Join us on September 11, 2019 for this unique meetup in the innovative atmosphere of WERK1 Munich – one of Europe's liveliest startup hubs.
The evening starts with a pre-meetup, hands-on workshop from 5-6:45 PM. This is perfect if you’re new to KNIME Analytics Platform. After a short introduction to the tool, you’ll build your first workflow, which reads and pre-processes some customer data before training a machine learning model. At the end of the workshop, we’ll show you different options to deploy the trained machine learning model using KNIME Server.
After the workshop, we invite you to refuel on some snacks and drinks before the talks start at 7:15 PM. Please note you are very welcome to join the meetup at any time during the evening.
Automated Retrieval of Market & Competitive Information using KNIME is the first presentation, by Dr. Matthias Stephan, who leads the Robotics Automation Team at DataVisions, Siemens Digital Industries. Based on a concrete competitive analysis example he will demonstrate how KNIME Analytics Platform can be used to automatically extract strategic competitive information from financial reports. Looking further ahead, he will introduce a number of text mining methods for analyzing web archives (such as a common crawl) to reveal company-relevant market trends.
The second presentation is by Simon Kraus, who manages operative logistics projects at GEMMACON GmbH and will present how he used KNIME in one of his projects to automate his data pipelines. The essential boundary restriction was to generate formatted documents that are used as working documents for his project members and in addition were exchanged with the client. The key challenge was to create these documents automatically from various inputs and highlight urgent tasks. This was not only possible to develop with KNIME and its integrations, but now the project manager himself can maintain and tweak the workflow without the help of our internal data scientists.
The last talk ‘Anomaly Detection for IoT Predictive Maintenance and Credit Card Fraud’ is by Kathrin Melcher (KNIME). An anomaly is a generic concept: it refers to any irregular or unexpected event, be it a mechanical tool failure, an arrhythmic heartbeat - or a fraudulent transaction. Therefore, the same techniques to predict anomalies in IoT data can also be used to detect credit card fraud and vice versa.
In this talk Kathrin shows three different approaches to discover the "unknowns": classical machine learning and deep learning neural auto-encoder for credit card fraud detection and a rule based technique for anomaly detection in IoT based time series data.
The meetup closes after a lively networking session. We look forward to seeing you!
Note: Please bring your own laptop with KNIME Analytics Platform pre-installed for the pre-meetup hands-on workshop. To install KNIME Analytics Platform download the installation files here.
You might find these videos helpful during installation.
- 5:00 PM – Registration
- 5:15 PM – Pre-meetup, hands-on KNIME Workshop by Kathrin Melcher (KNIME)
- 6:45 PM – Break: snacks and beverages
- 7:15 PM – Welcome by trusted KNIME Partner GEMMACON and KNIME
- 7:30 PM – Automated Retrieval of Market & Competitive Information using KNIME by Matthias Stephan (Siemens)
- 8:00 PM – Using KNIME Analytics Platform to Automate Data Pipelines for Formatted Working Documents by Simon Kraus (GEMMACON)
- 8:20 PM – Anomaly Detection for IoT Predictive Maintenance and Credit Card Frauds by Kathrin Melcher (KNIME)
- 8:40 PM – Networking