The internet consists of many networks connected by the BGP protocol and can easily be manipulated by a hacker. Every day, hackers reroute internet traffic and use that to impersonate entities such as companies, devices and humans. To detect this rerouting, which is also know as a route leak, a BGP monitor can be used. A BGP monitor checks whether the local route is correct by comparing the local route to the route from hundreds of devices. The lower the percentage of the routes that are the same, the bigger the chance there is a route leak. Hence, a user can tell whether a hacker manipulated the route. This project aims to build a BGP monitor application, that will ease the process of finding whether a hacker rerouted a part of the internet. To achieve this, the Internet Monitor application has been developed. In the Internet Monitor, an admin can add different types of tasks that will be run by the devices. This input will then be transmitted to the backend and to the database, where they will be stored. After some time, the backend will send out the tasks to the nodes. The nodes will execute the task and send back the results to the backend. These results can then be used to check whether the information from BGP servers or the local result is the same, and thus not manipulated. Apart from this, Internet Monitor also offers the possibility to see the map of all the real-time traffic between devices.

Analyzing and forecasting multivariate time series using networks is interesting in traffic, energy consumption, or financial forecasting applications. The main challenge is to capture both spatial and temporal dependencies in the data alongside the dynamics of the network itself. Graph Neural Networks (GNNs) have shown reliable performance in network-based data by modeling the connections as a graph. GNNs, using temporal and recurrent structures, are further developed as a prominent tool for processing multivariate time series. However, the dynamics of the network, the evolution of the network over time, and its effect on the data should be studied more. Recent works augment temporal GNNs with graph learning modules to account for the changes in the network; however, often, these methods do not consider the dynamics directly, and all of them stop the graph changing after the training phase. This thesis focuses on an online graph adapting approach to deal with the dynamic data over networks. The proposed method is a hybrid model consisting of a temporal GNN inspired by the physics of dynamic networks, trains its parameters during the training phase, and adapts the graph with the stream of temporal data online. The experiments study the graph adaptation module's role on various benchmark datasets in forecasting tasks, such as traffic, windmill energy, and financial forecasting.