Abstract: A method, system and computer-usable medium are disclosed for identifying unchecked criteria within a form. Natural Language Processing (NLP) is applied to unstructured data within a target form to identify elements of a form structure. Analytic analysis is then applied to the resulting form structure elements to identify a hierarchical structure and associated element placement. Implicit selections within the form are then identified by applying rules based upon other selections and their orientation to anchor terms to determine the completeness of the form, based upon aggregation of form elements. The form structure elements and the hierarchy metadata are then processed logically reassemble the form's hierarchy in flattened forms for multi-layer, sub-element anaphora resolution.

Abstract: A method, system and computer-usable medium are disclosed for identifying unchecked criteria within a form. Natural Language Processing (NLP) is applied to unstructured data within a target form to identify elements of a form structure. Analytic analysis is then applied to the resulting form structure elements to identify a hierarchical structure and associated element placement. Implicit selections within the form are then identified by applying rules based upon other selections and their orientation to anchor terms to determine the completeness of the form, based upon aggregation of form elements. The form structure elements and the hierarchy metadata are then processed logically reassemble the form's hierarchy in flattened forms for multi-layer, sub-element anaphora resolution.

Abstract: A method, system and a computer program product are provided for classifying elements in a ground truth training set by iteratively assigning machine-annotated training set elements to clusters which are analyzed to identify a prioritized cluster containing one or more elements which are frequently misclassified and display machine-annotated training set elements associated with the first prioritized cluster along with a warning that the first prioritized cluster contains one or more elements which are frequently misclassified to solicit verification or correction feedback from a human subject matter expert (SME) for inclusion in an accepted training set.

Abstract: Mechanisms are provided in which a first knowledge graph, comprising nodes representing entities and edges between nodes indicative of a relationship between the entities, is received. The mechanisms identify a candidate missing edge connecting a node of the first knowledge graph to another node not present in the first knowledge graph and evaluate the candidate missing edge to determine if the candidate missing edge should be added to the first knowledge graph. The mechanisms expand the first knowledge graph to include the candidate missing edge connecting the node to a newly added node that is newly added to the first knowledge graph, to thereby generate an expanded knowledge graph, in response to the evaluation indicating that the candidate missing edge should be added to the first knowledge graph. The mechanisms then perform an operation on the expanded knowledge graph to generate a knowledge output.

Abstract: A method, system and a computer program product are provided for verifying ground truth data by iteratively assigning machine-annotated training set examples to clusters which are prioritized based on verification scores to identify and display one or more prioritized review candidate training set examples grouped in a prioritized cluster in order to solicit verification or correction feedback from a human subject matter expert for inclusion in an accepted training set.

Abstract: A method, system and a computer program product are provided for verifying ground truth data by iteratively clustering machine-annotated training set examples with validation set examples to identify and display one or more prioritized review candidate training set examples grouped with validation set examples meeting a predetermined misclassification criteria in order to solicit verification or correction feedback from a human subject matter expert for inclusion in an accepted training set.

Abstract: A web service packages an XForm type form containing interactive fields used by the web service, in a RSS/ATOM feed item. The web service sends the RSS/ATOM feed item to a computerized client, wherein the computerized client is a subscriber to the web service's RSS/ATOM feed and the computerized client is in network communication with a server hosting the web service. The computerized client displays the form for use by a user. Responsive to the user of the computerized client manipulating interactive fields in the form, the computerized client packages the manipulated form in a return RSS/ATOM feed item. The computerized client sends the return RSS/ATOM feed item to the server hosting the web service, wherein the server is a subscriber to the computerized client's return RSS/ATOM feed.

Abstract: An embodiment of the present executes a sequence of a plurality of application programming interfaces. Request objects are generated for the application programming interfaces of the sequence. Each request object is generated in accordance with a common object model. Execution of each application programming interface is enabled in an order of the sequence to process a corresponding request object and generate a corresponding response object containing results. Each response object is generated in accordance with the common object model and includes information from the corresponding request object aggregated with the results. Further, each request object for a succeeding application programming interface in the sequence includes the aggregated information from a response object of a prior executed application programming interface in the sequence.

Abstract: A mechanism is provided for optimizing parsing of grammatical prose in a multi-prose electronic document. A received multi-prose electronic document is partitioned into a set of text segments of homogenous prose. A set of predefined Natural Language (NL) parsers is executed on each text segment in order to discern and measure a quality of parses by each NL parser throughout the multi-prose electronic document. A cross-parse analysis of the multi-prose electronic document and an aggregation of adjacent clauses or sentences of poor quality parses are performed. For each text segment, aggregation patterns yielding a best semantic role match are identified from one of the set of predefined NL parsers. Then, a parsed electronic document is generated using the set of predefined NL parsers such that each text segment is parsed by a NL parser that has the best semantic role match to that text segment.

Abstract: Candidate answers are generated by a question-answering system in response to a question from a user. One or more generated candidate answers are compared to previous question-answer sets. The previous question-answer sets are indexed and stored in a database which includes a knowledge graph on the previous questions-answer sets. A previous question-answer set is identified as correlating with a first generated candidate answers. The previous question-answer set is identified using relationships of the database. The first generated candidate answer is restructured using the question as a statement which is embedded with the first generated candidate answer. The restructured generated candidate answer which correlates with the previous question-answer set is supplemented with content from the previous question-answer set.

