Semantium provides consulting in Data Science, Machine Learning, and Web Technology.

Combining a rigorous academic background and extensive business knowledge, we enable customers to economically exploit state-of-the-art technology.

Currently based in the beautiful city of Munich, Germany, we operate on a global level.

How Semantium can help you:

  • Big Data starts when Excel quits. We provide encompassing services in the fields of analysis and visualization of large datasets. Our background in Web science allows us to integrate large-scale Web resources with customer data. In a tight feedback-loop, we create actionable intelligence. We combine excellent technical skills with a sound business understanding. We consult in the following areas:

    • Data cleansing
    • Data analysis
    • Graph analysis
    • Data visualization
    • Data-driven strategic consulting
  • For a long time, Artificial intelligence has failed to meet initial promises. Lately, modern algorithms from the subfield Machine Learning have shown a massive business potential, and can yield significant competitive advantage. Possible applications:

    • Recommender systems
    • Fraud detection
    • Customer classification
    • Computer Vision
    • Sentiment Analysis
    • Natural Language Processing
  • While the Web startup scene is very dynamic, unfortunately often there's a lack of technical expertise in teams. Frontend and backend technologies are moving at an astonishing rate, and technical design decisions often pave the way to success or defeat. From task-based consulting to interim management, we can close technical gaps. We consult in the following areas:

    • Frontend architecture consulting (Which JS framework should we use?)
    • Backend architecture consulting (Programming Language, Databases, NoSQL)
    • Semantic Web Technology
    • Virtualisation & cloud consulting
    • Content delivery networks, caching solutions, load balancing
    • Security audits, stress / load tests

    Suitable industries:

    • Web Startups
    • E-Commerce


Dipl.-Kfm. Kurt Uwe Stoll

Uwe Stoll submitted his PhD thesis in May 2014 in the area of Semantic E-Commerce / Machine Learning and expects a completion of the doctorate by early 2016. Based on academic education, experience in Data Science and Machine Learning, and solid business understanding, he meets the requirements of a large range of data-driven projects.

Uwe was a research scientist at Prof. Dr. Martin Hepp’s chair for E-Business and Web Science, with a special focus on E-Commerce. In the last years, he engaged in designing and building the software ecosystem for the GoodRelations Web vocabulary, which is a world standard for Semantic E-Commerce, since its adoption by the leading search engines. That included development and maintenance of mission-critical Web applications. He managed research projects in the area of Semantic Matchmaking and Ontology-based Product Data Management, and published and presented my research at international conferences.

Uwe achieved fundamental knowledge in business strategy in my MBA studies with the special subjects Marketing and Innovation Management. During the last four years, he consulted in the Munich E-Commerce scene in business model design and technology architectures.

During his PhD, he gained significant experience in Software Design. Architecture and implementation of the system helped to achieve a solid understanding of large-scale data processing. Since 2010 he lead the Open Source development of the MSemantic Magento extension, which has more than 9000 installations to date.

The central idea of the thesis was the combination of Machine Learning algorithms with Semantic Web data. Being fascinated with the power of automation Machine Learning provides, Uwe contributed a new way of using Semantic Web data as a learning set for Machine Learning. In the course of the implementation for the thesis, he evaluated a significant range of Machine Learning approaches. He analyzed performance and computational complexity as well as provided recommendations for large-scale usage in a peer reviewed paper. While he programmed mainly in the Python programming language with the Pandas Data Analysis Library, current projects are started in Clojure on the JVM.

Recently, Uwe worked as Senior Lead Machine Learning Engineer at, London, to further attack the Web Information extraction problem space with various Machine Learning approaches.

Since early 2015, Uwe grew the Deep Learning Munich meetup to more than 900 members.

Dr. Mouzhi Ge

Dr. Mouzhi Ge is the Head of Research and Development at Semantium. He is also affiliated as a Junior Professor of Computer Science at University Bolzano. Before joining University Bolzano, he had worked as an IT Manager in the Oxbridge Investment Ltd. in UK, as a Postdoctoral Researcher in the Technical University of Dortmund in Germany and as a Visiting Researcher in the University of Milano-Bicocca in Italy. Meanwhile as a Senior Researcher, he is also affiliated at the Business Informatics Group in the Dublin City University in Ireland.
Dr. Ge's research and development are focused on E-Commerce Applications, Web Site Architecture, Business Data Analysis, PHP and Java Frameworks. From 2002 – 2005, Dr. Ge was leading a software development group in DreamTech and finished more than 20 Web development projects, which includes the Website of China Consumer Association and the online examination system in Hebei Province, China. These Web applications have been used by more than 1 million users. From 2005 to 2011, he has researched and implemented various novel Web applications such in recommender systems, cloud computing, and Big Data.

Contact Us

On The Web: Twitter