Yahoo! Inc.

Adobe, Inc.

Palo Alto Research Center Inc.

Adchemy, Inc.

Xerox Corporation

Oath Inc.

Excalibur Ip, LLC

Verizon Media Inc.

Ayman Farahat

Tarun Bhatia

Francine R Chen

Richard Edward Chatwin

Eric Theodore Bax

Thorsten Brants

Darshan Vishwanath Kantak

Matthew Robert McEachen

Geoffrey D Nunberg

Dipika Bhatia

Charles R Mathis

David Ayman Shamma

Nicolas B Ducheneaut

Shyam Kapur

Lyndon Kennedy

Christian Holz

Gerry Pesavento

Ricardo Alberto Baeza-Yates

Jing Jin

Simon Kayode Osindero

Prasad Chalasani

Philippe Jean-Paul Golle

Ioannis Kalantidis

Mounia Lalmas

Frank Richard Bentley

Gabriele Tolomei

Pengyuan Wang

Yevgeniy Medynskiy

Nesreen Kamel Abuseree Ahmed

Aleksandra Korolova

Scott Campbell Johnston

Disclosed are embodiments for determining the impact of one or more latent factors on user interaction metrics based at least in part on an impact model. The embodiments identify a value for a user interaction metric, the user interaction metric measuring interaction with content and identify an impact for a latent factor on the user interaction metric, the impact determined based at least in part on a model providing a relationship between the user interaction metric and the latent factor. Additionally, embodiments may involve adjusting an attribute of the electronically provided content based at least in part on the impact of the latent factor on the user interaction metric.

Evaluating coincident interaction indicators

The present teaching relates to providing content supply adjustment. A first dataset associated with user interactions directed to one or more content items placed on a target property and a second dataset associated with user interactions directed to the one or more content items places on a reference property are received for evaluation. Further, a cost of placing the one or more content items on the target property is determined based on the first and second datasets.

Method and system for providing content supply adjustment

Methods and systems for predicting content consumption are provided herein. An application log of a user, comprising a user's application data, and a viewing log of the user, comprising the user's viewing data (e.g., television programs watched by the user), may be evaluated over a time period to construct a model. The model may comprise a correlation between the viewing log and the application log during the time period (e.g., what applications the user interacts with while watching a program). Second application data, regarding application usage of a second user, may be extracted. The model may be applied to the second application data to identify an expected viewing action of the second user (e.g., what program the second user is likely to watch during the time period based upon applications used by the second user). The second user may be provided with content related to the expected viewing action.

Predicting content consumption

Disclosure includes system, method and architecture for selecting supplemental digital content using visual appearance. Digital content that is visually similar, or dissimilar, to digital content requested by a user can he identified and provided for presentation with the requested content. The additional digital content is visually congruent, or visually incongruent, with content requested by a user, such that the additional content is similar, or dissimilar, to the requested content from a visual standpoint. In a presentation of the requested content, the presentation position of each additional content item relative to the presentation position of each requested content can be identified using visual congruence, or visual incongruence.

Method and system for selecting supplemental content using visual appearance

Systems and methods for tracking brand reputation and market share are provided. The system includes one or more devices having a processor and a non-transitory storage medium accessible to the hardware processor. The device is programmed to obtain an awareness index at a plurality of levels at least partially based on the brand data. The device is programmed to obtain a favorability index as a ratio of user numbers based on the brand data. The device is programmed to obtain a branding index by combining the awareness index and the favorability index. The device is programmed to obtain an affinity score for a group of users at least partially based on the brand data and recommend the group of users based on the affinity score to increase the branding index.

Systems and methods for tracking brand reputation and market share

Techniques are disclosed for using social networking resources to evaluate the influence of offline marketing assets. In certain embodiments data posted to a social network is analyzed to identify references to a particular marketer and/or a particular brand. Where such reference is identified, geographic location data associated with the posted data is stored in a database. As this location data is aggregated over a period of time, it can be cross-referenced with location data corresponding to offline marketing assets that belong to the particular marketer and/or that are otherwise associated with the particular brand. Offline marketing assets that geographically correspond to a cluster of social networking activity can be understood as wielding influence within social networks and being capable of generating valuable word-of-mouth.

