Ranking

From David's Wiki
Revision as of 20:47, 15 April 2024 by David (talk | contribs) (Created page with "Some notes on ranking techniques ==Basics== [https://medium.com/@mayurbhangale/pointwise-pairwise-and-listwise-learning-to-rank-baf0ad76203e Pointwise, Pairwise and Listwise Learning to Rank] ===Point-wise ranking=== In point-wise ranking, you have some scores for you document <math>y_i</math> so you can train your model <math>f</math> to predict such scores in a ==Metrics== ===Cumulative Gain=== Suppose you have a list of results <math>x_1,..., x_n</math> with rel...")
(diff) ← Older revision | Latest revision (diff) | Newer revision → (diff)

Some notes on ranking techniques

Basics

Pointwise, Pairwise and Listwise Learning to Rank

Point-wise ranking

In point-wise ranking, you have some scores for you document yi so you can train your model f to predict such scores in a


Metrics

Cumulative Gain

Suppose you have a list of results x1,...,xn with relevency r1,...,rn.
Then the cumulative gain at position p is the sum of the relevency of the first p results: (1)CGp=i=1pri

The discounted cumulative gain (DCG) takes the position into account, discounting lower-ranked results: (2)DCGp=i=1prilog2(i+1)

The normalized discounted cumulative gain (NDCG) is 1-normalized by dividing over the best possible ranking: (3)NCDGp=DCGg(r)maxrDCGp(r)