Avance Programming Topics
The Georgetown Analytics program gives an online course in programming preparation that covers R, Python, an comman line use in the summer prior to matriculation. The course is equivalent to three creits, is esigne for matriculating MS Analytics stuents, an is offere free of charge. It is require for incoming stuents who o not have a computer science egree an aequate preparation. Stuents amitte to the program will only have this requirement waive with the approval from the Program Director (Ami Gates) or Assistant Director (Heather Connor). This course will run uring Georgetown Summer Session II (July 9 – August 10). Stuents must complete this course to matriculate in the fall unless grante a waiver by the program.
ANLY-501: Introuction to Data Analytics
This course introuces stuents to several core ata science concepts. It teaches stuents how to synthesize isparate, possibly unstructure ata to better unerstan an characterize the worl, an in some cases, to raw meaningful inferences. Topics covere inclue: the history of ata science, successes an failures in ata analytics, the ata analytics life cycle, ata/web scraping an APIs, ata wrangling, ata characterization (correlations, ientifying clusters an associations), ata inference an basic machine learning, network analysis, ata ethics, an visual analytics. Stuents will work on a semester-long ata science project that starts with question formulation an ata collection, an goes through all the stages of the life cycle, culminating in ata storytelling. The course also maps ata science case stuies to topics presente throughout the semester. Prerequisites: Intermeiate coing experience in Python3, an knowlege of introuctory statistics, 3 creits.
ANLY-502: Massive Data Funamentals.
Toay’s ata scientists are commonly face with huge ata sets (Big Data) that may arrive at fantastic rates an in a broa variety of formats. This core course aresses the resulting challenges. The course will introuce stuents to the avantages an limitations of istribute computing an to methos of assessing its impact. Techniques for parallel processing (MapReuce) an their implementation (Haoop) will be covere, as well as techniques for accessing unstructure ata an for hanling streaming ata. These techniques will be applie to real worl examples, using clusters of computational cores an clou computing. Prerequisite: Working knowlege of Python an the Unix comman line, some knowlege of ata structures an ANLY-501, 3 creits.
ANLY-503: Scientific an Analytical Visualization.
Presenting quantitative information in visual form is an essential communication skill for ata professionals. This course introuces representation methos an visualization techniques for complex ata, rawing on insights from cognitive science an graphic esign. Stuents will obtain an overview of the human visual system, learn to use moels for ata an for images, an acquire goo esign practices, such as those using the “grammar of graphics.” Stuents will use common statistical esign tools such as graphic methos in Python3, interactive graphic methos such as Bokeh, Leaflet, an NetworkD3, the R package ggplot2, an Tableau. Prerequisites: ANLY-501,ANLY-511, ANLY-512 3 creits.
ANLY-511: Probabilistic Moeling an Statistical Computing.
Probabilistic moels are essential for the unerstaning of ata that are affecte by uncertainty. This course introuces stuents to the funamentals of probabilistic moeling an then covers computational techniques for the analysis of such ata. After introucing basic concepts an approaches such as probability istributions, ranom variables, an conitioning, the course covers basic probability istributions that are frequently use in practice an some of their properties, such as Laws of Large Numbers. In the secon half, stuents will learn about computational techniques for the use of probabilistic moels. This inclues methos for faithful simulation of ranom variables (Monte Carlo), the extraction of conense moels from observe ata (maximum likelihoo, Bayesian moels), methos for moels with hien or partially observe variables (latent variables, expectation-maximization, hien Markov moels), an some general ata science techniques that incorporate probabilistic moels (graphical moels, stochastic optimization). Prerequisites: Introuctory statistics, some coing experience (e.g. R), 3 creits.
ANLY-512: Statistical Learning
Statistical Learning is concerne with algorithms that use statistical techniques to fin structure or patterns in given ata (unsupervise learning) or use given instances of ata to preict outcomes in new cases (supervise learning). A well-known metho of this type is linear regression, an this will be covere early in the course. Statistical methos for making iscrete preictions (classification) such as logistic regression will also be covere. Special emphasis will be place on techniques for hanling high-imensional ata (i.e. instances with many attributes), incluing variable selection an imension reuction. The course will also cover ensemble methos such as bagging an boosting that are often use to improve the results of given classification methos. Unsupervise methos covere in this course inclue moel-base an hierarchical clustering. Prerequisites: ANLY-511, 3 creits.
