Research and Development in Information technology
With the exponential growth of sensor networks, web logs, social networks and interconnected application components, large collections of data are continuously generated with high speed and in different formats. Traditional technologies are not able to deal with the immense scale and dynamicity of these data. Thus, storage, information retrieval, data analysis and visualization must be redefined. The main topic of the RDI team is about large-scale data management.
The scientific and technological challenges to address are related to:
- Real-time and distributed processing of various data sources (streams, web content, linked open data, etc.)
- Use semantic technologies to add a semantic layer
- Distributive and collaborative data mining
- Modeling and validation of complex systems
- Optimizing resources in large scale systems
- Sensor networks generate heterogeneous and asynchronous data flows that need to be transmitted through constrained networks such as GPRS, which is typical of the growing demand around Machine-to-machine solutions. These kinds of system often require complex integration and data mining techniques. For example: military applications, e-commerce RFID-based applications, weather observation and forecast applications, and energy production supervision applications.
- Internet/mobile portals that need to integrate a great deal of different services in a consistent way for a large amount of users and profiles. The utilization context is important and QoS issues need to be taken into account for response time, availability, usability, etc. Furthermore, in terms of development, these are systems that require the use of heterogeneous languages and technologies. For example: banking portals, e-commerce portals, ERPs.