High Dimensional Big Data Engineering
Thanks to advances in monitoring devices and modelling techniques, modern big data grows in both its quantity and dimensionality. Many popular applications involve processing and understanding the information contained in high-dimensional datasets, for example, document classification, pattern recognition, intrusion detection, recommender systems, etc. The intelligence of these applications heavily relies on the efficacy of processing and extracting meaningful patterns out of the datasets and the accuracy of searching. In reality, the balance between efficiency and accuracy plays a key role in building scalable services.
The workshop, held at the Computer Laboratory, University of Cambridge on Friday 22 January 2016, brought together the experts in computer science, statistics and mathematics in the leading institutes within and beyond the UK. It focused on the state-of-the-art engineering and algorithmic solutions adopted in realistic and large-scale applications in the context of high-dimensional big data.
Presentations and speakers
Title |
Speaker |
Large-Volume, High-Dimensional Data Processing at Thomson Reuters |
Dr. Jochen Leidner (Director), Corporate Research & Development, Thomson Reuters, UK |
Random Projection Ensemble Classification |
Prof. Richard Samworth, Statistical Laboratory, University of Cambridge, UK |
Multiple Random Projection Tree (MRPT) on High-Dimensional Data |
Prof. Teemu Roos, Department of Computer Science, University of Helsinki, Finland |
High-Dimensional Big Data Analysis |
Dr. Dimitris Tasoulis (Senior Execution Researcher), Winton Capital Management, UK |
A Knowledge Graph for Education and Learning |
Mads Holmen (CEO), Bibblio Inc., UK |
Multi Scale Machine Learning Methodologies for Molecular Biology Data |
Dr. Pietro Lio’, Computer Laboratory, University of Cambridge, UK |
Statistical Calculations at Scale Using Decisions and Emulation |
Dr. Daniel Lawson, School of Social and Community Medicine, University of Bristol, UK |
Uncovering Multi-Modal Spread Modes using Joint Diagonalisation |
Dr. Eiko Yoneki, Computer Laboratory, University of Cambridge, UK |
Optimal Hyperplanes for Clustering. Early results from High Dimensional Genomics Data |
Dr. David Hofmeyr, Lancaster University |
Accurate estimation of breakouts in high-dimensional panel data |
Leonid Torgovitski Mathematical Institute, University of Cologne, Germany |