Vol 16, No 1 (2025)
Using the Mask R-CNN model for segmentation of real estate objects in aerial photographs
Abstract
The mass appearance of illegal and unregistered in the Unified State Register of Real Estate (USRRE) real estate objects complicates cadastral registration for many entities at the territorial and administrative levels. Traditional methods of identifying objects of this type, based on manual analysis of geospatial data, are labor-intensive and time-consuming.To improve the efficiency of this process, it is proposed to automate the detection of objects in aerial photographs by solving the instance segmentation problem using the Mask R-CNN deep learning model. The article describes the preparation of a dataset for this model, examines the main quality metrics, and analyzes the results obtained. The efficiency of the Mask R-CNN model in practice is shown for solving the problem of detecting construction projects that are not registered in the USRRE.
Program Systems: Theory and Applications. 2025;16(1):3-44
3-44
On the implementation of QR-decomposition on a three-dimensional systolic array
Abstract
Intensive data flows, formed systems of linear equations in real time, as well as systems of linear equations of large dimensionality cause the involvement of systolic arrays for their machine solution. In the presented systolic array, designed to reduce matrices to triangular form, the realization of orthogonal rotation transformations can be carried out both by two-dimensional vector rotation devices CORDIC, and its modifications. For the proposed systolic array, descriptions of its configuration, operation and technical characteristics, as well as the structure of input and output data flow are given.
Program Systems: Theory and Applications. 2025;16(1):45-59
45-59
The ontology complex as the model of intelligent system to support rehabilitation of patients after stroke
Abstract
The main incentive for the introduction of computer technologies into the healthcare system is the desire to significantly improve the quality of life of people. This includes improving the quality and speed of treatment, reducing the cost of medical services and acquiring effective means to comply with regulatory requirements.At the present stage of rehabilitation development, the need for active implementation of medical decision support systems and artificial intelligence technologies becomes obvious. These technologies can significantly improve the understanding of the clinical aspects of disorders, the level of activity and participation of stroke patients in the rehabilitation process. A key component of the successful application of these systems is the importance of formalizing knowledge and creating ontologies that provide a structured and connected presentation of medical information and define the rules for their interpretation.This paper presents a set of interrelated ontological models underlying the intellectual decision support system being developed in the rehabilitation of stroke patients. The IACPaaS cloud platform is used to implement the complex of ontologies. Ontologies and the target resources generated on their basis are the basic elements of the system being developed, which will soon be provided to healthcare professionals to solve urgent rehabilitation issues. Mechanisms are provided for the planned expansion and refinement of the knowledge base, which will allow the system to easily adapt to new medical research results and optimize its work as a whole.
Program Systems: Theory and Applications. 2025;16(1):61-82
61-82
Multimodal Stock Price Prediction: A Case Study of the Russian Securities Market
Abstract
Classical asset price forecasting methods primarily rely on numerical data, such as price time series, trading volumes, limit order book data, and technical analysis indicators. However, the news flow plays a significant role in price formation, making the development of multimodal approaches that combine textual and numerical data for improved prediction accuracy highly relevant.This paper addresses the problem of forecasting financial asset prices using the multimodal approach that combines candlestick time series and textual news flow data. A unique dataset was collected for the study, which includes time series for 176 Russian stocks traded on the Moscow Exchange and $79,555$ financial news articles in Russian.For processing textual data, pre-trained models RuBERT and Vikhr-Qwen2.5-0.5b-Instruct (a large language model) were used, while time series and vectorized text data were processed using an LSTM recurrent neural network. The experiments compared models based on a single modality (time series only) and two modalities, as well as various methods for aggregating text vector representations.Prediction quality was estimated using two key metrics: Accuracy (direction of price movement prediction: up or down) and Mean Absolute Percentage Error (MAPE), which measures the deviation of the predicted price from the true price. The experiments showed that incorporating textual modality reduced the MAPE value by 55%.The resulting multimodal dataset holds value for the further adaptation of language models in the financial sector. Future research directions include optimizing textual modality parameters, such as the time window, sentiment, and chronological order of news messages.
Program Systems: Theory and Applications. 2025;16(1):83-130
83-130