International Business Machines Corporation

Yahoo Assets LLC

Roie Melamed

Aharon Abadi

Gennady Laventman

Yoav Tock

Yair Koren

Guy Laden

Yohay Kaplan

Natalie Shapira

Aidan Shribman

Benjamin Michael Parees

Peter Bak

Moria Abadi

Avi Yaeli

Dean Har'el Lorenz

Alexey Roytman

Gregory Chockler

Eli Packer

Yuval Nardi

Matan Yitshak Ninio

Yishai Shimoni

Eli Luboshitz

Ehud Karavani

Sivan Ravid

Ashraf Haib

Hisham Yasin

Lior Ness

One or more computing devices, systems, and/or methods for content recommendations using historical future data are provided. A model serving delay time is computed as an average of training delays of events. A historical data time interval is determined based upon the model serving delay time. A model is trained for predicting user content preferences using historic user distribution data and historic content distribution data associated with the historic data time interval. The model is utilized to generate and provide content recommendations to users.

Content recommendations based upon historical future data

In some examples, a system for implementing a protocol for data transmission can include a processor to transmit a public key or an encrypted name to a ledger of a source computing device and receive a scoring code, training data, and data definitions from the source computing device. The processor can also transmit predicted labels encrypted with the public key to the source computing device, the predicted labels based on the training data and the data definitions, and receive labeled test data from the source computing device. The processor can also transmit decrypted predicted labels to the source computing device, receive a list of evaluation results from a set of remote computing devices, transmit scoring results based on the list of evaluation results to the set of remote computing devices, and execute an instruction based on a techniques corresponding to a highest score from the scoring results.

Distributed anonymous scoring technique

There is provided a computer implemented method for detection of likelihood of malignancy in an anatomical image of a patient for treatment planning, comprising: receiving an anatomical image, feeding the anatomical image into a global component of a model trained to output a global classification label, feeding the anatomical image into a local component of the model trained to output a localized boundary, feeding the anatomical image patch-wise into a patch component of the model trained to output a patch level classification label, extracting a respective set of regions of interest (ROIs) from each one of the components, each ROI indicative of a region of the anatomical image likely to include an indication of malignancy, aggregating the ROIs from each one of the components into an aggregated set of ROIs, and feeding the aggregated set of ROIs into an output component that outputs an indication of likelihood of malignancy.

Systems and methods for detection likelihood of malignancy in a medical image

A transaction manager can obtain a first lease that dedicates a set of virtual resources to the transaction manager for a first time interval. The transaction manager can send a commit request to one or more resource managers regarding a first transaction. The transaction manager can store respective responses from each respective resource manager. The transaction manager can determine if each response is affirmative, and, if each response is affirmative, the transaction manager can complete the first transaction.

Lease-based management for atomic commit protocols

Switching from primary to backup data storage by preparing a backup copy of multiple data sets, where, prior to the preparing, the backup copy is updated in accordance with a backup protocol specifying synchronously updating the backup copy to reflect changes made to one type of data stored in a primary copy of the data sets, and asynchronously updating the backup copy to reflect changes made to another type of data stored in the primary copy, and where the preparing includes identifying any inconsistency in any interdependent data in the data sets of the backup copy in accordance with a predefined schema of interdependent data in the data sets, and correcting any identified inconsistency in the data sets of the backup copy in accordance with a predefined inconsistency correction protocol, and causing the backup copy to be used in place of the primary copy for directly servicing data transactions.

Data consistency when switching from primary to backup data storage

One or more location updates from one or more user devices can be associated with one or more grid elements dividing a geographic area. Each location update corresponding to a set of user device data can be stored in a hashtable according to grid element. The hashtable can be resized based on the number of grid elements containing one or more sets of user device data. A set of grid elements can be determined to overlap a geofence, and zero or more sets of user device data can be retrieved from the hashtable corresponding to zero or more user devices located within the geofence.

Grid-based geofence data indexing

A document management system including a document manager connected to a document storage, and a backup coordinator connected to the document manager and adapted to: continuously receive a plurality of notifications from the document manager, each including information describing a change in a document stored in the document manager's document storage; and for each notification in the plurality of notifications: select a backup agent from a plurality of backup agents connected to the backup coordinator; send a backup request to the backup agent including the information, for the purpose of updating a copy of the document stored in a backup storage connected to the backup agent; wait for an acknowledgement message from the backup agent; and send another backup request to another backup agent selected from the plurality of backup agents upon a failure to receive the acknowledgement message within an identified amount of time after sending the backup request.

Managing big data on document based NoSQL databases

A system for producing secure data management software, comprising at least one hardware processor adapted to: receive a plurality of data patterns, each comprising at least one data field identifier selected from a set of protected data field identifiers of at least one data repository, at least one output target, and an access instruction; identify in a plurality of computer instructions of the data management software one or more forbidden output instructions by matching one or more reaching definitions of some of the plurality of computer instructions with one or more of the plurality of data patterns; and remove the one or more forbidden output instructions from the plurality of computer instructions.

System and method for producing secure data management software

A computerized method of improving position measurement of an indoor positioning system (IPS), comprising: gathering a plurality of position samples of a mobile device measured by an indoor wireless device, each of the position samples is indicative of a measured position and a measurement time; receiving a current position sample of the mobile device; for each of the plurality of position samples and the current position sample, determining a time weight, the time weight is higher for samples measured at a later time; for each of the plurality of position samples and the current position sample, determining a signal weight, the signal weight is higher for measured with stronger signal; and estimating a current position of the mobile device by weighting at least some of the position samples with the respective weights.

Methods and systems for improving accuracy of indoor positioning

A computerized method of improving position measurement of an indoor positioning system (IPS), comprising: gathering a plurality of position samples of a mobile device measured by an indoor wireless device, each of the position samples is indicative of a measured position and a measurement time; estimating a movement direction of the mobile device from the plurality of position samples; receiving a current position sample of the mobile device; estimating a temp movement direction of the mobile device from at least one of the plurality of position samples and the current position sample; and when the temp movement direction is identified as false, marking the current position sample as noise.

Methods and systems for noise removal in indoor positioning

Automatic determination of presence-related roles. Presence data of a plurality of users in a space comprised of one or more zones is received. For each user of the plurality of users a value of a target metric in a zone is computed based on the presence data, the target metric defining a measure of user presence. A presence-related role is assigned to a user with respect to the zone, the presence-related role being comprised in a set of one or more presence-related roles describing categories of spatiotemporal occupancy relation between the user and the zone, wherein said assigning comprises applying a mapping function on the value, the mapping function maps between values of the target metric and the one or more presence-related roles, the mapping function is affected by analytics of the presence data.

Growing community of inventors

Haifa, Israel

Roie Melamed