International Business Machines Corporation

Technion Research & Development Foundation Limited

Other

Hewlett-packard Company

Technion R&d Foundation Ltd.

Dealigence Inc.

Dataxel Corporation

Oded Shmueli

Luis Alfonso Lastras-Montano

Tin Kam Ho

Rajesh Ramkrishna Bordawekar

Lila Shnaiderman

Royi Ronen

Mukund Raghavachari

Benny Yehezkel

Livio B Soares

Gershon Elber

Ehud Behar

Orit Shaked

Michele M Franceschini

Sriram Kolinjivadi Padmanabhan

Carl Staelin

Michael George Burke

Darryl Greig

Itai Shaked

Ashish Kundu

George Andrei Mihaila

David Konopnicki

Vincent Kulandaisamy

Michael Elad

Marie-Anne Neimat

Lior Leiba

Yehoshua Sagiv

Sharon Krisher

Michael Gertelman

Rojeh Hayek

Tami Tamir

Kurt Shoens

Michael Leybovich

There is provided a method of computing lineage, comprising: managing a dataset of records, each record associated with set(s) of vectors of real numbers that encode an approximation of lineage of the respective record, the set(s) of vectors computed by an encoding process, obtaining result record(s) in response to executing a query on the dataset, computing set(s) of vectors for the result record(s), searching the set(s) of vectors on the records of the dataset to identify a record associated with a subset of vectors that are statistically similar to the set(s) of vectors for the result record(s), and providing a subset of the records corresponding to the identified subset of records, the subset of the records having a likelihood of contributing to the existence of the result record(s) in response to execution of the query.

Database record lineage and vector search

A method of applying proof of lottery to select block forgers in a blockchain, comprising performing the following at a certain one of a plurality of computing nodes connected to a blockchain network: (1) transmitting one or more of a plurality of participation transactions submitted by at least some of the plurality of computing nodes for participating in selection process conducted to select forgers from the plurality of computing nodes to forge blocks to be added to the blockchain; (2) determining a respective forger, during each selection process, by applying a selection function to an outcome of a hash function and a plurality of participation transactions extracted from a first subset of blocks preceding the respective block, the hash function is applied to a second subset of blocks preceding the respective block; and (3) forging the respective block in case the certain computing node is selected as the respective forger.

Proof of lottery (PoL) blockchain

A method comprising: receiving a plurality of pairs of queries associated with a database, wherein the queries in each pair in the plurality of pairs of queries have an identical FROM clause; at a training stage, training a machine learning model on a training set comprising: (i) the plurality of pairs of queries, and (ii) labels associated with containment rates between each of the pairs of queries over the database; and at an inference stage, applying the trained machine learning model to a pair of target queries, to estimate containment rates between the target pair of queries over the database.

Estimating query cardinality

A method, system, and non-transitory compute readable medium for vector representation of a sequence of items, including training a sequence using a first distributed representation, such that a new distributed representation is produced for which vector entries of the new distributed representation are amplified to create dominant dimensions for when the vector entries of each item correspond to a class of an item to be explained and fractionalizing vector entries of each item that do not correspond to the class of the item to be explained such that the dominant dimensions correspond to higher absolute value entries than the fractionalized vector entries in order to emphasize the dominant dimensions.

Vector representation of words in a language

A method comprising: prior to performing a planned i-th data movement in a distributed data system comprising a plurality of storage sites, generating (i) a first map of current locations of data item replicas following an immediately-preceding (i&#x2212;1)-th data movement, (b) a first timestamp associated with a completion of said (i&#x2212;1)-th data movement, and (c) a second map of expected locations of data item replicas following said i-th data movement; implementing said i-th data movement; generating, upon performing said i-th data movement, a second timestamp associated with a completion of said i-th data movement, wherein said second timestamp is indexed as i; and if a failure occurs with respect to at least one of said storage sites during said i-th data movement, determining a completion status of said i-th data movement, based on a comparison between said indexing of said second map and said indexing of said second timestamp.