Evaluating the influence of offline assets using social networking resources

Methods and apparatus for analysis of semi-structured social media are described. A method comprises classifying a plurality of user-generated content entries into a plurality of categories based at least in part on an analysis of respective structured components of at least a subset of the plurality of entries. The method further includes determining, based at least in part on an analysis of additional components of entries of a particular category, a set of representative content elements of entries of the particular category, and generating a report that comprises one or more representative content elements of the particular category.

Mining semi-structured social media

Methods and systems are provided for association of various online elements with emotional states. A set of emotional states may be generated or obtained, and particular online elements may be assigned to particular emotional states of the set, such as with emotional states that are reflected or represented by or in the online elements. Online elements can be of various types, such as, for example, Web sites, blogs, articles, online events, online activities, online messaging, etc. User association or interaction with online emotional state-associated online elements can be used in classifying users into particular emotional states.

Emotional web

The present invention provides techniques for use in audience-related measurements, including hybrid techniques which include use of panel-based audience-related measurement information and census-based audience-related measurement information. Techniques are provided in which panel-based audience-related measurement information is obtained from multiple providing entities, and census-based audience-related measurement information is obtained one or more providing entities. The obtained information is stored in a database. The stored information may be provided to the information-providing entities, and may be used in calibrating or improving the accuracy of audience-related measurements.

Multisided audience-related hybrid measurements

Methods and systems for determining the effect of a host on network latency while delivering online ads are disclosed. Observational test data relating to one or more network latency tests conducted on a plurality of hosts may be obtained. A first group and a second group of hosts may be created, wherein the first group includes the host and the second group doesn't include the host. At least one predetermined metric for each of the first group and the second group is computed based on the test data. A ratio of the at least one predetermined metric for the first group to the at least one predetermined metric for the second group is computed. The host may then be ranked based at least in part on the ratio.

Methods and systems for determining the effect of a host on network latency while delivering online ADS

Techniques for training and using linked event detection systems and transforming source-identified stopwords are provided. A training corpus of source identified stories and a reference language is determined. Optionally, stopwords for source-identified stories are transformed based on statistical analysis of parallel verified and un-verified transformations. Reference language and non-reference language terms are selectively included in source-pair term frequency-inverse story frequency models. Optionally, incremental source-identified term frequency-inverse story frequency models are determined. Selected terms are weighted and similarity metrics determined. Associated source-pair statistics, computed in part from a training corpus, are combined with the values of each similarity metric in the set of similarity metrics to form a similarity vector. Similarity vectors and verified link label information are used to determine a predictive model. Similarity vectors for story pairs are used with the predictive model to determine if the story-pairs are linked. Sources are arranged based on source inter-relationships into a source-hierarchy. Progressively more refined source-pair similarity statistics are also provided. New sources and associated source-pair similarity statistics are added by substituting related source-pair similarity statistics based on the source hierarchy and source characteristics. The source-pair similarity statistics are used to optionally normalize the similarity metrics.

Systems and methods for linked event detection

The present invention is directed towards systems and method for generating a bid amount for one or more keywords for a display of an advertisement associated with a given keyword. The method comprises calculating estimates of one or more advertising metrics for one or more keywords associated with one or more advertisements displayed at one or more positions. Target positions are identified for the one or more keywords at which a return on investment earned by the display of one or more advertisements in response to a query that matches a given keyword exceeds a threshold by a given probability on the basis of the estimates to select a given target position that provides a greatest estimated profit. Bid amounts are generated for the one or more keywords in order to obtain the display of associated advertisements at the given target position.

System and method for generating optimized bids for advertisement keywords

Methods and systems are provided for emotional mapping, such as of online users, based at least in part on online activities of users. Techniques are provided in which information including a set of emotional states is generated or obtained, such as a hierarchical network of emotional states representing a spectrum of human emotions. Information regarding user online activities and content choices of a user is obtained. Based at least in part on this information, the user is classified into an emotional state of the set of emotional states, and advertisements or content may be targeted to the user accordingly.

Emotional mapping

Growing community of inventors

San Francisco, CA, United States of America

Ayman Farahat