What is DSRC?

 

     Science and engineering progress is increasingly becoming dependent on data, yet traditional data technologies were not designed for the complexity of the modern world. The RPI Data Science Research Center acquires, processes, archives, analyzes, visualizes, simulates, and disseminates complex data to close the data-to-knowledge gap.

 

Data Science Research Center (DSRC) brings researchers from many different disciplines to model, analyze, simulate, visualize, and secure complex data acquired from diverse domains across multiple time and length scales.

 

DSRC serves as the melting pot of ideas and expertise in research areas such as computer science, biology, engineering, mathematics, physics, environmental science, library and social sciences. DSRC facilitates collaboration and interaction among not only RPI students, postdocs, and faculty but also investigators from external institutions both from academia and industrial research labs. The investigators in the center study data intensive complex problems in diverse application areas including medicine, oceanography, and networks (e.g., telecommunication, data, grid). One particular challenge the DSRC investigators take up is to bridge the gaps between mathematical sciences and life sciences by developing data driven models and algorithms.

 

Objective of DSRC is to become a center with national and international visibility, and provide support and infrastructure to its members for solving data centric and data intensive research problems by capitalizing on RPI’s super computer center (CCNI) and experimental media  and performing art center (EMPAC). Members of DSRC will collaborate and interact via workshops in specific topics, group meetings, seminars, student internships at industrial research labs. DSRC plans to offer an educational and training program for graduate students and postdocs to prepare the next generation data scientist and engineers.

                

The current scientific focus of the center is on multiscale approaches to complex data obtained from diverse domains including biomedical, environmental, engineering and social domains. The Center aims to vertically integrate solutions to several challenges of data science:

 

               1.     Data Acquisition and Preprocessing
                              Finding and accessing relevant data; discovery
                              Filtering, noise reduction
                              Semantic enhancements

               2.     Data Complexity
                              High dimensionality & multimodality
                              Heterogeneity
                              Inter- and multi-disciplinary
                              Large volume and rate

               3.     Modeling, Analysis, Learning and Knowledge extraction
                              Physics and informatics based multiscale models
                              Linear and nonlinear analysis methods
                              Supervised and unsupervised classification and learning
                               Missing data, uncertainty

               4.     Simulation, and Visualization
                              Sensory Analytics
                              Games
                              Dynamic interactivity

               5.     Security and Privacy
                              Integrity
                              Access control
                              Confidentiality
                              Anonymity