COVID-19 was declared a global pandemic by the World Health Organization in March 2020, and has infected more than 4 million people worldwide with over 300,000 deaths by early May 2020. Many researchers around the world incorporated various prediction techniques such as Susceptible–Infected–Recovered model, Susceptible–Exposed–Infected–Recovered model, and Auto Regressive Integrated Moving Average model (ARIMA) to forecast the spread of this pandemic. The ARIMA technique was not heavily used in forecasting COVID-19 by researchers due to the claim that it is not suitable for use in complex and dynamic contexts. The aim of this study is to test how accurate the ARIMA best-fit model predictions were with the actual values reported after the entire time of the prediction had elapsed. We investigate and validate the accuracy of an ARIMA model over a relatively long period of time using Kuwait as a case study. We started by optimizing the parameters of our model to find a best-fit through examining auto-correlation function and partial auto correlation function charts, as well as different accuracy measures. We then used the best-fit model to forecast confirmed and recovered cases of COVID-19 throughout the different phases of Kuwait’s gradual preventive plan. The results show that despite the dynamic nature of the disease and constant revisions made by the Kuwaiti government, the actual values for most of the time period observed were well within bounds of our selected ARIMA model prediction at 95% confidence interval. Pearson’s correlation coefficient for the forecast points with the actual recorded data was found to be 0.996. This indicates that the two sets are highly correlated. The accuracy of the prediction provided by our ARIMA model is both appropriate and satisfactory.

Cosine similarity is one of the most popular distance measures in text classification problems. In this paper, we used this important measure to investigate the performance of Arabic language text classification. For textual features, vector space model (VSM) is generally used as a model to represent textual information as numerical vectors. However, Latent Semantic Indexing (LSI) is a better textual representation technique as it maintains semantic information between the words. Hence, we used the singular value decomposition (SVD) method to extract textual features based on LSI. In our experiments, we conducted comparison between some of the well-known classification methods such as Naïve Bayes, k-Nearest Neighbors, Neural Network, Random Forest, Support Vector Machine, and classification tree. We used a corpus that contains 4,000 documents of ten topics (400 document for each topic). The corpus contains 2,127,197 words with about 139,168 unique words. The testing set contains 400 documents, 40 documents for each topics. As a weighing scheme, we used Term Frequency.Inverse Document Frequency (TF.IDF). This study reveals that the classification methods that use LSI features significantly outperform the TF.IDF-based methods. It also reveals that k-Nearest Neighbors (based on cosine measure) and support vector machine are the best performing classifiers