International Business Machines Corporation

Adobe, Inc.

Irgelkha Mejia

Michele Saad

Howard Neil Anglin

Yvonne M Young

Nicholas J Ruegger

Ronald Eduardo Oribio

Denise Bell

Robert W Burke, Jr

Pamela F Geiger

Oriana Jeannette Love

Leah A Lawrence

Robert Burke

Eunyee Koh

Jose Ignacio Echevarria Vallespi

Sushma Bharat Patel

Ajay Jain

Borna Safabakhsh

Lauren Dest

Dustin Ground

Anderson Edward Wolfe

Catherine Chiodo

Michelle Mee-June Lee

Joseph Arcuri Anthony

Jacob Benjamin Hanson-regalado

Andrew Thomson

Arjun Athreya

Anna Hammond

Nicholas Joseph Ruegger

Yvonne Marie Young

In accordance with the described techniques, a background generation system receives one or more images depicting an object, and textual information describing the object. A generative text model is employed to generate a prompt based on the one or more images and the textual information. Further, a generative image model is employed to generate an output image. To do so, the generative image model generates a background image based on the prompt, and the object is incorporated into the background image. Using a visual saliency model, the background generation system determines a visual saliency defining a degree of fixation on the object within the output image. The background generation system outputs the output image based on the visual saliency meeting a threshold.

Saliency-based background generation

In implementation of automated colorimetry techniques supporting color classification, a computing device implements a color coordination system to receive a digital image depicting a person. The color coordination system then identifies a color classification for the person based on the digital image, the color classification associated with a color recommendation that is represented as a color distribution. The color coordination system identifies an item associated with a color of the color recommendation by identifying a point of the color distribution associated with a color of the item that is within a threshold distance from a point associated with the color of the color recommendation. Then, the color coordination system displays a recommendation that includes the item in a user interface.

Automated colorimetry techniques supporting color classification

A search system employs arrival times with associated confidence scores as search facets for identifying items. The search system identifies a plurality of items based on search input. An arrival time and associated confidence score are determined for each item from the plurality of items. Search results are provided for the plurality of items in response to the search input. The search results are provided based at least in part on the arrival times and associated confidence scores for the plurality of items.

Delivery-related search and analytics

Systems and methods use machine learning models with content editing tools to prevent or mitigate inadvertent disclosure and dissemination of sensitive data. Entities associated with private information are identified by applying a trained machine learning model to a set of unstructured text data received via an input field of an interface. A privacy score is computed for the text data by identifying connections between the entities, the connections between the entities contributing to the privacy score according to a cumulative privacy risk, the privacy score indicating potential exposure of the private information. The interface is updated to include an indicator distinguishing a target portion of the set of unstructured text data within the input field from other portions of the set of unstructured text data within the input field, wherein a modification to the target portion changes the potential exposure of the private information indicated by the privacy score.

Machine learning modeling for protection against online disclosure of sensitive data

Methods, systems, and non-transitory computer readable media are disclosed for utilizing machine learning models to extract digital signals from low-results web queries and generate item demand deficiency predictions for digital item lists corresponding to websites. In one or more embodiments, the deficiency identification system identifies a low-results query submitted by client devices navigating a website. The deficiency identification system generates features for the low-results query and the digital item list to generate a deficiency prediction relative to demand indicated by the low-results query. In some embodiments, the deficiency identification system utilizes a deficiency prediction model to process the extracted signals and generate a deficiency confidence score corresponding to the low-results query. Based on the deficiency confidence score, the deficiency identification systemd can generate and provide demand notifications via one or more graphical user interfaces.

Utilizing machine learning models to process low-results web queries and generate web item deficiency predictions and corresponding user interfaces

Certain aspects and features of this disclosure relate to chromatic undertone detection. For example, a method involves receiving an image file and producing, using a color warmth classifier, an image warmth profile from the image file. The method further involves applying a surface-image-trained machine-learning model to the image warmth profile to produce an inferred undertone value for the image file. The method further involves comparing, using a recommendation module, and the inferred undertone value, an image color value to a plurality of pre-existing color values corresponding to a database of production images, and causing, in response to the comparing, interactive content including the at least one production image selection from the database of production images to be provided on a recipient device.

