NDCG
An information retrieval evaluation metric for assessing ranking quality. It jointly considers the ranking positions and relevance degrees of multiple relevant documents in a result list, and is one of the mainstream metrics used to evaluate the ranking quality of retrieval systems.
Timeline (last 90 days)
Success criteria are quantified thresholds agreed upon in advance, such as recall, MRR, or NDCG reaching a certain threshold.
The project introduces three standard retrieval evaluation metrics: Hit Rate, MRR (Mean Reciprocal Rank), and NDCG (Normalized Discounted Cumulative Gain).
Retrieval metrics commonly used in RAG evaluation include Hit Rate, Mean Reciprocal Rank (MRR), and Normalized Discounted Cumulative Gain (NDCG).
Offline ranking quality is measured by NDCG, MAP, and MRR, classification prediction accuracy is measured by AUC and LogLoss, and online performance must be validated through A/B testing.
All Facts (4)
Success criteria are quantified thresholds agreed upon in advance, such as recall, MRR, or NDCG reaching a certain threshold.
50%UnverifiedThe project introduces three standard retrieval evaluation metrics: Hit Rate, MRR (Mean Reciprocal Rank), and NDCG (Normalized Discounted Cumulative Gain).
50%UnverifiedRetrieval metrics commonly used in RAG evaluation include Hit Rate, Mean Reciprocal Rank (MRR), and Normalized Discounted Cumulative Gain (NDCG).
50%UnverifiedOffline ranking quality is measured by NDCG, MAP, and MRR, classification prediction accuracy is measured by AUC and LogLoss, and online performance must be validated through A/B testing.
50%