Institute of Informatics and Statistics

Autism spectrum disorder (ASD) affects many children and limits their social interaction, communication, and behavioral skills. Regular follow-ups with qualified professionals assist in the patients' development through sessions and progress evaluations. This progress is manually recorded by professionals, which can lead to errors in analysis. To assist with these annotations, an automation process is proposed using pose estimation and object detection techniques in computer vision to detect interaction between participants in therapy sessions with children with ASD. For this purpose, the Yolov8 object detector and Yolov8-Pose estimator are applied. Subsequently, interaction detection is performed using the results obtained from the previous predictions. Heuristics were introduced to compare these techniques. To enhance performance, solutions were explored to address the limitations encountered in object detection and pose estimation predictions. The results demonstrated a 15.8% improvement in precision for the pose estimation heuristic compared to the bounding box heuristic, achieving satisfactory performance for the proposed approach.

A couple bugfixes finish off the v0.5 series.

The paper studies the performance of public higher education institutions in Slovakia. Data envelopment analysis is applied to assess the efficiency of selected universities in their two core mission areas: providing education and doing research and development. We focus on how the size and the structure of the academic staff and the capital resources influence the educational and research outputs. Data from public sources for years 2014 – 2015 is used in the analysis. Efficiency scores and projections calculated by designed DEA models serve to benchmark assessed universities and provide recommendations for improvement of their performance

The consideration of environmental costs is essentially changing the transportation policy in developed countries. The reduction in total distance travelled will in itself provide environmental benefits due to the reduction in fuel consumed and in the resulting pollutants. The cash deployment strategy for a network of ATMs should take into account the analysis of inventory policies, logistics costs as well as of the routing of replenishment vehicles. The optimal strategy has to focus on the reduction of cash-related expenses while safeguarding that ATMs do not run out of cash. The problem, which can be tackled as a kind of rich vehicle routing problem is in the paper solved using parallel genetic algorithm. The proposed model is able to solve cases with simultaneous requirements of several replenishments for some of the customers (ATMs) daily and a single replenishment in several days for other groups of customers.

In this paper we evaluate eight well-known Information Extraction (IE) tools on a task of Named Entity Recognition (NER) in microposts. We have chosen six NLP tools and two Wikipedia concept extractors for the evaluation. Our intent was to see how these tools would perform on relatively short texts of microposts. Evaluation dataset has been adopted from the MSM 2013 IE Challenge. This dataset contained manually annotated microposts with classification restricted to four entity types: PER, LOC, ORG and MISC.

The Dolby Digital (AC-3) and the Dolby Digital Plus or Enhanced AC-3 (E-AC-3) systems are currently the key enabling technologies for high-quality compression of digital audio signals. For the time-to-frequency transformation of an audio data block and vice versa, both systems have adopted a long transform being the modified discrete cosine transform (MDCT). The AC-3 additionally defines two variants of cosine-modulated filter banks called the first and second short transforms. A fast conversion algorithm is presented to convert the frequency coefficients of the long (MDCT) transform to those of two short transforms and vice versa, directly in the frequency domain. It is based on a block sparse matrix factorization of a conversion matrix. The fast conversion algorithm is efficient in terms of the structural simplicity, arithmetic complexity and memory requirements compared to the obvious conversion methods. Moreover, the existing AC-3 fast computational modules may be simply re-used in the conversion procedures. Consequently, the E-AC-3 to AC-3 bit stream conversion and the AC-3 to E-AC-3 bit stream transcoding can be realized in a simplified and efficient way, thus minimizing the amount of partial decoding/encoding and memory requirements during the conversion and transcoding processes.

Abstract Problem solving methods, inspired biologically or by nature, are gaining in importance. Their potential for devising new, but also for increasing efficiency of some well-known methods for solving different kinds of problems, is becoming more and more obvious. The behaviour of social insects provides a lot of inspiration. We took inspiration from honey bees, in particular the way they behave in a hive. We proposed a modified model devised for searching parts of the web online. In a series of experiments, we attempted to follow a developing story with the aim to automatically find documents (web pages) bringing news on the story. This service, working as a beehive, delivered documents as expected.Keywordsbee hive modelbee swarm intelligenceweb searchon-line web crawlingdeveloping story

The economy of Slovakia experienced a turning point in the 1st half of 2008 and entered a phase of decline. The negative impacts of the global economic crisis became evident in the 2nd half of 2008 and led into a recession in the 1st quarter of 2009. The composite leading indicator was originally intended for forecasting of business cycle turning points between the decline and growth phases. The aim of this paper is to transform the qualitative information from composite leading indicator into quantitative forecast and verify whether the beginning of recession in Slovakia could have been identified in advance. The ARIMAX and error correction models are used for the composite reference series and GDP forecasts respectively. The final result shows that the composite leading indicator is useful not only for identifying turning points, but also for the prediction of recession phase.

The use of a new progressive method suitable for the exposure optimization is investigated for large and complex defect-free chips direct written by ZBA 21 electron beam pattern generator together with all corresponding microprocesses. Well controlled and resolved details (spaces) between individual quasi-square structures of the final large area neural holography chip were achieved in the range of about 50 nm.

A lot of attention is now being focused on connectionist models known under the name “reservoir computing”. The most prominent example of these approaches is a recurrent neural network architecture called an echo state network (ESN). ESNs were successfully applied in more real-valued time series modeling tasks and performed exceptionally well. Also using ESNs for processing symbolic sequences seems to be attractive. In this work we experimentally support the claim that the state space of ESN is organized according to the Markovian architectural bias principles when processing symbolic sequences. We compare performance of ESNs with connectionist models explicitly using Markovian architectural bias property, with variable length Markov models and with recurrent neural networks trained by advanced training algorithms. Moreover we show that the number of reservoir units plays a similar role as the number of contexts in variable length Markov models.KeywordsRecurrent Neural NetworkHide UnitSymbolic SequenceRecurrent NetworkEcho State NetworkThese keywords were added by machine and not by the authors. This process is experimental and the keywords may be updated as the learning algorithm improves.