Search

Naive Bayes (NB) is one of the top ten machine learning algorithms whereas its attribute independence assumption rarely holds in practice. A feasible and efficient approach to improving NB is relaxing the assumption by adding augmented edges to the restricted topology of NB. In this paper we prove theoretically that the generalized topology may be a suboptimal solution to model multivariate probability distributions if its fitness to data cannot be measured. Thus we propose to apply log-likelihood function as the scoring function, then introduce an efficient heuristic search strategy to explore high-dependence relationships, and for each iteration the learned topology will be improved to fit data better. The proposed algorithm, called log-likelihood Bayesian classifier (LLBC), can respectively learn two submodels from labeled training set and individual unlabeled testing instance, and then make them work jointly for classification in the framework of ensemble learning. Our extensive experimental evaluations on 36 benchmark datasets from the University of California at Irvine (UCI) machine learning repository reveal that, LLBC demonstrates excellent classification performance and provides a competitive approach to learn from labeled and unlabeled data. [ABSTRACT FROM AUTHOR]

Feature selection is an optional preprocessing procedure and is frequently used to improve the classification accuracy of a machine learning algorithm by removing irrelevant and/or redundant features. However, in many real-world applications, the test cost is also required for making optimal decisions, in addition to the classification accuracy. To the best of our knowledge, thus far, few studies have been conducted on test-cost-sensitive feature selection (TCSFS). In TCSFS, the objectives are twofold: 1) to improve the classification accuracy and 2) to decrease the test cost. Therefore, in fact, it constitutes a multiobjective optimization problem. In this paper, we transformed this multiobjective optimization problem into a single-objective optimization problem by utilizing a new evaluation function and in this paper, we propose a new general wrapper framework for TCSFS. Specifically, in our proposed framework, we add a new term to the evaluation function of a wrapper feature selection method so that the test cost of measuring features is taken into account. We experimentally tested our proposed framework, using 36 classification problems from the University of California at Irvine (UCI) repository, and compared it to some other state-of-the-art feature selection frameworks. The experimental results showed that our framework allows users to select an optimal feature subset with the minimal test cost, while simultaneously maintaining a high classification accuracy. [ABSTRACT FROM AUTHOR]

This paper presents a workbench to get simple neural classification models based on product evolutionary networks via a prior data preparation at attribute level by means of filter-based feature selection. Therefore, the computation to build the classifier is shorter, compared to a full model without data pre-processing, which is of utmost importance since the evolutionary neural models are stochastic and different classifiers with different seeds are required to get reliable results. Feature selection is one of the most common techniques for pre-processing the data within any kind of learning task. Six filters have been tested to assess the proposal. Fourteen (binary and multi-class) difficult classification data sets from the University of California repository at Irvine have been established as the test bed. An empirical study between the evolutionary neural network models obtained with and without feature selection has been included. The results have been contrasted with nonparametric statistical tests and show that the current proposal improves the test accuracy of the previous models significantly. Moreover, the current proposal is much more efficient than the previous methodology; the time reduction percentage is above 40%, on average. Our approach has also been compared with several classifiers both with and without feature selection in order to illustrate the performance of the different filters considered. Lastly, a statistical analysis for each feature selector has been performed providing a pairwise comparison between machine learning algorithms. [ABSTRACT FROM AUTHOR]

Averaged one-dependence estimators (AODE) is a state-of-the-art machine learning tool for classification due to its simplicity, high computational efficiency, and excellent classification accuracy. Weighting provides an effective mechanism to ensemble superparent one-dependence estimators (SPODEs) in AODE by linearly aggregating their weighted probability estimates. Supervised weighting and unsupervised weighting are proposed to learn weights from labeled or unlabeled data, whereas their interoperability has not previously been investigated. In this paper, we propose a novel weighting paradigm in the framework of semi-supervised learning, called semi-supervised weighting (SSW). Two different versions of weighted AODEs, supervised weighted AODE (SWAODE) which performs weighting at training time and unsupervised weighted AODE (UWAODE) which performs weighting at classification time, are built severally. Log likelihood function is introduced to linearly aggregate the outcomes of these two weighted AODEs. The proposed algorithm, called SSWAODE, is validated on 38 benchmark datasets from the University of California at Irvine (UCI) machine learning repository and the experimental results prove the effectiveness and robustness of SSW for weighting AODE in terms of zero-one loss, bias, variance and etc. SSWAODE well achieves the balance between the ground-truth dependencies approximation and the effectiveness of probability estimation. [ABSTRACT FROM AUTHOR]

