The UC San Diego Halıcıoğlu Data Science Institute (HDSI) is emerging as a world leader in data science research, and its mission is strongly focused on the training and education of the next generation of leaders in data science. One way in which the HDSI accomplishes this is by supporting students, scientists, and faculty in developing new data analytic methods and infrastructure, and to foster “data science thinking” in a variety of problems in science, technology, engineering, government, the arts, and humanities to solve some of the world’s most pressing problems.
HDSI Undergraduate Scholarship Presentations
What cognitive mechanisms underlie the ability to draw objects at different levels of specificity? In this study, we explore the hypothesis that the ability to draw objects at different levels of specificity jointly depends on the abstract, conceptual knowledge we have about objects and our immediate visual experience of the world around us. To this end, we recruited a sample of adults to draw familiar objects at the specificity of a category, and of an instance. We found a difference in the recognizability of drawings that could not be explained by the amount of ink used in each drawing.
Justin Kang & Shone Patil
Using NFL rushing data we built pursuit vector based simulations with the goal of predicting how many yards a running back will gain on a given play. We utilized feature engineering to build out an accurate physics engine to simulate plays. Merged Madden game data with original rushing data to enhance realism of simulations. Compared outcomes of our prediction method to others such as histogram and K-Nearest Neighbors.
Scalable systems for Machine Learning are largely siloed into dataflow systems for structured data and deep learning (DL) systems for unstructured data. This gap has left workloads that jointly analyze both forms of data with poor systems support, leading to both low system efficiency and grunt work. Vista is a new data system that resolves systems issues by elevating ent