NCC Montenegro representatives attended CASTIEL supported, presentation of White Paper “Development of a Maturity Assessment Tool to Improve SME HPC Capabilities” for EuroCC2 industry working group. Dr Özlem Sarı, NCC Turkey, presented industrial users acquisition process and development of HPC Maturity Assessment Tool to improve SME HPC capabilities.
During the presentation, Proof of Concept study, Course of Action, GAP analyses, as well as methodological approach and validation process were duly elaborated. In addition, there was a vibrant discussion on challenges faced and experiences were shared regarding SME collaboration and industry interaction, especially regarding assessment results/HPC maturity levels, identifying HPC-suitable problems and business benefits of HPC integration.
Researchers from UDG and NCC Montenegro published a paper “Machine Learning Models for Statistical Analysis” by M. Grebovic et al. in The International Arab Journal of Information Technology, Vol. 20, No. 3A, Special Issue 2023. This was a follow up effort on the paper previously presented at the ACIT2023 conference.
Click on image to open
ABSTRACT – Compared to traditional statistical models, Machine Learning (ML) algorithms provide the ability to interpret, understand and summarize patterns and regularities in observed data for making predictions in an advanced and more sophisticated way. The main reasons for the advantage of ML methods in making predictions are a small number of significant predictors of the statistical models, which means limited informative capability, and pseudo-correct regular statistical patterns, used without previous understanding of the used data causality. Also, some ML methods, like Artificial Neural Networks, use non-linear algorithms, considering links and associations between parameters. On the other hand, statistical models use one-step-ahead linear processes to improve only short-term prediction accuracy by minimizing a cost function. Although designing an optimal ML model can be a very complex process, it can be used as a potential solution for making improved prediction models compared to statistical ones. However, ML models will not automatically improve prediction accuracy, so it is necessary to evaluate and analyze several statistical and ML methods, including some artificial neural networks, through accuracy measures for prediction purposes in various fields of applications. A couple of techniques for improving suggested ML methods and artificial neural networks are proposed to get better accuracy results.
Researchers from UDG and NCC Montenegro published a paper “Exploratory analysis of text using available NLP technologies for Serbian language”. The authors are L. Lakovic, S. Cakic, I. Jovovic, and D. Babic. The paper was presented at 22nd International Symposium INFOTECH-JAHORINA, that took place on 15-17 March 2023. The paper is published in IEEE Xplore.
ABSTRACT – This paper combines available NLP technologies for Serbian languages and traditional data science methods in order to analyze collected dataset on the news headlines related to the COVID-19 pandemics. As an addition to NLP technologies for the Serbian language, a specialized database was created in an attempt to enhance the research within the field. Within the paper, the database was exploratory analyzed, and perspectives of the work with the data were thoroughly explored.
Researchers from the EuroCC NCC Montenegro team published a scientific paper in IOS press journal Technology and Health Care – Volume Pre-press. This article belongs to the “Special Issue for magazine Technology and health care: official journal of the European Society for Engineering and Medicine ” (link). The paper “Competence development as critical issue for successful performance in HPC technology environment: A case study of Montenegro” by B. Malisic, S. Tinaj provides an overview of the concepts and systems needed to develop the competencies needed to implement modern technology such as High-Performance Computing (HPC) in Montenegro. In work and in practice through the EuroCC project within the National Competence Center Montenegro, we start from a holistic approach to competences, which includes or integrates knowledge, skills, attitudes and values that influence the ability and competence for certain actions, in this case for the adoption and application of technology. Learn more at the following link.
Click on image to open
ABSTRACT: This paper provides an overview of the concepts and systems needed to develop the competencies needed to implement modern technology such as High-Performance Computing (HPC) in Montenegro. In this research paper competencies are viewed holistically. This paper will elaborate the defined competencies related to the HPC technology environment, identified during the implementation of the EuroCC project in Montenegro, but also based on market analyses, combined with the identified indicators of absorption capacity, generally at the national level. By identifying the innovative and business potential of representatives of the public, academic and economic sectors, with special reference to small and medium-sized companies and the IT cluster that make up the dominant segments in the structure of the market sample, as generators and accelerators of innovation, smart growth and the digital economy, we got a clear picture regarding the development of necessary competencies within the NCC team but also at the national level. There must be a systemic approach and sustainable, dynamic projects and tools for the development of human resources in the development of competences.
