Technologies for Big Data
Eleven research projects in computer science studied and invented technologies needed for harnessing current and future big data.
They covered infrastructure aspects, including data access, cleaning, indexing and pre-processing; and they covered analytics, including query processing, data mining and machine learning, to facilitate knowledge extraction from data. These advances can improve the functionality and performance of big data applications, for instance by enhancing privacy or reducing the computing and data resources needed for model training in machine learning.
All projects of the module “Technologies”
- Stream analytics: fast processing and privacy-preserving tools
- Machine learning models: robustness and generalisability
- Data centres: efficient performance monitoring
- Loosely structured data: new tools for integration
- Language models: new methods for conversational agents
- Coresets: big data with less data
- Scala programming language: enabling big data analytics
- In-network computing: solutions for graph analytics
- Data streams: monitoring in real-time
- Fast prediction algorithms
- Graph analytics and mining