Università della Svizzera italiana Home

The Practice of Data Science

The practice of Data Science is based on the recognition that large quantities of data are: (i) Available before a question is posed and (ii) Collected for purposes that may not be related to the specific questions of interest. 

Data Science extracts meaning from data revealing insights that lie hidden beneath the surface of "big data" increasingly produced by routine individual behavior.  Data Science is a fundamentally interdisciplinary enterprise because it integrates multiple areas of scientific inquiry, each characterized by different skills, training, and professional development trajectories. Data Science projects aim to make sense of data, whether it is big data, complex data, qualitative, hidden, incomplete or corrupted data.

The Data Science process generally includes the following phases:



Understanding the data generating process
social science, communication, behavioral and cognitive science;



data engineering, acquisition, storage, indexing, retrieval, pre-processing, quality assurance



visualization, validation



statistical modeling, estimation and prediction, profiling, pattern recognition



synthesizing analysis, supporting individuals and organizations to address specific problems or improve and re-engineer a process



Providing actionable knowledge for decision making


This Data Science process represents a value chain and is universal across many empirical domains of application.



    All events