DataBase and Data Mining Group


Research Activities

The DataBase and Data Mining Group is a research group of the Department of Control and Computer Engineering of the Politecnico di Torino. The interests of the group span over all aspects of Data Science and Machine Learning.

Relevant topics for the group include, but are not limited to, the following areas: Explainability and Fairness in Machine Learning, Finance and Quantitative Trading, Natural Language Processing, Concept Drift Detection, Unsupervised Learning, Time series analytics, and stream processing, Sensor-based data analytics, Smart Cities, Big Data Processing and Analytics, Data Warehousing and Data Mining.

Use the buttons below to discover more.

The group is also involved in several research activities conducted within funded project. Use the button below to jump to the dedicated section.

Explainability, Fairness and Bias

Recent data mining and machine learning models are often considered black-boxes, as the process that led to certain output is undisclosed. The group puts effort into developing novel techniques to unveil the “reasoning” behind the models’ decisions for both structured (tables, records, tabular data) and unstructured (images, texts) data.

Finance and Quantitative Trading

Short Bio of the Area…

Natural Language Processing

Short Bio of the Area…

Concept Drift Detection

Automatically detection of prediction-quality degradation of machine learning models due to class-based concept drift. 


  • Ventura et Al. – A new unsupervised predictive-model self-assessment approach that SCALEs.
  • Cerquitelli et Al. – Towards a real-time unsupervised estimation of predictive model degradation.
  • Cerquitelli et Al. – Automating concept-drift detection by self-evaluating predictive model degradation.

Data Science in Academy

We understand the importance of exposing students as early as possible to Data Science and Machine Learning topics. We thus propose several initiatives that let students learn by doing.

Machine Learning @ PolitTO

MAchine Learning At poliTO (MALTO) is a student team with the goal to take part in international data science shared tasks, projects, and competitions.

For more details, visit the team’s website.

Data Science Lab Environment

Data Science Lab Environment (DSLE) is a web platform to host data science competitions. It is currently used in the course Data Science Lab: process and methods to assess students’ abilities in solving classification, regression tasks.

Project code

Research Bites

Research Bites, a series of short research talks and seminaries held by PhD students and international faculty members for students of the course Data Science Lab: process and methods. The goal of RB is to disseminate cutting-edge research topics, in short, high-level pills.

Funded Projects

The group is focused on bringing high-quality research into funded projects of national and international nature.

SERENA – EU Project

A verSatilE plug-and-play platform enabling remote pREdictive mainteNAnce


  • Proto et Al. – PREMISES, a scalable data-driven service to predict alarms in slowly-degrading multi-cycle industrial processes.
  • Panicucci et Al. – A Cloud-to-Edge Approach to Support Predictive Analytics.
  • Apiletti et Al. – iSTEP, an integrated Self-Tuning Engine for Predictive maintenance in Industry 4.0.
  • Ventura et Al. – A new unsupervised predictive-model self-assessment approach that SCALEs.

Previous Research

The following is a comprehensive list of research and projects carried on during the years.