Learning to Rank
Learning to rank is a machine learning approach that trains a model to order search results directly from historical data — like past clicks and conversions — instead of relying on hand-set ranking weights.
Part of Ranking Optimization
Rather than a team manually deciding how much weight relevance, popularity, or recency should carry, a learning-to-rank model learns those tradeoffs from examples of what shoppers actually engaged with, and can adjust as behavior changes over time.