International Business Machines Corporation

Umasuthan Ramakrishnan

Albee Jhoney

Soma Shekar Naganna

Bhavani Kumar Eshwar

Anbazhagan Mani

Hariharan Krishna

Joseph Xaviour

Deepak M Srinivasa

Manu Kuchhal

Shashidhar R Yellareddy

Dattatreya S Vellal

Krithikaa Krishnamoorthy

S Venkatakrishnan

Amit Malla

A method, computer program product, and system performing a method that includes a processor(s) obtaining a digital document defining insurance parameters for utilization by a user of a portion of computing resources selected from a group of computing resources comprising the shared computing environment. The processor(s) generates, based on the digital document, parameters defining the portion of the computing resources and risk management rules reflecting guaranteed thresholds for utilization and performance of the portion of the computing resources defined in the digital document. The processor(s) configures a controller in the shared computing environment, based on the parameters and the risk management rules and obtains a notification of a failure in the portion of the computing resources in violation of the risk management rules and details of the failure to determine the extent and impact of the failure. The processor(s) automatically generates a claim package and transmits it to the administrator.

Automation and validation of insurance claims for infrastructure risks and failures in multi-processor computing environments

A computer-implemented method includes detecting an update to a record in an entity table of a database. At least one of an age score, a lineage score, and a completeness score for the record is calculated, responsive to the update. A trust factor is calculated, by a computer processor, based on the at least one of the age score, the lineage score, and the completeness score for the record. The trust factor indicates a level of trustworthiness of the record. It is decided whether to use data in the record based on the trust factor.

Generating a trust factor for data in non-relational or relational databases

A computer-implemented method includes receiving a column-oriented table comprising data for a column family, wherein the data for the column family comprises column names and corresponding column values, receiving a set of anonymous column names for the column family, receiving a set of synonymous column names for the column family, determining a weighting for each column name that is not an anonymous column name based on the count or frequency of occurrence of the column name and the synonymous column names within the column-oriented table, and processing the column-oriented table with a probabilistic matching engine using the weighting for each column name. A corresponding computer program product and computer system are also disclosed herein.

Data matching for column-oriented data tables

Granular event management for service platforms. First bundle information relating to the content of a bundle is received at a first time, wherein the bundle relates to one or more resources in an application of a distributed computing environment. The bundle is monitored for a bundle event, wherein the bundle event is generated from a change in a lifecycle state of the bundle. Whether the bundle event has occurred is determined. Responsive to determining that the bundle event has occurred, second bundle information relating to the content of the bundle is received at a second time. Responsive to receiving the second bundle information, a granular event associated with the bundle event is identified, wherein the granular event relates to a difference between the first bundle information and the second bundle information.

Granular event management for service platforms

Software that performs the following steps: (i) receiving first bundle information relating to the content of a bundle at a first time, wherein the bundle relates to one or more resources in an application of a distributed computing environment; (ii) monitoring the bundle for a bundle event, wherein the bundle event is generated from a change in a lifecycle state of the bundle; (iii) determining that the bundle event has occurred; (iv) responsive to determining that the bundle event has occurred, receiving second bundle information relating to the content of the bundle at a second time; and (v) responsive to receiving the second bundle information, identifying a granular event associated with the bundle event, wherein the granular event relates to a difference between the first bundle information and the second bundle information.

Determining and conveying availability information regarding a user may be accomplished by a variety of techniques. In one general implementation, determining and conveying availability information regarding a user may include determining an availability intent for a user of a user interface device, determining at least one availability desire for the user, and/or determining an inferred availability state for the user. The availability intent, the availability desire, and/or the inferred availability state for the user may be used to determine a composite availability state for the user, and the composite availability state may be conveyed to a remote server system.

Determining and conveying user availability

Random sequences are generated based on a stochastic generative model having multiple random variates. Inputs representative of the stochastic generative model are received. The inputs include a first random variate having a finite set of alphabets, a second random variate having a set of alphabets, and a third random variate having a finite set of alphabets. Outputs representative of the random sequences are generated based on the stochastic generative model. The outputs include a first random sequence that is a finite-length random sequence of alphabets randomly selected from the first random variate, a second random sequence having a set of alphabets selected from the second random variate, and a third random sequence having a set of alphabets randomly selected from the third random variate.

Generating random sequences based on stochastic generative model having multiple random variates

Provided are methods, apparatus or computer programs for collaborative problem solving and/or automated discovery and dynamic integration of capabilities in a distributed, multidisciplinary work environment. A set of software-implemented agents distributed across a network are used to identify agents and resources on the network that are capable of performing required tasks. The agents collaborate to generate a plan comprising a task allocation and task execution sequence in which tasks are allocated to selected ones of the agents and identified resources, and then the selected agents and resources execute the allocated tasks in accordance with the generated plan. Services (such as processing functions and knowledge) provided by hardware and software resources in a network are represented as 'capabilities' of an associated agent. The functions implemented by each agent are also represented as 'capabilities'. The abstract &#x2018;capabilities&#x2019; can be published and then discovered and dynamically integrated to solve domain specific problems.

Growing community of inventors

Bengaluru, India

Umasuthan Ramakrishnan