Google Inc.

Ebay, Incorporated

Qian Zhong

Guanglei Song

Zhaohui Chen

Ankur Jain

Muhammad Faisal Rehman

Mohammad Mahdian

Omkar Muralidharan

Christopher Davis Monkman

Yebin Tao

Claire McLeod

Ryan McDonald

Aranyak Mehta

Kent Andrew Edmonds

Joseph Anthony Beynon

Anjan Goswami

Gagan Aggarwal

Rehan Khan

Patrick Hummel

Jingzhou Hua

Shaohua Sun

Kshipra Uday Bhawalkar

Claire Mcleod

Methods, systems, and apparatus, including computer programs encoded on a computer storage medium, for obtaining cluster data, determining a portion of the search queries within the given query cluster that trigger a local search feature, assigning to the intent flag a value indicating that the given cluster is high local intent cluster when the portion of the search queries within the given query cluster that trigger the local search feature meets the local intent threshold, assigning to the intent flag a value indicating that the given cluster is a low local intent cluster when the portion of the search queries within the given query cluster that trigger the local search feature fails to meet the local intent threshold, and modifying a content item including adding a local content feature to the content item when the search query is determined to be included in the high local intent cluster.

Triggering local extensions based on inferred intent

Methods, systems, and apparatus, including computer programs encoded on a computer storage medium, for optimizing transmission of digital components. In one aspect a digital component distribution system performs operations including determining a minimum eligibility value required to maintain the eligibility of a given digital component based on a next highest eligibility value of one of the other digital components. Based on the next highest eligibility value, a baseline transmission value required to be submitted for distribution of the given digital component is determined. A normalized transmission value is determined based on the minimum eligibility value, and transmission of the given digital component to a client device presenting a particular electronic document is triggered based on the normalized transmission value.

Optimizing transmission of digital components

In various example embodiments, systems and methods for providing category aspect information by mining historical data is provided. In example embodiments, a table comprising joined data is accessed. The table includes historical data that comprises user behavior data based on actions performed with past queries by users, data describing publication, and a determined category for each publication. Demand scores based on the joined data are determined. The determined demand scores are used to determine most relevant aspect name and aspect value pairs for a category. Publications having at least one of the most relevant aspect name and aspect value pairs are displayed visually distinguished from less relevant publications that exclude the most relevant aspect name and aspect value pairs.

System and method for mining category aspect information

Methods, systems, and apparatus, including computer programs encoded on a computer storage medium. In some implementations, methods include determining multiple different configurations of multiple digital components eligible to be presented within a single digital component slot in a given electronic document being presented at a client device, selecting a particular configuration that has a highest configuration score and includes a given digital component, determining an offset factor that quantifies a negative impact of the given digital component on other digital components, optimizing a distribution amount applied to the given digital component based on the offset factor and a baseline distribution amount applied to each digital component in each of the multiple different configurations, and transmitting the given digital component to the client device based on the distribution amount for the given digital component, including causing the given digital component to be presented at the client device in the particular configuration.

Optimized digital component analysis system

In various example embodiments, a system and method for determining query aspects for appropriate categories is provided. In example embodiments, query data is collected. The query data indicates a query term and tracked user behavior associated with past queries involving the query term. An aspect demand ratio is calculated for each category of a category tree based on the collected query data. Using at least one processor, one or more lowest categories of the category tree that satisfies a category threshold is determined. The one or more lowest categories are then appended to the query term and stored in a database.

System and method for determining query aspects at appropriate category levels

Systems and methods for providing category aspect information by mining historical data is provided. A table comprising joined data is accessed. The table includes historical data that comprises user behavior data based on actions performed with past queries by users listing data that includes aspect data for each listing and a determined category for each listing. Demand scores based on the joined data are determined. The determined demand scores are used to determine at least one relevant aspect name for a category. A message is transmitted for display on a device of the user that indicates the at least one most relevant aspect name for the category and suggests the user provide an aspect value that corresponds to the at least one most relevant aspect name to be included in a new listing.

In various example embodiments, systems and methods for providing category aspect information by mining historical data is provided. In example embodiments, historical data is compiled. The historical data comprises user behavior data based on actions performed with past queries by users. Listing data is accessed. The listing data includes aspect data for each listing. The historical data is then joined with the listing data and a determined category of the listing to create joined data. Demand scores based on the joined data are determined. The determined demand scores is then sorted to determine at least one relevant aspect name for a category.

Methods, systems, and apparatus, including computer programs encoded on a computer storage medium, for determining short click measures and long click measures on a per-category basis. In one aspect, a method includes determining, for each of a set of categories, a respective first threshold click duration that is a long click duration that defines a period of time for which click durations resulting from selections of resources that belong to the category and that meet the first threshold click duration are classified as long clicks, and a respective second threshold click duration is a short click duration that defines a period of time for which click durations resulting from selections of resources that belong to the category and that do not meet the second threshold click duration are classified as short clicks.

Calibrating click duration according to context

Described herein are methods and systems for promoting item listings that satisfy a query based on the item listings being assigned to certain categories. A query to identify a set of item listings is processed where each item listing associated with an item or service being offered for sale and assigned to a leaf-level category. The scope of the query is determined based on a dictionary of queries or a length of the query when the query is not found in the dictionary. One or more categories are identified based on the scope of the query. A search results page is presented with the item listings from the identified one or more categories.

Method and system for classifying queries to improve relevance of search results

Various embodiments include a method and system for a seller conversion factor to ranking score for presented item listings. In an online marketplace with sellers and item listings, a first sum may be calculated by identifying a first set of item listings associated with the seller. A second sum may also be calculated by identifying a second set of item listings, where the second set is a subset of the first set and are successful item listings. Further, a conversion factor may be calculated from the first and second sums by dividing the second sum by the first sum. The conversion factor may then be used as a component in an item listing quality score for a third set of item listings associated with the seller.

Growing community of inventors

San Jose, CA, United States of America

Qian Zhong