Abstract: A multi-tenant system stores a hierarchy of machine-learned models, wherein each machine-learned model is configured to receive as input a set of search results and generate as output scores for ranking the set of search results. Each machine-learned model is associated with a set of dimensions. The system evaluates search query performance. Performance below a threshold causes a new model to be generated and added to the hierarchy of models. Upon execution of a new search query associated with the same set of dimensions as the newly created model, the new model is used to rank that search query's search results.

Abstract: A multi-tenant system stores a hierarchy of machine-learned models, wherein each machine-learned model is configured to receive as input a set of search results and generate as output scores for ranking the set of search results. Each machine-learned model is associated with a set of dimensions. The system evaluates search query performance. Performance below a threshold causes a new model to be generated and added to the hierarchy of models. Upon execution of a new search query associated with the same set of dimensions as the newly created model, the new model is used to rank that search query's search results.

Abstract: A document retrieval system tracks user selections of documents from query search results and uses the selections as proxies for manual user labeling of document relevance. The system trains a model representing the significance of different document features when calculating true document relevance for users. To factor in positional biases inherent in user selections in search results, the system learns positional bias values for different search result positions, such that the positional bias values are accounted for when computing document feature features that are used to compute true document relevance.

Abstract: A multi-tenant system stores a hierarchy of machine-learned models, wherein each machine-learned model is configured to receive as input a set of search results and generate as output scores for ranking the set of search results. Each machine-learned model is associated with a set of dimensions. The system evaluates search query performance. Performance below a threshold causes a new model to be generated and added to the hierarchy of models. Upon execution of a new search query associated with the same set of dimensions as the newly created model, the new model is used to rank that search query's search results.

Abstract: A multi-tenant system stores a hierarchy of machine-learned models, wherein each machine-learned model is configured to receive as input a set of search results and generate as output scores for ranking the set of search results. Each machine-learned model is associated with a set of dimensions. The system evaluates search query performance. Performance below a threshold causes a new model to be generated and added to the hierarchy of models. Upon execution of a new search query associated with the same set of dimensions as the newly created model, the new model is used to rank that search query's search results.

Abstract: A document retrieval system tracks user selections of documents from query search results and uses the selections as proxies for manual user labeling of document relevance. The system trains a model representing the significance of different document features when calculating true document relevance for users. To factor in positional biases inherent in user selections in search results, the system learns positional bias values for different search result positions, such that the positional bias values are accounted for when computing document feature features that are used to compute true document relevance.

Abstract: A system stores objects of different types and allows search over the objects. The system receives search requests and processes them to determine search results matching the search criteria. The system ranks the search results based on weighted aggregates of features describing objects represented by each search result. The system monitors search results that were accessed by user for further information and marks them as accessed results. The system adjusts the feature weights used for ranking search results to optimize the ranking of the search results. The system analyzes the result of using the adjusted feature weights on past searches that are stored in the system. The system determines an aggregate accessed results rank for each adjusted set of weights. The system selects a set of feature weights that optimizes the aggregate accessed results rank for past searches.