The main research track of KDDLab in the field of Complex Networks is Multidimensional Network Analysis. Traditionally, Complex Network Analysis has been monodimensional: researchers focused their attention to network with a single kind of relation represented. KDDLab is pushing the research over multidimensional networks, i.e. network with multiple kind of relations, since they are a better model to represent the complexity in reality (transportation, infrastructure and social networks are often multidimensional). In this novel scenario, we want to develop a basic theory and simple analytical measures, as well as more complex analysis and algorithms, such as community discovery, link prediction and the analysis of highly connected nodes (i.e. hubs). From the main multidimensional network research, different branches have emerged: we are investigating also mobility networks, using human trajectories as edges to connect different geographical areas and/or points of interest. Also, we have interest in analyze trust networks, particularly the problem of identifying possible privacy and/or security fallacies: is it possible that following a path of "trusted" individuals the information can be lead to an individual not trusted by the source of the information?
Analysis methods and tools to extract knowledge hidden in the data, including frequent patterns, clustering and classification.
Visual representation coupled with advanced analytics to comprehend and understand complex and large data.
A combination of analytic, machine learning, data mining and statistical skills as well as experience with algorithms and technological tools.
Acquiring strategies to manage and analyse large data sets and related tools such as MapReduce, Spark, Hive and Pigas well as NoSQL databases.
Identify hidden sub-structures within complex networks and exploit them to bound homophilic behaviors.
Understand hidden features of products and customers studying their position in the network built over the market.
Track, understand and forecast topological perturbations that affect complex networks as time goes by.
Understanding the patterns of success in several fields: sports performance, popularity of artistic items, emergence of new technologies.
Developing new methods of performance measurement by taking advantage of the huge growth of data collected during sport events.
Design algorithms for estimating the distribution of a population across different classes, and for tracking the changes in this distribution.
Design and develop useful tools for understanding, monitoring and signaling diffusion phenomena.
Developing of models to predict the well-being of territories based on Big Data on human behavior.
Design algorithms for discovering discrimination in socially sensitive decision data and for enforcing fairness in data mining models.
Extract meaningful knowledge from complex online and offline social contexts.
Correlate multiple data sources to build and thus understand semantic enriched descriptions of real world networked contexts.