Using Machine Learning to Find Dependencies in Data to Improve Working Conditions

This paper discusses the benefits and potential of machine learning in data analysis and its ability to discover hidden dependencies. The paper also discusses a case study on the application of machine learning in medical research to identify factors affecting heart failure mortality. An experiment was conducted investigating predictions using a decision tree, which showed higher accuracy after feature selection. After that, an experiment was conducted investigating the predictions using artificial neural network—multilayer perceptron, which also showed higher accuracy after feature selection. As a result, the most significant influencing factors are identified. Two classifiers were also proposed that can be used to predict mortality, which will improve the working conditions in manufacturing. The results of the study emphasize the importance of using machine learning in modern data analytics and their potential to improve predictive and analytical ability.