Algorithms For Classification Tasks Research Articles

Comparative analysis and prediction of coronary heart disease

Cardiovascular disease (CVD) <span>is now one of the leading causes of death worldwide and was also thought to be a serious illness in the mid and old ages. Artificial intelligence and machine learning have a huge impact on the healthcare areas. As a result, getting a familiar individual with data processing techniques suitable for numerical health data. Although, the most often used algorithms for classification tasks will be incredibly advantageous in terms of time management. In particular here, a common procedure has been proposed for predicting cardiovascular disease. Accordingly, we herein consider nine typical classifiers of both machine learning and deep learning technology for the comparative analysis and prediction of coronary heart failure. These models are computationally inexpensive and easy to build. Moreover, these classifiers are tested and compared using a confusion matrix in the Jupyter notebook, yielding classification measures such as accuracy, f1-score, recall, and precision. As a result, the logistic regression classifier gives the maximum possible accuracy, precision, and f1-score of 90.78%, 90.24%, and 91.35% respectively.</span>

VoPo leverages cellular heterogeneity for predictive modeling of single-cell data

High-throughput single-cell analysis technologies produce an abundance of data that is critical for profiling the heterogeneity of cellular systems. We introduce VoPo (https://github.com/stanleyn/VoPo), a machine learning algorithm for predictive modeling and comprehensive visualization of the heterogeneity captured in large single-cell datasets. In three mass cytometry datasets, with the largest measuring hundreds of millions of cells over hundreds of samples, VoPo defines phenotypically and functionally homogeneous cell populations. VoPo further outperforms state-of-the-art machine learning algorithms in classification tasks, and identified immune-correlates of clinically-relevant parameters.

Adaptive modelling of spatial diversification of soil classification units

AbstractThe article presents the results of attempts to use adaptive algorithms for classification tasks different soils units. The area of study was the Upper Silesian Industrial Region, which physiographic and soils parameters in the form of digitized was used in the calculation. The study used algorithms, self-organizing map (SOM) of Kohonen, and classifiers: deep neural network, and two types of decision trees: Distributed Random Forest and Gradient Boosting Machine. Especially distributed algorithm Random Forest (algorithm DRF) showed a very high degree of generalization capabilities in modeling complex diversity of soil. The obtained results indicate, that the digitization of topographic and thematic maps give you a fairly good basis for creating useful models of soil classification. However, the results also showed that it cannot be concluded that the best algorithm presented in this research can be regarded as a general principle of system design inference.

Parameters Selection of LLE Algorithm for Classification Tasks

The crux in the locally linear embedding algorithm or LLE is the selection of embedding dimensionality and neighborhood size. A method of parameters selection based on the normalized cut criterion or Ncut for classification tasks is proposed. Differing from current techniques based on the neighborhood topology preservation criterion, the proposed method capitalizes on class separability of embedding result. By taking it into consideration, the intrinsic capability of LLE can be more faithfully reflected, and hence more rational features for classification in real-life applications can be offered. The theoretical argument is supported by experimental results from synthetic and real data sets.

Blocked 3×2 Cross-Validated t-Test for Comparing Supervised Classification Learning Algorithms

In the research of machine learning algorithms for classification tasks, the comparison of the performances of algorithms is extremely important, and a statistical test of significance for generalization error is often used to perform it in the machine learning literature. In view of the randomness of partitions in cross-validation, a new blocked 3×2 cross-validation is proposed to estimate generalization error in this letter. We then conduct an analysis of variance of the blocked 3×2 cross-validated estimator. A relatively conservative variance estimator that considers the correlation between any two two-fold cross-validations, and was previously neglected in 5×2 cross-validated t and F-tests is put forward. A corresponding test using this variance estimator is presented to compare the performances of algorithms. Simulated results show that the performance of our test is comparable with that of 5×2 cross-validated tests but with less computation complexity.

An empirical evaluation of constructive neural network algorithms in classification tasks

Unlike conventional Neural Network (NN) algorithms that require the definition of the NN architecture before learning starts, Constructive Neural Network (CoNN) algorithms enable the network architecture to be constructed along with the learning process. This paper presents and discusses the results of an empirical evaluation of seven two-class CoNN algorithms, namely Tower, Pyramid, Tiling, Upstart, Shift, Perceptron Cascade (PC) and Partial Target Inversion (PTI) in 12 knowledge domains. The way each particular algorithm approaches the growing of the network determines their differences. This paper also presents and analyses empirical results of five multiclass CoNN algorithms in five knowledge domains, namely MTower, MPyramid, MTiling, MUpstart and MPerceptron Cascade, which can be considered extensions of their two-class counterparts. Results obtained with the Pocket with the Ratchet Modification (PRM) algorithm, with its multiclass version, the PRMWTA algorithm and with the back propagation algorithm, are presented for comparison.

Efficient adaptive learning for classification tasks with binary units.

This article presents a new incremental learning algorithm for classification tasks, called NetLines, which is well adapted for both binary and real-valued input patterns. It generates small, compact feedforward neural networks with one hidden layer of binary units and binary output units. A convergence theorem ensures that solutions with a finite number of hidden units exist for both binary and real-valued input patterns. An implementation for problems with more than two classes, valid for any binary classifier, is proposed. The generalization error and the size of the resulting networks are compared to the best published results on well-known classification benchmarks. Early stopping is shown to decrease overfitting, without improving the generalization performance.