The thesis reconstructed and analysed transaction-level data of a particular darknet market. Moreover, the thesis reveals what kind of vendors are active on this darknet market, based on their characteristics. Finally, this research identified what the relative importance is of different vendor characteristics for vendor performance on darknet markets.

Side-channel attacks (SCA) aim to extract a secret cryptographic key from a device, based on unintended leakage. Profiled attacks are the most powerful SCAs, as they assume the attacker has a perfect copy of the target device under his control. In recent years, machine learning (ML) and deep learning (DL) techniques have became popular as profiling tools in SCA. Still, there are many settings for which their performance is far from expected. In such occasions, it is very important to understand the difficulty of the problem and the behavior of the learning algorithm. To that end, one needs to investigate not only the performance of machine learning but also to provide insights into its explainability.

In this work, we look at various ways to explain the behaviour of ML and DL techniques. We study the bias-variance decomposition, where the predictive error in various scenarios is split in bias, variance and noise. While the results shed some light on the underlying difficulty of the problem, existing decompositions are not tuned for SCA. We propose the Guessing Entropy (GE) bias-variance decomposition, incorporating the domain-specific GE metric in a tool to analyse attack characteristics. Additionally, we show the relation between the mean squared error and guessing entropy. Our experiments show this decomposition is a useful tool in trade-offs such as model complexity.

To dive deeper into the inner representations of neural networks (NNs), we use Singular Vector Canonical Correlation Analysis (SVCCA) to compare models used in SCA. We find that different datasets, or even leakage models, are represented very differently by neural networks. We apply SVCCA to a recent portability study, which shows one should be careful to overtrain their networks with too much data.

Finally, do we even need complicated neural networks to conduct an efficient attack? We demonstrate that a small network can perform much better by mimicking the outputs of a large network, compared to learning from the original dataset.

Web applications have been gaining increased popularity around the globe, in such a way that a growing number of users are attracted to make use of the functionality and information provided by these applications. While providing solutions to complicated problems in a fast and reliable way is one of the most advantages of using web applications, these platforms can cause adverse effect on user’s life if controlled in unauthorized way by malicious people. A platform with more vulnerabilities are more likely to be attacked. This research is focusing on building a prediction model for detecting vulnerabilities of web applications at eBay. Based on the analysis of important features, we dig deeper to find decisive factors of web application vulnerabilities. Making use of data on GitHub, we extract features related to source code files and developer networks, such as modification frequency, number of involved developers and duration between two commits. By applying machine learning techniques in the field of vulnerability prediction, we are able to provide reasonable suggestions for developers in the beginning phase. This can help develop relative defect-free and well-documented software. In this paper, we will explain the prediction model in detail from the aspects of code complexity, developers' behaviors and their networks. Moreover, according to results of various classifiers, we offer possible causes of vulnerabilities and reasonable suggestions for avoiding vulnerabilities in the future. To conclude, main contributions of this thesis are valuable feature engineering, the machine learning model and applicable suggestions for predicting vulnerabilities effectively at eBay.