Research and knowlodge exchange projects with both national and international industry and/or academic partners. My current active project is BigDataStack, where we are examining how to use real-time log data from cloud cluster deployments to produce effective machine-learned models for ranking candidate ways a user's application might be deployed on the cloud.
The aim of BigDataStack is to deliver a set of technologies that cover the complete stack of what is required for data operations and applications in a cloud computing environment. While the majority of the approaches for data operations and data-intensive applications (e.g. Hadoop, Spark, Hive, etc) "run on top" on typical infrastructure management systems (e.g. OpenStack, vCloud, CloudStack, etc), BigDataStack aims to develop a frontrunner data-driven cluster management system that is runtime adaptable, fully efficient and optimized for data operations, managing resources according to data-based decisions.
IMX is an industry collaboration project funded by the DataLab. It builds upon my knowledge of data mining and machine learning to develop techniques for comparing finance assets with model portfolios, such that alternative financial packages can be developed that fulfil the same role as an existing package, but at lower cost/risk.
SUPER is a project funded by the European Commission within the 7th framework programme under the SEC-2013-1 security call. SUPER explored a holistic integrated framework for tracking and evaluating the reactions of different types of stakeholders (e.g. victims, volunteers, citizens) to emergencies using social media, while at the same time empowering security forces and civil protection agencies to fully leverage social media in their operations.
The ReDites project was a follow-up project to CROSS (as well as other projects from the same funding call), and aimed to develop a demonstrator system for detecting, interpreting and monitoring events in social media.
TRIP was a knowledge exchange project with TRIPDatabase.com, which is a commercial SME that specializes in medical search. The aim was to research new approaches to personalize medical search results based on the type of the user and similar users for use within the TRIPDatabase.com search product.
The CROSS project examined how to find novel events in social streams such as Twitter, Wikipedia or Newswire, in real time. In this project we developing the Twitter event detection system at the core of the project based on locality sensitive hashing.