An analysis on security vulnerabilities of blockchain technology: Examination with topic modeling

Blockchain technology was introduced with Bitcoin in 2008 and has been used in many sectors, especially in finance. This technology, which offers innovative solutions in various fields with its secure, transparent and decentralized structure, has difficulties in the adoption process due to security vulnerabilities. In the literature, studies on blockchain vulnerabilities are generally limited to theoretical level or case studies. In this study, security vulnerabilities in blockchain-based applications were examined and 8705 data collected from the social media platform Twitter were analyzed with natural language processing (NLP) techniques. By using the Latent Dirichlet Allocation (LDA) method, the types and prevalence of vulnerabilities were determined and these findings have provided guidance for the security of future blockchain applications. The aim of the study is to better understand the security risks of blockchain systems and develop strategies to mitigate these risks. The thesis consists of data collection, cleaning, analysis and interpretation of the results. The results obtained are expected to contribute to the security of blockchain-based technologies.