Research Experience

Army Research Lab

Worked as a Research Assistant on a project with the US Army on analysis of Physiological data related to stress for inducing adaptations to enhance regulation of brain states and cognitive task performance.

In order to evaluate the effects of neurofeedback training on attention and task performance, the were asked to complete five neuro-feedback training sessions (on 5 different days) and to perform a simulated shooting task in a virtual reality environment under different attention-demanding conditions. The shooting task was be performed prior to beginning neurofeedback training, following each training session, and following the completion of five training sessions. In total, there were six neurofeedback training/shooting task experimental sessions. Each session was take approximately 2.5 hr and the seven sessions (17-24 hrs total) were be completed within an approximately three-week period.

I was responsible for recording, collection and analysis of Electroencephalography (EEG) and Neuro-feedback data. I interacted with the participants setting up and maintaining the equipment setup. The data was used to develop Machine learning model to predict the stress-level and brain states on performing difficult tasks.

CHMPR Lab (sponsored by NASA)

Responsible for analyzing OCO-2 satellite data provided by NASA related to Carbon Dioxide global measurements in UMBC’s CHMPR lab in collaboration with NASA.Gathered latitude and longitude points located in predetermined regions to calculate and compare the predicted CO2- fluxes. Results employing a Feed Forward Backward Propagation Neural Network model on two architectures, an IBM Minsky Computer node and a hybrid version of the ARC D-Wave quantum annealing computer.

Data Mining

There are two steps in Data Classification- One is supervised learning where Training data is analyzed by classification Algorithm and another is Classification where test data are used to estimate the exactness of classification rule. Clustering is the method of categorizing the data into clusters in such a way that objects inside a cluster have very high connection in comparison to one another but are very
much different to objects in other cluster. We used Weka tool on several algorithms related to ARM, Clustering and
Classification and have also compared them on the basis of their characteristics.

The paper was published in the International Journal of Emerging Research in Management &Technology