International Business Machines Corporation

Erin M Farr

Eli M Dow

Michael E Gildein

Moses J Vaughan

Yu-Cheng Hsu

Douglas Michael Zobre

Pratik Gupta

An embodiment identifies a fileset used by an executing application. An embodiment generates, at a first time, a first snapshot of the fileset, the first snapshot comprising a first state of the fileset at the first time and a first indicator corresponding to the first state of the fileset. An embodiment selects, at a recovery time later than the first time, from a sequence of snapshots of the fileset including the first snapshot, a recovery snapshot, the recovery snapshot comprising a recovery indicator, wherein the recovery indicator has a difference from the first indicator that is higher than a threshold value. An embodiment restores, using the recovery snapshot, the fileset to a state other than the first state, the restoring resulting in a restored application.

Application dependency based malware recovery

Embodiments include method, systems and computer program products for performing summarization of a recording based on biometric and business process data. Aspects include receiving a recording of a customer service representative performing a task, receiving a biometric data from a sensor configured to monitor the customer service representative during the performance of the task and obtaining a business process data relating to one or more of the task and the customer service representative. Aspects also include identifying one or more portions of interest from the recording based on one or more of the biometric data and the business process data and compiling the one or more portions of interest into a summarization of the task.

Summarization of a recording for quality control

An aspect of providing augmented reality model comparison and deviation detection includes receiving a three-dimensional (3D) model of an object that is associated with a domain, determining a set of characteristics of the object from the 3D model, and searching a domain-specific database for data matching the set of characteristics. The domain-specific database corresponds to the domain associated with the object. An aspect also includes determining an identification of the object from data in the domain-specific database that matches the set of characteristics. For each characteristic in the set of characteristics, and aspect further includes comparing each value of the characteristic to a corresponding value in the matching data of the domain-specific database, and calculating any variation between the corresponding value of the characteristic and the value of the matching data. Another aspect includes creating a searchable index of the matching data and corresponding variations.

Augmented reality model comparison and deviation detection

A method for managing log messages in a system includes identifying a log message having a data value, filtering a first data value from a historical log record for a first interval, predicting whether any particular system events should occur when the first data value is filtered from a received log record and identifying the corresponding system events, initiating a second interval while filtering the first data value from a received log record, determining whether any non-predicted system events have occurred, and removing the filter for the first data value responsive to determining that a non-predicted system event has occurred.

Growing community of inventors

Fishkill, NY, United States of America

Erin M Farr