As a widely known disease diabetes mellitus makes the human body produce quite less hormone and also tend to cause increased glucose that results in abnormal metabolism of varied organs in the body like eyes, kidneys, etc. Diabetic analysis has attracted the research community to treat some missing values and class imbalance issues. The performance of diabetes mellitus classification by the usage of machine learning techniques is comparatively low. We suggest this paper on imbalanced dataset with missing values, an efficient prediction algorithm for diabetes mellitus classification using Deep 1D-Convolutional Neural Network values. The outlier detection is used for removing missing values first. Then, oversampling method (SMOTE) is used to reduce the influence of imbalance class on prediction performance. Finally, predictions are produced using a DCNN classifier and are evaluated using a selective set of evaluation indicators. Experiments on the Pima Indian diabetes dataset (PIDD) from UCI Repository (University of California at Irvine) have yielded positive results. Our proposed DCNN algorithm has been shown to be successful and superior. [ABSTRACT FROM AUTHOR]

Multigranulation rough set (MGRS) theory provides an effective manner for the problem solving by making use of multiple equivalence relations. As the information systems always dynamically change over time due to the addition or deletion of multiple objects, how to efficiently update the approximations in multigranulation spaces by making fully utilize the previous results becomes a crucial challenge. Incremental learning provides an efficient manner because of the incorporation of both the current information and previously obtained knowledge. In spite of the success of incremental learning, well-studied findings performed to update approximations in multigranulation spaces have relatively been scarce. To address this issue, in this paper, we propose matrix-based incremental approaches for updating approximations from the perspective of multigranulation when multiple objects vary over time. Based on the matrix characterization of multigranulation approximations, the incremental mechanisms for relevant matrices are systematically investigated while adding or deleting multiple objects. Subsequently, in accordance with the incremental mechanisms, the corresponding incremental algorithms for maintaining multigranulation approximations are developed to reduce the redundant computations. Finally, extensive experiments on eight datasets available from the University of California at Irvine (UCI) are conducted to verify the effectiveness and efficiency of the proposed incremental algorithms in comparison with the existing non-incremental algorithm. [ABSTRACT FROM AUTHOR]

Scores of machine learning algorithms depend on a good distance measure to achieve high performance. The inverted specific-class distance measure, simply ISCDM, is proposed to find reasonable distance measure between each pair of instances with nominal attributes only. ISCDM does not depend on the attribute value of the training instance, which makes it less sensitive to missing values in the training set and more robust to non-class attribute noise. However, in ISCDM, all attributes are assumed to be fully independent. It is obvious that the attribute independence assumption in ISCDM is rarely true in reality, which would harm its performance in the applications with complex attribute dependencies. In this paper, we single out an improved inverted specific-class distance measure by relaxing its unrealistic attribute independence assumption. We call it averaged one-dependence inverted specific-class distance measure, simply AODISCDM. We experimentally tested it on 29 classification problems from the University of California at Irvine (UCI) repository and found that it significantly outperforms ISCDM in terms of the negative conditional log likelihood (-CLL) and the root relative squared error (RRSE). Besides, the proposed AODISCDM maintains the computational simplicity (no search involved) and robustness that characterise ISCDM. [ABSTRACT FROM AUTHOR]

• We have developed a clustering algorithm for an unknown number of clusters by MOGA. • It works with continuous and categorical featured data sets. • It can work with data sets having missing values. • The final solution is selected by majority vote by all non-dominated solutions. • Context-sensitive and cluster-orient genetic operators are designed. Many clustering algorithms categorized as K -clustering algorithm require the user to predict the number of clusters (K) to do clustering. Due to lack of domain knowledge an accurate value of K is difficult to predict. The problem becomes critical when the dimensionality of data points is large; clusters differ widely in shape, size, and density; and when clusters are overlapping in nature. Determining the suitable K is an optimization problem. Automatic clustering algorithms can discover the optimal K. This paper presents an automatic clustering algorithm which is superior to K -clustering algorithm as it can discover an optimal value of K. Iterative hill-climbing algorithms like K -Means work on a single solution and converge to a local optimum solution. Here, Genetic Algorithms (GA s) find out near global optimum solutions, i.e. optimal K as well as the optimal cluster centroids. Single-objective clustering algorithms are adequate for efficiently grouping linearly separable clusters. For non-linearly separable clusters they are not so good. So for grouping non-linearly separable clusters, we apply Multi-Objective Genetic Algorithm (MOGA) by minimizing the intra-cluster distance and maximizing inter-cluster distance. Many existing MOGA based clustering algorithms are suitable for either numeric or categorical features. This paper pioneered employing MOGA for automatic clustering with mixed types of features. Statistical testing on experimental results on real-life benchmark data sets from the University of California at Irvine (UCI) machine learning repository proves the superiority of the proposed algorithm. [ABSTRACT FROM AUTHOR]

