Persistent growth of information in recent decades, along with the development of new information technologies for its management, have made it essential to develop systems that allow to synthesize this massive information or better known as big data. In this article, a feedback based system for massive processing of digital newspapers is presented. This system synthesizes the most relevant information from different news stories obtained from several sources. System is fed with information from the Internet using web scraping techniques. All this information is stored in a data lake which has been implemented using NoSQL databases. Next, data processing is performed, focusing on words, their relevance, and their correlation with other words from related content groups or headlines. In order to perform this aggrupation, machine learning Large Language Model (LLM), K Nearest Neighbors (KNN) and text mining techniques are used. New text mining algorithms are also developed to adjust thresholds during content aggregation and synthesis. Finally, the results visualization mechanism is presented which allow users to give a punctuation to the news stories. This mechanism represents a feedback punctuation for the system which will be considered into the global punctuation, which is the basis to show the results. This system can be useful to summarize all the information contained in the news stories which are stored in Internet, providing users a fast way to be informed.

Download article

This paper explores the application of various machine learning techniques to model the optimal measurement time required after machining with a probe on CNC machine tools. Specifically, the research employs four different machine learning models: Elastic Net, Neural Networks, Decision Trees, and Support Vector Machines, each chosen for their unique strengths in addressing different aspects of predictive modeling in an industrial context. The study examines the input parameters such as material type, post-processing wall thickness, cutting depth, and rotational speed over measurement time. This approach ensures that the models account for the variables that significantly affect CNC machine operations. Regression value, mean square error, root mean square error, mean absolute percentage error, and mean absolute error were used to evaluate the quality of the obtained models. As a result of the analyses, the best modeling results were obtained using neural networks. Their ability to accurately predict measurement times can significantly increase operational efficiency by optimizing schedules and reducing downtime in machining processes.

Download article

The International Classification of Diseases (ICD) is utilized by member countries of the World Health Organization (WHO). It is a critical system to ensure worldwide standardization of diagnosis codes, which enables data comparison and analysis across various nations. The ICD system is essential in supporting payment systems, healthcare research, service planning, and quality and safety management. However, the sophisticated and intricate structure of the ICD system can sometimes cause issues such as longer examination times, increased training expenses, a greater need for human resources, problems with payment systems due to inaccurate coding, and unreliable data in health research. Additionally, machine learning models that use automated ICD systems face difficulties with lengthy medical notes. To tackle this challenge, the present study aims to utilize Medical Information Mart for Intensive Care (MIMIC-III) medical notes that have been summarized using the term frequency-inverse document frequency (TF-IDF) method. These notes are further analyzed using deep learning, specifically bidirectional encoder representations from transformers (BERT), to classify disease diagnoses based on ICD codes. Even though the proposed methodology using summarized data provides lower accuracy performance than state-of-the-art methods, the performance results obtained are promising in terms of continuing the study of extracting summary input and more important features, as it provides real-time ICD code classification and more explainable inputs.

Download article

The sixth-generation (6G) communication technology has potential in various applications, for instance, industrial automation, intelligent transportation, healthcare systems, and energy consumption prediction. On the other hand, the concerns of privacy measures and security measures in 6G-enabled networks are considered critical issues and challenges. The integration of 6G with advanced technologies for example computing, Artificial Intelligence (AI), and Internet of Things (IoT) is a common theme in this paper. Additionally, the paper discusses the challenges and advancements required for 6G technology to be utilized with other technologies, involving edge technology, big data analytics, and deep learning. In this review paper, the authors overview the integration of 6G with cutting-edge technologies like IoT, IoMT, AI, and edge computing that address human requirements and issues. In addition, to make values for new technologies like Big data, federated learning machine learning, deep learning, and multiple aspects are merged collectively to offer a network for the machine and human growing era. These integrations can be utilized for monitoring energy consumption in real-time, intelligent transportation solutions, improved security in industrial applications, signal reconstruction, and industrial automation. Additionally, the authors illustrate the critical considerations and challenges that face the integration of 6G for instance, performance requirements, security, and privacy concerns. Overall, this paper suggests that 6G communication technology can revolutionize different sides of our society, and enhance efficiency and accuracy in various future industrial automation and sectors.

Download article

Osteoarthritis is one of the leading causes of disability around the globe. Up to this date there is no definite cure for cartilage lesions. Only fast and accurate diagnosis enables prolonging joint survivor time. Available diagnostic methods have disadvantages such as high price, radiation, need for experienced radiologists or low availability in some regions. The present study evaluates the use of vibroarthorgraphy as a method of cartilage lesion detection. 47 patients with diagnosed cartilage lesions, and 51 healthy control group patients have been enrolled in this study. The cartilage in the study group was evaluated intraoperatively by experienced orthopaedic surgeon. Signal acquisition was performed in open and closed kinematic chain based on 10 knee joint movements from 0-90 degrees. By using EEMD-DFA algorithms, reducing classifier inputs using ANOVA and then classifying using artificial neural networks (ANN), a classification accuracy of almost 93% was achieved. A sensitivity of 0.93 and a specificity of 0.93 with an AUC of 0.942 were obtained for the multilayer perceptron network. These results allow to apply this testing protocol in a clinical setting in the future. 

