09 / THE NETWORK
Semantic Knowledge Graph
OBSIDIAN GRAPH VIEW SIMULATION
vector_search.py
import numpy as np
from sklearn.metrics.pairwise import cosine_similarity
# Find semantic neighbors
def find_links(current_note, vault):
query_vec = embed(current_note)
scores = cosine_similarity([query_vec], vault.vectors)
# Threshold for serendipity
links = [vault.ids[i] for i in scores > 0.85]
return links
from sklearn.metrics.pairwise import cosine_similarity
# Find semantic neighbors
def find_links(current_note, vault):
query_vec = embed(current_note)
scores = cosine_similarity([query_vec], vault.vectors)
# Threshold for serendipity
links = [vault.ids[i] for i in scores > 0.85]
return links
語義連結 (Semantic Linking)：
傳統檔案總管靠「檔名」分類，這是死胡同。
AI 靠 Vector Embeddings (向量嵌入) 理解內容的「語義」。它能發現：「這篇講 AI 架構的筆記，跟那篇講生物神經網絡的筆記，相似度高達
89%。」
於是，一條你從未想過的連結誕生了。