公开(公告)号：US20160292591A1

公开(公告)日：2016-10-06

申请号：US14972578

申请日：2015-12-17

Applicant: SAS Institute Inc.

Inventor： Georges H. Guirguis ,  Dominique J. Latour ,  Robert S. Ray

IPC: G06N99/00

CPC classification number: G06F16/212 ,  G06N20/00

Abstract: A computing device creates and stores a state of an analytic model. A first indicator of a dataset is received. A second indicator of an analytic model type of a plurality of analytic model types to train using the dataset is received. An analytic engine of an analytic model of the analytic model type is instantiated. The analytic model of the analytic model type is trained using the dataset and the instantiated analytic engine. A third indicator to save a state of the analytic model is received. The state of the trained analytic model is generated. The generated state of the trained analytic model is written to an output file. The generated state includes a reentry point name of a function of the analytic model type called to instantiate the trained analytic mode.

Abstract translation: 计算设备创建并存储分析模型的状态。 接收到数据集的第一个指示符。 接收使用数据集训练的多个分析模型类型的分析模型类型的第二指示符。 分析模型类型的分析模型的分析引擎被实例化。 使用数据集和实例化分析引擎训练分析模型类型的分析模型。 接收到保存分析模型状态的第三个指标。 生成训练有素的分析模型的状态。 训练有素的分析模型的生成状态被写入输出文件。 所生成的状态包括被称为实例化经过训练的分析模式的分析模型类型的函数的折返点名称。

公开(公告)号：US20160283715A1

公开(公告)日：2016-09-29

申请号：US15043012

申请日：2016-02-12

Applicant: SAS Institute Inc.

Inventor： Brian Lee Duke ,  Paul C. Dulany ,  Kannan Shashank Shah

IPC: G06F21/56 ,  G06F21/62 ,  G06N99/00

CPC classification number: G06F21/566 ,  G06F21/6218 ,  G06F2221/0735 ,  G06N99/005 ,  G06Q20/4016

Abstract: Systems and methods are provided for identifying and detecting unauthorized user activity and for decreasing the rate of false-positives. The disclosed systems and techniques may involve analysis of users' past activity data so that individual classifications and authorization decisions with respect to requested user activity are based on activity data associated with a user's use of multiple services.

Abstract translation: 提供了系统和方法，用于识别和检测未经授权的用户活动以及降低假阳性率。 所公开的系统和技术可以涉及对用户过去的活动数据的分析，使得关于所请求的用户活动的个人分类和授权决定基于与用户对多个服务的使用相关联的活动数据。

公开(公告)号：US09443336B2

公开(公告)日：2016-09-13

申请号：US14205586

申请日：2014-03-12

Applicant: SAS Institute Inc.

Inventor： Jordan Riley Benson ,  Joseph Oliver Hines, Jr. ,  David J. Caira ,  Douglas R. Dotson ,  Frank Lee Wimmer ,  David Langton Clarke ,  Ernest C. Pasour, III ,  Nascif A. Abousalh-Neto ,  Ravinder Devarajan ,  Rajiv Ramarajan ,  Himesh G. Patel

IPC: G09G5/00 ,  G06T11/60 ,  G06T11/20

CPC classification number: G06T11/60 ,  G06T11/206

Abstract: A method of proportional highlighting of data is provided. A graph presented on a display includes a first axis, a second axis, and a first value marker that indicates a value determined from data selected for presentation. The first axis includes a minimum value and a maximum value. The second axis includes a plurality of category values. An indicator identifying a subset of the data is received. A proportional value is determined for the first value marker based on the received indicator. A second value marker indicating the proportional value is presented on the graph overlaid on the first value marker when the determined proportional value is between the minimum value and the maximum value. A scale adjustment marker is presented on the graph without adjusting the first axis when the determined proportional value is not between the minimum value and the maximum value.

Abstract translation: 提供了一种比例突出显示数据的方法。 呈现在显示器上的图形包括第一轴，第二轴和第一值标记，其指示根据为呈现选择的数据确定的值。 第一轴包括最小值和最大值。 第二轴包括多个类别值。 接收到识别数据子集的指示符。 基于接收到的指示符，确定第一值标记的比例值。 当所确定的比例值在最小值和最大值之间时，指示比例值的第二值标记被显示在叠加在第一值标记上的图上。 当确定的比例值不在最小值和最大值之间时，图表上显示比例调整标记，而不调整第一轴。

公开(公告)号：US20160248693A1

公开(公告)日：2016-08-25

申请号：US15145140

申请日：2016-05-03

Applicant: SAS Institute Inc.

