Robustness of Support Vector Regression and Random Forest Models: A Simulation Study

Classical statistics are usually based on parametric models, where the performance depends heavily on assumptions and is not robust in the presence of outliers in the data. Due to the COVID-19 pandemic, our daily lives have changed significantly, including slowing economic growth. These extreme changes can manifest as an outlier in time series studies and adversely affect the results of data analysis. Many classical methods of official statistics are prone to outliers. In this work, we evaluate machine learning methods: Support Vector Regression (SVR) and Random Forest (RF) and compare it with ARIMA to determine the robustness through simulation studies. Robustness is measured by the sensitivity of the SVR and Random Forest hyperparameter and the model's error in the presence of outliers. Simulations show that more outliers lead to higher RMSE values, and conversely, more samples lead to lower RMSE values. The type of outliers significantly impacts the RMSE value of the ARIMA model, where additional outliers (AO) have a worse impact than temporary change (TC). Consecutive outliers produce a smaller RMSE mean than non-consecutive outliers. Based on the sensitivity of hyperparameters, SVR and Random Forest models are relatively robust to the presence of outliers in the data. Based on the simulation results of 100 iterations, we find that SVR is more robust than ARIMA and Random Forest in modeling time series data with outliers. © 2023, The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd.

From Heroes to Scoundrels: Exploring the effects of online campaigns celebrating frontline workers on COVID-19 outcomes.

This paper examines the effects of online campaigns celebrating frontline workers on COVID-19 outcomes regarding new cases, deaths, and vaccinations, using the United Kingdom as a case study. We implement text and sentiment analysis on Twitter data and feed the result into random regression forests and cointegration analysis. Our combined machine learning and econometric approach shows very weak effects of both the volume and the sentiment of Twitter discussions on new cases, deaths, and vaccinations. On the other hand, established relationships (such as between stringency measures and cases/deaths and between vaccinations and deaths) are confirmed. On the contrary, we find adverse lagged effects from negative sentiment to vaccinations and from new cases to negative sentiment posts. As we assess the knowledge acquired from the COVID-19 crisis, our findings can be used by policy makers, particularly in public health, and prepare for the next pandemic.

Analysis of Prediction Data for the Third Wave of COVID-19 in Bogor Regency

The COVID-19 pandemic is far from over. The government has carried out several policies to suppress the development of COVID-19 is no exception in Bogor Regency. However, the public still has to be vigilant especially now we will face a year-end holiday that can certainly be a trigger for the third wave of COVID-19. Therefore, researchers aim to make predictions of the increase in positive cases, especially in the Bogor Regency area to help the government in making policies related to COVID-19. The algorithms used are Gaussian Process, Linear Regression, and Random Forest. Each Algorithm is used to predict the total number of COVID-19 cases for the next 21 days. Researchers approached the Time Series Forecasting model using datasets taken from the COVID-19 Information Center Coordinationn Center website. The results obtained in this study, the method that has the highest probability of accurate and appropriate data contained in the Gaussian Process method. Prediction data on the Linear Regression method has accurate results with actual data that occur with Root Mean Square Error 1202.6262. © 2021 IEEE.

Analysis of Mental Health During the Covid-19 Pandemic in Indonesia using Twitter Data

Covid-19, which has infected Indonesia, has had a significant impact on Indonesia in various sectors and has a direct psychological impact on the entire community, such as a fear attack, anxiety, stress, and depression. Not being able to meet friends, study and work from home, the existence of the PSBB policy, the large number of news and hoaxes about Covid-19, and worrying about being infected are some of the factors that can cause psychological problems. At this time, social media was helpful to get the latest information, share various content, tell stories, and express opinions or thoughts. This study will conduct a classification and analysis related to mental health during the pandemic using tweets shared by Indonesian users and then compare the algorithms, which are Naïve Bayes, SVM, Logistic Regression, and Random Forest. From the labeling process, 612 tweets indicate psychological problems, and 168 tweets indicate anxiety problems. This study succeeded in building two classification models to detect psychological problems and anxiety problems. Model 1 was built using the Naïve Bayes because Naïve Bayes algorithm has the highest results of all evaluations with 74.36% accuracy, 74.28% precision, 74.35% recall, and 74.30% f1-score. While model 2 was built using SVM algorithm because SVM has the highest score for accuracy with 76.42%, precision with 74.91%, and f1-score with 75.19%. © 2021 IEEE.