A Low-Cost System for Moisture Content Detection of Bagasse upon a Conveyor Belt with Multispectral Image and Various Machine Learning Methods

Development of an Active Training Method for Belt Conveyor

The global situation related to the COVID-19 pandemic has forced employers to find an adequate way to conduct training in order to ensure work safety. The underground mining industry is one of the industries which, due to its nature, was not able to switch to remote work. Conducting traditional training risked spreading the virus among workers. For this purpose, it was necessary to start a search for a form of training that would be safe and would not cause additional stress for employees. Research on the development of an active employee training method and testing of the method itself was conducted online. In order to develop a method of active training, one of the most important workstations was selected, which is the operation of the conveyor belt. The training method comprises four training modules. The modules cover questions related to the operation of the conveyor belt, emergencies, its assembly and disassembly, repair and maintenance. The developed issues also take into account questions concerning natural hazards and work safety. The entire training course lasts 10 days. Every day, an employee receives a set of eight questions sent to their email address, which they must answer before starting work. The article describes the methodology and implementation of the training.

Improving Sports Outcome Prediction Process Using Integrating Adaptive Weighted Features and Machine Learning Techniques

Developing an effective sports performance analysis process is an attractive issue in sports team management. This study proposed an improved sports outcome prediction process by integrating adaptive weighted features and machine learning algorithms for basketball game score prediction. The feature engineering method is used to construct designed features based on game-lag information and adaptive weighting of variables in the proposed prediction process. These designed features are then applied to the five machine learning methods, including classification and regression trees (CART), random forest (RF), stochastic gradient boosting (SGB), eXtreme gradient boosting (XGBoost), and extreme learning machine (ELM) for constructing effective prediction models. The empirical results from National Basketball Association (NBA) data revealed that the proposed sports outcome prediction process could generate a promising prediction result compared to the competing models without adaptive weighting features. Our results also showed that the machine learning models with four game-lags information and adaptive weighting of power could generate better prediction performance.