Abstract
In this project, we explore designing a new application to analyse market manip ulation in cryptocurrency-related news. At the core of this application is a novel framework, which leverages deep learning and natural language processing (NLP) techniques to identify persuasive techniques in a cryptocurrency-related article from four perspectives: (1) Propaganda value, (2) Sentiment value, (3) Financial spec ulation value and (4) Topic distribution. The designed application, FINsight, can be deployed to help cryptocurrency investors to review the trustworthiness of news and aid them in making more informed investment decisions.