Google Trends Data about Mental Health during COVID-19 Pandemic Using Time Series Regression

Along with the increasing number of COVID-19 sufferers during the Pandemic period, there was also an increase in searches related to mental health. Researchers have used a lot of Google Trends (GT) data to predict disease. However, researchers dissatisfied with the normalized index of GT began turning to Google Extended Trends for Health (GETH). Permissions and coding skills are needed to be able to access data from GETH. We have made one of the more friendly user interfaces for users without qualified coding skills. Using Google application programming interface (API), the data needed can quickly be taken according to the date parameter and the keywords required. We used 13 keywords using Indonesian to get search data on Google, as well as the number of positive COVID-19 sufferers in Indonesia released by the government. The regression analysis results show that the influence of the thirteen variables related to mental health on the positive cases of COVID-19 is 68.1%. In comparison, the most significant variables of the regression coefficient are cemas (anxiety), bunuh diri (suicide), and insomnia. The most partial variable is insomnia. © 2022 IEEE.

Comparison of Supervised Learning Methods for COVID-19 Classification on Chest X-Ray Image

The Coronavirus (COVID-19) pandemic is still ongoing in almost all countries in the world. The spread of the virus is very fast because the transmission process is through air contaminated with viruses from COVID-19 patients’ droplets. Several previous studies have suggested that the use of chest X-Ray images can detect the presence of this virus. Detection of COVID-19 using chest X-Ray images can use deep learning techniques, but it has the disadvantage that the training process takes too long. Therefore, the research uses machine learning techniques hoping that the accuracy results are not too different from deep learning and result in fast training time. The research evaluates three supervised learning methods, namely Support Vector Machine (SVM), K-Nearest Neighbor (K-NN), and Random Forest, to detect COVID-19. The experimental results show that the accuracy of the SVM method using a polynomial kernel can reach 90% accuracy, and the training time is only 462 ms. Through these results, machine learning techniques can compensate for the results of the deep learning technique in terms of accuracy, and the training process is faster than the deep learning technique. The research provides insight into the early detection of COVID-19 patients through chest X-Ray images so that further medical treatment can be carried out immediately. © 2022 CommIT Journal. All rights reserved.