Detection of Radicalisation and Extremism Online: A Survey

Introduction: Due to the lack of regulation, the large volume of user-generated online content reflects more closely the offline world than official news sources. Therefore, social media platforms have become an attractive space for anyone seeking independent information. One of the main goals of this work is to clarify concepts such as Extremism and Collective Radicalisation, Social Media, Sentiments/Emotions/Opinions Analysis, as well as the combinations of all of them. Methods: The automatic identification of extremism and collective radicalisation requires sophisticated Natural Language Processing (NLP) methods and resources, especially those dealing with opinions, emotions or sentiment analysis. Text mining and knowledge extraction are also crucial, in particular, directed toward social media and micro-blogging. Results: The present document comprehends a study on theoretical material, focusing on the main concepts of the subject, including the main problems and challenges, from the different areas that compose online radicalisation research. Understanding and detecting extremism and collective radicalism online has a connection to sentiment analysis and opinion mining. There are many barriers to understanding extremism and collective radicalisation; one is to differentiate between who is really engaged in the process and who is just eventually talking about it. Conclusions: The other focus of this work is to find the best ways to identify extremism and collective radicalisation on the internet, using sentiment analysis and focusing on probabilistic methods to create an unsupervised and language-independent approach.

On the classification of motive power failures

The Aim of the article is to develop a motive power failure classification to enable substantiated definition of dependability requirements for motive power as a part of a railway transportation system, as well as for organizing systematic measures to ensure a required level of its dependability over the life cycle. Methods. The terminology of interstate dependability-related standards was analysed and the two classifications used by OJSC “RZD” for estimating the dependability of technical systems and motive power were compared. The dependability of railway transportation systems is studied using structural and logical and logical and probabilistic methods of dependability analysis, while railway lines are examined using the graph theory and the Markov chains. Results. An analysis of the existing failure classifications identified shortcomings that prevent the use of such classifications for studying the structural dependability of such railway transportation systems as motive power. A classification was developed that combines two failure classifications (“category-based” for the transportation process and technical systems and “type-based” for the motive power), but this time with new definitions. The proposed classification of the types of failures involves stricter definitions of the conditions and assumptions required for evaluating the dependability and technical condition of an item, which ensures correlation between the characteristics of motive power and its dependability throughout the life cycle in the context of the above tasks. The two classifications could be used simultaneously while researching structural problems of dependability using logical and probabilistic methods and Markov chains. The developed classification is included in the provisions of the draft interstate standard “Dependability of motive power. Procedure for the definition, calculation methods and supervision of dependability indicators throughout the life cycle” that is being prepared by JSC “VNIKTI” in accordance with the OJSC “RZD” research and development plan. Conclusion. The article’s findings will be useful to experts involved in the evaluation of motive power dependability.

Declarative Approaches to Counterfactual Explanations for Classification

We propose answer-set programs that specify and compute counterfactual interventions on entities that are input on a classification model. In relation to the outcome of the model, the resulting counterfactual entities serve as a basis for the definition and computation of causality-based explanation scores for the feature values in the entity under classification, namely responsibility scores. The approach and the programs can be applied with black-box models, and also with models that can be specified as logic programs, such as rule-based classifiers. The main focus of this study is on the specification and computation of best counterfactual entities, that is, those that lead to maximum responsibility scores. From them one can read off the explanations as maximum responsibility feature values in the original entity. We also extend the programs to bring into the picture semantic or domain knowledge. We show how the approach could be extended by means of probabilistic methods, and how the underlying probability distributions could be modified through the use of constraints. Several examples of programs written in the syntax of the DLV ASP-solver, and run with it, are shown.

TOWARDS A NEW SOLAS CONVENTION: A TRANSFORMA TION OF THE SHIP SAFETY REGULATORY FRAMEWORK

I congratulate the author of the article with his successful publication. The necessity to change the SOLAS is timely. Suggestions concerning simplifications in the application of the Convention, the use of more simple methods and delegation of greater freedom of choice for ship owners and designers have the right. In my opinion, the increasing use of probabilistic methods in the rules of the Convention is not always justified. For example, the Graph - analytical method that has been used previously to assess the unsinkable ship, gives the designer more opportunity than the current Probabilistic standard.

Methods of Learning the Structure of the Bayesian Network

Sometimes in practice it is necessary to calculate the probability of an uncertain cause, taking into account some observed evidence. For example, we would like to know the probability of a particular disease when we observe the patient’s symptoms. Such problems are often complex with many interrelated variables. There may be many symptoms and even more potential causes. In practice, it is usually possible to obtain only the inverse conditional probability, the probability of evidence giving the cause, the probability of observing the symptoms if the patient has the disease.Intelligent systems must think about their environment. For example, a robot needs to know about the possible outcomes of its actions, and the system of medical experts needs to know what causes what consequences. Intelligent systems began to use probabilistic methods to deal with the uncertainty of the real world. Instead of building a special system of probabilistic reasoning for each new program, we would like a common framework that would allow probabilistic reasoning in any new program without restoring everything from scratch. This justifies the relevance of the developed genetic algorithm. Bayesian networks, which first appeared in the work of Judas Pearl and his colleagues in the late 1980s, offer just such an independent basis for plausible reasoning.This article presents the genetic algorithm for learning the structure of the Bayesian network that searches the space of the graph, uses mutation and crossover operators. The algorithm can be used as a quick way to learn the structure of a Bayesian network with as few constraints as possible.learn the structure of a Bayesian network with as few constraints as possible.

