How It Works

Data Upload and Vectorization

1. Upload Data: Users upload their datasets to the Evolve Network.

2. Vectorization Process: The data is automatically vectorized using embedding models hosted by peer compute resources. This process converts the data into vectors that represent its semantic context.

Query Processing

1. User Query: A user submits a query through the Evolve Network platform.

2. Embedding the Query: The query is embedded into the vector space using the same embedding models that were used for the data.

3. Vector Matching: The embedded query is compared to the vectorized data in the Vector DB. The system identifies vectors that are closest to the query vector in the vector space, indicating high semantic relevance.

4. Contextual Clustering: The system uses clusters of relevant information to ensure the context of the query is accurately captured.

Re-Ranking Mechanism

1. Initial Retrieval: The initial set of relevant vectors is retrieved based on their proximity to the query vector.

2. Re-Ranking: The re-ranker mechanism re-evaluates these vectors, further refining the search results to ensure the most relevant and useful information is presented to the user.

3. The development of the re-ranker mechanism will further enhance the accuracy and relevance of search results over time.