Inventor： Michael Thomas ,  Gerald Donald Baulier

IPC: H04L12/911 ,  G06F3/06

CPC classification number: H04L67/2852 ,  G06F3/0608 ,  G06F3/0619 ,  G06F3/0661 ,  G06F3/067 ,  G06F9/5027 ,  G06F9/544 ,  H04L41/509 ,  H04L47/70 ,  H04L47/827 ,  H04L69/03

Abstract: A computer-readable medium is provided that causes a computing device to serve data resources. A received block is compressed with previously compressed blocks to create a new compressed block stored in a pre-allocated block of memory. A reference to the selected pre-allocated block of memory is stored in a tree map using a unique identifier. A second block is received. The pre-allocated block of memory is identified from the tree map using the unique identifier. The received block and at least one of the previously compressed blocks is read from the block of memory. The received second block is compressed with the at least one of the one or more previously compressed blocks to create a second new compressed block stored in the selected second pre-allocated block of memory. A reference to the selected second pre-allocated block of memory is stored in the tree map based on the unique identifier.

Abstract translation: 提供了一种使计算设备提供数据资源的计算机可读介质。 接收的块用先前压缩的块压缩以创建存储在预先分配的存储块中的新的压缩块。 对所选择的预先分配的存储块的引用使用唯一标识符存储在树形图中。 接收到第二个块。 使用唯一标识符从树形图中识别预先分配的存储块。 从存储器块读取所接收的块和至少一个先前压缩的块。 所接收的第二块被一个或多个先前压缩的块中的至少一个压缩，以产生存储在所选择的第二预先分配的存储块中的第二新的压缩块。 基于唯一标识符，将所选择的第二预分配存储块的引用存储在树形图中。

公开(公告)号：US09419886B2

公开(公告)日：2016-08-16

申请号：US14228666

申请日：2014-03-28

Applicant: SAS Institute Inc.

Inventor： Brian Oneal Miles

IPC: H04L12/24 ,  H04L12/751 ,  H04L12/26 ,  G06F17/30 ,  G06N5/04

CPC classification number: H04L45/02 ,  G06F17/30345 ,  G06N5/048 ,  H04L41/142 ,  H04L43/04

Abstract: A method of creating an object store is provided. Node table information reading and link table information are read. The node table information includes node information for a plurality of nodes. The link table information includes link information between pairs of nodes of the plurality of nodes. An anchored network record is created for each node of the plurality of nodes based on the node information and the link information and a defined maximum degree of separation. The anchored network record includes anchor node information associated with an anchor node of the anchored network record and a node record for each node of the plurality of nodes that is within the defined maximum degree of separation from the anchor node of the anchored network record. The created anchored network record is stored for each node of the plurality of nodes.

Abstract translation: 提供了一种创建对象存储的方法。 读取节点表信息读取和链接表信息。 节点表信息包括多个节点的节点信息。 链路表信息包括多个节点的节点对之间的链路信息。 基于节点信息和链接信息以及定义的最大分离度，为多个节点中的每个节点创建锚定的网络记录。 锚定的网络记录包括与锚定的网络记录的锚点节点相关联的锚点节点信息，以及与所锚定的网络记录的锚点节点在所定义的最大分离度之内的多个节点中的每个节点的节点记录。 为多个节点中的每个节点存储所创建的锚定网络记录。

公开(公告)号：US20160203490A1

公开(公告)日：2016-07-14

申请号：US15009475

申请日：2016-01-28

Applicant: SAS Institute Inc.

Inventor： Ankur Gupta ,  Brian Lee Duke ,  Binbin Li ,  Prathaban Mookiah

IPC: G06Q20/40 ,  G06Q20/38

CPC classification number: G06Q20/4016

Abstract: A fraud score for a transaction in connection with an account is computed from retrieved data to indicate a probability of the account being in a compromised condition. A travel score is computed, wherein the computed travel score indicates a likelihood that a user of the account is traveling from a user home location at the time of the received transaction. A self-similarity score may be computed if the computed fraud score is above a predetermined threshold to indicate similarity of the received transaction to other transactions of the account in the set of prior transactions. A suggested action is determined, based on a fraud decisioning operation (and optionally the self-similarity score) and a travel decisioning operation using the fraud score and travel score, respectively.

