International Business Machines Corporation

Oath Inc.

Yahoo! Inc.

Excalibur Ip, LLC

Yahoo Assets LLC

Oleg Rokhlenko

Eli Arbel

Idan Szpektor

David Carmel

Ronny Lempel

Nadav Golbandi

Oren Shlomo Somekh

Dan S Pelleg

Karen Frida Yorav

Shirin Oskooi

Yuval David Pinter

Somesh Jain

Archit Shrivastava

Zohar Karnin

Cynthia Rae Eisner

Shahar Golan

Cindy Eisner

Edward Bornikov

Oren Anava

Adi Omari

Methods and devices for assessing the quality of user-generated content are described. In one embodiment, a method is disclosed for measuring the quality of a user-generated answer to a question by combining various factors, including question-answer surface word vector similarity, question-answer explicit semantic analysis vector similarity, answer-answer explicit sematic analysis vector similarity, query performance predictor, sentiment analysis, textual analysis of the answer, and reputation of the answerer. The method uses a learning procedure to determine the best algorithm for measuring the overall quality of the answer based on these factors.

Quality-based scoring and inhibiting of user-generated content

Method, system, and programs for estimating interests of a plurality of users with respect to a new piece of information are disclosed. In one example, historical interests of the plurality of users are obtained with respect to one or more existing pieces of information. One or more users are selected from the plurality of users. Historical interests of the one or more users can minimize an objective function over the plurality of users. Interests of the one or more users are obtained with respect to the new piece of information. Estimated interests of the plurality of users are generated with respect to the new piece of information based on the obtained interests of the one or more users.

Method and system for cold-start item recommendation

Briefly, example methods, apparatuses, and/or articles of manufacture are disclosed that may be implemented, in whole or in part, using one or more computing devices to facilitate and/or support one or more operations and/or techniques for in part, to facilitate and/or support one or more operations and/or techniques for ranking answers for on-line community questions.

Ranking answers for on-line community questions

A content item presented to a user may involve two or more topics, such as news articles mentioning athletes or actors. Users may enjoy expressing opinions about a comparison of the topics, such as personal preferences or an evaluation of talent. Presented herein are techniques for automatically generating, for the content item, a poll among the topics of the content item. The respective comparisons of a comparison set are evaluated for the topics of the content item, and a comparison relevance score is computed identifying the relevance of the comparison of the topics. The comparison having the highest relevance for the topics is selected, and a comparison question is formulated and presented to the user with the content item. Results may be tabulated and presented for the topics of the content item. These techniques facilitate user expression about the content item without depending upon user-authored polls for each content item.

Automatically generated comparison polls

A method for estimating model parameters. The method comprises receiving a data set related to a plurality of users and associated content, partitioning the data set into a plurality of sub data sets in accordance with the users so that data associated with each user are not partitioned into more than one sub data set, storing each of the sub data sets in a separate one of a plurality of user data storages, each of said data storages being coupled with a separate one of a plurality of estimators, storing content associated with the plurality of users in a content storage, where the content storage is coupled to the plurality of estimators so that the content in the content storage is shared by the estimators, and estimating, asynchronously by each estimator, one or more parameters associated with a model based on data from one of the sub data sets.

Efficient and fault-tolerant distributed algorithm for learning latent factor models through matrix factorization

A method for adjusting one or more parameters associated with a model. The method comprises obtaining, from a first source, first information related to activity of a user. The method further comprises adjusting one or more parameters associated with a model based on the first information collected within a first length of time, and obtaining, from a second source, second information related to activity of the user. The method further comprises adjusting the one or more parameters associated with the model based on the second information collected within a second length of time and a measure indicative of performance of the model, wherein the second length of time is larger than the first length of time.

Almost online large scale collaborative filtering based recommendation system

A circuit design, responsive the input signals, may be obtained and processed. The circuit design may define connections between combinational elements, memory elements, and input signals. Identification of cut-off points may be performed with respect to predetermined combinational logic input signals. The cut-off points may be connections whose values are not dependant on the value of the predetermined combinational logic input signals. An approximated circuit design may be synthesized by relaxing the logic associated with the cut-off points. Based on the approximated circuit design, processing may be performed. In some exemplary embodiments, a clock gating function of a memory element may be determined by approximating the circuit design with respect to the output signal of the memory element. The clock gating function may be determined based on the approximated circuit design and introduced to the circuit design, with or without additional refinement.

Circuit design approximation

A method, apparatus and computer program product for modifying a circuit design. The method comprising: obtaining a design of a circuit, the design comprising at least a first memory element and a second memory element. The method further comprising selecting the second memory element to be a dominant memory element over the first memory element. The method further comprising modifying the design of the circuit by replacing usage of an output signal of the first memory element with usage of an output signal of the dominant memory element in one or more cycles in which values of the output signals of the first memory element and the dominant memory element are equal. Whereby a reduction in observabiltiy of the first memory element in the design is achieved.

Reducing observability of memory elements in circuits

Clock gating circuit is determined by transforming a clock gating opportunity function to a non-Boolean function and constraining inputs of the non-Boolean function. The non-Boolean function may be a ternary function. Constraining the inputs may be achieved by introducing control variables and a cardinality constraint associated with their values. The non-Boolean function may be utilized to approximate universal quantification of an input assigned with a non-Boolean value, such as 'don't care' value. The non-Boolean function may be utilized to provide an ALL SAT solution of a Boolean function using a SAT solver.

SAT-based synthesis of a clock gating function

A method for reducing a depth of binary decision diagram includes identifying, with a processing device, one or more less probable positive paths within the binary decision diagram, the less probable positive paths within the binary decision diagram include a number of decision nodes from a starting variable to a positive, binary logic '1' terminal node within the binary decision diagram, and the identified less probable positive paths are eliminated. Identifying the less probable positive paths within the binary decision diagram includes: assigning &#x192; to the binary decision diagram, assigning K to a depth threshold of the binary decision diagram, and constructing a second binary decision diagram g by logically ORing together positive paths in &#x192; that have a length above K. Eliminating the identified less probable positive paths in &#x192; comprises obtaining an approximated function &#x192;', by conjuncting &#x192; and the negation of g, such that &#x192;&#x2032;&#x2261;&#x192;^.

Approximation of a clock gating function via BDD path elimination

An initial clock gating function is introduced to an original circuit design. Using abstraction-refinement, the initial clock gating function is modified such that the gated circuit design is equivalent to the original circuit design. A model checker, such as a SAT solver, may be utilized to determine equivalency of two circuit designs. A counter-example may be determined by the model checker to negate equivalency. The counter-example may be utilized to modify the initial clock gating function to determine a modified gated circuit design that is equivalent to the original circuit design.

Growing community of inventors

Haifa, Israel

Oleg Rokhlenko