Mindmaps using ChatGPT and PlantUML

In my previous 2 part series on using Mermaid.js with ChatGPT to build system diagrams (part 1 and part 2), we looked at the capabilities of ChatGPT to build sequence, activity, state and C4 models.

I got interested to explore this further, and here is my next series on building more sophisticated diagrams with ChatGPT using PlantUML

PlantUML lets you create a wide range of diagrams from text descriptions. In its current state, it can support Gantt charts, mindmaps, concept diagrams and more, in addition to all UML diagrams

I was particularly interested in building MindMaps, so here's my first post of the PlantUML series. My interaction with ChatGPT and using it to create MindMaps with PlantUML

Introduction to Mindmaps

Prompt: What's a mindmap ChatGPT: A mind map represents ideas or information organized around a central concept or key idea. Mind maps are often used to visually organize information and create a clear structure for thoughts and concepts. The nodes in a mind map can branch out to show relationships between ideas. They can be colour-coded, labelled with icons, or annotated with text to clarify the relationships further. Mind maps are commonly used in note-taking, brainstorming, problem-solving, and decision-making, as they help to capture and organize thoughts visually and intuitively.

Basic MindMap

Prompt: Okay, Let's start by creating a simple MindMap on the topic of Artificial Intelligence Can you give the mindmap in PlantUML format. Keep it structured from the core central topic branching out to other domains and sub-domains. Let's go to 4 levels to begin with. Add the start and end mindmap tags and keep it expanding on one side for now ChatGPT: Sure, here's a simple mind map of the topic "Artificial Intelligence" in PlantUML format, with four levels

@startmindmap
* Artificial Intelligence
** AI Technologies
*** Machine Learning
**** Supervised Learning
**** Unsupervised Learning
**** Reinforcement Learning
*** Natural Language Processing (NLP)
**** Sentiment Analysis
**** Part-of-Speech Tagging
**** Named Entity Recognition
@endmindmap

and to visualize it, let's open draw.io or diagrams.net and then go to Arrange -> Insert -> Advanced -> PlantUML

and paste the above snippet

and here's the visualization

Cool! We have a Mind map!!

Let's try tweaking a few settings.

Increasing the Levels

Prompt: Can you increase it by two more levels and add further domains ChatGPT: Sure, here's the expanded mind map of the topic "Artificial Intelligence" in PlantUML format, with six levels:

@startmindmap
* Artificial Intelligence
** AI Technologies
*** Machine Learning
**** Supervised Learning
***** Regression
***** Classification
**** Unsupervised Learning
***** Clustering
***** Association Rules
**** Reinforcement Learning
*** Natural Language Processing (NLP)
**** Sentiment Analysis
***** Polarity Analysis
***** Emotion Analysis
**** Part-of-Speech Tagging
**** Named Entity Recognition
** AI Applications
*** Computer Vision
**** Object Recognition
**** Image Segmentation
*** Robotics
**** Navigation
**** Manipulation
*** Speech Recognition
**** Speaker Verification
**** Speech Synthesis
@endmindmap

Adding Colour Codes

Next, let's add colour codes to our MindMaps.

note: I had to provide the syntax for providing colour codes as ChatGPT is probably using an older syntax which breaks the plantUML code on pre-processing

Prompt: Can you add colour codes to the above mind map? Use the format *[#FFFF00] Artificial Intelligence. The above is just an example; use your colour codes for each node based on the complexity of each node in terms of the time it takes to learn that topic for a beginner

ChatGPT: Here's an updated version of the mindmap, with colour codes indicating the complexity of each topic

@startmindmap
*[#red] Artificial Intelligence
**[#red] AI Technologies
***[#orange] Machine Learning
****[#orange] Supervised Learning
*****[#yellow] Regression
*****[#yellow] Classification
****[#orange] Unsupervised Learning
*****[#yellow] Clustering
*****[#yellow] Association Rules
****[#orange] Reinforcement Learning
***[#orange] Natural Language Processing (NLP)
****[#yellow] Sentiment Analysis
*****[#green] Polarity Analysis
*****[#green] Emotion Analysis
****[#yellow] Part-of-Speech Tagging
****[#yellow] Named Entity Recognition
**[#red] AI Applications
***[#orange] Computer Vision
****[#yellow] Object Recognition
****[#yellow] Image Segmentation
***[#orange] Robotics
****[#yellow] Navigation
****[#red] Manipulation
***[#orange] Speech Recognition
****[#yellow] Speaker Verification
****[#yellow] Speech Synthesis
@endmindmap

and the representation

But what do these colours mean? Let me try and get a legend

Prompt: can you create a PlantUML legend for the colour codes above? Mark them against the time duration based on which the colours were selected ChatGPT: Here is a PlantUML legend for the colour codes, indicating the time duration for a beginner to learn each topic:

legend 
-   Green: Topics that can be learned in less than a week (around 10-20 hours of study)
-   Yellow: Topics that take a few weeks to a month to learn (around 30-50 hours of study)
-   Orange: Topics that take a few months to learn (around 100-150 hours of study)
-   Red: Topics that require substantial effort and time to learn (around 200-300 hours or more of study) 
endlegend

and importing in draw.io

Great, we have a mind map with colour codes and a corresponding legend with the description of each colour code.

