Machine Learning Methods for Detecting and Monitoring Extremist Information on the Internet

Programming and Computer Software - In this paper, we employ machine learning methods to solve the problem of countering terrorism and extremism by using information from the Internet. This problem...     

Predicting COVID-19-Induced Lung Damage Based on Machine Learning Methods

Programming and Computer Software - In this paper, we consider the course of the coronavirus disease (COVID-19) in human patients. We investigate anamnesis, examination, and clinical analysis data,...     

Peculiarities of the Microstructure and Properties of Parts Produced by the Direct Laser Deposition of 316L Steel Powder

Russian Journal of Non-Ferrous Metals - The direct laser deposition of metal powders is one additive method of producing functional materials. It consists of the melting of metallic powders by a...     

Optimization of Novikov gears: Numerical solution

The load capacity of Novikov gears proves higher than that of involute gears. The conditions of existence and quality of the Novikov gear system are represented by a system of equations and constraints. Numerical solution yields three-dimensional arrays consisting of the numbers of teeth of the interlocking gears and the corresponding geometric parameters.     

Automatic text summarization using latent semantic analysis

In the paper, the most state-of-the-art methods of automatic text summarization, which build summaries in the form of generic extracts, are considered. The original text is represented in the form of a numerical matrix. Matrix columns correspond to text sentences, and each sentence is represented in the form of a vector in the term space. Further, latent semantic analysis is applied to the matrix obtained to construct sentences representation in the topic space. The dimensionality of the topic space is much less than the dimensionality of the initial term space. The choice of the most important sentences is carried out on the basis of sentences representation in the topic space. The number of important sentences is defined by the length of the demanded summary. This paper also presents a new generic text summarization method that uses nonnegative matrix factorization to estimate sentence relevance. Proposed sentence relevance estimation is based on normalization of topic space and further weighting of each topic using sentences representation in topic space. The proposed method shows better summarization quality and performance than state-of-the-art methods on the DUC 2001 and DUC 2002 standard data sets.     

Outlier Detection Algorithms in Data Mining Systems

The paper discusses outlier detection algorithms used in data mining systems. Basic approaches currently used for solving this problem are considered, and their advantages and disadvantages are discussed. A new outlier detection algorithm is suggested. It is based on methods of fuzzy set theory and the use of kernel functions and possesses a number of advantages compared to the existing methods. The performance of the algorithm suggested is studied by the example of the applied problem of anomaly detection arising in computer protection systems, the so-called intrusion detection systems.     

Turbulent pulsations of the microelectrode limiting diffusion current

Flat mounted microelectrodes, for which the longitudinal length,L, is small in comparison with the diffusion entrance region are used to measure wall shear stress fluctuations. However, microelectrode turbulent noise depends both on wall shear stress fluctuations (i.e. longitudinal velocity pulsations) and on normal pressure gradient fluctuations (i.e. normal velocity pulsations). The relative value of these two factors depends on the microelectrode length. The influence of normal and longitudinal velocity pulsations on the microelectrode turbulent noise are the same order of magnitude if the lengthL is approximately 10 or 20 times the thickness of the viscous sublayer. Consequently it is possible to determine statistical characteristics of normal and longitudinal velocity fluctuations by using a number of microelectrodes of different lengths in the flow direction.This paper was presented at the Workshop on Electrodiffusion Flow Diagnostics, CHISA, Prague, August 1990.     

Design for a Hydrocyclone with a Disk Deflector Offering Improved Efficiency for Separating Mechanical Impurities

Chemical and Petroleum Engineering - A novel design of a hydrocyclone incorporating a disk deflector is proposed. This component allows the separation efficiency to be increased by reducing the...