Forecasting Web Page Popularity using Artificial Neural Networks

Understanding a web page's popularity is crucial for content producers, marketers, and search engine optimisation experts in the current digital era. The use of artificial neural networks (ANNs) as a prediction tool to estimate web page popularity is examined in this research article. Our goal is to use ANNs, which have shown themselves to be effective and adaptable tools across a range of applications, to help tackle this problem. Gathering a thorough dataset of web pages with different variables such page content, keywords, social media activity, and past traffic trends is the first step in our research. Next, in order to provide a reliable input dataset for our neural network models, we pre-process and normalise this data. To choose the best model for predicting web page popularity, we create and train many ANNs with different topologies in the following steps. We include convolutional neural networks (CNNs) to analyse embedded picture and video material on web sites and recurrent neural networks (RNNs) to capture temporal relationships in web page traffic patterns in order to increase the complexity of our model. To increase the forecast accuracy of our model, we also take into account a number of other variables, including domain authority, external backlinks, and user engagement metrics.