Abstract: A method, system and computer-usable medium are disclosed for automating the generation of an incorrect answer to a question suitable for a multiple choice exam. An input corpus of human-readable text associated with a subject domain is provided to a question generation system, where it is processed to generate a set of question-answer (QA) pairs. The set of QA pairs is then processed with the corpus of input text to extract a set of input keywords and concepts. A concept dependency graph is then used to perform disambiguation operations on the set of input keywords and concepts, and the reference keywords and concepts it contains, to generate a set of distractor words. The resulting set of distractor words is then processed with the set of QA pairs to generate a set of multiple choice question-answers that include various distractor answers.

Abstract: Technology for determining an amount of time to wait to retry requests to a representational state transfer (REST) server system for a REST resource, where the time to wait is always chosen to be a prime number of time units (for example, slots, milliseconds). While currently conventional systems will sometimes use a prime number of time units to wait for a retry request, various embodiments of the present invention will always, and invariably, use a prime number of time units. The REST resource may be, for example, a REST application programming interface (API) that is requested by and delivered to a client system using hypertext transfer protocol (HTTP).

Abstract: A computer-implemented method according to one embodiment includes identifying a first classifier training data element and a second classifier training data element, calculating a similarity metric between the first classifier training data element and the second classifier training data element, and determining a classification for the first classifier training data element and the second classifier training data element, utilizing the similarity metric between the first classifier training data element and the second classifier training data element.

Abstract: An approach is provided in which an information handling system analyzes a set of historical form data corresponding to a set of completed forms. The set of historical form data includes information indicating a historical order at which a set of fields were completed in the set of completed forms. Next, the information handling system determines, based on the set of historical form data, a current order at which to complete a set of incomplete fields included in a current form. In turn, the information handling system displays the current form and a user interface on a display that indicates the current order at which to complete the incomplete fields on the current form.

Abstract: A prescriptive meeting resource recommendation engine automatically learns participant and resource preferences in the context of given meeting input data using natural language features, and automatically recommends all relevant participants and resources (teleconferences, web meetings, links, etc.) to the meeting creator. The engine uses a feature data store to associate historical persons and historical resources with various natural language features, e.g., chargrams. As the host enters text in an invitation template (such as in the title field), the engine extracts current natural language features and computes current participant scores and current resource scores based on the current natural language features. A “forgetfulness” routine is applied to the feature data store to phase out the influence of stale data.

Abstract: Technology for determining an amount of time to wait to retry requests to a representational state transfer (REST) server system for a REST resource, where the time to wait is always chosen to be a prime number of time units (for example, slots, milliseconds). While currently conventional systems will sometimes use a prime number of time units to wait for a retry request, various embodiments of the present invention will always, and invariably, use a prime number of time units. The REST resource may be, for example, a REST application programming interface (API) that is requested by and delivered to a client system using hypertext transfer protocol (HTTP).

Abstract: A method, system, and computer program product are provided for classifying spoken audio content with a cognitive audio classifier by applying a set of distorted audio resources through a set of speech-to-text models STTi (STT1 . . . STTn) to get a set of interference coherence scores based on the transcript for each speech-to-text model STTi, thereby generating a measured baseline Mi (M1 . . . Mn) and a practical baseline Pi (P1 . . . Pn) that is associated with a coherence matrix for the audio effects AEj (AE1 . . . AEk) that were used to generate the distorted audio resources, thereby generating training data for use in training a cognitive audio classifier which classifies input spoken audio content to measure a quality of detected vocabulary elements from the spoken audio content under the set of audio distortion effects for each speech-to-text model STTi.

Abstract: A method, system and computer-usable medium are disclosed for using a context dependency graph to automate the generation of an incorrect answer to a question suitable for a multiple choice exam. A reference corpus is used to generate a concept dependency graph that contains reference keywords and concepts associated with the subject domain of an input corpus. Relationships between the reference keywords and concepts within the concept dependency graph are identified. Once identified, they are used to process a set of input keywords and concepts extracted from the input corpus, and the reference keywords and concepts, to generate a set of distractor words. The resulting set of distractor words is then processed with a set of QA pairs associated with the input corpus to generate a set of multiple choice question-answers that include various distractor answers.

Abstract: Embodiments provide a computer implemented method in a data processing system including a processor and memory storing instructions, which are executed by the processor to cause the processor to implement the method for providing an annotation policy for annotating a corpus including a plurality of electronic documents. The method includes: annotating an occurrence of a first term with a class in an electronic document; recommending a new annotation policy based on at least one annotation for the occurrence of first term; and storing the new annotation policy in a storage device.

Abstract: An approach is provided in which an information handling system analyzes a set of historical form data corresponding to a set of completed forms. The set of historical form data includes information indicating a historical order at which a set of fields were completed in the set of completed forms. Next, the information handling system determines, based on the set of historical form data, a current order at which to complete a set of incomplete fields included in a current form. In turn, the information handling system displays the current form and a user interface on a display that indicates the current order at which to complete the incomplete fields on the current form.

