The objective of the LACODAM team is to considerably facilitate the process of making sense from large quantities of data, either for deriving new knowledge or for taking better actions. Nowadays, this process is mostly manual, and relies on the analyst understanding of the domain, the data at hand and a plethora of complex computational tools. We envision a novel generation of data analysis approaches where the many different ways of discovering structure in data are automatically explored, and only the most relevant structures are shown to the analyst. Such notion of relevance is highly dependent on domain knowledge and analyst’s own knowledge: such knowledge will be central to our approach. The solutions we envision requires to bridge data mining techniques with artificial intelligence approaches, both for taking knowledge into account in a principled way, and to introduce automated reasoning techniques in knowledge discovery workflows. Moreover, in order to acquire as much knowledge as possible, we will investigate community-based approaches, designed for communities of analysts and practitioners working on a given domain, sharing datasets, knowledge and results, and giving feedback.

Research directions

  • Automating the exploration of the KDD search space
  • Pattern mining operators
  • Scaling up through in-memory approaches
  • User/system interactions
  • Collaborative knowledge and feedback management