// MODEL OPTIMIZATION AND PROMPT SYNTAX TERM
Cosine Similarity
A common mathematical way to measure how similar two vectors are by looking at the angle between them; a smaller angle means more similarity.
TECHNICAL DEFINITION
A metric used to quantify the similarity between two non-zero vectors by measuring the cosine of the angle between them, ranging from -1 (opposite) to 1 (identical direction), commonly applied to vector embeddings to assess semantic similarity.
BACKGROUND
A recommender system, also called a recommendation algorithm, recommendation engine, or recommendation platform, is a type of information filtering system that suggests items most relevant to a particular user. The value of these systems becomes particularly evident in scenarios where users must select from a large number of options, such as products, media, or content. Major social media platforms and streaming services rely on recommender systems that employ machine learning to analyze user behavior and preferences, thereby enabling personalized content feeds.
READ MORE ON WIKIPEDIASYNONYMS & ALIASES
- Cosine distance (inverse)
- vector similarity
USAGE NOTE
Cosine similarity is the most widely used metric for comparing vector embeddings in semantic search and RAG systems.
DEVELOPERS
Organizations developing technology related to Cosine Similarity.
Develops a cloud-native vector database optimized for similarity search (including cosine similarity) of high-dimensional vectors, crucial for Retrieval Augmented Generation (RAG) and semantic search in AI engineering and prompt design.
An open-source vector database that allows developers to store data objects and vector embeddings and perform lightning-fast similarity searches using algorithms like cosine similarity, vital for AI applications and prompt engineering.
The company behind Milvus, an open-source vector database, and offers Zilliz Cloud, a managed service. Milvus is designed for efficient similarity search of massive vector datasets, frequently using cosine similarity for AI workloads.
Provides powerful embedding models (e.g., text-embedding-ada-002) whose output vectors are commonly compared using cosine similarity for tasks like semantic search, content moderation, and RAG architectures in AI engineering.
Offers a platform, libraries (like transformers and sentence-transformers), and models that are extensively used for generating and comparing text embeddings, where cosine similarity is a fundamental metric for understanding semantic relationships.
Specializes in large language models and highly performant embedding models. Their platform facilitates building RAG systems and semantic search, heavily relying on cosine similarity for retrieving relevant information.
Google's Vertex AI platform provides services for managing and deploying machine learning models, including vector search capabilities (e.g., Matching Engine) which leverage cosine similarity for recommendations, search, and RAG.
Offers a unified platform for data, analytics, and AI. Their solutions include vector search and tools for building and deploying LLM applications, where vector embeddings and similarity metrics like cosine similarity are critical.
An open-source, embeddable vector database that facilitates adding knowledge, facts, and skills to LLMs. It uses similarity search (often cosine similarity) to retrieve relevant contextual information for prompt augmentation.