International Business Machines Corporation

Microsoft Technology Licensing, LLC

Robert K Tucker

Corville O Allen

Darryl Maurice Adderly

Scott N Gerard

Robert Kerr

Dominique Martindale

Martin Nathan

Natasa Milic-Frayling

Ralph Sommerer

Ian P Kitchen

Jurij Leskovec

Mechanisms for generating different lines of questioning based on evaluation of a previous line of questioning are provided. A set of evidential data specifying a plurality of concept entities and input specifying a scenario to be evaluated are received. The scenario specifies a hypothetical link between at least two of the concept entities. A first set of questions corresponding to the at least two information concept entities are evaluated based on the set of evidential data. Based on results of evaluating the first set of questions, a second set of questions is automatically generated to further expand upon and investigate the results of evaluating the first set of questions. The second set of questions are processed and an indication of the scenario and a corresponding measure of support for or against the scenario being a valid scenario is output.

Generating additional lines of questioning based on evaluation of a hypothetical link between concept entities in evidential data

Mechanisms for evaluating an evidential statement in a corpus of evidence are provided. A first evidential statement for which corroboration is sought is received and a corpus of evidence data is processed to determine a measure of corroboration of the first evidential statement by other evidence data in the corpus of evidence data. An indication of trustworthiness of the first evidential statement is generated based on the measure of corroboration of the first evidential statement by the other evidence data in the corpus of evidence data. A representation of the indication of the trustworthiness of the first evidential statement is output in association with the first evidential statement.

Evaluating evidential links based on corroboration for intelligence analysis

Mechanisms for evaluating an evidential statement in a corpus of evidence are provided. An evidential statement is received for determining a level of confidence in a hypothetical ontological link of an ontology. A source of the evidential statement is identified and a grading of the source of the evidential statement is determined based on a source grading measurement value indicative of a degree of reliability and credibility of the source. An indication of trustworthiness of the evidential statement is generated based on the source grading measurement value. A representation of the indication of trustworthiness of the evidential statement is output in association with the evidential statement.

Grading sources and managing evidence for intelligence analysis

Mechanisms, in a Question Answering (QA) system comprising a processor and a memory, for evaluating a hypothetical link in an ontology are provided. An initial analysis of the ontology is performed to identify a set of information concept entities and links between information concept entities in the ontology. The hypothetical link between a first information concept entity and a second information concept entity in the ontology is generated based on the initial analysis of the ontology. Natural language questions corresponding to the hypothetical link are processed to generate answer results directed to a plurality of links between a plurality of information concept entities. The answer results are aggregated across the plurality of links to determine an aggregate answer result for the hypothetical link. An indication of whether or not the hypothetical link is a valid link is output based on the aggregate answer result for the hypothetical link.

Evidence aggregation across heterogeneous links for intelligence gathering using a question answering system

A query comprising a graph pattern comprising a plurality of graph triples of node-edge-node is received. A graph database comprising a plurality of graph patterns is accessed and a plurality of graph patterns in the graph database that match the received query are identified. An inverse frequency for each graph triple of the received query in the accessed graph database is calculated and from this a score for each graph pattern in the graph database that matches the received query is also calculated, the score comprising a sum of the inverse frequencies for each graph triple contained within the respective graph pattern. The plurality of graph patterns in the graph database that match the received query are then ranked according to their respective calculated scores.

Ranking of graph patterns

A method, computer system, and a computer program product for entity selection in a visualization of a network graph is provided. The present invention may include receiving as input a selection interaction. The present invention may then include generating a profile based on one or more entity characteristics of one or more entities selected in the selection interaction. The present invention may then include generating a weighting for one or more entities based on the generated profile, wherein the generated weighting is a sum of weights of one or more entity characteristics. The present invention may lastly include partitioning the user interface into one or more selectable areas, wherein a size of the selectable area is based on the generated weighting for a displayed entity.

Entity selection in a visualization of a network graph

A system, method, computer program product and computer program for evaluating links between objects are provided. A receive ontology component receives an ontology and an identify component identifies, from the ontology, semantic feature types within the ontology that can be used to measure the links between the objects. A data receive component receives instance information and maps the instance information into an ontological form of the instance information. An analyze component analyzes the ontological form to generate an ontological mapping of the instance information. A match component analyzes the mapping to identify matches with semantic patterns. A strength component analyzes the associated semantic features associated with the objects of the matches to determine weightings for the links of the matches. An alert component provides the links and associated weightings.

Generating derived links

An approach is provided to identify permanence data corresponding to terms included in a corpus of a question answering (QA) system. Based on the identified permanence, a time-based confidence of each of the terms is established. Terms are identified as a plurality of candidate answers to a question posed to the QA system. Each of the plurality of candidate answers are scored with the scoring being at least partially based on the time-based confidence established for each of the terms.

Applying level of permanence to statements to influence confidence ranking

Mechanisms are provided for identifying and evaluating hypothetical ontological links in an ontology and outputting intelligence information. An ontology is received that comprises a plurality of information concept objects and one or more actual links between the information concept objects. An indication of a selected information concept object for which a hypothetical ontological link is to be evaluated is received. One or more natural language questions are generated for processing by a Question Answering (QA) system pipeline based on at least an identification of a type of the selected information concept object. The one or more natural language questions are processed to generate answer results and a score for the hypothetical ontological link is calculated based on the answer results. Intelligence information associated with the hypothetical ontological link is output based on the score for the hypothetical ontological link.

Intelligence gathering and analysis using a question answering system

Mechanisms for generating a set of questions to evaluate a link between concept entities are provided. A hypothetical link between at least two information concept entities is generated. A set of questions corresponding to the hypothetical link is retrieved and pruned into a subset of questions based on at least one of characteristics of the hypothetical link or characteristics of the at least two information concept entities. The pruned set of questions is processed based on a corpus of evidence to generate a measure of support for or against the hypothetical link being an actual link between the at least two information concept entities. A validity indication for the hypothetical link indicating whether or not the hypothetical link is an actual link between the at least two information concept entities is output.

Question pruning for evaluating a hypothetical ontological link

Mechanisms for evaluating a link between information concept entities are provided. A set of evidential data specifying a plurality of information concept entities is received and a link between at least two information concept entities in the set of evidential data is generated. The set of evidential data is evaluated with regard to whether or not the set of evidential data supports or refutes the link. The evaluation of the set of evidential data comprises analyzing language of natural language statements in the set of evidential data to identify certainty terms within the natural language statements. A confidence value for the link is calculated based on results of the evaluation of the set of evidential data and a knowledge output is generated based on the link and the confidence value associated with the link.

Grading ontological links based on certainty of evidential statements

A browser sessions search tool allows a user to search through records of previously viewed resource pages, even across multiple browser sessions. Each visit to a resource page results in creation of a visit data structure that references the resource page content. A rich assortment of data is recorded for each visited resource page, such as full text and layout data, time of access and navigation type that lead to the page. In one embodiment, color aspects of each resource page are stored to assist in a search, although other recorded resource page characteristics may additionally or alternatively include without limitation other visual, audio, navigational, or metadata aspects.

Growing community of inventors

Cambridge, United Kingdom

Robert K Tucker