Jul 16, 2013 · 7:00 PM
This location is shown only to members
30 min - I will be presenting on data manipulation with python and the pandas library. Pandas is an excellent option for quick and easy data analysis. It provides R-package-like functionality and integrates seamlessly with other Python data tools. This talk will offer an introduction and will provide practical examples of common data manipulation tasks.
10 min - Martin Burch will present a project involving public geospatial datasets
10 min - John Cardinale: R offers a free and fully functional structural equational modeling package call lavaan (latent variable analysis). This tool allows for modeling relationships between correlated variables, which uncovers both the underlying data structure as well as potential causal paths between overarching unobservable constructs. Lavaan rivals commercial SEM packages in it's flexibility.
10 min - Jia Zhang: This talk is a visualization project based on the membership data taken from historical archives of the Royal Society of London - the oldest and longest existing scientific society in the world. The goal is to use network measurements and visualizations to better understand the evolution of disciplinary distinctions within the sciences and through this pilot project establish new models for historians of science to conduct research.
10 min - Mike Schoeffler: Data crushes, data love - how to avoid falling in love with your models too quickly. A discussion of some real-life pitfalls in moving from hypothesis to useful predictions.
Free pizza & beer provided by Meetup