The study's main goal is to identify if a person has chronic renal illness or not. The medical condition known as chronic kidney failure damages overall kidneys' ability to collect harmful substances from your blood and keep good health in general. Some causes for chronic kidney disease contain lower blood pressure, anemia, weaker bones, poor diet, and trauma. Some causes for chronic kidney disease contain lower blood pressure, anemia, weaker bones, poor diet, and trauma. The value was estimated by the classification method using machine learning technologies. The patient's status of non-chronic kidney failure & chronic kidney failure will be estimated using classification models constructed using various classification techniques. These models were put to the test using a collection of data on chronic kidney illness from the University of California at Irvine, this had four hundred entries of data with twenty-five attributes. The outcomes of various models are examined. The model built with Decision tree method performed the best in terms of correctness using 14 attributes for the small dataset, based on the comparison. The final phase of this study will examine how effectively the machine learning system predicts chronic kidney failure with respect to precision, recall, accuracy, & F1-Score. [ABSTRACT FROM AUTHOR]

Data splitting is an important step in artificial neural network (ANN) models, which is found in the form of training and testing subsets. In general, a random data splitting method is favored to divide a pool of samples into subsets, without considering the quality of data for the training step of a neural network. The drawback of poor data splitting methods is that they poses ill effects to the performance of the neural network when the data involves complex matrices or multivariate modeling. In order to overcome this drawback, the current paper presents our proposed M-SPXY method. It is based on a modified version of Sample Set Partitioning, which relies on a joint X-y distances (SPXY) method. The proposed method has resulted in better performance, compared to the modified Kennard-Stone (KS) method, using Mahalanobis distances (MDKS). In our experiments, the proposed approach was employed to compare various data splitting methods using data sets from the repository of the University of California in Irvine (UCI), processed through an Extreme Learning Machine (ELM) neural network. Performance was measured in terms of classification accuracy. The results indicate that the classification accuracy of the proposed M-SPXY process is superior to that of the MDKS data splitting method. [ABSTRACT FROM AUTHOR]

This paper presents a new learning algorithm for multitask pattern recognition (MTPR) problems. We consider learning multiple multiclass classification tasks online where no information is ever provided about the task category of a training example. The algorithm thus needs an automated task recognition capability to properly learn the different classification tasks. The learning mode is "online" where training examples for different tasks are mixed in a random fashion and given sequentially one after another. We assume that the classification tasks are related to each other and that both the tasks and their training examples appear in random during "online training." Thus, the learning algorithm has to continually switch from learning one task to another whenever the training examples change to a different task. This also implies that the learning algorithm has to detect task changes automatically and utilize knowledge of previous tasks for learning new tasks fast. The performance of the algorithm is evaluated for ten MTPR problems using five University of California at Irvine (UCI) data sets. The experiments verify that the proposed algorithm can indeed acquire and accumulate task knowledge and that the transfer of knowledge from tasks already learned enhances the speed of knowledge acquisition on new tasks and the final classification accuracy. In addition, the task categorization accuracy is greatly improved for all MTPR problems by introducing the reorganization process even if the presentation order of class training examples is fairly biased. [ABSTRACT FROM AUTHOR]

Predictive emission monitoring systems (PEMS) are software solutions for the validation and supplementation of costly continuous emission monitoring systems for natural gas electrical generation turbines. The basis of PEMS is that of predictive models trained on past data to estimate emission components. The gas turbine process dataset from the University of California at Irvine open data repository has initiated a challenge of sorts to investigate the quality of models of various machine learning methods to build a model for predicting CO and NOx emissions depending on ambient variables and the parameters of the technological process. The novelty and features of this paper are: (i) a contribution to the study of the features of the open dataset on CO and NOx emissions for gas turbines, which will enable one to more objectively compare different machine learning methods for further research; (ii) for the first time for the CO and NOx emissions, a model based on symbolic regression and a genetic algorithm is presented—the advantage of this being the transparency of the influence of factors and the interpretability of the model; (iii) a new classification model based on the symbolic regression model and fuzzy inference system is proposed. The coefficients of determination of the developed models are: R 2 = 0.83 for NOx emissions, R 2 = 0.89 for CO emissions. [ABSTRACT FROM AUTHOR]