Abstract: A method, system, and computer program product are provided for classifying spoken audio content with a cognitive audio classifier by applying a set of distorted audio resources through a set of speech-to-text models STTi (STT1 . . . STTn) to get a set of interference coherence scores based on the transcript for each speech-to-text model STTi, thereby generating a measured baseline Mi (M1 . . . Mn) and a practical baseline Pi (P1 . . . Pn) that is associated with a coherence matrix for the audio effects AEj (AE1 . . . AEk) that were used to generate the distorted audio resources, thereby generating training data for use in training a cognitive audio classifier which classifies input spoken audio content to measure a quality of detected vocabulary elements from the spoken audio content under the set of audio distortion effects for each speech-to-text model STTi.

Abstract: A prescriptive meeting resource recommendation engine automatically learns participant and resource preferences in the context of given meeting input data using natural language features, and automatically recommends all relevant participants and resources (teleconferences, web meetings, links, etc.) to the meeting creator. The engine uses a feature data store to associate historical persons and historical resources with various natural language features, e.g., chargrams. As the host enters text in an invitation template (such as in the title field), the engine extracts current natural language features and computes current participant scores and current resource scores based on the current natural language features. A “forgetfulness” routine is applied to the feature data store to phase out the influence of stale data.

Abstract: Embodiments provide a computer implemented method in a data processing system including a processor and memory storing instructions, which are executed by the processor to cause the processor to implement the method for providing an annotation policy for annotating a corpus including a plurality of electronic documents. The method includes: annotating an occurrence of a first term with a class in an electronic document; recommending a new annotation policy based on at least one annotation for the occurrence of first term; and storing the new annotation policy in a storage device.

Abstract: A computer-implemented method according to one embodiment includes identifying a first classifier training data element and a second classifier training data element, calculating a similarity metric between the first classifier training data element and the second classifier training data element, and determining a classification for the first classifier training data element and the second classifier training data element, utilizing the similarity metric between the first classifier training data element and the second classifier training data element.

Abstract: In an approach for providing a recommendation for human interaction in an environment, a processor receives information from one or more devices in an area. A processor analyzes the information to identify at least two people and a context of an interaction, wherein a first person has the interaction with a second person. A processor applies a behavioral model to the context and the interaction to identify a recommendation, wherein the behavioral model is a model of a plurality of previous interactions in a plurality of areas, a plurality of contexts associated to the plurality of previous interactions, and a plurality of recommendations associated to the plurality of previous interactions. A processor provides the recommendation to the first person.

Abstract: The disclosed embodiments include a computer-implemented method executed by a cognitive system. In one embodiment, the computer-implemented method includes the step of receiving a first image as part of a visual communication session. The method identifies metadata associated with the first image. The method identifies objects and their surroundings depicted in the first image. The method identifies properties associated with the objects and their surroundings depicted in the first image. The method retrieves information associated with the objects and their surroundings depicted in the first image based on the metadata and the properties associated with the objects and their surroundings. The method interacts with a user based in part on the information associated with the objects and their surroundings depicted in the first image.

Abstract: Accurate and concise summarization of a media production is achieved using cognitive analysis which groups segments of the production into clusters based on extracted features, selects a representative segment for each cluster, and combines the representative segments to form a summary. The production is separated into a video stream, a speech stream and an audio stream, from which the cognitive analysis extracts visual features, textual features, and aural features. The clustering groups segments together whose visual and textual features most closely match. Selection of the representative segments derives a score for each segment based on factors including a distance to a centroid of the cluster, an emotion level, an audio uniqueness, and a video uniqueness. Each of these factors can be weighted, and the weights can be adjusted in accordance with user input. The factors can have initial weights which are based on statistical attributes of historical media productions.

Abstract: Accurate and concise summarization of a media production is achieved using cognitive analysis which groups segments of the production into clusters based on extracted features, selects a representative segment for each cluster, and combines the representative segments to form a summary. The production is separated into a video stream, a speech stream and an audio stream, from which the cognitive analysis extracts visual features, textual features, and aural features. The clustering groups segments together whose visual and textual features most closely match. Selection of the representative segments derives a score for each segment based on factors including a distance to a centroid of the cluster, an emotion level, an audio uniqueness, and a video uniqueness. Each of these factors can be weighted, and the weights can be adjusted in accordance with user input. The factors can have initial weights which are based on statistical attributes of historical media productions.

Abstract: A method, system and computer-usable medium are disclosed for automating the generation of an incorrect answer to a question suitable for a multiple choice exam. An input corpus of human-readable text associated with a subject domain is provided to a question generation system, where it is processed to generate a set of question-answer (QA) pairs. The set of QA pairs is then processed with the corpus of input text to extract a set of input keywords and concepts. A concept dependency graph is then used to perform disambiguation operations on the set of input keywords and concepts, and the reference keywords and concepts it contains, to generate a set of distractor words. The resulting set of distractor words is then processed with the set of QA pairs to generate a set of multiple choice question-answers that include various distractor answers.

Abstract: A method, system and computer-usable medium are disclosed for using a context dependency graph to automate the generation of an incorrect answer to a question suitable for a multiple choice exam. A reference corpus is used to generate a concept dependency graph that contains reference keywords and concepts associated with the subject domain of an input corpus. Relationships between the reference keywords and concepts within the concept dependency graph are identified. Once identified, they are used to process a set of input keywords and concepts extracted from the input corpus, and the reference keywords and concepts, to generate a set of distractor words. The resulting set of distractor words is then processed with a set of QA pairs associated with the input corpus to generate a set of multiple choice question-answers that include various distractor answers.