Karunakar Thammadi

Fraud in financial systems is a very significant issue in the modern financial systems and it is becoming difficult to identify fraudulent transactions. Old ways of fraud detection usually look at the transactions one at a time and do not look at how accounts relate to each other. This project uses a network of financial transactions with nodes corresponding to accounts or transactions and edges corresponding to money transfers. Patterns of these transaction networks are learned using Graph Neural Networks in the form of a Directed Graph Convolutional Network (Directed GCN). The datasets used in the evaluation of the model included PaySim, Elliptic Bitcoin, and synthetic fraud dataset. Graph features were also added which helped in understanding the patterns of the transactions better. Precision, recall, F1 score and AUC are some of the metrics used to measure the performance of the model. Besides, a Streamlit dashboard was created in order to visualize the fraud predictions and transaction trends. These findings indicate that the graph models can be useful to model the relationships between transactions and assist in better fraud detection.

Due to the environmental systems of various parts of the world, especially that of the Amazon rainforest, Africa, and Southeast Asia, illegal mining has been a key environmental problem in many areas. Illegal mining leads to massive destruction of the environment such as deforestation, river contamination, and wildlife habitat destruction. Late notice on illegal mining sites is part of the greatest problems that government and environmental agencies have to experience. When perilous activities may be noted, it is usually too late as serious harm to the environment may have been done. This project aims to create an artificial intelligence-based early warning system to detect destabilising illegal mining activities with the help of Sentinel-2 satellite images. It uses satellite data to detect environmental anomalies, vegetation changes and abnormal land-use patterns, related to illegal mining. Through the use of machine learning algorithms, anomalies in land development are detected and an early warning system is developed to assist environment managers to identify and monitor these areas.