Your Name and Title: 

Eric Silberberg 

Instructional design and education librarian 

 

Library, School, or Organization Name: 

Queens College, City University of New York 

 

Co-Presenter Name(s):  

None 

 

Area of the World from Which You Will Present:  

New York City 

 

Language in Which You Will Present:  

English 
 

Target Audience(s):  

Librarians and archivists with a basic understanding of Python or another programming language. 

 

Short Session Description (one line): 

Go from AI consumer to creator: start building library apps that use Google's PaLM LLM. 

 

Full Session Description (as long as you would like): 

Given the volatility of the AI ecosystem, the tool you use this week may disappear by the next. To address this, we have the option to create our own custom tools. 

This presentation offers a step-by-step tutorial on how to use Google's PaLM LLM to build our own library tools to automate tasks such as collection assessment, citation analysis, data clean up, etc. I will begin with how to customize PaLM’s options and then demonstrate how to incorporate AI into a Python script in order to automate library activities. 

The insights shared in this presentation were acquired through a broader initiative aimed at increasing faculty engagement with a university's institutional repository. The idea was that if faculty were aware of the open access (OA) policies of the journals in which they publish, they would be empowered to self-deposit. This led to the development of a program that automated the CV review process: I used AI to extract the journal names from citations and then pass those journal names to Sherpa Romeo to produce a report of OA policies per citation. This application will serve as the object lesson for the presentation. 

 

Websites / URLs Associated with Your Session: 

https://github.com/esilberberg/scholcom_tools 

You need to be a member of Library 2.0 to add comments!

Join Library 2.0

Votes: 0
Email me when people reply –