Why is normalization of vectors important before indexing in a hybrid search system?

Comprehensive and Detailed In-Depth Explanation:

Normalization scales vectors to unit length, ensuring comparisons (e.g., cosine similarity) reflect directional similarity, not magnitude differences, critical for hybrid search accuracy. This makes Option C correct. Option A is false—vectors represent semantics, not just keywords. Option B (size reduction) isn’t the goal. Option D (sparse to dense) is unrelated—normalization adjusts length. Normalized vectors ensure fair similarity metrics.

OCI 2025 Generative AI documentation likely explains normalization under vector preprocessing.