We focus on analytics that offer huge potential to transform the efficiency and productivity of industries such as health, logistics, urban development, transport, environment and security.
Examples of the application of our research could include:
- Data Analytics
- Smart cities – understanding the footprint of the city with respect to energy consumption and movement of people
- Transport – inferring road risk from simple smartphone data
- User Analytics
- Smart cities – understanding and inferring a person’s activity based on their spatio-temporal footprint in the city
- Text Analytics
- Tweet data modelling and analytics – data management and query language frameworks to capture, store, query, and analyse tweets for performing analysis (like opinion mining, e.g. what do people think of a company, or event detection, e.g. understanding what is going on in the city)
Research and Innovation
The Lab’s research emphasis includes:
- Complex networks and data analytics: Data analytics involves getting insights from data by identifying trends and presenting these findings in an easy and simplified way that everyone can understand. Through more complex data analytics we aim to provide not only analysis, but also to develop algorithms and techniques to handle large and interconnected data sets.
- Highly streaming data: Processing data that arrives at high speed from sensors, networks, or log files is a key challenge for modern information systems. As multiple information systems are required to provide a fast response and are often also limited by specific hardware constraints, we aim to design algorithms that efficiently manage these data streams. The group has experience in developing learning methods and real-time recognition from smartphone sensor inputs. Such specialist expertise delivers a crucial advantage in developing applications requiring rapid responses and monitoring.
- Data fusion: Data is stored by organisations with different purposes. This fact forces data to take different formats or levels of detail (specificity), while at the same time often leaving little reusable value or rendering the data incompatible with other organisations. In order to get insights from data among different repositories and formats, the group will focus on seeking novel approaches to unify these sources of information. Academics and current PhD students are investigating methods to handle structured and semi-structured data from social network data, scientific papers and geospatial data.
- Data access: Security and privacy have always been a primary concern for centralised or distributed information systems, and in the era of Big Data this is not going to change. We seek to study how to safely treat information, how to trace the source of data and ensure its integrity (provenance), how to safely preserve data, and how to efficiently query data sources.
- Infrastructure and architectures: Data is distributed in different locations and new technologies are being developed to store and communicate large repositories. We have conducted research in cloud computing so this can directly benefit and complement other research groups at the School of Science.