Chromatic undertone detection

The present technology provides for facilitating removal of undesired search results. In one embodiment, a search request including a search term(s) to use for performing a search is obtained. Thereafter, a search query is generated to execute the search. The search query includes the search terms and a removal parameter indicating a particular search result to exclude from search results returned in response to the search request. A set of search results are provided for display via a user device. Such a set of search results can be identified based on execution of the search query and exclude the particular search result.

Facilitating search result removal

Some example embodiments include a method for recharging a number of battery electric vehicles. The method include receiving (by a control module configured to control an electrical grid system that include a number of recharging stations that are configured to recharge the number of battery electric vehicles and from the number of battery electric vehicles) usage data that comprises a current charge level, a current location, and a planned itinerary that includes a destination. The method includes determining anticipated electrical loads in the number of sectors of the electrical grid system based on the usage data of the number of battery electric vehicles. The method also includes redistributing the electrical supply on the electrical grid system to at least one recharging station of the number of recharging stations based on the anticipated electrical loads, prior to actual usage defined by the usage data by the number of battery electric vehicles.

Recharging of battery electric vehicles on a smart electrical grid system

A method for recharging an electric vehicle includes receiving, by a control module and from the vehicle in transit along a route having a number of stops for passenger pickup, a current charge level and a current location. The stops include recharging stations for recharging the vehicle. The method includes receiving from a next stop, an anticipated stop time at the next stop for the vehicle. The method includes determining a power output to be supplied to the vehicle by a recharging station at the next stop based on the current charge level. The power output comprises an amount of power to be supplied at the next stop, the amount of power for satisfying a minimum amount of charge to enable the vehicle to arrive at a subsequent stop after the next stop. The method includes transmitting, to the recharging station at the next stop, the power output.

A computer program product for recharging a number of battery electric vehicles includes computer usable program code. The computer usable program code is configured to receive, from the number of battery electric vehicles that are to recharge at a number of recharging stations, usage data. The usage data includes a current charge level, a current location, and a planned itinerary that includes a destination. The computer usable program code is configured to determine anticipated electrical loads in the number of sectors of the electrical grid system based on the usage data of the number of battery electric vehicles. The computer usable program code is configured to redistribute the electrical supply on the electrical grid system to at least one recharging station of the number of recharging stations based on the anticipated electrical loads, prior to actual usage defined by the usage data by the number of battery electrical vehicles.

Provided are techniques for smart posting with data analytics. A message is received before the message is posted to a social media service. The message is analyzed using data analytics to obtain analysis results. The obtained analysis results are compared to similar analysis results stored for at least one pre-existing message. For one or more correlations between the message and at least one pre-existing message, one or more contributing terms that have semantic meaning within a context of the social media service are determined. Based on the one or more contributing terms, one or more suggestions for improving the message are generated. The message is modified based on the one or more suggestions.

Smart posting with data analytics and semantic analysis to improve a message posted to a social media service

The computer manages media content by receiving, at a first memory storage location, a first file having one or more associated properties. The computer identifies a second file, at a second memory storage location, having one or more associated properties that are substantially similar to the first file. The computer compares the one or more properties of the first file to the one or more properties of the second file. The computer determines whether to replace the first file with the second file, based on the comparison of the one or more properties of the first file to the one or more properties of the second file. Responsive to determining to replace the first file with the second file, the computer stores at the first memory storage location a link to the second memory storage location.

Media file replacement

Some example embodiments include a method for recharging a number of battery electric vehicles. The method include receiving (by a control module configured to control an electrical grid system that include a number of recharging stations that are configured to recharge the number of battery electric vehicles and from the number of battery electric vehicles) usage data that comprises a current charge level, a current location, and a planned itinerary that includes a destination. The method includes determining anticipated electrical loads in the number of sectors of the electrical grid system based on the usage data of the number of battery electric vehicles. The method also includes redistributing the electrical supply on the electrical grid system to at least one recharging station of the number of recharging stations based on the anticipated electrical loads, prior to actual usage defined by the usage data by the number of battery electrical vehicles.

Growing community of inventors

Austin, TX, United States of America

Irgelkha Mejia