Handling failures in distributed data system

A method and apparatus are provided for automatically generating and processing first and second concept vector sets extracted, respectively, from a first set of concept sequences and from a second, temporally separated, concept sequences by performing a natural language processing (NLP) analysis of the first concept vector set and second concept vector set to detect changes in the corpus over time by identifying changes for one or more concepts included in the first and/or second set of concept sequences.

Analyzing concepts over time

A method and apparatus are provided for recommending concepts from a first concept set in response to user selection of a first concept Ci by performing a natural language processing (NLP) analysis comparison of vector representations of user concepts contained in written content authored by the user and candidate concepts in a first concept set to determine a similarity measure for each candidate concept, and to select therefrom one or more of the candidate concepts for display as recommended concepts which are related to the user concepts contained in written content authored by the user based on the similarity measure between each candidate concept and each user concept.

Content authoring

A method, system, and non-transitory compute readable medium determining and discerning items with multiple meanings in a sequence of items including producing a distributed representation for each item of the sequence of items including a word vector and a context vector, partitioning the sequence of items into classes, for an item using a representative word vector of each class, calculating a cosine distance between the word vector of said item and the class representative vector, and producing a new sequence of items by modifying the distributed representation in the producing by replacing each occurrence of an item depending on the cosine distance calculated by the calculating.

Determining and discerning items with multiple meanings

A nanosatellite with an illumination element, and an arrangement of nanosatellites in low earth orbit (LEO) arranged to controllably apply their illumination to be visible on the ground. The nanosatellites may be coordinated to provide illumination events visible on the ground at particular locations and particular times. Each nanosatellite has an illumination element that provides a sustained external illumination event being of at least momentary duration and to be visible from the ground. The event forms at least part of a display for viewing from the ground.

Terrestrially observable displays from space

A database system, computer program product, and a method for evaluating aggregates in database systems includes hashing of aggregation keys on a per bucket basis, and depending on a number of hashed tuples per bucket, sorting said tuples. Additionally, depending on the number of hashed tuples per bucket, the bucket is kept without change. Moreover, depending on the number of hashed tuples per bucket, maintaining a secondary hash table for a particular bucket, map tuples to it, aggregate as you map.

Method and system for hybrid sort and hash-based query execution

Systems and methods are provided for moving replicas of a target data object between a plurality of storage nodes that store multiple replicated data objects, by setting new applications to a move mode, waiting for prior applications to exit before moving the target data object, determining that an indication of access to the target data object is done, and responsively copying the replicas of the target data object from first locations among the plurality of data storage nodes to second locations among the plurality of storage nodes.

Moving replicated data in a cloud environment

A method comprising using one or more hardware processors for automatically receiving two or more data objects and two or more data storage addresses, wherein each data storage address directs to one of two or more non-transitory computer readable storage medium locations. The method comprises the action of computing two or more parameter values for each of the data objects. The method comprises the action of grouping the data objects based on the parameter values and two or more predefined rules, thereby producing two or more groups. The method comprises the action of assigning each of the groups to one of the data storage addresses based on the predefined rules. The method comprises the action of storing each of the objects of the groups at one of the non-transitory computer readable storage medium locations based on the assigning.

Optimized data distribution system

A method of processing a query to a graph database using processors. The method comprises providing threads to be executed on processors, each the thread is associated with one of a plurality of unique thread identifiers, providing a graph database having a plurality of graph database nodes and a plurality of graph database edges, each the graph database edge represents a relationship between two of the plurality of graph database nodes, receiving a query tree that defines a tree comprising plurality of query nodes connected by a plurality of query tree edges, and searching at least part of the graph database for a match with the query tree, wherein the searching is executed by the processors, and wherein each processor searches one of a plurality of sub-graphs of the graph database, each the sub-graph is defined by one of the plurality of thread identifiers.

Processing query to graph database

An approach is provided for automatically generating and processing concept vectors by extracting concept sequences from one or more content sources and generating a first concept vector for a first concept by supplying the concept sequences as inputs to a vector learning component, such that the first concept vector comprises information interrelating the first concept to other concepts in the concept sequences which is inferred from the concept sequences.

Growing community of inventors

Nofit, Israel

Oded Shmueli