Predicting Cyber Threats From Twitter Using Codeless LSTM Knime Model

Abstract

Cybersecurity threats pose significant risks in the increasingly interconnected digital world. Traditional security measures struggle to keep pace with modern cyberattacks, necessitating innovative approaches for proactive threat detection. This paper explores a codeless approach using Long Short-Term Memory (LSTM) model within the Knime analytics platform to predict emerging cyber threats from Twitter data to avoid the complexity and hassle of writing and debugging code. Based on the implementation results, the proposed model achieved accuracy on the prediction around 74%.