Personal Knowledge Graphs in Obsidian
Personal Knowledge Graphs
A personal knowledge graph represents an individual’s knowledge, ideas, and connections between different pieces of information. It’s a way to organize and visualize one’s understanding of various topics, concepts, and relationships.
Here’s a breakdown of the components of a personal knowledge graph:
1. Nodes: Nodes represent individual pieces of information or concepts. These can include facts, ideas, terms, or any other knowledge units. Each node typically contains specific attributes, such as a title, description, source, or metadata.
2. Edges: Edges represent relationships between nodes. They signify how different pieces of information are connected or related. These relationships can be hierarchical, associative, causal, or any other type that reflects the connections between concepts.
3. Attributes: Attributes provide additional information about nodes or edges. They can include metadata such as creation date, authorship, importance, relevance, or any other properties that help contextualize the information within the graph.
4. Hierarchical Structure: Personal knowledge graphs are often hierarchical, with nodes organized into categories or topics. This helps users navigate and explore their knowledge in a structured manner, moving from broader concepts to more specific details.
5. Semantic Meaning: Nodes and edges in a personal knowledge graph are not just isolated pieces of information but carry semantic meaning. This means that the connections between nodes represent meaningful relationships, allowing users to infer new insights or understandings by traversing the graph.
Overall, a personal knowledge graph is a powerful tool for knowledge management, sensemaking, and lifelong learning, empowering individuals to organize, connect, and leverage their understanding of the world.
To learn more, you can read my in-depth article
Personal Knowledge Graphs is about small data
Soft data and small data sets is overlooked hidden gems of personal ai assistance systems and personal knowledge management
Soft data is a data set that:
- are brain and human-friendly
- data that you care
- data that help you to take a decisions
- data that help you create
Source of small data
- your personal notes
- your reflection of common knowledge
- your values , ideas and dreams
In-depth tour to small data
Meta Frameworks for Idea Management and Personal Knowledge Graphs
The most common tool-agnostic framework for soft data based on personal note-taking is a Zettelkasten.
Notes is addressing a Notes in a graph . The ontology of this graph is simple: you have different kinds of nodes with completely relaxed and free ontologies on relations. Zettelkasten idea is to postpone a categorisation and maximise a reuse of nodes in completely new relations and context.
Nodes
- Fleeting Notes The Zettelkasten method starts with what are called “fleeting notes.” These are quick, temporary notes that you jot down when an idea strikes you. They can be as simple as a sentence or a keyword. The purpose is to capture the thought before it escapes your mind.
- Literature Notes When you read a book, article, or any other source of information, you create “literature notes.” These are summaries or paraphrases of the material, written in your own words. The aim is to distill the essence of what you’ve read, making it easier to review later.
- Permanent Notes The next step is to convert these literature notes into “permanent notes.” These are well-thought-out notes that you integrate into your Zettelkasten system. They are written in a way that makes them understandable even when taken out of context. Each permanent note should focus on a single idea or concept.
Edges
Linking and Indexing The real magic of Zettelkasten comes into play when you start linking these permanent notes together. By creating links between related notes, you’re essentially building a web of interconnected ideas. This enables you to see the relationships between different pieces of information, facilitating deeper understanding and creative thinking.
Structure
Free structure with no categories
As we can see this framework benefits only from nodes and edges
Toolings
As we can see, there are no user-friendly tools for full-scale personal knowledge graph management on the market. The majority of tools are oriented toward Professional oncologists in the enterprise domain.
The best on-user space in my opinion is Obsidian and Logseq
Obsidian won a competition because of community and extensibility via plugins, but log set has quite interesting properties like an embedded datalog engine and a more data-driven approach
more about logseq
What we need — typed Links
I wrote my take on the Obsidian community a long time ago
One simple missed feature that could turn Obsidian into a full-scale Personal Knowledge Graph is typed Links. We need the ability to attach metadata to relations itself.
MD files have a concept of data for links.
[obsidian](obsidian.md){caption: "likes"}As I wrote on my prev article we could have a partial solution of this problems with plugins
Obsidian Plugins for Personal Graphs
Excalidraw & ExcaliBrain
Make notes visual and allow the creation of your own naive ontologies mapped to a visual ontology that the excalibrain will understand.
juggl
It allows you to customize Graph rendering, but what's more interesting is that it allows you to introduce typed links and create captions on graph edges.
One of down sides of the plugin — performs quite low on big amounts of nodes
Graph-Link-Types
More simple plugin that just add caption s and typed links to already existing graph interface
What does the future look like?
As you can see, it is still difficult to manage Personal Knowledge Graphs and to apply reasoning and other powerful algorithms already used by big tech to user personal notes. Obsidian is a good start, but we need help.
One way is to use AI assent, GNN, and graph-oriented models that will help users co-create a personal Knowledge graph out of conversations and notes.
The same, more specific models, together with graph reasoning, could be applied to discover new knowledge and help users be creative.
MyKin.ia
Local first and privacy first AI assistant that co-creates a Personal Knowledge Graph with a user while the user is chatting with the app.
Mykin teams empower users with Knowledge Gpaphs and Personal Knowledge Graphs in quite an Obsidian Like manner
- local first
- privacy first
- full user ownership on data
Now user is not alone on his Knowledge journey. User co-create with an AI assistant that does a heavy lifting of Personal Knowledge Graph creation, management, and maintenance.
