Data Acquisition and Management

Catalog description

This course focuses on the data structures, data design patterns, algorithms, methods, and best practices for the pre-modeling phases of data science workflows, including problem formulation, gather, analyze, explore, model, and communicate, analytics programming focuses on the gather, analyze, and explore workflow steps. This comprises the 'data wrangling' work which is where most data scientists spend the majority of their time. Because data science is iterative, this preparatory work informs the modeling phase. Often, the creation and validation of new models requires going back for additional data, different data transformations, and exploration of data distributions. In short, every effective data scientist needs to master analytics programming. Course topics include reading from or writing to databases, text files, and the web; shaping data into 'tidy' data frames, exploratory data analysis, data imputations, feature engineering, and feature scaling.