DocumentationComputer Science資料庫Vector DatabaseSemantic SearchSemantic Search語意搜尋透過 embedding 向量表示文字、圖片等資料格式，實現基於語意的相似度搜尋。Vector Search 向量搜尋是語意搜尋的核心技術，透過計算向量間的距離來衡量相似度。Distance Metrics 三種常見的距離計算方式 Euclidean distance - 歐幾里得距離Internal product - 內積Cosine similarity - 餘弦相似度Euclidean distance d(p,q)2=(q1−p1)2+(q2−p2)2 d(p, q)^2 = (q_1-p_1)^2 + (q_2-p_2)^2 d(p,q)2=(q1−p1)2+(q2−p2)2Internal product a×b=∑i=1nai×bi a \times b = \sum_{i=1}^n a_i \times b_i a×b=i=1∑nai×biCosine similarity cos(a)=a⋅b∣∣a∣∣⋅∣∣b∣∣=∑i=1nai×bi∑i=1nai22×∑i=1nbi22 \cos(a) = \frac{a \cdot b}{||a||\cdot||b||} = \frac{\sum_{i=1}^n a_i \times b_i}{\sqrt[2]{\sum_{i=1}^n a_i^2} \times \sqrt[2]{\sum_{i=1}^n b_i^2}} cos(a)=∣∣a∣∣⋅∣∣b∣∣a⋅b=2∑i=1nai2×2∑i=1nbi2∑i=1nai×bi相關主題 RAG (Retrieval-Augmented Generation)Redis Vector DatabaseRedis Vector Database
