Abstract: Disclosed are various embodiments for performing data search queries for descriptive semantics extracted from item reviews. Search queries that resulted in a selection of a browse node of an item taxonomy are identified and various items associated with the browse node are determined. Customer reviews for the various items are used to phrases used in the customer reviews that also appear in the search queries. The phrases may be ranked based at least in part on a polarity metric that describes a subjectivity, sentiment, and/or coverage of the phrase. Phrases with a rank exceeding a predefined threshold may be made available for use by a search component of a network site.

Abstract: Systems and methods involve filling missing attribute values from unstructured text. A computing device may provide a plurality of items, such as an item catalog for an electronic marketplace. When an item is found to have a missing attribute value, a plurality of existing values for that attribute is compiled by mining other items. Text associated with the item is parsed to determine possible values for the attribute. From those possible values, the most likely value is identified and the missing attribute value is populated with that value.

Abstract: Systems and methods involve filling missing attribute values from unstructured text. A computing device may provide a plurality of items, such as an item catalog for an electronic marketplace. When an item is found to have a missing attribute value, a plurality of existing values for that attribute is compiled by mining other items. Text associated with the item is parsed to determine possible values for the attribute. From those possible values, the most likely value is identified and the missing attribute value is populated with that value.