Skip to content

Latest commit

 

History

History
42 lines (33 loc) · 2.22 KB

evaluation-metrics.md

File metadata and controls

42 lines (33 loc) · 2.22 KB

  1. Precision at k (P@k):

    • Measures the number of relevant documents in the top k results.
    • Formula: P@k = (Number of relevant documents in top k results) / k

  2. Recall:

    • Measures the number of relevant documents retrieved out of the total number of relevant documents available.
    • Formula: Recall = (Number of relevant documents retrieved) / (Total number of relevant documents)

  3. Mean Average Precision (MAP):

    • Computes the average precision for each query and then averages these values over all queries.
    • Formula: MAP = (1 / |Q|) * Σ (Average Precision(q)) for q in Q

  4. Normalized Discounted Cumulative Gain (NDCG):

    • Measures the usefulness, or gain, of a document based on its position in the result list.
    • Formula: NDCG = DCG / IDCG
      • DCG = Σ ((2^rel_i - 1) / log2(i + 1)) for i = 1 to p
      • IDCG is the ideal DCG, where documents are perfectly ranked by relevance.

  5. Mean Reciprocal Rank (MRR):

    • Evaluates the rank position of the first relevant document.
    • Formula: MRR = (1 / |Q|) * Σ (1 / rank_i) for i = 1 to |Q|

  6. F1 Score:

    • Harmonic mean of precision and recall.
    • Formula: F1 = 2 * (Precision * Recall) / (Precision + Recall)

  7. Area Under the ROC Curve (AUC-ROC):

    • Measures the ability of the model to distinguish between relevant and non-relevant documents.
    • AUC is the area under the Receiver Operating Characteristic (ROC) curve, which plots true positive rate (TPR) against false positive rate (FPR).

  8. Mean Rank (MR):

    • The average rank of the first relevant document across all queries.
    • Lower values indicate better performance.

  9. Hit Rate (HR) or Recall at k:

    • Measures the proportion of queries for which at least one relevant document is retrieved in the top k results.
    • Formula: HR@k = (Number of queries with at least one relevant document in top k) / |Q|

  10. Expected Reciprocal Rank (ERR):

    • Measures the probability that a user finds a relevant document at each position in the ranked list, assuming a cascading model of user behavior.
    • Formula: ERR = Σ (1 / i) * Π (1 - r_j) * r_i for j = 1 to i-1
      • Where r_i is the relevance probability of the document at position i.