Accurate identification of COVID-19 is now a critical task since it has seriously damaged daily life, public health, and the economy. It is essential to identify the infected people to prevent the further spread of the pandemic and to treat infected patients quickly. Machine learning techniques have a significant role in predicting of COVID-19. In this study, we performed binary classification (COVID-19 vs. other types of coronavirus) by extracting features from genome sequences. Support vector machines, naive Bayes, K-nearest neighbor, and random forest methods were used for classification. We used viral gene sequences from the 2019 Novel Coronavirus Resource Database. Experimental results presented show that a decision tree method achieved 93% accuracy.

The prognosis for pancreatic cancer (PC) is poor, with a 5-year survival rate of approximately 10%. Methods such as machine learning (ML) can facilitate prognostic assessments by examining complex patterns in patient data that may not be discernible with traditional methods. The aim of this study was to analyze prognostic factors that may influence overall survival in advanced-stage PC using ML methods, and to evaluate the performance of various ML algorithms in predicting patient survival outcomes. A total of 315 patients with inoperable locally advanced or metastatic PC between 2005 and 2023 were included in the study. MATLAB software was used for feature selection. Overall survival was defined as the time from diagnosis to death or last follow-up, and was used as the primary parameter for analysis. The power of 19 clinical and laboratory features of the patients to predict whether patients were deceased, as reflected by importance scores (F-scores), was evaluated using the minimum redundancy–maximum relevance, chi-square, analysis of variance, and Kruskal–Wallis tests as feature selection methods. 24 ML methods were evaluated with these feature selection methods and the results regarding the most effective features were used to predict whether patients were deceased or not. The median age of the patients was 62 years, and 30.5% were women while 69.5% were men. As a result of the analysis of the feature selection methods, the first-line chemotherapy a patient received had the highest F-score in predicting that patient’s survival. Among ML methods, the support vector machine (SVM) kernel method had the highest accuracy rate (87%) in predicting whether patients were deceased. When the feature selection methods were combined with the SVM kernel ML method, patients’ survival statuses could be predicted with an accuracy rate of 87.9%. The SVM kernel method has been demonstrated to show potential as a means of predicting survival for patients with advanced PC. The integration of feature selection with this method yielded high accuracy, thereby underscoring its significance. The findings emphasize the pivotal function of first-line chemotherapy and indicate that ML models have the potential to enhance clinical decision-making and patient care.

PID controllers are important control methods that are widely used in industrial processes. Proper tuning of PID gains is critical for achieving the state-of-the-art system performance. Therefore, optimizing PID gains is an important research topic in the field of control engineering. In this study, PID controller gains are automatically tuned using metaheuristic optimization methods. These methods use an iterative approach to calculate optimal values of PID controller gains based on different optimization techniques. The interaction between artificial intelligence and control systems requires a multidimensional approach across different disciplines. In the study, we perform Particle Swarm Optimization, Gray Wolf Optimization, Whale Optimization Algorithm, Firefly Algorithm, Harris Hawks Optimization, Artificial Hummingbird Algorithm and African Vulture Optimization Algorithm to determine PID gains. In the simulation, step input is applied to the dynamic equation of the unmanned free-swimming submersible vehicle. The fitness function is determined with respect to controller integral square error, settling time value, and maximum percent overshoot value. We also evaluate the optimization time of the selected algorithms based on the fitness function. Experimental results present that Artificial Hummingbird Algorithm, Gray Wolf Optimization and Particle Swarm Optimization achieve significant performance. This underlines that using metaheuristic methods in PID gain optimization increase overall system performance.