RiskDataScience identifies and develops effective solutions for the financial services industry using proven data science / machine learning techniques.
Customers in the B2B sector can use this to improve their information base, to recognize relationships more effectively and to increase the quality of forecasts. Processes can be optimized and costs reduced through better analysis, greater flexibility, and a more adequate presentation of results. Especially in the currently tight financial environment, the active development of digitization procedures can generate decisive competitive advantages.
In addition, we offer open source financial risk management tools that enable companies outside the financial services sector to identify credit, market and operational risks and integrate them with the benefits and needs of the Internet of Things.
We offer our solutions according to customer requirements in the form of consulting projects or as software packages. For all approaches presented in the use cases and blogs, we have developed working prototypes (mostly on the basis of Python and R). The solutions described in the User Area can also be operated online.
Our solutions include various fields, among others
- for text analysis, in which we offer in particular the following two application categories:
- Classification: Based on suitable training data, our tools enable the classification of “new” texts, for a detection e.g. of bankruptcies or relevant trading news.
- Semantic analysis: Texts can be grouped by similarity using our Regulytics® application or even compared at a granular level in terms of content. An application of this concerns the efficient analysis of regulatory texts, which is also available online and available for the building regulations of the German federal states as a free demo version.
- for the (quantitative) Risk Management:
- By combining methods from the areas of Artificial Intelligence and Financial Risk Management, we have developed methods for the automated portfolio optimization in terms of risk/return aspects
- We offer extensive tools for credit, market and operational risk on a VBA and Python basis. In the case of exchange rate risks, we have also developed a free web application.
- In this context, we also offer methods that can be used to classify market phases using data science methods and possibly identify looming crises.
- Our proprietary “Classification VaR” methodology quantifies the risks associated with the use of automated classification techniques.
- Other methods – such as reputation tracking via social media, identifying operational risk correlations or grouping geographic addresses – complement our risk offer.
We continuously develop our methods and solutions according to the requirements of our customers.
For creating sustainable value for its customers, RiskDataScience relies in particular on
- Experience: the CEO of RiskDataScience has more than 12 years of experience in the financial industry. He worked as a banker, consultant and auditor, among others at Deutsche Pfandbriefbank and Deloitte.
- Pragmatism: RiskDataScience has developed executable prototypes for all use cases presented on the website as well as all other offered use cases. For all concepts offered, the focus is always on application and feasibility.
- Flexibility: as a small and independent company, RiskDataScience offers solutions adapted to the respective customer requirements and develops them together with the customers.
- Efficiency: RiskDataScience relies on solutions that are as simple as possible and as complex as necessary. As far as possible and desired, open source tools are used, which considerably reduces costs.
- Transparency: RiskDataScience does not offer black box tools, but gives customers direct insight into how offline solutions work. Thus, the tools can be further developed independently and used sustainably.
Dr. Dimitrios Geromichalos
Founder / CEO
RiskDataScience UG (haftungsbeschränkt)
Theresienhöhe 28, 80339 Munich, Germany
Phone: +4989244407277, Fax: +4989244407001