Adding Icons

Next, let's try and add icons to our MindMap

note: I had to provide the syntax for providing icons as ChatGPT is probably using an older syntax which breaks the plantUML code on pre-processing

Prompt: Can you add icons to the above MindMap using plantUML-supported icons from the open iconic library? The format for usage is as *[#red] <&star> Manipulation. The above is an example of a red colour node and start icon from the open iconic library. Keep the same colour codes as above but use relevant icons as appropriate ChatGPT:

@startmindmap
*[#red] <&home> Artificial Intelligence
**[#red] <&aperture> AI Technologies
***[#orange] <&puzzle-piece> Machine Learning
****[#orange] <&tag> Supervised Learning
*****[#yellow] <&code> Regression
*****[#yellow] <&circle-check> Classification
****[#orange] <&dashboard> Unsupervised Learning
*****[#yellow] <&list> Clustering
*****[#yellow] <&tag> Association Rules
****[#orange] <&pulse> Reinforcement Learning
***[#orange] <&cog> Natural Language Processing (NLP)
****[#yellow] <&pulse> Sentiment Analysis
*****[#green] <&arrow-top> Polarity Analysis
*****[#green] <&eye> Emotion Analysis
****[#yellow] <&tag> Part-of-Speech Tagging
****[#yellow] <&bookmark> Named Entity Recognition
**[#red] <&aperture> AI Applications
***[#orange] <&image> Computer Vision
****[#yellow] <&target> Object Recognition
****[#yellow] <&crop> Image Segmentation
***[#orange] <&code> Robotics
****[#yellow] <&map-marker> Navigation
****[#red] <&star> Manipulation
***[#orange] <&microphone> Speech Recognition
****[#yellow] <&person> Speaker Verification
****[#yellow] <&audio> Speech Synthesis
@endmindmap

and the representation

icons from the open iconic library in plantUML

The icons don't all look appropriate, but at least a very good attempt by ChatGPT to map close icons based on topics.

Onto the next step

Adding Description

Prompt: Can you add a definition to only the leaf nodes? These definitions should be word-wrapped using PlantUML format and not have a surrounding box. Keep the Icons and Colors of the nodes as-is, and add additional description nodes to all leaf nodes. The format for adding leaf nodes is ****_ description ChatGPT:

@startmindmap
*[#red] <&home> Artificial Intelligence
**[#red] <&aperture> AI Technologies
***[#orange] <&puzzle-piece> Machine Learning
****[#orange] <&tag> Supervised Learning
*****[#yellow] <&code> Regression
******_ A statistical approach to finding the relationship between a dependent and one or more independent variables.
*****[#yellow] <&circle-check> Classification
******_ The task of categorizing a set of data into classes based on their features.
****[#orange] <&dashboard> Unsupervised Learning
*****[#yellow] <&list> Clustering
******_ A method of organizing a set of data into groups (clusters) based on their similarities.
*****[#yellow] <&tag> Association Rules
******_ A statistical approach for finding relationships between variables in large datasets.
****[#orange] <&pulse> Reinforcement Learning
*****_ An area of machine learning that focuses on training an agent to make a sequence of decisions by maximizing a reward.
***[#orange] <&cog> Natural Language Processing (NLP)
****[#yellow] <&pulse> Sentiment Analysis
*****[#green] <&arrow-top> Polarity Analysis
******_ A method of identifying the sentiment expressed in a text as positive, negative, or neutral.
*****[#green] <&eye> Emotion Analysis
******_ The task of detecting and classifying emotions in text or speech.
****[#yellow] <&tag> Part-of-Speech Tagging
*****_ The task of marking each word in a sentence with its corresponding part of speech, such as noun, verb, adjective, etc.
****[#yellow] <&bookmark> Named Entity Recognition
*****_ A NLP technique that identifies named entities in a text, such as organizations, people, locations, etc.
**[#red] <&aperture> AI Applications
***[#orange] <&image> Computer Vision
****[#yellow] <&target> Object Recognition
*****_ The task of recognizing objects in images or videos based on their features and attributes.
****[#yellow] <&crop> Image Segmentation
*****_ The process of dividing an image into multiple segments or regions, each of which corresponds to a different object or part of the image.
***[#orange] <&code> Robotics
****[#yellow] <&map-marker> Navigation
*****_ The task of guiding a robot from one place to another.
****[#red] <&star> Manipulation
*****_ The task of manipulating objects in the environment using a robot.
***[#orange] <&microphone> Speech Recognition
****[#yellow] <&person> Speaker Verification
*****_ The task of verifying the identity of a speaker based on their speech.
****[#yellow] <&audio> Speech Synthesis
*****_ The task of generating speech from text.
@endmindmap

and the representation

and there you have it. Sophisticated Mind Maps made with ChatGPT, colour-coded with legend, icon enabled and descriptive.

Thanks for reading!

Please subscribe here, always check our new posts, and remember to clap if you enjoyed reading this post.