Welcome to our first ever international Meetup, a result of VDSGs ongoing effort of internationalization. Meet the Data Science Alliance, where speakers from Belgrade, Zagreb, and Vienna will share their knowledge with the international community.
Talks will be held in 3 cities in parallel on May 15th. The concept of the event is to have one talk in-live and two talks over the online streams at every location.
The theme of the meetup will be the application of Machine Learning & Deep Learning on real-life examples. You will get a chance to listen to 3 distinguished speakers from Belgrade, Vienna, and Zagreb.
Oppening talk – 17:30-17:45
Speaker -Marko Knezivic
Topic: Recommender systems for personalized content in video games
Marko Knežević holds PhD title at the Department of Computing and Control Engineering at the Faculty of Technical Sciences, University of Novi Sad.
In this talk, we’ll give light overview over recommendation techniques and how we allowed model to capture some general behavior patterns regarding buying and how we tackled common obstacles such as data dimensionality and class imbalance.
Second Talk- 18:15- 18:40
Topic: Exploiting Data in Medicine
Allan Hanbury is Professor for Data Intelligence at the TU Wien and faculty member of the Complexity Science Hub Vienna.
He was a scientific coordinator of the EU-funded Khresmoi Integrated Project on medical and health information search and analysis and is a co-founder of contextflow, the spin-off company commercializing the radiology image search technology developed in the Khresmoi project.
This talk will present examples of approaches to extracting value from medical data, in particular from research articles and radiology images, as well as paths toward more effective exploitation of data in medicine.
Third Talk- 18:50- 19:15
Topic: AI in FinTech service: Predicting Credit Debt Bankruptcy
Tomislav plays with data for a long time on every possible and impossible way. First Big Data project was on account-tickets in a book keeping service (low tech approach). Where the most see just numbers and letters, he finds a purpose and information. For a while now he is playing on a field of DM/ML models for business purposes, with the last couple of years having a spotlight on text analytics and NLP (unstructured datasets)
This talk is going to be about how we develop models which can tell us with high accuracy rate if the debt user will go into the personal bankruptcy. Our algorithms are self-learned, so with every new iteration result are showing greater accuracy.
In the first part of the talk we shall speak about how we used Gradient Boosting Machines (GBM) & Random Forest on the past and current data of credit debts of the users (such as how much are the user in the minus on bank and credit cards, how long they were late with payments and cetera) to predict personal bankruptcy. We have followed short term loans (up to 60 months), which were granted in the US. Project was also done for several Europeans countries, but with a bit lesser accuracy – caused by harder access to the data on customer placements and the level of indebtedness.
In the second part of the talk we shall explain how our US clients used this for their business optimization, for cash flow improvement and risk reduce, but also to find out exactly what is troubling the user who cannot repay the loan and offer ways to help him repay (and thus reduce his own loss).
Stay tuned and RSVP on our Meetup:
Looking forward to meeting you !