Google Inc.

Radu Jurca

Matsvei Zhdanovich

Luuk Van Dijk

Steffen Gerd Meschkat

Bruce Winston Murphy

David Lecomte

Anja Hauth

Ivan Zauharodneu

Pascal Fleury

Ashutosh Kulshreshtha

Luca De Alfaro

Chiu Wah Kelvin So

Mahesh Keralapura Manjunatha

Shen Wang

Xi Xiong

Daniel Svonava

David Fan

Yao Liu

Andrei Dragus

Jakub Ocwieja

Vinay Vyas Vemuri

Md Mahbubul Hasan

Muruo Liu

Methods, systems, and apparatus, including computer programs encoded on a computer storage medium, for a content platform that receives a request to provide a digital component. The request includes information about users to which the digital component is directed. Futurized queries are obtained from serving logs of a serving system that is configured to execute an existing digital component using serving code that directs digital content to the set of users. The futurized queries are loaded as data structures in memory of a forecasting system. The system uses an instruction set derived from the serving code to determine that similarity between the particular futurized query and the request exceeds a threshold similarity. The system then generates a forecast output as a response to the request based on futurized queries that exceed the threshold similarity. The forecast output includes data describing future provision of the digital component.

Modeling distribution of digital components over a network

Systems, methods, and computer-readable media for interpreting geographical search queries are provided. Query templates for geographical search queries are generated from search logs of previously executed geographical search queries. Unique queries are extracted from the search logs and interpretations for these queries are obtained. Based on these interpretations, query templates having an arrangement of term types are extracted. Additionally, probability distributions for the query templates for a context, such as a locale, language, client type, etc., are determined.

Systems, methods, and computer-readable media for interpreting geographical search queries

A spelling system derives a language model for a particular domain of structured data, the language model enabling determinations of alternative spellings of queries or other strings of text from that domain. More specifically, the spelling system calculates (a) probabilities that the various query entity types&#x2014;such as STREET, CITY, or STATE for queries in the geographical domain&#x2014;are arranged in each of the various possible orders, and (b) probabilities that an arbitrary query references given particular ones of the entities, such as the street 'El Camino Real.' Based on the calculated probabilities, the spelling system generates a language model that has associated scores (e.g., probabilities) for each of a set of probable entity name orderings, where the total number of entity name orderings is substantially less than the number of all possible orderings. The language model can be applied to determine probabilities of arbitrary queries, and thus to suggest alternative queries more likely to represent what a user intended.

Training a probabilistic spelling checker from structured data

Methods, systems, and apparatus, including computer programs encoded on a computer storage medium, for identifying geographic features. In one aspect, a method includes receiving a query. Geographic features are identified, each geographic feature being associated with one or more names, each geographic feature being associated with at least one name that includes the query. A feature-query score is computed for each geographic feature, including: for each name of the geographic feature that includes the query, identifying a computed feature-name score, wherein the feature-name score is computed based on a count of a number of occurrences of the name in a query log, wherein each occurrence is attributed to the feature; and computing the feature-query score based on the identified feature-name scores. The geographic features are ranked according to the feature-query scores.

Identifying geographic features from query prefixes

A system and method for updating and correcting facts that receives proposed values for facts from users and determines a consensus value for the fact based on a distribution of the probabilities that the proposed values are the correct value for the fact wherein the probability that the proposed value is the correct value for the fact is based on a score indicative of the probability that the user proposing the value proposes accurate values.

Crowdsensus: deriving consensus information from statements made by a crowd of users

Methods, systems, and apparatus, including computer programs encoded on a computer storage medium, for provided expanded geographical query suggestions. In one aspect, a method includes receiving one or more characters input into a user interface. A node in a graph is identified based on the one or more received characters, in which the node represents an entity having a name whose prefix contains the one or more received characters. A candidate node is identified in the graph that is related to the identified node, wherein the candidate node represents an entity having a name whose prefix does not contain the one or more received characters. The name of the entity represented by the candidate node is provided as an expanded query suggestion.

Query suggestions

In one embodiment, a map server provides online maps with advertisements corresponding to map features included in the online maps. The map server holds an auction to determine which advertisers can display their associated advertisements in an online map. In one embodiment, the map server selects which advertisers with collocated advertisements in the online map can display their advertisements and calculates a price paid by the selected advertisers for the display of the advertisements.

Allocating and pricing geo-located advertisements for display on maps

Growing community of inventors

Adliswil, Switzerland

Radu Jurca