Creating replicable metadata analyses with Python, Pandas, and Jupyter notebooks

Description

Many Alma users know that saved reports in Alma Analytics can be rerun on demand to replicate the same conditions and transformations. But what if you need to create a report that combines multiple data sources, relies on data not available in Analytics, or requires spot-checking the results of multiple complex data transformations? In this presentation, we will introduce the benefits and potential of building replicable analyses using the free, cross-platform, and well-documented combination of Python, Pandas, and Jupyter Notebooks. We will describe and demonstrate several replicable analyses we’ve created for various ongoing workstreams, including processing HathiTrust holdings, determining post-cancellation access for serials, identifying existing holdings for publisher ebook offers, and others. The examples we will share rely on data from Alma, but the techniques and tools are relevant for processing data from any source.

Period	Oct 13 2021
Event title	ELUNA Learns 2021: Analytics
Event type	Conference
Degree of Recognition	National