VIDEO
HIGHLIGHT
Log InSign up
Visual AI Research Tool: Google Scholar, ChatGPT and Elicit Combined
SummaryTranscriptChat

Visual AI Research Tool: Google Scholar, ChatGPT and Elicit Combined

Introduction to InfraNodus AI Research Tool

In this section, the speaker introduces InfraNodus, an AI research tool that uses text analysis, network science and latest AI models to visually explore any topic in a scientific discourse.

How InfraNodus Works

  • InfraNodus enables users to visually explore any topic in a scientific discourse.
  • Users can import data from scientific journals using Google Scholar.
  • The imported data is visualized as a graph where the texts analyzed are the Google Scholar search results.
  • The graph helps identify main ideas and how they connect with each other. It also reveals gaps between them.

Using InfraNodus for Qualitative Analysis

In this section, the speaker demonstrates how to use InfraNodus for qualitative analysis by exploring a topic he doesn't know much about.

Analyzing Text Network Visualization

  • The speaker removes search terms used in Google Scholar so that they don't take too much attention.
  • He looks at the graph and identifies some interesting patterns or ideas that he finds important.
  • He writes down some insights he got from analyzing the graph such as qualitative research and thematic analysis being used in Psychology and health.
  • He recommends looking at peripheral nodes to see if there's some interesting stuff.

Exploring Genderqueer Narratives

  • The speaker selects nodes related to genderqueer narratives and discovers it's also used in sociology and anthropology.

Overview of Analytics Panel

In this section, the speaker introduces the analytics panel and explains how it can be used to identify main topics based on clustering of terms. The built-in AI can also be used to interpret these clusters.

Identifying Main Topics

  • The analytics panel identifies main topics based on clustering of terms.
  • Clicking "Review High Level Ideas" sends clusters to the AI for interpretation.
  • The AI generates names for clusters, giving an understanding of what they are about.

Exploring Interesting Tangents

  • Focus on exploring interesting tangents related to personal interests.
  • Clicking on a term shows in which context it was used and provides more information.
  • Using manual AI query with GPT-4 generates responses explaining concepts.

Importing New Content

  • Import new content by performing a search in Google Scholar and adding relevant terms back into the graph.
  • Filtering results by last search allows for easier analysis of new content.

Analytical Interviewing and Cognitive Reflexivity

The speaker discusses the use of InfraNodus to explore topics related to analytical interviewing and cognitive reflexivity. They demonstrate how the AI model can generate content in relation to selected terms, and how the system can identify gaps in a discourse and propose connections between them.

Using InfraNodus for Idea Generation

  • The speaker finds analytical interviewing and cognitive reflexivity interesting topics.
  • The AI model generates meaningful responses that connect ideas not well connected.
  • The speaker notes that using AI models is an interesting way to connect ideas that are not well connected.
  • Ideas generated by the AI model can be used to see how InfraNodus could be helpful for studying these topics.

Identifying Structural Gaps

  • InfraNodus identifies structural gaps between topics that exist in a discourse but are not yet well connected.
  • The speaker demonstrates how they use the AI model to generate research questions based on identified structural gaps.
  • Generated research questions are relevant because they touch upon existing topics in the discourse but connect them in new ways.

Human-in-the-loop Work with AI

  • The speaker uses manual AI queries to feed generated research questions back into the system.
  • This approach allows for human steering of conversations while using insights from the graph and structure of the discourse.

Design Principles and Thematic Analysis

In this section, the speaker discusses design principles and how they can be used in interdisciplinary problem-solving. They also explore thematic analysis and how it can be used for reflexive practices.

Design Principles

  • Design principles involve an interdisciplinary approach to problem-solving, incorporating human-centeredness, collaboration, creativity, experimentation, and adaptability.
  • The combination of critical thinking with design principles can be helpful for thematic analysis and reflexive practices.
  • Using the structural Gap inside feature allows for filtering statements on a graph to develop discourse further.
  • Discourse connector points are derived from social sciences and identify concepts with high influence but a small number of connections.

Thematic Analysis

  • Interview analysis in the field of psychology and health could be an easy entry strategy into scientific discourse.
  • Thematic analysis can uncover essential health-related processes pertaining to personal well-being.
  • Generating content using AI can link ideas together in an interesting way.
  • A summary of notes can provide insight into the synthesis of all ideas studied.

Conclusion

  • The speaker demonstrates how to import content from Google Scholar or other research paper archives, visualize graphs, identify main topics visually or using AI, zoom in on interesting topics, add project nodes based on research ideas, use Gap insights to generate gaps in networks to connect ideas more interestingly.

Using AI as an Assistant in Research Processes

In this section, the speaker explains how AI can be used as an assistant in research processes. The speaker demonstrates a general workflow and encourages viewers to apply it to their own research processes.

General Workflow for Using AI as an Assistant

  • In a real-life process, using AI would take more time and involve importing more data.
  • Instead of using AI to generate content, it can be used to guide thought through existing discourse.
  • This approach can lead to interesting discoveries and ideas along the way.
  • The workflow demonstrated by the speaker is available on the right-hand side of the screen for future reference.

Conclusion

  • Viewers are encouraged to apply this workflow to their own research processes.
Video description

In this video, I demonstrate how you can combine text network analysis with GPT-4 AI using #InfraNodus research tool in order to get an overview of any topic. Unlike other tools, like ChatGPT and Elicit, you can use the text visualization to reveal the most relevant patterns and — more importantly — the gaps between them and then use the built-in AI to retrieve the main topics and research questions that can help you move this field further. This makes it much easier to understand the context first to know what questions to ask. In this video, I'm using the results from Google Scholar (although you can also use PubMed, Arxiv, and PLOS data or your own list of abstracts or papers). The basic workflow is to visualize the main ideas around a certain topic (in our case: thematic analysis and qualitative research) and then use the graph in order to know what questions we can send to the built-in GPT-4 AI to generate interesting research ideas. Try it on https://infranodus.com Timecodes: 0:00 Introduction 0:41 Scientific discourse analysis via Google Scholar 1:11 InfraNodus algorithm explanation 2:26 Analyzing a topic using a text network 2:40 💡Hint #1: Remove the search terms to see the context around 3:27 💡Hint #2: Reveal interesting patterns and concepts in a text graph 4:20 💡Hint #3: Writing down project notes (coding) 4:54 💡Hint #4: Look at the periphery of ideas for interesting concepts 5:39 💡Hint #5: Reveal high-level ideas using GPT-4 AI 6:44 💡Hint #6: Focus on high-level ideas at the periphery to avoid generic insights 7:36 💡Hint #7: Ask GPT-4 AI a freeform question to clarify your insights 9:53 💡Hint #8: Import more specific data from Google scholar 10:54 Filtering search results by specific search query 11:42 💡Hint #9: Zooming Out / Zooming In — Finding a new topic to focus on 11:56 Finding relations between concepts 12:25 💡Hint #10: Using GPT-4 AI model to connect ideas in a new way 13:20 The mushroom analogy — picking ideas in the forest of science 14:04 💡Hint #11: Structural gaps — connecting interesting topics 15:58 Human-in-the-loop AI workflow: feeding AI questions back to itself 18:08 💡Hint #12: Discourse connector points — how to embed your ideas into an existing discourse 20:30 Generating a summary of your research notes using the AI 21:23 Generating an article outline 21:55 Conclusion: summary of the workflow Try it on https://infranodus.com Support channel: https://support.noduslabs.com

Visual AI Research Tool: Google Scholar, ChatGPT and Elicit Combined | YouTube Video Summary | Video Highlight