Abstract translation: 根据检索到的数据计算与帐户相关的交易的欺诈分数，以指示帐户处于受损状态的概率。 计算旅行分数，其中所计算的行进分数表示在接收到的交易时该帐户的用户正在从用户本地位置行进的可能性。 如果所计算的欺诈分数高于预定阈值，则可以计算自相似度分数，以指示所接收的交易与先前交易集合中的帐户的其他交易的相似性。 基于欺诈决策操作（以及可选地，自相似度得分）以及使用欺诈评分和旅行得分的旅行决策操作来确定建议的动作。

公开(公告)号：US20160171089A1

公开(公告)日：2016-06-16

申请号：US14750886

申请日：2015-06-25

Applicant: SAS Institute Inc.

Inventor： Victor H. Richard ,  Christian Haxholdt ,  Glenn Good

IPC: G06F17/30

CPC classification number: G06Q10/00 ,  G06F17/30961 ,  G06Q10/063 ,  G06Q10/067 ,  G06Q30/0202

Abstract: The present disclosure relates to resolving over multiple hierarchies. Specifically, various techniques and systems are provided for adjusting multiple hierarchies for consistency within levels of the hierarchies, using an optimization-based approach that results in an accurate projection across dimensions and levels in hierarchies. The systems and methods may include receiving data associated with nodes of two or more hierarchies, wherein nodes are associated with original node values, identifying a common level node and a target level node for each of the hierarchies, identifying a linking constraint, wherein the linking constraint includes a rule to a node from a hierarchy to make it consistent with a node from another hierarchy, applying the linking constraint to the common level node of each of the hierarchies, wherein applying the linking constraint includes generating updated common node values associated with the common level nodes, and wherein updated common node values are the same node values, applying the updated common node values to the target level node of each of the hierarchies, wherein applying the updated common node values includes generating updated target node values, and generating a resolved hierarchy using the updated target node values.

Abstract translation: 本公开涉及解决多个层次结构。 具体地，提供了各种技术和系统，用于使用基于优化的方法来调整多个层次以便在层次结构的层次内的一致性，这导致跨层次中的维度和级别的准确投影。 系统和方法可以包括接收与两个或更多层次的节点相关联的数据，其中节点与原始节点值相关联，识别每个层级的公共级节点和目标级节点，识别链接约束，其中链接 约束包括从层次结构到节点的规则以使其与来自另一分层结构的节点一致，将所述链接约束应用于每个所述层次结构的公共级节点，其中应用所述链接约束包括生成与所述层级相关联的更新的公共节点值 公共级节点，并且其中更新的公共节点值是相同的节点值，将更新的公共节点值应用于每个层级的目标级别节点，其中应用更新的公共节点值包括生成更新的目标节点值，并且生成 使用更新的目标节点值解析层次结构。

公开(公告)号：US09367799B2

公开(公告)日：2016-06-14

申请号：US14924929

申请日：2015-10-28

Applicant: SAS Institute Inc.

Inventor： Patrick Hall ,  Ilknur Kaynar Kabul ,  Jared Langford Dean ,  Ralph Abbey ,  Susan Haller ,  Jorge Silva

IPC: G06F17/00 ,  G06N3/08 ,  G06N5/02 ,  G06F17/30

CPC classification number: G06F17/30395 ,  G06F17/18 ,  G06F17/30312 ,  G06F17/30554 ,  G06F17/30595 ,  G06F17/30598 ,  G06F17/30716 ,  G06F17/30958 ,  G06K9/622 ,  G06K9/6256 ,  G06K9/6272 ,  G06N3/0454 ,  G06N3/08 ,  G06N3/088 ,  G06N5/02 ,  G06N99/005

Abstract: A computing device presents a cluster visualization based on a neural network computation. First centroid locations are computed for first clusters. Second centroid locations are computed for second clusters. Each centroid location includes a plurality of coordinate values where each coordinate value relates to a single variable of a plurality of variables. Distances are computed pairwise between each centroid location. An optimum pairing is selected based on a minimum distance of the computed pairwise distances where each pair is associated with a different cluster of a set of composite clusters. Noised centroid location data is created. A multi-layer neural network is trained with the noised centroid location data. A projected centroid location is determined in a multidimensional space for each centroid location as values of hidden units of a middle layer of the multi-layer neural network. A graph is presented for display that indicates the determined, projected centroid locations.

