Abstract: Aspects of the present disclosure relate to interactive search training. A training canvas comprises results associated with a search query. The training canvas may be used as part of a training session that occurs during normal use of a search platform. When the search platform is first used, the results may be provided based on an existing model. An irrelevant result may be removed from the training canvas, such that a replacement result is added in its place. Additionally, results may be reordered, thereby indicating a ranking with which results should be displayed. Such interactions with the training canvas may be used to generate training data, such that a new model is trained accordingly. Thus, interactions with the training canvas yield high-quality training data that is usable to generate a model having equal or greater performance than a model that was trained using an equivalent amount of implicit training data.

Abstract: Aspects of the present disclosure relate to interactive search training. A training canvas comprises results associated with a search query. The training canvas may be used as part of a training session that occurs during normal use of a search platform. When the search platform is first used, the results may be provided based on an existing model. An irrelevant result may be removed from the training canvas, such that a replacement result is added in its place. Additionally, results may be reordered, thereby indicating a ranking with which results should be displayed. Such interactions with the training canvas may be used to generate training data, such that a new model is trained accordingly. Thus, interactions with the training canvas yield high-quality training data that is usable to generate a model having equal or greater performance than a model that was trained using an equivalent amount of implicit training data.

Abstract: One embodiment selects from a set of query-suggestion pairs a first query and a subset of query-suggestion pairs that each has the first query as its query; computes a Log Likelihood Ratio (LLR) value for each query-suggestion pair from the subset of query-suggestion pairs; ranks the subset of query-suggestion pairs according to their respective LLR values; removes from the subset of query-suggestion pairs all query-suggestion pairs whose LLR values are below a predetermined LLR threshold; computes a Pointwise Mutual Information (PMI) value for each remaining query suggestion pair from the subset of query-suggestion pairs; removes from the subset of query-suggestion pairs all query-suggestion pairs whose PMI values are below a predetermine PMI threshold; and constructs a ranked set of suggestions for the first query, wherein the ranked set of suggestions comprises one or more suggestions of the remaining query-suggestion pairs from the subset of query-suggestion pairs.

Abstract: One embodiment selects from a set of query-suggestion pairs a first query and a subset of query-suggestion pairs that each has the first query as its query; computes a Log Likelihood Ratio (LLR) value for each query-suggestion pair from the subset of query-suggestion pairs; ranks the subset of query-suggestion pairs according to their respective LLR values; removes from the subset of query-suggestion pairs all query-suggestion pairs whose LLR values are below a predetermined LLR threshold; computes a Pointwise Mutual Information (PMI) value for each remaining query suggestion pair from the subset of query-suggestion pairs; removes from the subset of query-suggestion pairs all query-suggestion pairs whose PMI values are below a predetermine PMI threshold; and constructs a ranked set of suggestions for the first query, wherein the ranked set of suggestions comprises one or more suggestions of the remaining query-suggestion pairs from the subset of query-suggestion pairs.