Suchergebnisse

Relative to the number of casualties, COVID-19 ranks among the ten most devastating plagues in history. The pandemic hit the South Asian nation of Bangladesh in early March 2020 and has greatly impacted the socio-economic status of the country. In this article, we propose a compartmental model for COVID-19 dynamics, introducing a separate class for quarantined susceptibles, synonymous to isolation of individuals who have been exposed and are suspected of being infected. The current model assumes a perfect quarantine based on contact with infectious individuals. Numerical simulation is conducted to investigate the efficiency of disease control by segregating suspected individuals and other non-pharmaceutical interventions. In addition, we assort quantitatively the importance of parameters that influence the dynamics of the system. Fitting the system to the early phase of COVID-19 outbreaks in Bangladesh, by taking into account the cumulative number of cases with the data of the first 17-week period, the basic reproduction number is estimated as 1.69.

In this article, relevant approaches to creating and using mathematical models in linguistics are examined. Analyzed are the types of mathematical methods used to solve linguistic tasks. Also investigated are the fundamental principles for creating mathematical models. Stressed in particular is that the approach usually applied in solving physical tasks can be effective. Its typical feature is that a universal theory explaining the interaction among a system's functional elements on the most general level is the basis of the model. In this case, the necessary approximating dependencies are not conjectured, as is often the case, but obtained or calculated on the basis of the fundamental theory. The prospects for applying this approach in linguistics are defined. Examples of the modeling of linguistic systems and a comparative analysis of their various methods are provided. Much attention is paid to the universality of the mathematical models used to analyze linguistic material. It is noted that if the approximate dependencies used are based on a system's general universal properties, this can significantly complicate the quantitative analysis of the data. The advantages and shortcomings of mathematical modeling in solving applied linguistic tasks are illustrated. Besides "classic" models (such as Zipf's law for the rank distribution of words in a text), other modeling approaches are described. In particular, attention is paid to the methods of constructing mathematical models based on neural networks. In this case, the approximation of dependence is realized in the form of sequential essentially nonlinear transformations. The main disadvantage of this approach is connected with the technical complications of execution and with the absence of overt analytical dependence. The main advantage is hidden in the potential possibility of creating approximate dependencies of practically any type. As a comparison of various modeling approaches, the article provides examples of solving the same linguistic tasks using several methods.

The U.S. Department of Defense (DoD) has experienced persistent challenges with software development across various acquisition programs. Attempts to address those challenges through workforce initiatives (such as hiring, training, and professional development) have been severely hampered by an inability to identify and characterize the presumably tens of thousands of software professionals working within DoD's programs. In this report, the authors attempt to identify and characterize DoD's software workforce, the types of software developed within DoD, and the variety of methods DoD programs use when employing that workforce to develop software. The research team's effort to develop a comprehensive picture of the software workforce was hindered by a lack of data in existing DoD databases and by internal DoD organizational issues that limited the team's ability to gather data. Although this report provides a snapshot of DoD's software workforce, the types of software developed within DoD, and the variety of methods DoD programs use when employing that workforce to develop software, there are significant limitations in the underlying data. The research team's recommendations, using lessons learned from this effort, focus on what defense leaders will need to better understand, train, and employ DoD's software acquisition workforce in the future

"The difficult and continually evolving operations in Iraq and Afghanistan show the complexities of what is now termed irregular warfare and highlight the need for new approaches to the security challenges with which the United States is now contending and will likely confront in the future. The research reported in this monograph focused on answering a rather straightforward, but thus far largely unanswered, question: What can the U.S. military learn from other militaries about how better to prepare for full-spectrum operations and deployments? To this end, RAND was asked by the OSD for Personnel and Readiness to examine the militaries of China, France, the UK, India, and Israel."--P. xiii

The Swedish National Program to Develop Suicide Prevention was created through collaboration between the National Board of Health and Welfare, the National Institute of Public Health, and the Centre for Suicide Research and Prevention. These institutions aim to provide joint support for the development of suicide prevention in Sweden by, for example, encouraging educational and development project. The program was signed for the National Council for Suicide Prevention by Agneta Dreber, Director-General, National Institute of Public Health; Danuta Wasserman, Professor of Psychiatry and Suicidology, Head, National Centre for Suicide Research and Prevention; and Claes Örtendahl, Director-General, National Board of Health and Welfare. This article, edited by Jan Beskow, Professor of Psychiatry at the Centre for Suicide Research and Prevention, is an abridged version submitted to Crisis at the request of the Editors.