![Munich Datageeks cover photo](https://secure.meetupstatic.com/photos/event/9/2/9/6/clean_467017526.webp)
What we’re about
Let's come together and discuss the latest advancements in machine learning and natural language processing, as well as data analytics. We'll explore the use of various open-source technologies for distributed computing and data processing, building and deploying machine learning models, data storage and management, and data visualization. We welcome participation from anyone who wants to share their experiences or ask questions on topics related to smart and large-scale data processing. The meetup is open to students, academic researchers, software engineers, statisticians, data scientists, and entrepreneurs who want to make an impact in the digital world while networking with like-minded individuals over a drink.
Our Meetup is once a month, typically scheduled and announced two weeks in advance. All Meetups will be conducted in English.
More information can be found at:
http://www.munich-datageeks.de/
Slides: https://speakerdeck.com/munichdata
Follow us on LinkedIn:
https://www.linkedin.com/company/munich-datageeks
Upcoming events (1)
See all- Munich Datageeks July EditionAllianz Beratungs- und Vertriebs-AG München, Unterföhring
We are thrilled to announce our next Meetup on July 30th at Allianz.
Format:- 2 talks (each ca. 40 min incl. discussion)
- Time for networking + food + drinks before, in between and after the presentations
- Talks are held in English
- We will be taking photos and/or film footage at the event. These will be used to share news about our meetups and to publicize upcoming events.
The lineup:
First talk:
Christian Lahr and Richard Vergin - The Voice of the Customer: Leveraging Large Language Models for Deeper Customer Analysis in InsuranceAbstract:
Another LLM use case...![img](https://mail.google.com/mail/u/0?ui=2&ik=5d89717325&attid=0.1&permmsgid=msg-f:1803744484716999924&th=1908300d5d2074f4&view=fimg&fur=ip&sz=s0-l75-ft&attbid=ANGjdJ9Qrx-HQh-pB19Ueckn4imC0RxvJC6g6A2dwDfROTs8c-3IUogeJCU46OH-YW4DLI35WN2dWhwlJFGhraoV5pca-HezLujcp2ODv7-fDhWvYA1_mXQ28zopLag&disp=emb)
In this presentation, we will share insights into how Allianz is using LLMs to automatically analyze its customer communications. Our ambition is to enhance customer understanding, detect dissatisfaction early, use insights to improve service quality, streamline internal processes, and maintain customer loyalty. At the DataGeeks event, we will be demonstrating our tool stack on AWS, leveraging named entity recognition, OCR and large language models to extract valuable features at scale as a baseline for several upcoming use cases.Bio:
Christian is Data Scientist and Cloud Engineer at metafinanz with a background in financial mathematics. He is assisting clients in designing and building the whole value chain of intelligent systems.
Richard is a Data Scientist at metafinanz with a background in engineering and applied research in engineering sciences. He is assisting clients in developing and implementing machine learning and AI solutions.Second talk:
Viktor Begun - Cost-optimized probabilistic maintenance for condition monitoring of wind turbines with rare failuresAbstract: We solved a very difficult but important task – predicting
failures of reliable, expensive machines that almost never fail, but
sometimes they still do. We built a model that works knowing only 2
different failures in 2 out of 5 offshore wind turbines and predicts 6
other failures in these 2 and in 1 previously healthy turbine, see
arxiv.org/abs/2407.09385. Practitioners may find the paper valuable
for its efficiency. The proposed approach saves more than 60,000 Euro
on average per 5 turbines in 4 months, and 85,000 Euro on average per 5 turbines per year.Bio: Viktor is a researcher at Technische Hochschule Ingolstadt with a
background in Theoretical Physics and Data Science. He also created
the “Happy Machine Learning” group on Facebook and LinkedIn to
provide a supportive environment for those who want to learn Machine
Learning and its applications.