Hepatitis C is a liver infection caused by a virus, which results in mild to severe inflammation of the liver. Over many years, hepatitis C gradually damages the liver, often leading to permanent scarring, known as cirrhosis. Patients sometimes have moderate or no symptoms of liver illness for decades before developing cirrhosis. Cirrhosis typically worsens to the point of liver failure. Patients with cirrhosis may also experience brain and nerve system damage, as well as gastrointestinal hemorrhage. Treatment for cirrhosis focuses on preventing further progression of the disease. Detecting cirrhosis earlier is therefore crucial for avoiding complications. Machine learning (ML) has been shown to be effective at providing precise and accurate information for use in diagnosing several diseases. Despite this, no studies have so far used ML to detect cirrhosis in patients with hepatitis C. This study obtained a dataset consisting of 28 attributes of 2038 Egyptian patients from the ML Repository of the University of California at Irvine. Four ML algorithms were trained on the dataset to diagnose cirrhosis in hepatitis C patients: a Random Forest, a Gradient Boosting Machine, an Extreme Gradient Boosting, and an Extra Trees model. The Extra Trees model outperformed the other models achieving an accuracy of 96.92%, a recall of 94.00%, a precision of 99.81%, and an area under the receiver operating characteristic curve of 96% using only 16 of the 28 features. [ABSTRACT FROM AUTHOR]

Breast cancer is one of the largest second causes of dead among women. So, computer aided diagnosis system will be working on mammograms. In early stage breast cancer can be identified by through CAD algorithms. But, this algorithms accuracy of existing system unsatisfied result. Later, they are using to detect this cancer like microscopic images, mammography to ultrasonography and MRI. This type of prediction gave only false prediction then it took more time and cost effects. In proposed system of in this project, it will be working on big data and machine learning based on nine features in breast cancer data set from UCI Irvine Machine Learning Repository database. Big data is used to pre-processing from various fields from datasets and used to accurate detection. Machine Learning can be used to implementing the supervised algorithm’s to detect cancer type based on benign and malignant breast masses. These algorithms will give us additional accurate results for detecting the breast cancer. [ABSTRACT FROM AUTHOR]

Swarm intelligence techniques have a vast range of real world applications.Some applications are in the domain of medical data mining where, main attention is on structure models for the classification and expectation of numerous diseases. These biomedical applications have grabbed the interest of numerous researchers because these are most serious and prevalent causes of death among the human whole world out of which breast cancer is the most serious issue. Mammography is the initial screening assessment of breast cancer. In this study, an enhanced version of Harris Hawks Optimization (HHO) approach has been developed for biomedical databases, known as DLHO. This approach has been introduced by integrating the merits of dimension learning-based hunting (DLH) search strategy with HHO. The main objective of this study is to alleviate the lack of crowd diversity, premature convergence of the HHO and the imbalance amid the exploration and exploitation. DLH search strategy utilizes a dissimilar method to paradigm a neighborhood for each search member in which the neighboring information can be shared amid search agents. This strategy helps in maintaining the diversity and the balance amid global and local search. To evaluate the DLHO lot of experiments have been taken such as (i) the performance of optimizers have analysed by using 29-CEC -2017 test suites, (ii) to demonstrate the effectiveness of the DLHO it has been tested on different biomedical databases out of which we have used two different databases for Breast i.e. MIAS and second database has been taken from the University of California at Irvine (UCI) Machine Learning Repository.Also to test the robustness of the proposed method its been tested on two other databases of such as Balloon and Heart taken from the UCI Machine Learning Repository. All the results are in the favour of the proposed technique. [ABSTRACT FROM AUTHOR]

Trained radial basis function networks are well-suited for use in extracting rules and explanations because they contain a set of locally tuned units. However, for rule extraction to be useful, these networks must first be pruned to eliminate unnecessary weights. The pruning algorithm cannot search the network exhaustively because of the computational effort involved. It is shown that using multiple pruning methods with smart ordering of the pruning candidates, the number of weights in a radial basis function network can be reduced to a small fraction of the original number. The complexity of the pruning algorithm is quadratic (instead of exponential) in the number of network weights. Pruning performance is shown using a variety of benchmark problems from the University of California, Irvine machine learning database. [ABSTRACT FROM AUTHOR]