All issues

Modern big science projects are becoming highly data intensive to the point where offline processing of stored data is infeasible. High data throughput computing, or Data Stream Computing, for future projects is required to deal with terabytes of data per second which cannot be stored in long-term storage elements. Conventional data-centres based on typical server-grade hardware are expensive and are biased towards processing power. The overall I/O bandwidth can be increased with massive parallelism, usually at the expense of excessive processing power and high energy consumption. An ARM System on Chip (SoC) based processing unit may address the issue of system I/O and CPU balance, affordability and energy efficiency since ARM SoCs are mass produced and designed to be energy efficient for use in mobile devices. Such a processing unit is currently in development, with a design goal of 20 Gb/s I/O throughput and significant processing power. The I/O capabilities of consumer ARM System on Chips are discussed along with to-date performance and I/O throughput tests.

Detecting controversy in online discussions is critical for managing public relations, as it helps inform various processes from policymaking to business. This work aims to expand approaches to online controversy detection based on the expressed emotions. Controversy was defined as an online content phenomenon of provoking disagreements and conflict. This study builds upon prior semantic methods by analyzing estimates of emotional connotations of messages. Modern language models for emotion recognition and named entity recognition are explored as tools of controversy detection. The outputs of these models were aggregated by entity to estimate the entity’s emotional connotation. The emotional divergence score based on the dispersion of emotions was proposed to quantify controversy in user content. Then, entities with sufficiently high emotional divergence relative to the domain of discussions were selected as markers of controversy. A case study of Reddit data related to Sri-Lankan 2022 political crisis was conducted, showing the capabilities of emotional divergence score in controversy detection. A total of two datasets were collected with different methodologies: one aimed at collecting earlier messages and another aimed at collecting more recent ones. The collected data contained discussions of policy, public figures, organizations and locations tied to the crisis. When measured on manually annotated data samples, the proposed method achieved a recall value of 0.705 and a precision value close to 0.496 for the first dataset, while recall of 0.716 and precision of 0.436 were recorded for the second dataset. The main factors that limit the precision were found to be the quality of underlying models and false positives: highly discussed non-controversial markers. Lastly, it was identified that a study of regular emotional distribution of social media content may be helpful for improving controversy detection quality.

The purpose of this study is to develop a methodology for change points detection in time series, including financial data. The theoretical basis of the study is based on the pieces of research devoted to the analysis of structural changes in financial markets, description of the proposed algorithms for detecting change points and peculiarities of building classical and deep machine learning models for solving this type of problems. The development of such tools is of interest to investors and other stakeholders, providing them with additional approaches to the effective analysis of financial markets and interpretation of available data.

To address the research objective, a neural network was trained. In the course of the study several ways of training sample formation were considered, differing in the nature of statistical parameters. In order to improve the quality of training and obtain more accurate results, a methodology for feature generation was developed for the formation of features that serve as input data for the neural network. These features, in turn, were derived from an analysis of mathematical expectations and standard deviations of time series data over specific intervals. The potential for combining these features to achieve more stable results is also under investigation.

The results of model experiments were analyzed to compare the effectiveness of the proposed model with other existing changepoint detection algorithms that have gained widespread usage in practical applications. A specially generated dataset, developed using proprietary methods, was utilized as both training and testing data. Furthermore, the model, trained on various features, was tested on daily data from the S&P 500 index to assess its effectiveness in a real financial context.

As the principles of the model’s operation are described, possibilities for its further improvement are considered, including the modernization of the proposed model’s structure, optimization of training data generation, and feature formation. Additionally, the authors are tasked with advancing existing concepts for real-time changepoint detection.

The modern digitalization of cities occurs according to laws that are not yet clearly defined. The difficulty in understanding these laws is due to the high uncertainty of urban events and the difficulties in developing adequate analytical models for them. To substantiate management decisions on city digitalization in these conditions, effective methods for processing multidimensional time series are necessary. The objective is to develop a promising method. The proposed method is based on a new model of a neural network control system and algorithms for implementing its main functions. Distinctive features of the model include both structural aspects and new rules for neural network-based substantiation of management decisions on city digitalization, using unsupervised training of the applied recurrent neural network. This model provides for adaptive neural network management of city digitalization according to a sliding program. New rules are proposed for the forward and backward transformation of multidimensional time series, modeling the operation of a streaming recurrent neural network, and identifying feasible programs for city digitalization. By adhering to the proposed rules, it is possible not only to eliminate information loss in such a network but also to ensure its stable operation. The features of this neural network’s operation in substantiating feasible urban digitalization programs are described. The requirements for the software implementation of the proposed method are substantiated. The results of experiments using data for St. Petersburg are presented. These data included monthly volumes of shipped computers, electronic and optical products, as well as monthly volumes of shipped products for all manufacturing industries minus the first-time series values ??for the period from January 2021 to December 2025. The effectiveness of management decisions on the city’s digitalization was predicted based on the total volumes of shipped products to the horizon from January to December 2026. It is shown that the application of the proposed method allows for expanded capabilities for adaptively substantiating feasible management decisions on the city’s digitalization while reducing computer modeling costs.

High frequency algorithmic trading became is a subclass of trading which is focused on gaining basis-point like profitability on sub-second time frames. Such trading strategies do not depend on most of the factors eligible for the longer-term trading and require specific approach. There were many attempts to utilize machine learning techniques to both high and low frequency trading. However, it is still having limited application in the real world trading due to high exposure to overfitting, requirements for rapid adaptation to new market regimes and overall instability of the results. We conducted a comprehensive research on combination of known quantitative theory and reinforcement learning methods in order derive more effective and robust approach at construction of automated trading system in an attempt to create a support for a known algorithmic trading techniques. Using classical price behavior theories as well as modern application cases in sub-millisecond trading, we utilized the Reinforcement Learning models in order to improve quality of the algorithms. As a result, we derived a robust model which utilize Deep Reinforcement learning in order to optimise static market making trading algorithms’ parameters capable of online learning on live data. More specifically, we explored the system in the derivatives cryptocurrency market which mostly not dependent on external factors in short terms. Our research was implemented in high-frequency environment and the final models showed capability to operate within accepted high-frequency trading time-frames. We compared various combinations of Deep Reinforcement Learning approaches and the classic algorithms and evaluated robustness and effectiveness of improvements for each combination.