Researchers from the AIMHiGH team published a scientific paper in MDPI journal Sensors, This article belongs to the “Special Issue Novel Architectures and Applications for Artificial Intelligent and Internet of Things Ecosystems” (link). The paper “Developing Edge AI Computer Vision for Smart Poultry Farms Using Deep Learning and HPC” by S. Cakic, T. Popovic, S. Krco, D. Nedic, D. Babic, and I. Jovovic reports on the approach, experiences, and results of using HPC and AI to develop advanced Edge AI computer vision solutions for smart agriculture systems. AIMHiGH project is implemented as an experiment done in the context of FF4EuroHPC project. FF4EuroHPC is a European initiative that helps facilitate access to all high-performance computing-related technologies for SMEs and thus increases the innovation potential of European industry. Whether it is running high-resolution simulations, doing large-scale data analyses, or incorporating AI applications into SMEs’ workflows, FF4EuroHPC connects business with cutting-edge technologies. Learn more at: link.
ABSTRACT – This research describes the use of high-performance computing (HPC) and deep learning to create prediction models that could be deployed on edge AI devices equipped with camera and installed in poultry farms. The main idea is to leverage an existing IoT farming platform and use HPC offline to run deep learning to train the models for object detection and object segmentation, where the objects are chickens in images taken on farm. The models can be ported from HPC to edge AI devices to create a new type of computer vision kit to enhance the existing digital poultry farm platform. Such new sensors enable implementing functions such as counting chickens, detection of dead chickens, and even assessing their weight or detecting uneven growth. These functions combined with the monitoring of environmental parameters, could enable early disease detection and improve the decision-making process. The experiment focused on Faster R-CNN architectures and AutoML was used to identify the most suitable architecture for chicken detection and segmentation for the given dataset. For the selected architectures, further hyperparameter optimization was carried out and we achieved the accuracy of AP = 85%, AP50 = 98%, and AP75 = 96% for object detection and AP = 90%, AP50 = 98%, and AP75 = 96% for instance segmentation. These models were installed on edge AI devices and evaluated in the online mode on actual poultry farms. Initial results are promising, but further development of the dataset and improvements in prediction models is needed.
Researchers from NCC Montenegro presented a paper at the 27th IEEE Conference on Information Technology IT2023 on 17th February 2023. The paper is titled “Disease Prediction Using Machine Learning Algorithms” and authored by I. Jovovic, D. Babic, T. Popovic, S. Cakic and I. Katnic.
ABSTRACT – This study aimed to investigate the application of machine learning techniques for disease prediction. Three popular machine learning algorithms, Random Forest, Support Vector Machines and Naive Bayes, were employed and their performance was evaluated. Results showed that the best performing model was based on Random Forest algorithm with the average accuracy of 87%. This model has been additionally tuned in order to achieve even better performance, which resulted with 90% accuracy. This study highlights the potential of AI in disease prediction and provides insights into the importance of algorithm selection and tuning for optimal performance.
Mr. Ivan Jovovic presenting at IEEE 2023 conference
The paper will soon be available through IEEE Xplore
Researchers from NCC Montenegro presented a paper at the 27th IEEE Conference on Information Technology IT2023. The paper is titled “Vision-based Vehicle Speed Estimation Using the YOLO Detector and RNN” and authored by Andrija Peruničić, Slobodan Djukanović and Andrej Cvijetić
ABSTRACT : The paper deals with vehicle speed estimation using video data obtained from a single camera. We propose a speed estimation method which uses the YOLO algorithm for vehicle detection and tracking, and a recurrent neural network (RNN) for speed estimation. As input features for speed estimation, we use the position and size of bounding boxes around the vehicles, extracted by the YOLO detector. The proposed method is trained and tested on the recently proposed VS13 dataset. The experimental results show that the box position does not bring any improvement in the speed estimation performance. The proposed RNN-based estimator gives an average error of 4.08 km/h using only the area of bounding box as input feature, which significantly outperforms audio-based approaches on the same dataset.
