Yahoo! Inc.

Nilesh Dalvi

Shanmugasundaram Ravikumar

Bo Pang

Andrew Stephen Tomkins

Philip Lewis Bohannon

Vinay Kakade

Sathiya Keerthi Selvaraj

Arun Sundaresan Iyer

Raghu Ramakrishnan

David Mark Ciemiewicz

Arup Kumar Choudhury

Vipul Agarwal

Mani Abrol

Alok S Kirpal

Fei Sha

Disclosed are methods and apparatus for matching sets of text to objects are disclosed. In accordance with one embodiment, a set of text is obtained. For instance, the set of text may include a review. A numerical value is determined for each of a plurality of objects, where the numerical value indicates a likelihood that the corresponding one of the plurality of objects is a subject of the set of text. Each of the plurality of objects has an object type defined by a set of one or more attributes, each of the set of one or more attributes having associated therewith a corresponding set of one or more parameters, wherein the numerical value is determined using the set of text and a value of each of the set of one or more parameters for each of the set of one or more attributes. One of the plurality of objects that is most likely to be the subject of the set of text is identified based upon the numerical value that has been determined for each of the plurality of objects.

Translation model and method for matching reviews to objects

Techniques are provided for the efficient location, processing, and retrieval of local product information derived from web pages generally locatable through form queries submitted to web pages often referred to as the 'deep' or 'hidden' web. In an embodiment, information such as product information and dealer-location information is located on a web page form such as a dealer-locator form. After location of a suitable web page form, editorial wrapping is performed to create an automated information extraction process. Using the automated information extractor, deep-web crawling is performed. A grid-based extraction of individual business records is performed, and matching and ingestion are performed in conjunction with a business listing database. Finally, metadata tags are added to entries in the business listing database. Metadata tags also may be added to entries in other databases.

Method and system for form-filling crawl and associating rich keywords

A computer-implemented method to determine a robust wrapper includes developing a model indicative of the temporal history of a document, such as a web document written in a markup language. Based on the developed model, robustness characteristics are determined for a plurality of different wrappers representing associated paths to the data item in a representation of the document. Based on a result of the determining operation, a result wrapper of the plurality of wrappers is provided. The result wrapper has a desired robustness characteristic.

Robust wrappers for web extraction

A method, apparatus, and computer-readable medium are provided for matching items of user-generated content to entities is provided. Items of user-generated content, such as status updates, are gathered. For each of the items, a machine determines a degree to which the item is associated with an entity. In one aspect, items are matched to an entity by matching the content of the items to attributes of the entity. In another aspect, items are matched to an entity by predicting attributes of an author of the items and determining a distance between the predicted attributes of the author and the attributes of the entity. The distance may be a physical distance between locations of the entity and user or a contextual distance between categories for the entity and posts by the author. Items matched to the entity may be displayed on an interface concurrently with information about the entity.

Matching items of user-generated content to entities

A method is provided to associate reviews that have unknown correspondences to tangible entities to structured objects that have known correspondences to tangible entities comprising: transforming a respective review and text from a respective structured object to a collection of words that intersect the respective review and text from the respective structured object; determining a measure of a likelihood of a match as a function of respective probabilities of occurrences of respective words of such intersecting collection within generic review text and respective probabilities of occurrences of respective words of such intersecting collection within structured object text.

Matching reviews to objects using a language model

Growing community of inventors

Menlo Park, CA, United States of America

Nilesh Dalvi