Data Science in Telecom: 100 billion Transactions Per Day
Getting insights from 100 billion DNS transactions per day: challenges of applying data science to big data. Examples of passive DNS analysis discoveries.
Yuriy Yuzifovich, Nominum, San Francisco
EPAM journey: from data science to production (via videoconferencing)
Growing data science practice at EPAM: examples and future outlook.
Val Tsitlik, Vice President and the Head of the Big Data Competency Center, EPAM, Washington DC
Deep Learning and Convolutional Networks
Fully connected vs convolutional neural nets: machine learning in image processing. Scaling up with high performance computing techniques.
Artem Nikonorov, SSAU, Samara
Modern Medical Neuro-Imaging
Frontiers in neuroinformatics: Reverse engineering the brain from fMRI brain scans. Distributed processing pipeline using Apache Spark.
Konstantin Bychenkov, sMedx, Samara
Big Data Analytics, Data Science & Machine Learning with GPU
Anton Dzhoraev, Nvidia
Data engineering boost: how to make Python crawl faster
Speeding up data engineering tasks: Scaling data processing pipelines both horizontally and vertically to achieve better performance with Python.
Nikolay Markov, Aligned Research, San Francisco
Deep Learning for 3D Shapes Analysis
Alexandr Notchenko, Teatrall, Moscow
XGBoosting TV ratings: using data science for viewership prediction (via videoconferencing)
Using eXtreme Gradient Boosting (XGBoost) to forecast TV ratings: mixing viewership data with external data sources to improve prediction accuracy.
Dmitry Larko, Data Scientist, EPAM, San Francisco
Deep learning for video classification
Convolutional and recurrent neural networks: how it works, how tounite them, examples and results.
Michail Patkin, Deep learning engineer, Gentleminds, Samara