Data Science Conference 5.0
2019-11-19..2019-11-20
Belgrade, Serbia
| title | author | length |
|---|---|---|
| After Movie | 03m | |
A two‑minute highlight video condenses Data Science Conference 5.0 held 19–20 Nov 2019 at Hotel Hyatt Belgrade into fast cuts and upbeat music, showing… | ||
| WTF is Open Source free hardware and how is it influencing ML Data Centers | John Laban | 32m |
Open Compute Project designs and open-source hardware are concrete components used inside Facebook, Microsoft Azure, Goldman Sachs and Rackspace and this gear… | ||
| Responsible AI | Tarry Singh | 36m |
Describes building applied medical AI that actually ships: a breast‑cancer pipeline for Afghan clinics using heavy preprocessing, augmentation and custom… | ||
| How to make your CEO believer in Data Science | Natali Delic | 28m |
A telco builds a revenue-safe product called home internet / mobile network by adding a data-driven sales gate that predicts base-station capacity so agents… | ||
| Data Discussion - Digital transformation: Success stories of big corporation | Nikolay Zalevskiy, Milan Bukorovic, Dmitar Dilov | 50m |
Claims concrete outcomes: a machine learning pilot finds "digital oil" by identifying productive layers, predictive models run on refinery sensors to schedule… | ||
| The Coding Portion of Data Science | Milos Milovanovic | 28m |
Argues that data scientists must code beyond models, showing concrete practices: feature engineering in SQL rather than slow pandas, automated CI/CD with git… | ||
| Complex AI Forecasting Methods for Investments Portfolio Optimization | Pawel Skrzypek | 29m |
Describes the first complete AI investment platform that claims real‑time portfolio optimization using a hybrid of supervised and reinforcement learning, with… | ||
| The Dark Side of AI | Alessandro Palladini | 31m |
Argues that accuracy numbers hide real harm: a model at 95% accuracy still produces the 1% error that triggers actions and user rejection. Describes a released… | ||
| Recommender Systems for Personalized Financial Advice | Aleksandra Chirkina | 29m |
Production recommender uses matrix factorization collaborative filtering trained on implicit holdings turned into confidence-weighted labels where trading… | ||
| Machine Learning Driven Injury Prediction | Thomas Huijskens | 31m |
Consulting team at QuantumBlack builds a daily risk score to predict lower-body non-contact injuries for an elite football club, training models on… | ||
| AI for Waste Water Treatment | Robert van Straalen | 30m |
Only one of seven airflow sensors works, so a TensorFlow neural net is trained on one lane plus one year of one-minute samples to predict per-lane air flow… | ||
| Applied Data Science in decision making in Sales Department | Filip Jekic | 25m |
In this talk you will see how asw:maximus, real-time decision engine, can be applied in the decision making in sales departments. You will get a chance to see… | ||
| Security of Machine Learning | Jelena Milosevic | 28m |
Overview of concrete risks and tools: poisoning corrupts training data to inject backdoors while evasion crafts FGSM or Carlini‑Wagner perturbations at… | ||
| From 0 to a crowd sourced neural network in a jiffy | Jess Alfredsen, Sebastian Steenssøn | 21m |
A working production-grade NLP pipeline is built on stage: Doccano annotation feeds an RDS-backed dataset that is packaged to S3, sent to an AWS GPU training… | ||
| Scalable Automatic Machine Learning with H2O | Erin Ledell | 30m |
H2O AutoML runs distributed, Java‑based algorithms via Python/R/Scala, performs automatic target encoding for high‑cardinality categoricals (example: ~40,000… | ||
| Data Discussion - AI for Good | Tarry Singh, Alisa Piterova, Peter Morgan | 46m |
Concrete projects include using GANs and geomancer to synthesize realistic Amazon-2030 imagery and detect small fires, combining satellite imagery with drone… | ||
| Further enhancements of recommender systems using deep learning | Mladen Jovanovic | 29m |
Overview of concrete methods and results for recommender systems: compares matrix factorization and factorization machines to neural nets, shows where matrix… | ||
| Estimating Traffic Flow to find New Drive Thru Locations | Neil Farricker | 28m |
A pipeline produces countrywide link-level traffic estimates for South Korea by fusing OpenStreetMap road graphs, betweenness centrality scoring and car counts… | ||
| Trends and Practical Applications of AI & ML in FinTech Industry | Milos Kosanovic | 34m |
Claims that PSD2 open banking (PSD2) unlocks transaction-level data for new products, with Europe leading while China’s Alipay already serves ~1 billion users… | ||
| nothing matches | ||