Data engineering, analysis and processing

Head: Michela Papandrea

To be operational, data science applications require an efficient workflow of data access, process and storage that ensures the preservation of the data value and is not affected by the source, format or volume of the data. The research area in Data Engineering, Analysis and Processing focuses on the development of practical data science solutions and the process of making them operational, both from the data architecture and data manipulation points of view. The research area is skilled in the design and implementation of data science solutions from an infrastructural perspective, as well as in terms of data modeling, analysis, interpretation and visualization.

Specialty sectors:

  • Human behavioural analysis - Michela Papandrea
  • Big data processing - Massimo Coluzzi

Human behavioural analysis

Contact person: Michela Papandrea

The research sector in Human Behavioral Analysis aims at building human behavioral models to further understand the human mind. Our approach analyzes a wide array of heterogeneous factors that can affect the human brain, ranging from environmental ones to societal and biological ones. Key applications include behavior inference, recommendation systems, and personalization services.


  • Human behavioral sensing and modeling
  • Recommendation systems
  • Personalisation services
  • Sentiment analysis and emotion recognition

Big Data Processing

Contact Person: Massimo Coluzzi

Big data processing solutions face several challenges, including the design of immutable, distributed, reliable, and continuously-growing data storages, involving the design of scalable architectures and elastic data pipelines, as well as the development of software to analyze and process the data. The goals of the research sector in Big Data Processing are the development of distributed and elastic architectures to store and retrieve huge amounts of data and the development of data pipelines that leverage data mining methodologies (centralized, decentralized, and federated approaches).

  • Data lakes and data meshes
  • Infrastructures for big data analysis and processing
  • Lambda architectures
  • Unsupervised methodologies for topic extraction
  • Semi-supervised modeling
  • Collaborative learning