The research interests of the group span over the many challenges of heterogeneous, possibly multi-media, data management, with special attention to the so called “big data” challenges.
Data collected in most application domains are rich in volume and diversity. For example, in the medical domain, patients’ records include structured data (e.g., blood pressure values, temperature values), images (e.g., cat-scans), unstructured text (e.g., the reports written by the specialists), images and videos (e.g., recordings of invasive exams, such as a coronography), time series (e.g., ecg), and other forms of media. Consequently, generating value out of these rich and diverse data sets shares the 3V challenges ([V]olume, [V]elocity, and[V]ariety) of the so called “Big Data” applications. We note that to 3Vs are not sufficient and in order to support effective knowledge discovery, we must tackle additional, more specific, challenges, including those posed by the [H]igh-dimensional, [M]ulti-modal (temporal, spatial, hierarchical, and graph-structured), and inter-[L]inked nature of most multimedia data as well as the [I]mprecision of the media features and [S]parsity of the observations in the real-world. Moreover, since the end-users for most multimedia data exploration tasks are[H]uman beings, we need to consider additional fundamental constraints from the difficulties they face in providing unambiguous specifications of interest or preference, subjectivity in their interpretations of results, and their limitations in perception and memory. Last, but not the least, since a large portion of multimedia data is human-centered, we also need to account for the users’ (and others’) needs for [P]rivacy.
Interestingly, we observe that different domains and disciplines, apparently far from each other (such as Building Energy Consumption Analysis and Study of the Infectious Disease Propagation) can be seen as sharing common underlying models and can benefit from similar fundamental technological innovations.
The group works in tight collaboration with the EMITLAB (Enterprise, Media, and Information Technologies Labs) at the Arizona State University, also sharing research projects and the organization of scientific events.
More specifically, the group is active on the following research topics:
- Scalable techniques for tensor-based data analysis.
- Time series (possibly muti-variate) indexing, classification and querying algorithms.
- “Smart”technological solutions for large scale heterogeneous data management.
- Modeling users’ activity on social media,in collaboration RAI-CRIT (RAI Television Research Center), to leverage users’ social activity as an information source for television and radio programs recommendations systems.
- Definition of software instruments for improved accessibility to digital documents for visually impaired users.
Prof. Maria Luisa Sapino
Tel. (+39) 011 6706745