medical data extract patterns which are useful for diagnosis, prognoses and treatment of diseases. This extraction of patterns allows doctors and hospitals to be more effective and more efficient. The huge volume of data is the barrier in the detection of patterns. Feature selection techniques mainly used in data preprocessing for data mining. Methods/Statistical Analysis: Classification task leads to reduction of the dimensionality of feature space, feature selection process is used for selecting large set of features. The Ant Colony Optimization based feature selection method is applied on cancer datasets. Findings: This research work proposes about feature selection mechanism based on Ant Colony Optimization. In an ACO algorithm, the activities of ants have significance for solving different combinatorial optimization problem which selects most relevant features. Through several iterations filter based method finds the optimal feature subset. Based on the similarity between features the feature relevance will be computed, that shows to the minimization of the redundancy. To validate the proposed feature selection method Support Vector Machine classification is applied. The accuracy of classification for whole feature set and the reduced feature subset are compared. The improved accuracy proves that the proposed feature selection approach has selected informative feature of the cancer datasets. Applications/Improvements: The possibilities of using PSO algorithm is applied for finding the best features in future. Other algorithms are also considered for further implementation.

Pattern Matching based Classification using Ant Colony Optimization based Feature Selection

Classification is a method of accurately predicting the target class for an unlabelled sample by learning from instances described by a set of attributes and a class label. Instance based classifiers are attractive due to their simplicity and performance. However, many of these are susceptible to noise and become unsuitable for real world problems. This paper proposes a novel instance based classification algorithm called Pattern Matching based Classification (PMC). The underlying principle of PMC is that it classifies unlabelled samples by matching for patterns in the training dataset. The advantage of PMC in comparison with other instance based methods is its simple classification procedure together with high performance. To improve the classification accuracy of PMC, an Ant Colony Optimization based Feature Selection algorithm based on the idea of PMC has been proposed. The classifier is evaluated on 35 datasets. Experimental results demonstrate that PMC is competent with many instance based classifiers. The results are also validated using nonparametric statistical tests. Also, the evaluation time of PMC is less when compared to the gravitation based methods used for classification.

Ant Colony Optimization Based Feature Selection Method for QEEG Data Classification

ObjectiveMany applications such as biomedical signals require selecting a subset of the input features in order to represent the whole set of features. A feature selection algorithm has recently been proposed as a new approach for feature subset selection.MethodsFeature selection process using ant colony optimization (ACO) for 6 channel pre-treatment electroencephalogram (EEG) data from theta and delta frequency bands is combined with back propagation neural network (BPNN) classification method for 147 major depressive disorder (MDD) subjects.ResultsBPNN classified R subjects with 91.83% overall accuracy and 95.55% subjects detection sensitivity. Area under ROC curve (AUC) value after feature selection increased from 0.8531 to 0.911. The features selected by the optimization algorithm were Fp1, Fp2, F7, F8, F3 for theta frequency band and eliminated 7 features from 12 to 5 feature subset.ConclusionACO feature selection algorithm improves the classification accuracy of BPNN. Using other feature selection algorithms or classifiers to compare the performance for each approach is important to underline the validity and versatility of the designed combination.

An ant colony optimization based feature selection for web page classification.

The increased popularity of the web has caused the inclusion of huge amount of information to the web, and as a result of this explosive information growth, automated web page classification systems are needed to improve search engines' performance. Web pages have a large number of features such as HTML/XML tags, URLs, hyperlinks, and text contents that should be considered during an automated classification process. The aim of this study is to reduce the number of features to be used to improve runtime and accuracy of the classification of web pages. In this study, we used an ant colony optimization (ACO) algorithm to select the best features, and then we applied the well-known C4.5, naive Bayes, and k nearest neighbor classifiers to assign class labels to web pages. We used the WebKB and Conference datasets in our experiments, and we showed that using the ACO for feature selection improves both accuracy and runtime performance of classification. We also showed that the proposed ACO based algorithm can select better features with respect to the well-known information gain and chi square feature selection methods.