Primo Research Assistant is a tool powered by GenAI (specifically LLM) that supports exploring academic content by asking questions in natural language.
"Primo Research Assistant: Ask research questions. Explore new topics. Discover credible sources."
The Primo Research Assistant is a component of OneSearch. It uses content found in your library to identify five documents that may help answer your question by extracting relevant information from the description/abstracts of each source.
The five documents will include articles, eBooks, or other resources available in OneSearch. Each document provides an overview and supplies information about the source. An option for viewing related results will expand your query into OneSearch.
Your question/search is converted into a query that the search engine understands with the help of a Large Language Model (currently GPT 4o-mini). It identifies the "most relevant documents" in the index and ranks them according to how well they can answer the question.
With the help of the LLM, an answer - the top five documents is presented. Due to the nature of LLMs, answers to the same question are not always the same. There may be more than one possible answer and different resources that are relevant. If you are not satisfied with your answers, use the “Try again” button.
Clarivate / Primo tools "utilize LLMs to process your input and Clarivate’s licensed content for real-time responses. Neither your input nor the licensed content are stored by the large language model or used for any other purpose than the immediate interaction session. Clarivate does not use your data to directly or indirectly train LLMs" (Clarivate AI Policy).
"Your data is protected and processed securely to uphold its integrity and confidentiality. We strictly adhere to applicable data privacy policies and regulations, ensuring secure handling of your personal data and information. Our tools utilize LLMs to process your input and Clarivate’s licensed content for real-time responses" (Clarivate Data Privacy).
Keep in mind:
It is possible a search will return saying it can not generate any results. It does not necessarily follow that there is nothing available, but that the way the search was entered did not meet criteria set by the LLM - or that the LLM configured is set to "flag" topics or terms.
Take time to restate your query - or return to the main OneSearch page.
Explore how to locate Research Assistant, manage a search, and review an example search for defining social media filter bubbles.
From the navigation/quick link bar at the top of every OneSearch page; select the ellipses ... to view the extended menu options. Click on Research Assistant at the bottom of the page.
If prompted to login, select AU User Login to proceed.
Enter your research question in the search bar, or select one of the example questions to review suggested topics.
Research history is available; it's possible to revisit previous questions. User may opt to keep or delete any search query.
Research Question:How do I define social media filter bubbles?
When this search was conducted, the five sources included three articles and two conference proceedings. It also provided a link to View More Results from a library search.
Overview of Sources:
"Social media filter bubbles can be defined as phenomena where algorithmic personalization on platforms such as Facebook, Twitter, and YouTube creates a unique informational environment for each user by selectively presenting content that aligns with their previous behaviors and preferences. This process often reinforces pre-existing beliefs and limits exposure to diverse perspectives, potentially leading to ideological homogeneity and affecting public discourse and political polarization" (Research Assistant overview).