Download article

In this study, the authors present and scrutinize two deep learning models designed for predicting the states of epilepsy patients by utilizing extracted data from their brain's electrical activities recorded in electroencephalography (EEG) signals. The proposed models leverage deep learning networks, with the first being a recurrent neural network known as Long Short-Term Memory (LSTM), and the second a non-recurrent network in the form of a Deep Feedforward Network (DFN) architecture. To construct and execute the DFN and LSTM architectures, the authors rely on 22 characteristics extracted from diverse EEG signals, forming a comprehensive dataset from five patients. The primary goal is to forecast impending epilepsy seizures and categorize three distinct states of brain activity in epilepsy patients. The models put forward yield promising results, particularly in terms of classification rates, across various preceding seizure timeframes ranging from 5 to 50 minutes.

Download article

Modern technology has become a vital part of our daily lives, and the world has undergone remarkable advancements in various scientific and technological fields. The advancement of technology presents a variety of opportunities for students to promote academic development and make it easier to access education through online learning systems. The most difficult and most demanding task during learning is to be aware of and support the emotional side of students. Recognizing one's emotions is easy for humans, but it is a challenging task for computers due to the specific features of the human face. However, recent advances in computing and image processing have made it possible and easy to detect and categorize emotions in images and videos. This paper focuses on detecting learners' emotions in real time during synchronous learning. In this regard, a video/chat application has been developed for the tutor to detect the emotions of the learners while presenting his lesson. The emotions detected are separated into three states (Satisfied, Neutral and Unsatisfied); each state is made up of two or three distinct emotions. The objective is to assist teachers in adapting teaching methods in virtual learning settings according to the emotions of learners.

Download article

The function of Artificial Intelligence (AI) in Human-Robot Cooperation (HRC) in Industry 4.0 is unequivocally important and cannot be undervalued. It uses Machine Learning (ML) and Deep Learning (DL) to enhance collaboration between humans and robots in smart manufacturing. These algorithms effectively manage and analyze data from sensors, machinery, and other associated entities. As an outcome, they can extract significant insights that can be beneficial in optimizing the manufacturing process overall. Because dumb manufacturing systems hinder coordination, collaboration, and communication among various manufacturing process components. Consequently, efficiency, quality, and productivity all suffer as a whole. Additionally, Artificial Intelligence (AI) makes it possible to implement sophisticated learning processes that enhance human-robot collaboration and effectiveness when it comes to assembly tasks in the manufacturing domain by enabling learning at a level that is comparable to human-human interactions. When Artificial Intelligence (AI) is widely applied in Human-Robot Cooperation (HRC), a new and dynamic environment for human-robot collaboration is created and responsibilities are divided and distributed throughout social and physical spaces. In conclusion, Artificial Intelligence (AI) plays a crucial and indispensable role in facilitating effective and efficient Human-Robot Cooperation (HRC) within the framework of Industry 4.0. The implementation of Artificial Intelligence (AI)-based algorithms, encompassing deep learning, machine learning, and reinforcement learning, is highly consequential as it enhances human-robot collaboration, streamlines production procedures, and boosts overall productivity, quality, and efficiency in the manufacturing industry.

Download article

The article presents the so-called coin bag problem, which is modeled by linear Diophantine equations. The problem in question involves assessing the contents of a set of coins based on its weight. Since this type of problem is undecidable, a special variant of the problem was proposed for which effective problem-solving algorithms can be developed. In this paper, an original heuristic is presented (an algorithm based on problem decomposition) which allows to solve the coin bag problem in fewer steps compared to a brute force algorithm. The proposed approach was verified in a series of computational experiments. Additionally, an authentication scheme making use of the approach was proposed as an example of potential practical use.

Download article

Mental illness is a physical condition that significantly changes a person’s thoughts, emotions, and capacity to interact with others. The purpose of this study was to explore the application of Artificial Intelligence (AI) and Machine Learning (ML) algorithms in predicting behaviour regarding seeking treatment for mental illnesses, to support healthcare providers in reaching out to and supporting individuals more likely to seek treatment, leading to early detection, enhanced outcomes. The Open Sourcing Mental Illness (OSMI) dataset contains 1259 samples used for research and experiment. The study uses several classifiers (Random Forest, Gradient Boosting, SVM, KNN, and Logistic Regression) to take advantage on their unique capabilities and applicability for various parts of the prediction task. Experiments performed in Jupiter notebook and the major findings revealed varying levels of accuracy among the classifiers, with the Random Forest and 0.81 and Gradient Boosting classifiers 0.83 achieving highest accuracy, while the accuracy for SVM 0.82 and KNN 0.83 also give good result but Logistic Regression classifier had a lower accuracy 0.8. In conclusion, this research demonstrates the potential of AI and machine learning in predicting individual behaviour and offers valuable insights into mental health treatment-seeking behaviour.

Download article

This article presents a study aimed at using machine learning to automate the analysis of ultrasound images in the diagnosis of polycystic ovary syndrome (PCOS). Today, various laboratory and instrumental methods are used to diagnose PCOS, including the analysis of ultrasound images performed by medical professionals. The peculiarity of such analysis is that it requires high qualification of medical professionals and can be subjective. The aim of this work is to develop a software module based on convolutional neural networks (CNN), which will improve the accuracy and objectivity of diagnosing polycystic disease as one of the clinical manifestations of PCOS. By using CNNs, which have proven to be effective in image processing and classification, it becomes possible to automate the analysis process and reduce the influence of the human factor on the diagnosis result. The article describes a machine learning model based on CNN architecture, which was proposed by the authors for analyzing ultrasound images in order to determine polycystic disease. In addition, the article emphasizes the importance of the interpretability of the CNN model. For this purpose, the Gradient-weighted Class Activation Mapping (Grad-CAM) visualization method was used, which allows to identify the image areas that most affect the model's decision and provides clear explanations for each individual prediction.

Download article