Air pollution particulate matter (PM2.5) prediction in South African cities using machine learning techniques.

Background: Air pollution contributes to the most severe environmental and health problems due to industrial emissions and atmosphere contamination, produced by climate and traffic factors, fossil fuel combustion, and industrial characteristics. Because this is a global issue, several nations have established control of air pollution stations in various cities to monitor pollutants like Nitrogen Dioxide (NO2), Ozone (O3), Sulfur Dioxide (SO2), Carbon Monoxide (CO), Particulate Matter (PM2.5, PM10), to notify inhabitants when pollution levels surpass the quality threshold. With the rise in air pollution, it is necessary to construct models to capture data on air pollutant concentrations. Compared to other parts of the world, Africa has a scarcity of reliable air quality sensors for monitoring and predicting Particulate Matter (PM2.5). This demonstrates the possibility of extending research in air pollution control. Methods: Machine learning techniques were utilized in this study to identify air pollution in terms of time, cost, and efficiency so that different scenarios and systems may select the optimal way for their needs. To assess and forecast the behavior of Particulate Matter (PM2.5), this study presented a Machine Learning approach that includes Cat Boost Regressor, Extreme Gradient Boosting Regressor, Random Forest Classifier, Logistic Regression, Support Vector Machine, K-Nearest Neighbor, and Decision Tree. Results: Cat Boost Regressor and Extreme Gradient Boosting Regressor were implemented to predict the latest PM2.5 concentrations for South African Cities with recording stations using past dated recordings, then the best performing model between the two is used to predict PM2.5 concentrations for South African Cities with no recording stations and also to predict future PM2.5 concentrations for South African Cities. K-Nearest Neighbor, Logistic Regression, Support Vector Machine, Decision Tree, and Random Forest Classifier were implemented to create a system predicting the Air Quality Index (AQI) Status. Conclusion: This study investigated various machine learning techniques for air pollution to analyze and predict air pollution behavior regarding air quality and air pollutants, detecting which areas are most affected in South African cities.

Supervised machine learning models for depression sentiment analysis.

Introduction: Globally, the prevalence of mental health problems, especially depression, is at an all-time high. The objective of this study is to utilize machine learning models and sentiment analysis techniques to predict the level of depression earlier in social media users' posts. Methods: The datasets used in this research were obtained from Twitter posts. Four machine learning models, namely extreme gradient boost (XGB) Classifier, Random Forest, Logistic Regression, and support vector machine (SVM), were employed for the prediction task. Results: The SVM and Logistic Regression models yielded the most accurate results when applied to the provided datasets. However, the Logistic Regression model exhibited a slightly higher level of accuracy compared to SVM. Importantly, the logistic regression model demonstrated the advantage of requiring less execution time. Discussion: The findings of this study highlight the potential of utilizing machine learning models and sentiment analysis techniques for early detection of depression in social media users. The effectiveness of SVM and Logistic Regression models, with Logistic Regression being more efficient in terms of execution time, suggests their suitability for practical implementation in real-world scenarios.

Bootstrapping random forest and CHAID for prediction of white spot disease among shrimp farmers.

Technology is playing an important role is healthcare particularly as it relates to disease prevention and detection. This is evident in the COVID-19 era as different technologies were deployed to test, detect and track patients and ensure COVID-19 protocol compliance. The White Spot Disease (WSD) is a very contagious disease caused by virus. It is widespread among shrimp farmers due to its mode of transmission and source. Considering the growing concern about the severity of the disease, this study provides a predictive model for diagnosis and detection of WSD among shrimp farmers using visualization and machine learning algorithms. The study made use of dataset from Mendeley repository. Machine learning algorithms; Random Forest classification and CHAID were applied for the study, while Python was used for implementation of algorithms and for visualization of results. The results achieved showed high prediction accuracy (98.28%) which is an indication of the suitability of the model for accurate prediction of the disease. The study would add to growing knowledge about use of technology to manage White Spot Disease among shrimp farmers and ensure real-time prediction during and post COVID-19.