International Journal of Scientific Research and Engineering Development

Traditional methods of monitoring cryptocurrency transactions rely heavily on manual analysis and static rule-based systems, which are inefficient in handling the high volume, velocity, and complexity of blockchain data. These limitations hinder timely detection of suspicious activities such as abnormal transaction patterns, potential fraud, and illicit fund movements. This paper presents CryptoSentinel: Cryptocurrency Monitoring and Analysis System, an artificial intelligence and machine learning (AI/ML)-based solution designed to automate real-time monitoring and anomaly detection in Bitcoin transactions. The system integrates real-time blockchain data streams using APIs such as Blockchain.com and Blockstream, extracting key transaction attributes including transaction amount, fee, number of inputs and outputs, timestamp, block height, and sender–receiver addresses. A feature engineering pipeline processes these attributes into structured numerical representations, which are then analysed using the Isolation Forest algorithm for unsupervised anomaly detection. The model identifies deviations from normal transaction behaviour and assigns a risk score, enabling classification into severity levels such as Low, Medium, High, and Critical. The system is implemented with an interactive web-based dashboard using Streamlit, providing real-time visualization of transactions, anomaly distributions, and a dedicated interface for flagged transactions with complete details.

Shifa Bilal Tamboli, Simeen Phiroj Mulani, Arman Tajuddin Shiakh,"Cryptosentinel: Crypto Currency Monitoring and Analysing System" International Journal of Scientific Research and Engineering Development, V9(2): Page(2081-2089) Mar-Apr 2026. ISSN: 2581-7175. www.ijsred.com. Published by Scientific and Academic Research Publishing.