Nearest Neighbor Search
A technique that finds items most similar to a query item by computing distance in a multi-dimensional space. Common approach in vector search-based RAG, but can produce inaccurate results due to chunking artifacts.
As defined in When Vectors Break Down: Graph-Based RAG for Dense Enterprise Knowledge - Sam Julien, Writer
Other definitions
A classic machine learning algorithm that classifies a new data point by the labels of its nearest neighbors in feature space.
336 | Anil Ananthaswamy on the Mathematics of Neural Nets and AI
Where it comes up · 2
- When Vectors Break Down: Graph-Based RAG for Dense Enterprise Knowledge - Sam Julien, Writer3 claims
- 336 | Anil Ananthaswamy on the Mathematics of Neural Nets and AI1 claim
Most important claims · 4 of 4Search all claims for “Nearest Neighbor Search” →
- As the number of dimensions (features) grows, distance-based methods like k-nearest-neighbor fail because in high dimensions everything is roughly as far from everything else, so the notion that similar data points are closer than dissimilar ones collapses—this is the curse of dimensionality, predating and orthogonal to neural networks.336 | Anil Ananthaswamy on the Mathematics of Neural Nets and AI
- Vector retrieval with naive chunking and nearest-neighbor search produces inaccurate answers because relevant information can be split across chunks in ways that confuse similarity metrics, such as dating different Apple products incorrectly.When Vectors Break Down: Graph-Based RAG for Dense Enterprise Knowledge - Sam Julien, Writer
- Vector retrieval with naive chunking and nearest neighbor search produces inaccurate answers on dense text because semantic boundaries of chunks don't align with logical units—the example shows chunking the Apple timeline causes confusion between Lisa (1983) and Macintosh (1984) dates.When Vectors Break Down: Graph-Based RAG for Dense Enterprise Knowledge - Sam Julien, Writer
- Vector embeddings with chunking and nearest-neighbor search was Writer's second major retrieval approach before moving to graph-based systems.When Vectors Break Down: Graph-Based RAG for Dense Enterprise Knowledge - Sam Julien, Writer