发明名称 Dynamic recommendation framework for information technology management
摘要 A method, system, and article are provided for managing performance of a computer system. Both implicit and explicit recommendations for processing of tasks are provided. System performance is tracked and evaluated based upon the actions associated with the task. Future recommendations of the same or other tasks are provided based upon implicit feedback pertaining to system performance, and explicit feedback solicited from a system administrator.
申请公布号 US9323639(B2) 申请公布日期 2016.04.26
申请号 US200912556629 申请日期 2009.09.10
申请人 International Business Machines Corporation 发明人 Haber Eben M.;Matthews Tara L.;Tan Chung-Hao;Uttamchandani Sandeep M.
分类号 G06F11/34;H04L12/24;G06F19/00 主分类号 G06F11/34
代理机构 Lieberman & Brandsdorfer, LLC 代理人 Lieberman & Brandsdorfer, LLC
主权项 1. A method for managing system performance, comprising: a processor in communication with memory; dynamically recommending, by the processor, a set of tasks having at least one task recommendation for management of at least one computer system component, wherein the recommendation is based on a predicted impact on the computer system; responding to a selection of a task recommendation from the set, including producing explicit feedback data; tracking a measured impact of the selected task recommendation on computer system parameters, and obtaining implicit feedback data based on the measured impact; accumulating the implicit and explicit feedback, and categorizing the accumulated feedback; employing the categorized feedback to improve future task recommendations, including determining a variance of the categorized feedback, wherein the categorized feedback is averaged with prior categorized feedback data in response to the variance being less than or equal to a threshold, and the categorized feedback is re-categorized and averaged with prior categorized feedback data in response to the variance exceeding the threshold; and generating a set of future task recommendations comprising at least one future task recommendation based on the averaged feedback data.
地址 Armonk NY US