Abstract translation: 计算设备基于神经网络计算呈现集群可视化。 为第一个聚类计算第一个质心位置。 计算第二个聚类的第二个质心位置。 每个质心位置包括多个坐标值，其中每个坐标值与多个变量的单个变量相关。 距离在每个质心位置之间成对计算。 基于计算的成对距离的最小距离来选择最佳配对，其中每对与一组复合聚类的不同聚类相关联。 创建信号质心位置数据。 多层神经网络用信号质心位置数据进行训练。 在每个质心位置的多维空间中确定投影质心位置作为多层神经网络的中间层的隐藏单位的值。 呈现用于显示的图形，其指示确定的投影质心位置。

公开(公告)号：US09317594B2

公开(公告)日：2016-04-19

申请号：US13727951

申请日：2012-12-27

Applicant: SAS Institute Inc.

Inventor： Barry de Ville ,  Gurpreet Bawa

IPC: G06F17/30

CPC classification number: G06F17/3071

Abstract: Systems and methods for identifying data files that have a common characteristic are provided. A plurality of data files are received. The plurality of data files include one or more data files having the common characteristic. A list of key terms is generated from the plurality of data files. Data files from the plurality of data files that have an association with a social community are identified, where the social community is defined by one or more features. The list of key terms is updated based on an analysis of the identified features. The updated list of key terms is used to identify other data files that have the common characteristic.

Abstract translation: 提供了用于识别具有共同特征的数据文件的系统和方法。 接收多个数据文件。 多个数据文件包括具有共同特征的一个或多个数据文件。 从多个数据文件生成关键术语的列表。 识别与社会社区有关联的多个数据文件中的数据文件，其中社会群体由一个或多个特征定义。 基于所识别的特征的分析来更新关键术语的列表。 关键术语的更新列表用于识别具有共同特征的其他数据文件。

公开(公告)号：US20160048756A1

公开(公告)日：2016-02-18

申请号：US14924929

申请日：2015-10-28

Applicant: SAS Institute Inc.

Inventor： Patrick Hall ,  Ilknur Kaynar Kabul ,  Jared Langford Dean ,  Ralph Abbey ,  Susan Haller ,  Jorge Silva

IPC: G06N3/08 ,  G06F17/30 ,  G06N5/02

CPC classification number: G06F17/30395 ,  G06F17/18 ,  G06F17/30312 ,  G06F17/30554 ,  G06F17/30595 ,  G06F17/30598 ,  G06F17/30716 ,  G06F17/30958 ,  G06K9/622 ,  G06K9/6256 ,  G06K9/6272 ,  G06N3/0454 ,  G06N3/08 ,  G06N3/088 ,  G06N5/02 ,  G06N99/005

Abstract: A computing device presents a cluster visualization based on a neural network computation. First centroid locations are computed for first clusters. Second centroid locations are computed for second clusters. Each centroid location includes a plurality of coordinate values where each coordinate value relates to a single variable of a plurality of variables. Distances are computed pairwise between each centroid location. An optimum pairing is selected based on a minimum distance of the computed pairwise distances where each pair is associated with a different cluster of a set of composite clusters. Noised centroid location data is created. A multi-layer neural network is trained with the noised centroid location data. A projected centroid location is determined in a multidimensional space for each centroid location as values of hidden units of a middle layer of the multi-layer neural network. A graph is presented for display that indicates the determined, projected centroid locations.

Abstract translation: 计算设备基于神经网络计算呈现集群可视化。 为第一个聚类计算第一个质心位置。 计算第二个聚类的第二个质心位置。 每个质心位置包括多个坐标值，其中每个坐标值与多个变量的单个变量相关。 距离在每个质心位置之间成对计算。 基于计算的成对距离的最小距离来选择最佳配对，其中每对与一组复合聚类的不同聚类相关联。 创建信号质心位置数据。 多层神经网络用信号质心位置数据进行训练。 在每个质心位置的多维空间中确定投影质心位置作为多层神经网络的中间层的隐藏单位的值。 呈现用于显示的图形，其指示确定的投影质心位置。