Data Mining Smart Start Workshop – BNB-Qualitätstatistik und Training

Data Mining and Machine Learning in a Production Environment

Objective/Benefits: Important decisions should always be fact-based and thus data-driven. Especially in today’s world, with the ever-pr

esent topics surrounding digitalization, Industry 4.0, Big Data, and IoT, the use of data is not only becoming increasingly important but is also becoming indispensable for a company’s success.

But where and how do you start? What opportunities arise from technological developments in computing power and algorithms? Which data, platforms, and software fit your own goals? What about the Return on Investment (ROI)?

Especially at the beginning, many questions arise that can only be focused on and answered with an enormous amount of time and expertise. We would like to support you in this!

Contents of the Smart Start Workshop

Understanding and being able to apply terms, possibilities, and correlations (in my opinion) of Big Data, Industry 4.0, IoT, Predictive Analytics, Machine Learning (ML), Neural Networks, Fuzzy Logic, and Artificial Intelligence (AI)
Motivation for complex data analysis and predictive analytics
Multivariate approaches to reducing data complexity
Typical data mining approaches, methods, and analysis strategies
Data quality – structures and verification of their representativeness
Data access – context-oriented data storage, purposefully prepared for analysis
Best-practice approaches from the initial question to the solution (also within the “Digitalization” project team)
Real-world questions – technical data translation of the question – answering questions with data mining – real-world solutions and data-driven decisions
Presentation and comparison of selected data mining software packages
Tips and exchange of experience throughout the entire workshop
Exchange of experience and open discourse

Workshop Motivation:

Classical statistical methods can help identify business-relevant information in existing data to make appropriate decisions on this basis. However, simple statistical methods are often no longer suitable given the increasing complexity of data.
With more than ten influencing factors (predictors), linear models can only be used to a limited extent; significances are no longer recognized. However, they are completely overwhelmed at the latest when data from the Industry 4.0 context, often with over 100 predictors, needs to be evaluated. At the same time, only a few of this large number of variables are actually significant and “technically” relevant for explaining correlations. So, how can these be identified quickly and reliably?
Methods of multivariate analysis and data mining offer a solution that can meet these complex requirements. It is important that the results and insights from data mining can subsequently be transferred to and verified within a stringent propositional logic of classical statistics, such as in the Six Sigma methodology.

This results in the following objective for this workshop:

Participants will understand the possibilities and differences between classical statistical methods versus data mining algorithms and methods, and how they complement each other. They will be familiar with several structured approaches to complex data situations and data analysis and can immediately begin focusing on their own company’s specific questions.

This offer is specifically aimed at individuals and companies who want a quick and practice-oriented introduction to topics surrounding digitalization and data mining. During the two-day workshop, methodological and structural approaches will be presented, discussed, and applied. In addition, several complex datasets will be analyzed together using data mining (e.g., with neural networks and machine learning), and predictive models will be created.

In this way, the team experiences the various facets of the entire process from data acquisition to data utilization – starting from collection, storage, and preparation, through statistical methods and analysis, to the planning and recommendation of concrete measures.

Target Audience: Employees of a company with regular tasks in the digitalization process and/or data analysis

Workshop Duration: 2 days

Certification: Certificate of participation

Number of Participants: Minimum 7 people, maximum 12 people (in-person meetings) and 20 (online meetings)

Course Language: German

Speaker Team: Prof. Dr. Andreas Thümmel, QuaRisMa GmbH,
Björn Noreik, BNB-Qualitätsstatistik and Training

Costs: Course fee: €980 net per person for both days (Please note the minimum number of participants.)

Dates:

Individual for you and your team. Please contact us

Note: The focus is on the end-to-end process from data collection, through various statistical approaches to finding solutions, to the recommendation and validation of measures. The insights from the workshop can be implemented with various tools. Installation of the statistical software KNIME® and Minitab® is recommended for the workshop.