Embedding

Embedding is a point or a list of numbers in a multi-dimensional space. Its function is to compress complex, high-dimensional data into a fixed-dimensional vector while preserving some semantic or feature relationships between the data. The proximity between two vectors can be used as a measure of their similarity. The smaller the distance, the higher the correlation, and the larger the distance, the lower the correlation. Embedding can be applied in various applications, including search, clustering, recommendation, anomaly detection, diversity measurement, and classification.