Visual Qualitative Analysis using AI and Knowledge Graphs

Posted

InfraNodus can be used to conduct qualitative analysis on free-form questionnaire responses, open-ended survey answers, and other types of qualitative data. It can also be used to perform thematic analysis on the texts from scientific papers, corporate websites, and other sources. In this tutorial, we will demonstrate how to use text network visualization and GPT AI to perform qualitative research and thematic analysis.

Unlike other tools, like MaxQDA or Atlas.Ti, InfraNodus has a strong visual component. Humans are primarily visual thinkers, we are very good in identifying patterns in data when we see them. Most of the other tools only show tables and lists. InfraNodus shows the patterns of concepts that form topical clusters, so you can quickly reveal the main patterns, identify the most prominent themes, and also see how they are connected.

More importantly, you can also reveal the structural gaps in the discourse: the topics that should be connected but are not yet. This allows you to identify the themes that could be linked with new ideas or solutions to the problems that are currently not addressed.

Qualitative analysis of structural gaps shows which topics could be better connected
 


 

Qualitative Analysis Workflow


The basic qualitative analysis workflow in InfraNodus consists of the following steps:

  • 1. Upload the data (anything from plain text to SRT audio transcripts, CSV spreadsheets, text files, etc.)

  • 2. Clean the knowledge graph to remove unnecessary data (for example, selecting and removing the interviewer's questions)

  • 3. Inspect the knowledge graph to reveal the main concepts, topical clusters, and relations between them

    • Basic representation of text network for thematic analysis

  • 4. Use the built-in GPT-4 AI to retrieve high-level topics and ideas from text:

    • Use GPT-4 AI to reveal high-level ideas in your discourse

  • 5. Zoom into important topics, explore them in context:

    • Exploring relations between different themes in thematic analysis

  • 6. Find interesting relations between the terms and their relative weight in discourse

  • 7. Generate summaries for particular themes and topical clusters

  • 8. Detect the structural gaps in a discourse: find what's missing

  • 9. Find the discourse's entry points: themes that make it easier to relate

 

If you're interested to learn more about how it works, please, check our qualitative research and thematic analysis tutorial or try it directly in InfraNodus.Com.


 

Try It Yourself


Try this approach yourself using InfraNodus and your own data:


Sign Up     Log In   
 

 

Consulting Services


Contact us if you are interested to hire InfraNodus experts.


Contact Us