# Knowledge Graph

## From Files to Networks

Traditional knowledge management is **hierarchical**:

```
Folders → Subfolders → Files → Content
```

ATLAS knowledge management is **networked**:

```
Concepts ↔ Concepts ↔ Content ↔ Insights ↔ Actions
```

The graph structure reflects how ideas actually relate — not in neat folders, but in messy, beautiful webs of connection.

***

## Graph Structure

### Nodes (Entities)

| Node Type   | Description              | Example                            |
| ----------- | ------------------------ | ---------------------------------- |
| **Content** | Original source material | A tweet, note, or article          |
| **Concept** | Key idea or topic        | "Antifragility", "Network Effects" |
| **Insight** | Actionable learning      | "Systems fail at boundaries"       |
| **Action**  | Task or exploration      | "Research agent architectures"     |
| **Person**  | Referenced individual    | Authors, experts, contacts         |
| **Source**  | Origin platform          | Twitter, Apple Notes, etc.         |

### Edges (Relationships)

| Relationship   | Meaning                     |
| -------------- | --------------------------- |
| `CONTAINS`     | Content contains Concept    |
| `RELATES_TO`   | Concept relates to Concept  |
| `IMPLIES`      | Insight implies Action      |
| `REFERENCES`   | Content references Person   |
| `DERIVED_FROM` | Entity derived from Content |

***

## Graph Benefits

### 1. Connection Discovery

When you search for a concept, you don't just get direct matches. You see:

* Related concepts (1 hop away)
* Supporting content
* Derived insights
* Suggested actions
* Connected people

**Example Search: "Network Effects"**

```
Direct Results:
├── 12 content items mentioning network effects
├── 8 related concepts (viral growth, marketplaces, winner-take-all)
├── 5 insights about network effects
└── 3 action items

Graph Exploration:
├── "Platform Strategy" (strongly related)
├── "Aggregation Theory" (moderately related)
├── "Ben Thompson" (referenced author)
└── "Research: Metcalfe's Law applications" (suggested action)
```

### 2. Serendipitous Retrieval

The graph surfaces connections you didn't explicitly create:

> *"I searched for 'writing systems' and ATLAS showed me a connection to a tweet about Zettelkasten I saved two years ago. I'd completely forgotten about it."*

### 3. Knowledge Density Maps

Some areas of your knowledge are denser than others. The graph reveals:

* Where your expertise concentrates
* Knowledge gaps to fill
* Emerging interest clusters

### 4. Temporal Evolution

Watch your knowledge grow over time:

* When did you first encounter an idea?
* How has your understanding evolved?
* Which concepts are you actively building on?

***

## Visualization

### Interactive Graph View

The ATLAS dashboard includes a D3.js-powered graph visualization:

* **Zoom and pan** through your knowledge network
* **Click nodes** to explore connections
* **Filter by type** (concepts, content, insights)
* **Search highlights** matching nodes
* **Cluster detection** shows knowledge domains

### Graph Metrics

| Metric                     | What It Shows               |
| -------------------------- | --------------------------- |
| **Node Count**             | Size of your knowledge base |
| **Edge Count**             | Connectedness of ideas      |
| **Average Degree**         | How linked each concept is  |
| **Clustering Coefficient** | How ideas cluster together  |
| **Central Concepts**       | Most connected ideas        |

***

## Technical Architecture

### Storage Layer

```
SQLite + FTS5
├── Full-text search with relevance ranking
├── Efficient graph traversal via indexes
└── Portable, single-file database
```

### Graph Schema

```sql
-- Nodes
content(id, source, text, metadata, created)
concepts(id, name, description, importance, category)
insights(id, text, concept_id, importance)
actions(id, text, status, priority)

-- Edges
relationships(source_id, target_id, type, weight)
```

### Query Capabilities

ATLAS supports multiple query patterns:

**Full-Text Search**

```
"Find all content mentioning 'machine learning'"
```

**Graph Traversal**

```
"Find concepts related to 'productivity' within 2 hops"
```

**Aggregation**

```
"Show my top 10 concepts by importance score"
```

**Temporal**

```
"Show concepts added in the last 30 days"
```

***

## The Network Effect of Knowledge

Here's the key insight:

**Individual notes have linear value. Connected knowledge has exponential value.**

| Notes  | Isolated Value | Connected Value           |
| ------ | -------------- | ------------------------- |
| 100    | 100            | 100 + connections         |
| 1,000  | 1,000          | 1,000 + 10x connections   |
| 10,000 | 10,000         | 10,000 + 100x connections |

As your graph grows, each new piece of knowledge:

* Enriches existing nodes
* Creates new connection possibilities
* Increases the value of everything already there

**Your knowledge graph exhibits network effects.**


---

# Agent Instructions: Querying This Documentation

If you need additional information that is not directly available in this page, you can query the documentation dynamically by asking a question.

Perform an HTTP GET request on the current page URL with the `ask` query parameter:

```
GET https://docs.atlas-ai.org/features/knowledge-graph.md?ask=<question>
```

The question should be specific, self-contained, and written in natural language.
The response will contain a direct answer to the question and relevant excerpts and sources from the documentation.

Use this mechanism when the answer is not explicitly present in the current page, you need clarification or additional context, or you want to retrieve related documentation sections.