公开(公告)号：US20170068484A1

公开(公告)日：2017-03-09

申请号：US15257545

申请日：2016-09-06

Applicant: SAS Institute Inc.

Inventor： KEVIN L. SCOTT ,  METIN BALIKCIOGLU ,  BINGFENG DING ,  SHENGKUEI LIN ,  TUGRUL SANLI

IPC: G06F3/06

CPC classification number: G06Q10/0631 ,  G06Q10/02 ,  G06Q10/06 ,  G06Q10/06312 ,  G06Q30/0202 ,  G06Q50/12 ,  H04L47/782 ,  H04L47/823

Abstract: Exemplary embodiments are generally directed to methods, mediums, and systems for correcting censored or constrained historical data with various possible types of computing devices, including cloud-based devices, personal computing devices, and edge-based devices. The corrected data may be used in forecasting, for example to forecast demand for a limited resource. In some embodiments, the data is modeled at a higher level of granularity than an individual record. The aggregated demand may then be pro-rated over a group of categories or users where a given category of users that might be small or nonexistent over a certain time frame may be better accommodated. Moreover, it may be easier or more efficient to make assumptions and employ computing resources at the aggregate level.

Abstract translation: 示例性实施例通常涉及用于使用各种可能类型的计算设备（包括基于云的设备，个人计算设备和基于边缘的设备）来校正被检查或约束的历史数据的方法，介质和系统。 校正的数据可用于预测，例如用于预测有限资源的需求。 在一些实施例中，数据以比单个记录更高的粒度级进行建模。 然后可以对一组类别或用户评估总需求，其中在某一时间范围内可能小或不存在的给定类别的用户可能更好地适应。 此外，在总体水平上进行假设和采用计算资源可能更容易或更有效。

公开(公告)号：US20170078221A1

公开(公告)日：2017-03-16

申请号：US15264256

申请日：2016-09-13

Applicant: SAS Institute Inc.

Inventor： BINGFENG DING ,  FENG CHEN ,  KEVIN L. SCOTT ,  METIN BALIKCIOGLU ,  TUGRUL SANLI ,  XIAODONG YAO

IPC: H04L12/911

CPC classification number: G06Q10/06312 ,  G06Q10/04 ,  G06Q10/06 ,  G06Q10/0631 ,  G06Q30/0202 ,  H04L47/823

Abstract: Exemplary embodiments are generally directed to methods, mediums, and systems for accounting for extensions or reductions of the period for which a resource (e.g., computer processor time, scientific apparatus, storage units, devices, etc.) is allocated. According to exemplary embodiments, allocation-based aggregated effects of extension and relinquishment are modeled. The modeled effects are used to offset allocation forecasts based on historical data. As a result, the dimensionality of the problem of incorporating in-house data is greatly reduced as compared to other techniques, and allocation forecasts can be made more accurately and efficiently.

Abstract translation: 示例性实施例通常涉及用于计算分配资源（例如，计算机处理器时间，科学装置，存储单元，设备等）的周期的扩展或减少的方法，介质和系统。 根据示例性实施例，对扩展和放弃的基于分配的聚合效果进行建模。 模型效应用于根据历史数据抵消分配预测。 因此，与其他技术相比，内置数据的问题的维度大大降低，并且可以更准确和有效地进行分配预测。