Primo is the discovery search tool that enables you to search for books and full-text resources that the Library subscribes to, (e-books, articles, videos, etc.) from a single search box on the Library’s homepage.

Primo Research Assistant (Beta) is an AI-powered tool designed to streamline academic research by providing users with relevant sources and summaries. It allows you to search the Library’s catalogue using natural language questions. It analyses descriptions and abstracts from trusted academic resources to generate an overview of the tool’s top five ranked relevant sources to answer your question. Developed by Ex Libris, this assistant leverages generative AI to enhance your research, making it more efficient and accessible.

Research Assistant utilises a Retrieval Augmented Generation (RAG) architecture, combining large language model capabilities with indexed academic materials. When you submit a query, the system converts the query into a Boolean search format and then retrieves the most relevant results from Primo’s Central Discovery Index. Using AI-driven embeddings, it ranks the top results and then generates an overview based on the abstracts of the top five sources, while also providing inline citations and references.

Users can input natural language queries, making research more intuitive and it has multilingual support, which allows you to submit your queries in various languages and it will provide results in the same language as well. It can give you direct access to available full-text sources, but it is important to note that the results are not always available through Stellenbosch University and may have to be requested via Interlibrary Loans. Clients should also keep in mind that Research Assistant is still in beta development.

Below is a short demonstration of how Research Assistant works:

 

By integrating AI into academic research, Primo Research Assistant can greatly aid how students and scholars access and analyse information. Its ability to provide accurate, well-referenced summaries makes it a valuable asset for our staff and students.

Author: Kirchner van Deventer