Skip to Main Content

Using Generative AI To Do Research

Tools

This section of the guide is intended for students who want to engage with AI-Tools at a more advanced level for research purposes, e.g., where they are interested in seeing how these tools support literature mapping, systematic literature reviews, or analyzing scholarly documents in detailed ways. If we have you curious read our introduction below and check out the tools section.

 

The tools described in this section of the guide are a sub-genre of generative AI tools that are often described as research discovery and workflow tools because they are based on models that draw on large datasets of scholarly information. 

These tools work with scholarly resources like articles to search and analyze them, extract key information, generate literature maps and summaries, or provide features common with citation tools. In a nutshell, they are often billed as helping to automate research workflows.

Keep in Mind

How Library E-Resources Compare
For assignments or research projects that require use of scholarly sources, start by searching the library's academic discovery layer, OMNI, or the library's databases. The coverage is more comprehensive than the tools found here.

Access Entitlements (Free or Fee?)
As long as you are a registered current student at York University Libraries, access to the library's rich suite of scholarly e-resources will be available to you, while not all AI research discovery and workflow tools are free, though in the tabbed boxes above we do distinguish between free and freemium sources.

Always apply critical thinking
When using AI tools, apply critical thinking to ensure accuracy, responsible use, and reliability. Critical thinking skills are used when you corroborate AI-generated information with other, reliable evidence, when you question the results, and when you remain mindful of any ethical considerations, such as bias. Get into the habit of treating AI tools as helpful starting points, but not as the final authority on any topic.

We endeavour to share with you a selected range of AI research discovery and workflow tools here and give you pointers on what they can do.

We are not endorsing any specific tools, and as with any type of technology or information source, we encourage you to critically evaluate what it offers or generates and to use it in tandem with the library's subscription e-resources.

WHAT DO WE MEAN BY RESEARCH DISCOVERY & WORKFLOW TOOLS?

These are generative AI tools that promote research effectiveness and workflows by searching scholarly records and literature using Retrieval Augmented Generation (RAG) which enhances a standard Large Language Model by searching and analyzing scholarly data sets and more. 

In our introductory section, we highlight some points to keep in mind, and encourage you above all to think critically about what each tool can and can't do, and how it compares to library databases and e-resources.

WHAT TYPES OF TASKS CAN RESEARCH DISCOVERY & WORKFLOW TOOLS HELP WITH?

Research discovery and workflow tools like the ones listed below can typically help you to engage in one or more of the following types of tasks: 

  • Search scholarly literature.
  • Engage in a literature review.
  • Set up alerts or tap in to recommendation features.
  • Carry out a systematic literature review.
  • Generate a literature map and visualizations of research data.
  • Summarize or analyze articles/documents. 
    • Typically extends to user-uploaded documents/articles.
  • Extract specific data from articles/documents including user-uploaded documents.
  • Import records from a database to support screening, including for systematic review workflows.
  • Interface with citation management tools like Zotero.
FREELY AVAILABLE RESEARCH DISCOVERY & WORKFLOW TOOLS: SELECTED TOOLS WITH KEY FEATURES
Semantic Scholar
  • Developed by the Allen Institute for AI (non-profit organization)
  • Primarily a research search tool for scientific literature
  • Supports literature search and highlights influential citations
  • Supports storing and organizing of papers in an online library (with sign-in)
  • Provides AI-powered paper recommendations based on what's in your online library
  • Ask this Paper feature gives AI-generated answers to questions about papers 
Research Rabbit
  • Uses Semantic Scholar and PubMed as data sources
  • Supports searching for scientific literature
  • Generates literature maps including interactive visualizations of papers and co-author networks
  • Provides AI-powered paper recommendations based on what's in your online library
Inciteful
  • Uses Semantic Scholar and OpenAlex as data sources, combined with Crossref and OpenCitations
  • Focus on scientific literature
  • Paper discovery tool facilitates building a network of academic papers with an analysis feature that helps you discover relevant literature
  • Literature connector tool allows the user to choose two papers and shows you how the literature connects them together
Open Knowledge Maps
  • Developed by a charitable non-profit organization
  • Uses PubMed (medicine, life sciences) and BASE (all disciplines) as its data sources
  • Supports searching for academic literature
  • Generates knowledge maps of research topics by clustering related papers thematically
  • Indicates where papers are open access
Catchii
  • Bills itself as a freely available systematic review (SR) screener
  • States it can help with all main stages of an SR, by supporting database record imports, identifying duplicates and creating PRISMA flowcharts
  • Supports export of screened data to reference manager programs like EndNote and Mendeley
FREEMIUM RESEARCH DISCOVERY & WORKFLOW TOOLS (FREE AND PAID FEATURES): SELECTED TOOLS WITH KEY FEATURES
Elicit
  • Uses Semantic Scholar as search engine
  • A research tool for scientific literature 
  • Provides AI-powered paper recommendations based on what's in your online library
  • Offers summaries of papers and chat with papers (limits apply on free version)
  • Helps with article analysis by extracting details from papers into an organized table
  • Supports synthesis and extraction of data from papers
Iris AI
  • Focuses on open access scientific literature including PubMed and CORE
  • Supports academic literature search
  • Offers summaries of papers and chat with papers 
  • Supports systematic review tasks 
  • Uses RSpace with user-uploaded documents
  • Chats with uploaded documents 
  • Helps with paper analysis by extracting details from papers
  • Supports synthesis /summarization of data from papers
SciSpace
  • Combines research discovery and research workflows tools features
  • Bills itself for its focus on Science
  • Offers a chat with pdf feature
  • Helps with paper analysis by extracting details from papers
  • Supports synthesis /summarization of data from papers
Perplexity
  • Chatbot style tool
  • Crawls the internet for information in real time, so has similarities with standard grounded text-generation tools defined under our Types of Tools section, but sets itself apart by offering an option to limit search to academic literature
  • When outputs generated in response to prompts, comes with citations
  • Users can query this tool in multimodal ways, i.e., using images, PDFs or text files
  • Offers a quick search and a pro search (for more detailed, tailored responses)
Connected Papers
  • Uses Semantic Scholar and PubMed as its data source
  • Generates visualizations or graphs of literature based on metrics drawing on co-citation and bibliographic coupling
  • Includes prior and derivative works
Consensus
  • Uses Semantic Scholar as its data source
  • Supports searching for scientific literature
  • Extracts key insights and answers at the paper level
  • Has useful filters like methodology, open access and more
  • Includes quality indicators for papers
Keenious
  • Uses OpenAlex as data source, an open catalogue of scholarly papers
  • Supports literature searching
  • Offers chatting with pdf papers and data extraction
  • Offers paper summarization
  • Integrates with MS Word or Google Docs to support easy integration while writing
Scinapse
  • Similar to Elicit (listed above)
  • Uses data from Semantic ScholarPubMed and OpenAlex
  • Aggregates and analyzes academic papers
  • Supports synthesis and extraction of data from papers
  • Also bills its tool as providing industry insights through research papers analysis, i.e. not only more scholarly data
Humata
  • Primarily billed as a PDF AI
  • Offers chat with PDF feature to generate answers
  • Summarizes PDFs
  • Compares PDFs and supports analysis across them
Docalysis
  • Primary function is as PDF AI, though TXT,  CSV and DOCX files are also supported
  • Chats with your documents to generate answers 
  • Supports reading, analyzing, summarizing documents
ChatPDF
  • A PDF AI chat tool
  • Extracts information or answers questions from large PDF files
  • Supports reading, analyzing, summarizing PDF documents