发明名称 Presenting discriminant change history records on topology graphs
摘要 The described herein related to a method executable by a processor of a computing environment, where the processor is communicatively coupled to a database of the computing environment. The method includes identifying a feature set from a plurality of incident tickets available on the database and selecting a first subset of the feature set as a learning set. The method further includes generating information from the learning set, where the information indicates unique changes to the computing environment corresponding to a failure. The method further includes selecting a second subset of the feature set as a classification set, validating the information via the classification set to determine whether the learning set is rendering correct values; and sourcing the failure using the validated information based on determining that the learning set is rendering correct values.
申请公布号 US9524475(B1) 申请公布日期 2016.12.20
申请号 US201614986974 申请日期 2016.01.04
申请人 INTERNATIONAL BUSINESS MACHINES CORPORATION 发明人 Cmielowski Lukasz G.;Wrobel Andrzej J.
分类号 G06N99/00;G06F17/30;G06F11/07 主分类号 G06N99/00
代理机构 Cantor Colburn LLP 代理人 Cantor Colburn LLP ;Dobson Scott
主权项 1. A method executable by a processor of a computing environment, the processor communicatively coupled to a database of the computing environment, comprising: identifying, by the processor, a feature set from a plurality of incident tickets available on the database; selecting, by the processor, a first subset of the feature set as a learning set; generating, by the processor, information from the learning set, the information indicating unique changes to the computing environment corresponding to a failure; selecting, by the processor, a second subset of the feature set as a classification set; validating, by the processor, the information via the classification set to determine whether the learning set is rendering correct values; sourcing, by the processor, the failure using the validated information based on determining that the learning set is rendering correct values; and presenting, by the processor, on a topology graph of the computing environment the failure, the validated information, and the feature set, wherein the topology graph is utilized to identify components of the computing environment that have priority based on changes associated with the failure, the validated information, and the feature set, wherein the priority is based on a prioritization rank added to incident tickets of the feature set.
地址 Armonk NY US