Online harassment detection faces significant challenges due to its expansive reach and anonymity. Addressing this issue demands effective detection mechanisms capable of processing vast data streams and adapting to evolving online languages. Leveraging advancements in artificial intelligence, we propose a novel approach grounded in natural language processing (NLP) and metaheuristic algorithms. Our methodology integrates term frequency-inverse document frequency (TF-IDF) encoding and the AdaBoost algorithm for classification. To tackle the NP-hard problem of hyperparameter selection, we introduce a modified crayfish optimization algorithm (COA), termed GI-COA. This paper represents a pioneering effort in utilizing metaheuristic algorithms for hyperparameter selection in harassment detection models. Through experimentation, we demonstrate the efficacy of our approach in fostering a safer online environment. The best-performing optimized models demonstrate an accuracy exceeding 77%.

The power grid stands as a cornerstone of modern living, facilitating infrastructure, industry, and daily conveniences through a reliable supply of electricity. However, the transmission of power is beset by challenges such as maintenance costs, malfunctions, and losses due to resistance. Of particular concern in high voltage transmission lines is the occurrence of electrical arcing, stemming from environmental factors, debris, or grid damage. This study delves into the comparative analysis of various architectures of the YOLOv8 model for arc detection. The results showcase the high accuracy of the YOLOv8 m model, demonstrating its robust performance in accurately detecting and localizing objects in complex and dynamic environments. Future research prospects include the utilization of larger datasets to enhance model performance for broader applicability across diverse scenarios in power grid operations.

Main goal of the research presented in this manuscript is to analyze employees satisfaction by employing hybrid methods between machine learning and metaheuristics. Furthermore, the proposed work utilizes term frequency-inverse document frequency (TF-IDF), natural language processing (NLP) method, for the identification of important segments of the employees' statements towards their workplaces, while the extreme gradient boosting (XGBoost) model is used for predicting the outcomes. Since each machine learning technique should be calibrated to particular problem, the XGBoost has been optimized by a modified swarm metaheuristics based on red fox optimizer (RFO) algorithm and its performance is evaluated against other high-performing swarm intelligence approaches. The results indicate the superiority of proposed solution, which has been further evaluated with the Shapley additive explanations (SHAP) analysis.

Online harassment detection faces significant challenges due to its expansive reach and anonymity. Addressing this issue demands effective detection mechanisms capable of processing vast data streams and adapting to evolving online language. Leveraging advancements in artificial intelligence, we propose a novel approach grounded in natural language processing (NLP) and metaheuristic algorithms. Our methodology integrates term frequency-inverse document frequency (TF-IDF) encoding and the AdaBoost algorithm for classification. To tackle the NP-hard problem of hyperparameter selection, we introduce a modified crayfish optimization algorithm (COA), termed GI-COA. This paper represents a pioneering effort in utilizing metaheuristic algorithms for hyperparameter selection in harassment detection models. Through experimentation, we demonstrate the efficacy of our approach in fostering a safer online environment. The best performing optimize models demonstrate and accuracy exceeding 77%.