Can we predict the future of crypto currencies exchange rates with artificial intelligence and social media data? In this talk, Dr. Frey will take you on a journey and talk about a project where they used Twitter KPIs from in conjunction with exchange rates to evaluate machine learning generated coin forecasts. Thereby, he’ll reveal the underlying software architecture and how microservices, reactive Java and stream processing platforms got loosely coupled together with machine learning to achieve predictive capabilities. Throughout this journey, the most interesting findings, challenges and insights of the project will be shared.
Target Audience: Data Scientists, Big Data Analysts, Machine Learning/Model developers; Decision Makers, Architects
Prerequisites: Basic familiarity with machine learning is helpful, but not required. Interest in architectural stream processing concepts.
In a recent project, iunera sought to determine if it is possible to predict crypto currency exchange rates by utilizing social data from Twitter. I’ll talk about our experiences and describe how we leveraged online learning in conjunction with social data to determine if we are able to predict future currency exchange rates. I’ll point out the general architecture and describe the most interesting findings.
The audience can learn from our experiences and pitfalls. The new possibilities that opens up with more and more Kappa architectures in enterprises leverage new possibilities for real online learning. Therefore, the aspiration is to bring this field in general forward, by sharing insights.