File size: 8,424 Bytes
8f05ad1 | 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255 256 257 258 259 260 261 262 263 264 265 266 267 268 269 270 271 272 273 274 275 276 277 278 279 280 281 282 283 284 285 286 287 | """
Knowledge Graph Implementation
Graph-based knowledge representation using networkx.
"""
from typing import List, Dict, Optional, Any, Set, Tuple
import networkx as nx
import numpy as np
from datetime import datetime
import json
from pathlib import Path
class KnowledgeGraph:
"""Graph-based knowledge representation with entities and relationships."""
def __init__(
self,
max_nodes: int = 10000,
max_edges: int = 50000,
):
"""
Initialize the knowledge graph.
Args:
max_nodes: Maximum number of nodes
max_edges: Maximum number of edges
"""
self.max_nodes = max_nodes
self.max_edges = max_edges
self.graph = nx.MultiDiGraph()
self._node_attributes: Dict[str, Dict[str, Any]] = {}
self._edge_attributes: Dict[Tuple[str, str], Dict[str, Any]] = {}
def add_entity(
self,
entity_id: str,
entity_type: str,
properties: Optional[Dict[str, Any]] = None,
) -> bool:
"""
Add an entity to the knowledge graph.
Args:
entity_id: Unique identifier for the entity
entity_type: Type of entity (e.g., 'person', 'concept', 'code')
properties: Additional properties
Returns:
True if added, False if limit reached
"""
if self.graph.number_of_nodes() >= self.max_nodes:
return False
if entity_id not in self.graph:
self.graph.add_node(entity_id, type=entity_type)
self._node_attributes[entity_id] = {
"type": entity_type,
"created_at": datetime.now().isoformat(),
"properties": properties or {},
}
return True
def add_relationship(
self,
source_id: str,
target_id: str,
relationship_type: str,
properties: Optional[Dict[str, Any]] = None,
) -> bool:
"""
Add a relationship between entities.
Args:
source_id: Source entity ID
target_id: Target entity ID
relationship_type: Type of relationship
properties: Additional properties
Returns:
True if added, False if limit reached or entities don't exist
"""
if self.graph.number_of_edges() >= self.max_edges:
return False
# Ensure entities exist
if source_id not in self.graph:
self.add_entity(source_id, "unknown")
if target_id not in self.graph:
self.add_entity(target_id, "unknown")
self.graph.add_edge(source_id, target_id, type=relationship_type)
edge_key = (source_id, target_id)
self._edge_attributes[edge_key] = {
"type": relationship_type,
"created_at": datetime.now().isoformat(),
"properties": properties or {},
}
return True
def get_entity(self, entity_id: str) -> Optional[Dict[str, Any]]:
"""Get entity information."""
if entity_id not in self.graph:
return None
return {
"id": entity_id,
**self._node_attributes.get(entity_id, {}),
}
def get_relationships(
self,
entity_id: str,
relationship_type: Optional[str] = None,
) -> List[Dict[str, Any]]:
"""Get relationships for an entity."""
if entity_id not in self.graph:
return []
relationships = []
for source, target, data in self.graph.edges(data=True):
if source == entity_id or target == entity_id:
rel_type = data.get("type", "unknown")
if relationship_type and rel_type != relationship_type:
continue
relationships.append({
"source": source,
"target": target,
"type": rel_type,
})
return relationships
def find_similar_entities(
self,
entity_id: str,
max_results: int = 5,
) -> List[Tuple[str, float]]:
"""
Find similar entities using graph-based similarity.
Args:
entity_id: Entity to find similar
max_results: Maximum number of results
Returns:
List of (entity_id, similarity_score) tuples
"""
if entity_id not in self.graph:
return []
# Use common neighbors as simple similarity
neighbors = set(self.graph.neighbors(entity_id))
scores = []
for node in self.graph.nodes():
if node == entity_id:
continue
node_neighbors = set(self.graph.neighbors(node))
common = len(neighbors & node_neighbors)
if common > 0:
# Jaccard-like similarity
union = len(neighbors | node_neighbors)
score = common / union if union > 0 else 0
scores.append((node, score))
scores.sort(key=lambda x: -x[1])
return scores[:max_results]
def search_entities(
self,
entity_type: Optional[str] = None,
property_filter: Optional[Dict[str, Any]] = None,
) -> List[str]:
"""
Search for entities.
Args:
entity_type: Filter by entity type
property_filter: Filter by properties
Returns:
List of matching entity IDs
"""
results = []
for node in self.graph.nodes():
attrs = self._node_attributes.get(node, {})
# Check type filter
if entity_type and attrs.get("type") != entity_type:
continue
# Check property filter
if property_filter:
props = attrs.get("properties", {})
if not all(props.get(k) == v for k, v in property_filter.items()):
continue
results.append(node)
return results
def get_subgraph(
self,
entity_ids: List[str],
depth: int = 1,
) -> nx.MultiDiGraph:
"""
Get a subgraph around specified entities.
Args:
entity_ids: Center entities
depth: How many hops to include
Returns:
Subgraph
"""
nodes = set(entity_ids)
for _ in range(depth):
for entity in list(nodes):
nodes.update(self.graph.neighbors(entity))
return self.graph.subgraph(nodes).copy()
def export_json(self, filepath: str) -> None:
"""Export graph to JSON."""
data = {
"nodes": [
{
"id": node,
**self._node_attributes.get(node, {}),
}
for node in self.graph.nodes()
],
"edges": [
{
"source": source,
"target": target,
"type": data.get("type", "unknown"),
}
for source, target, data in self.graph.edges(data=True)
],
}
Path(filepath).write_text(json.dumps(data, indent=2))
def import_json(self, filepath: str) -> None:
"""Import graph from JSON."""
data = json.loads(Path(filepath).read_text())
for node_data in data.get("nodes", []):
node_id = node_data.pop("id")
self.add_entity(node_id, node_data.get("type", "unknown"), node_data.get("properties"))
for edge_data in data.get("edges", []):
self.add_relationship(
edge_data["source"],
edge_data["target"],
edge_data.get("type", "unknown"),
)
def get_stats(self) -> Dict[str, Any]:
"""Get graph statistics."""
return {
"num_nodes": self.graph.number_of_nodes(),
"num_edges": self.graph.number_of_edges(),
"num_node_types": len(set(
attrs.get("type") for attrs in self._node_attributes.values()
)),
"num_edge_types": len(set(
data.get("type") for _, _, data in self.graph.edges(data=True)
)),
}
def __repr__(self) -> str:
stats = self.get_stats()
return f"KnowledgeGraph(nodes={stats['num_nodes']}, edges={stats['num_edges']})" |