Web Traffic Time Series Forecasting Using ARIMA Model

Abstract

Abstract: Web traffic prediction is a major concern since it has the potential to produce severe snags in the working of websites. It is one of the most difficult tasks to make predictions about future time series values, so been a hot topic for research. The increase in web traffic may encounter a crashed site or very slow loading time. Such disturbances may cause many disturbances for the users, consequently decreased users rating of the site and user move to another site that affects the business. We have implemented a forecasting model to predict web traffic. ARIMA model is used for Web traffic time series forecasting. We have used some of the features like page name, date visited, and the number of visits for prediction with higher accuracy. Keywords: Web traffic prediction, ARIMA model, Time series forecasting, Data Collection and Feature Understanding.