Skip to Main Content

AI and Academic Research

A guide exploring the intersection between AI and academic research.

Scopus AI

Using Scopus AI

Use the link above to access Scopus.  Once there, look just above the search box for the Scopus AI search option.

 

Once in Scopus AI, use natural language to enter a research question.  Scopus AI will retrieve the most current, relevant documents from its corpus, analyze and summarize them, and provide links out to the originals.

See more about how Scopus AI works and how to use it.

Dissertations & Theses Research Assistant

The Research Assistant can help you:

  • Revise your search query
  • Gain a quick overview of a full-text dissertation or thesis
  • Brainstorm related research questions

If you run a search in a ProQuest Dissertations & Theses database and then access a full-text document in your results, its Key Takeaways will be influenced by your search to make them more relevant to your query, and a "Relationship to your search terms" section will be included in the Key Takeaways. The Key Takeaways, Important Concepts, and Related Research Topics are all extracted from the full text of the dissertation or thesis using Retrieval-Augmented Generation (RAG) techniques; no data from other documents or from public LLMs is used to generate them. ProQuest is not using the dissertations and theses to train LLMs, and ProQuest is only using generative AI with them within a secure environment.

LibrarySearch Research Assistant

The Research Assistant pinpoints five articles that can aid in answering your research question using the the descriptions/abstracts of each source. It provides in-line references to the sources to provide transparency in how each source contributed to the overview. Users can also refine their search by resource type (books, journal articles, and peer-reviewed) and date. To access the tool you can either use the direct link to it above, or look for the Research Assistant button in the top of any LibrarySearch results page:

 

To read more about how the tool works and what it is searching, including its corpus of documents, method of returning results, and which Large Language Model it uses in generating responses, look for the "How can I use Research Assistant?" link on the Research Assistant search page:

JSTOR Interactive Research Tool

The beta version of the JSTOR Interactive Research Tool is now available. To access the tool, you must create an individual JSTOR account and be logged in. Currently, the tool appears on content pages for journal articles, book chapters, research reports, and as an addition to JSTOR's standard keyword search. JSTOR has compiled additional information and FAQs regarding the new tool. 

Statista Research AI

Statista Research AI crafts responses using proprietary Statista data and analysis. This includes detailed descriptions and datasets from all available statistics, alongside information from topic pages, report pages, industry pages, and infographics. 

SciFinder-n: SearchSense

What does it do?

SciFinder uses AI to enhance search relevancy, to summarize results, and to suggest additional resources or search terms under the umbrella term of the "SearchSense" service.  

Natural Language Searching

Using human-curated search strings and phrases, SciFinder searching now supports natural language queries or searching on any form of a substance name.  

AI Summary

On the All and References pages for optimized searches with at least 5 results, SciFinder will present an AI summary and key findings of the search topic, based on top results.  

How does it work?

Natural Language Searching 

CAS uses AI to recognize the fields being searched, such as author, patent number, or reactions, and to provide the appropriate citations matching your search terms. Through concept-indexing functionality, SciFinder will recognize common phrases and redirect your search to controlled vocabulary matching that concept or phrase.  

AI Summary

In natural language or structure searches of at least 5 matching records, an AI Summary appears at the top of the All and References pages.  "The summary presents a quick overview of the literature space, providing both high-level overviews and more detailed information, which is generated by AI models trained by scientists focusing on scientific accuracy and traceable citations."  The AI model is trained only on SciFinder content.  

Clicking View All will display "Key Findings" with linked results to relevant publications.  

AI Summary in SciFinder