Human health risk assessment of lead, cadmium, and mercury co-exposure from agricultural soils in the Tuzla Canton (Bosnia and Herzegovina)

The aim of this study was to assess the risk of human exposure to lead (Pb), cadmium (Cd), and mercury (Hg) through agricultural soil by considering both uncertainty and variability in key exposure parameters. For this reason we collected soil samples from 29 locations in the Tuzla Canton (Bosnia and Herzegovina) and measured their metal levels with inductively coupled plasma atomic emission or absorption spectrometry (ICP-AES and ICP-AAS, respectively). The levels of Pb ranged from 13.33 to 1692.33 mg/kg, of Cd from 0.05 to 3.67 mg/kg, and of Hg from 0.02 to 2.73 mg/kg. To estimate cancer and non-cancer risks we used deterministic and semi-probabilistic methods. Lead was found to involve higher health risk than the other two heavy metals. Its hazard index (HI) decreased between population groups (children>women>men) and exposure routes (ingestion>skin contact>inhalation). Our Monte Carlo simulations indicated that Pb HIs for both adult populations had a 0.6 % probability to exceed the threshold value of 1, while in children this probability was 14.2 %. Cd and Hg showed no probability to exceed the threshold in any scenario. Our simulation results raise concern about possible adverse health effects of heavy metals from soil, especially in children. It is very important to continue monitoring environmental pollution and assess human health risk, not only with respect to soil, but also with other important environmental compartments, such as air and water.

Modeling the integral risk assessment for air pollution in the areas of highways by probabilistic methods

A methodology for calculating the integral risk of atmospheric pollution using Bayes’s theorem is proposed to take into account the action of mobile and stationary emission sources in the influence zones of highways, the response to the impact in the form of accumulation of emission components in depositing media and biological media of the population. At the first stage, the clustering of experimental data arrays was carried out, homogeneous road sections (clusters) were identified. The integral risk was calculated for the selected clusters. The risks of contamination of the investigated media have been calculated. A multiple regression model has been built to assess the level of integral risk with a high degree of reliability when compared with experimental data. The significance of the aerogenic factor in the formation of the level of integral risk is shown. A reduced model for assessing the integral risk by the level of risk of atmospheric air pollution is proposed. Grades of risk levels are given according to the degree of acceptability. It is possible to determine the contribution of the road transport component to the level of integral risk based on the obtained values of the final risk.

Assessing and predicting individual occupational risk and determining its exact categories using probabilistic methods

Existing approaches to occupational risk assessment more often involve evaluating its group levels and individual risks are assessed less frequently. These approaches provide deterministic risk assessment which doesn’t take into account uncertainty in risk categorizing when its values are close to boundaries between adjoining risks categories. It substantiates the necessity to assess occupational risk levels using probabilistic methods. Our research object was occupational risk and the basic subject was distribution of individual occupational risk levels among workers. Our test group was made up of oil and gas extraction operators exposed to noise equal to 80–85 dBA at their workplaces (173 people). Our control group included oil and gas extraction operators and engineering and technical personnel occupationally exposed to noise equal to 60–77.8 dBA (259 people). We performed a priori assessment of occupational health risks; accomplished epidemiologic analysis of a cause-effect relation between health disorders and work; calculated group occupational health risks; calculated and predicted individual occupational risk using mathematical modeling of dependence between probable negative responses and working conditions, age, and period of employment; determined risk categories more precisely using fuzzy sets by calculating the membership function. As a result, we established that proven individual risk levels were distributed unevenly (1.06•10-4–1.47•10-2) as per categories within a group characterized with a suspected average risk level. A category of proven individual risk levels was determined more precisely using fuzzy sets; after that distribution of probability of their membership was evaluated to detect that at the moment of the research a share of workers with their proven individual occupational risks falling into lower risk categories (p > 0.5) amounted to 89.6 %. We attempted to predict risks for the whole employment period given that working conditions remained the same and no prevention activities were provided. Our prediction revealed that individual occupational risks would remain unacceptable for all workers in the test group and would amount to 2.53•10-2–3.51•10-2; a risk category was also expected to become higher. In-dividual occupational risk would be categorized as average for most workers and as high for 23 % of them (p < 0.5).

PROBABILITY METHODS OF ANALYSIS OF ASSESSMENT OF THE CONDITION OF ROAD CLOTHES IN USE

This article is a continuation of the analysis of methods and criteria for assessing the condition of non-rigid pavement, which contains in its structural layers such hidden defects as cracks, material disintegration, violation of the structure of intermediate layers of monolithic material and others. The variety of models for assessing the condition of the pavement with destruction is explained by: variability of soil-geological, climatic conditions of operation of pavement; the variety of physical and mechanical characteristics of the materials of the structural layers of the pavement and the soil of the ground; heterogeneity of geometric parameters of pavements in the longitudinal and transverse profiles. According to many scientists to assess the condition of the pavement is a necessary condition for the use of methods of probabilistic analysis. Methods of designing and assessing the condition of the pavement structure must take into account the principles of reliability, including probabilistic methods of reliability analysis. Most of the considered probabilistic models for assessing the condition of pavement are focused on the design of new construction, which allows not to take into account the patterns of changes in physical and mechanical properties of materials of structural layers and their geometric parameters during operation. However, the most relevant is to assess the condition of pavement with damage, including hidden cracks. Much of the article is devoted to the analysis of this area of research. According to the results of the analysis, it is concluded that probabilistic methods allow to take into account the heterogeneity of parameters that characterize the stress-strain state of the pavement structure and can be used in models to assess the condition of pavement with cracks in layers of monolithic materials.