International Business Machines Corporation

Khoi-Nguyen Dao Tran

Jorge Andres Moros Ortiz

Mukesh Kumar Mohania

Timothy Michael Lynar

John Michael Wagner

Nebula Alam

Kimiko Wilson

Suman Sedai

Shaila Pervin

Simon Harris

Ying Xu

Christopher John Butler

Corey Francis

Jason Jingshi Li

Jey Han Lau

Will Dubyak

Jaysen Ollerenshaw

Jingshi Li

In an approach to generating conversational image representations, one or more computer processors detect one or more utterances by a user, wherein utterances are either textual or acoustic. The one or more computer processors generate one or more image representations of the one or more detected utterances utilizing a generative adversarial network restricted by one or more user privacy parameters, wherein the generative adversarial network is fed with an extracted sentiment, a generated avatar, an identified topic, an extracted location, and one or more user preferences. The one or more computer processors display the generated one or more image representations on one or more devices associated with respective one or more recipients of the one or more utterances.

Autocreation of conversational image representation

A method, computer system, and a computer program product for payload stabilization is provided. The present invention may include, in response to receiving at least one sensor data associated with a suspended payload, detecting an unstable movement in the suspended payload during a transport of the suspended payload. The present invention may also include implementing at least one sail coupled to the suspended payload to stabilize the detected unstable movement of the suspended payload.

Crane payload stabilization

In some examples, a user, either a customer or potential customer of a business, engages in conversations with a virtual assistant (VA) provided by the business. The virtual assistant (VA) is further supported by one or more human assistants (HA), if needed. In embodiments, to facilitate seamless transitions between a VA and a HA, when needed, an intelligent decision maker (IDM) is provided. The IDM receives a user question and a proposed answer to the question from a VA, evaluates the proposed answer in the context of the conversation, and determines if the proposed answer requires further review by an HA. In response to a determination that the proposed answer requires further review, the IDM sends the proposed answer to an HA, and, in response to an indication by the HA, takes further action in the conversation.

Hybrid conversations with human and virtual assistants

Location information associated with a parking area is received. Information about a plurality of vehicles parked in the parking area is determined. A first space located between a first parked vehicle and a first non-movable object located in front of the first parked vehicle is determined. The first space is comprised of a first required space and a first non-required space. A second space located between the first parked vehicle and a second non-movable object located behind the first parked vehicle is determined. The second space is comprised of a second required space and a second non-required space. Responsive to determining that the first non-required space is greater than the second non-required space, the first parked vehicle is moved towards the first space. After moving the first parked vehicle, the first space is equivalent to the first required space.

Optimized vehicle parking

A method for providing adaptive vehicle-proximity guidance may include monitoring transit-condition data of a roadway traversed by a user-vehicle. The method may further include determining, based at least in part on the transit-condition data, a proximity zone. The method may further include generating a first visual representation that corresponds to the proximity zone. The method may further include displaying the first visual representation onto the roadway proximate the proximity zone and updating the first visual representation in response to detected changes in the transit-condition data.

Adaptive vehicle-proximity guidance

Aspects of the invention include a computer-implemented method for quantifying tester sentiment. The computer-implemented method includes receiving a first feedback data from a first tester, where the first feedback data represents feedback generated by the first tester in connection with a first instance of a first software-under-development. The first feedback data is correlated to an aspect of the first instance of the first software-under-development to derive a tester sentiment data. The tester sentiment data is used to determine a tester sentiment rating of the first instance of the software-under-development. Based on the tester sentiment rating, a resource is applied to the first instance of the first software-under-development.

Quantifying tester sentiment during a development process

A method, a computer program product, and a computer system warn workers of a hazardous condition on a worksite. The method includes determining at least one outcome involving a worker and at least one worksite object based on conditions of the worksite. The method includes predicting whether a select one of the at least one outcome presents a hazardous condition to the worker. As a result of predicting that the select outcome presents the hazardous condition, the method includes transmitting a signal to a device associated with the worker. The signal triggers a warning to be received by the worker where the device is in a proximity of the worker.

Predictive warning system

A method for displaying a visual representation corresponding to a carriage capacity may include obtaining passenger data of passengers onboard a carriage of a multi-carriage vehicle. The method may further include predicting, based at least in part on the first passenger data, a number of passengers who will exit the carriage at a destination. The method may further include determining, based at least in part on the predicting, an available capacity for the carriage and generating a visual representation that corresponds to the available capacity. The method may further include displaying the visual representation at the destination of the multi-carriage vehicle.

Distribution management for public transit vehicles

A computer-implemented method includes detecting, using a processor system, a first delay in a conversation system. Initial contextual data is received at a first orchestration layer of the processor system, where the initial contextual data describes a context of a user. The first orchestration layer queries a first data source, based on the initial contextual data. Based on querying the first data source, the first orchestration layer augments the initial contextual data to determine updated contextual data describing the context of the user. Filler content personalized to the user is generated based on the updated contextual data describing the context of the user. The filler content is presented through a user interface during the first delay, responsive to detecting the first delay in the conversation system.

Natural language processor for cognitively generating contextual filler designed to keep users engaged

A caching method, system, and computer program product include compiling a combined action of a social interaction with a sharing of an object via a Uniform Resource Locator (URL) in real-time and from an offline database, extracting semantic information from the combined action, describing the semantic information of the combined action as an embedding of the object in a general object vector space, and creating a caching policy for access to the object based on a prediction of an access time derived from the general object vector space and the description of the combined action.

Growing community of inventors

Southbank, Australia

Khoi-Nguyen Dao Tran