Abstract: A recommendation engine for works of literature uses patterns of flow and element similarities for scoring a first user-rated work of literature against one or more recommendation candidate works of literature. Cluster models are created using meta-data modeling the works of literature, the meta-data having literary element categories and instances within each category. Each instance is described by an index value (position in the literature) and significance value (e.g. weight or significance). Cluster finding process(es) invoked for each instance in each category find Similarity Concept clusters and Consistency Trend clusters, which are recorded into the cluster models representing each work of literature. The cluster model can be printed or displayed so that a user can visually understand the ebb and flow of each literary element in the literature, and may be digitally compared to other cluster models of other works of literature for potential recommendation to a user.

Abstract: A technique for transforming questions for a question answering (QA) system includes clustering received questions by contextual accordance into a question set. A base lexical answer type (LAT) for the question set is determined. LATs of discordant questions in the question set are transformed to the base LAT. Answers for the questions in the question set are aggregated. The aggregated answers are then displayed.

Abstract: An approach is provided to correct natural language processing (NLP) annotators. The approach operates by receiving a set of supporting text noted by a user in response to the user identifying an error to a user question in a question answering (QA) system. The set of supporting text includes one or more text passages from which a correct answer should have been generated by the QA system. The QA system generates one or more scored candidate corrections with each of the scored candidate corrections is based on the identified error and the set of supporting text. The user can then select one or more of the scored candidate corrections as a confirmed correction to the error. The confirmed corrections are then applied to a corpus that is utilized by the QA system when answering questions.

Abstract: Visualizable subject matter contained within data is detected. A context in which the visualizable subject matter appears is determined. A visualization engine of a plurality of visualization engines is selected based, at least in part, on the context in which the visualizable subject matter appears. The visualization engine is configured to generate a visualization of the visualizable subject matter.

Abstract: Candidate answers are generated by a question-answering system in response to a question from a user. One or more generated candidate answers are compared to previous question-answer sets. The previous question-answer sets are indexed and stored in a database which includes a knowledge graph on the previous questions-answer sets. A previous question-answer set is identified as correlating with a first generated candidate answers. The previous question-answer set is identified using relationships of the database. The first generated candidate answer is restructured using the question as a statement which is embedded with the first generated candidate answer. The restructured generated candidate answer which correlates with the previous question-answer set is supplemented with content from the previous question-answer set.

Abstract: A method, system and computer-usable medium are disclosed for identifying unchecked criteria in unstructured and semi-structured data within a form. Text spans representing unchecked criteria within unstructured text in a form are detected and classified to facilitate accurate interpretation of the text. Section identification and annotation operations are then performed to identify and categorize sections within the form. Checklist sections within the form, along with associated checkmarks and boxes, are then identified, followed by the identification of checked item, criteria scope, and previously undetected checklist sections. Once all checklist sections and checked criteria have been identified, remaining text spans within a checklist section are annotated as unchecked criteria.

Abstract: Machine logic that automatically detects natural language processing (NLP) system annotation errors and correspondingly updates NLP annotators to prevent future erroneous annotations by performing the following steps: (i) determining that a first annotation error has occurred in an annotation of a corpus by the natural language processing system; (ii) generating a candidate set of annotation correction actions, where each annotation correction action of the set is adapted to prevent an occurrence of an error similar to the first annotation error by the natural language processing system; (iii) selecting an annotation correction action from the candidate set of annotation correction actions, based, at least in part, on a set of annotation correction confidence characteristics; and (iv) automatically applying the selected annotation correction action to the natural language processing system.

Abstract: Machine logic that automatically detects natural language processing (NLP) system annotation errors and correspondingly updates NLP annotators to prevent future erroneous annotations by performing the following steps: (i) determining that a first annotation error has occurred in an annotation of a corpus by the natural language processing system; (ii) generating a candidate set of annotation correction actions, where each annotation correction action of the set is adapted to prevent an occurrence of an error similar to the first annotation error by the natural language processing system; (iii) selecting an annotation correction action from the candidate set of annotation correction actions, based, at least in part, on a set of annotation correction confidence characteristics; and (iv) automatically applying the selected annotation correction action to the natural language processing system.

Abstract: A recommendation engine for works of literature uses patterns of flow and element similarities for scoring a first user-rated work of literature against one or more recommendation candidate works of literature. Cluster models are created using meta-data modeling the works of literature, the meta-data having literary element categories and instances within each category. Each instance is described by an index value (position in the literature) and significance value (e.g. weight or significance). Cluster finding process(es) invoked for each instance in each category find Similarity Concept clusters and Consistency Trend clusters, which are recorded into the cluster models representing each work of literature. The cluster model can be printed or displayed so that a user can visually understand the ebb and flow of each literary element in the literature, and may be digitally compared to other cluster models of other works of literature for potential recommendation to a user.

Abstract: A recommendation engine for works of literature uses patterns of flow and element similarities for scoring a first user-rated work of literature against one or more recommendation candidate works of literature. Cluster models are created using meta-data modeling the works of literature, the meta-data having literary element categories and instances within each category. Each instance is described by an index value (position in the literature) and significance value (e.g. weight or significance). Cluster finding process(es) invoked for each instance in each category find Similarity Concept clusters and Consistency Trend clusters, which are recorded into the cluster models representing each work of literature. The cluster model can be printed or displayed so that a user can visually understand the ebb and flow of each literary element in the literature, and may be digitally compared to other cluster models of other works of literature for potential recommendation to a user.