利索-全球专利检索

  1. 登录
  2. 注册

搜索结果

257 条记录 (耗时 0.032 秒)

    Method and system for generating a decision-tree classifier in parallel in a multi-processor system
    71.
    发明授权
    Method and system for generating a decision-tree classifier in parallel in a multi-processor system 有权
    在多处理器系统中并行生成决策树分类器的方法和系统

    公开(公告)号：US6138115A

    公开(公告)日：2000-10-24

    申请号：US245765

    申请日：1999-02-05

    申请人： Rakesh Agrawal ,  Manish Mehta ,  John Christopher Shafer

    发明人： Rakesh Agrawal ,  Manish Mehta ,  John Christopher Shafer

    IPC分类号： G06F17/30

    CPC分类号： G06F17/30705 ,  G06F17/30625 ,  Y10S707/962 ,  Y10S707/966 ,  Y10S707/968 ,  Y10S707/99933 ,  Y10S707/99936 ,  Y10S707/99937 ,  Y10S707/99944

    摘要： A method and system are disclosed for generating a decision-tree classifier in parallel in a multi-processor system, from a training set of records. The method comprises the steps of: partitioning the records among the processors, each processor generating an attribute list for each attribute, and the processors cooperatively generating a decision tree by repeatedly partitioning the records using the attribute lists. For each node, each processor determines its best split test and, along with other processors, selects the best overall split for the records at that node. Preferably, the gini-index and class histograms are used in determining the best splits. Also, each processor builds a hash table using the attribute list of the split attribute and shares it with other processors. The hash tables are used for splitting the remaining attribute lists. The created tree is then pruned based on the MDL principle, which encodes the tree and split tests in an MDL-based code, and determines whether to prune and how to prune each node based on the code length of the node.

    摘要翻译： 公开了一种用于在多处理器系统中从培训记录集并行生成决策树分类器的方法和系统。 该方法包括以下步骤：在处理器之间划分记录，每个处理器为每个属性生成属性列表，并且处理器通过使用属性列表重复分割记录来协同地生成决策树。 对于每个节点，每个处理器确定其最佳分割测试，并与其他处理器一起为该节点上的记录选择最佳的整体分割。 优选地，使用基尼系数索引和类别直方图来确定最佳分割。 此外，每个处理器使用split属性列表构建哈希表，并与其他处理器共享。 散列表用于分割剩余的属性列表。 然后，基于MDL原理修剪创建的树，MDL原理在基于MDL的代码中对树进行编码和分割测试，并根据节点的代码长度确定是否修剪和如何修剪每个节点。

    Dual mixed refrigerant cycle for gas liquefaction
    72.
    发明授权
    Dual mixed refrigerant cycle for gas liquefaction 有权
    双重混合制冷剂循环用于气体液化

    公开(公告)号：US6119479A

    公开(公告)日：2000-09-19

    申请号：US208562

    申请日：1998-12-09

    申请人： Mark Julian Roberts ,  Rakesh Agrawal

    发明人： Mark Julian Roberts ,  Rakesh Agrawal

    IPC分类号： F25J1/00 ,  F25B7/00 ,  F25J1/02 ,  F25J3/02 ,  F25J5/00

    CPC分类号： F25J1/0022 ,  F25J1/0052 ,  F25J1/0055 ,  F25J1/0097 ,  F25J1/0214 ,  F25J1/0231 ,  F25J1/0267 ,  F25J1/0278 ,  F25J1/0291 ,  F25J1/0292 ,  F25J2220/62 ,  F25J2220/64 ,  F25J2245/02

    摘要： Process and system for liquefying a pressurized gas, especially natural gas. The process is carried out in two heat exchangers, and cooling for each heat exchanger is provided by a mixed refrigerant which is vaporized at a single essentially constant pressure. Feed precooling, low level refrigerant precooling, and high level liquid refrigerant subcooling are effected in one of the heat exchanger against low level refrigerant vaporizing at a single essentially constant pressure. The process and system of the invention are especially well-suited for installation on ships, barges, and offshore platforms.

    摘要翻译： 用于液化加压气体，特别是天然气的方法和系统。 该方法在两个热交换器中进行，每个热交换器的冷却由在单个基本恒定的压力下汽化的混合制冷剂提供。 进料预冷却，低水平制冷剂预冷却和高液位制冷剂过冷却在一个热交换器中进行，以低压制冷剂在单个基本恒定的压力下蒸发。 本发明的方法和系统特别适用于船舶，驳船和海上平台上的安装。

    Operable and efficient distillation schemes for multicomponent separations
    73.
    发明授权
    Operable and efficient distillation schemes for multicomponent separations 失效
    多组分分离的可操作和有效的蒸馏方案

    公开(公告)号：US6106674A

    公开(公告)日：2000-08-22

    申请号：US84722

    申请日：1998-05-26

    申请人： Rakesh Agrawal ,  Zbigniew Tadeusz Fidkowski

    发明人： Rakesh Agrawal ,  Zbigniew Tadeusz Fidkowski

    IPC分类号： F25J3/00 ,  B01D3/14 ,  B01D3/42 ,  F25J3/02 ,  F25J3/04 ,  B01D3/32

    CPC分类号： B01D3/146 ,  F25J3/0219 ,  F25J3/0233 ,  F25J3/0257 ,  F25J3/0261 ,  F25J3/0295 ,  F25J3/044 ,  F25J3/0466 ,  F25J3/04793 ,  F25J2200/02 ,  F25J2200/30 ,  F25J2200/38 ,  F25J2200/50 ,  F25J2200/72 ,  F25J2210/12 ,  F25J2290/40 ,  Y10S203/20

    摘要： The present invention teaches an efficient and easier to operate distillation system to separate mixtures containing three or more components into streams enriched in one of the components. In this invention, a liquid stream enriched in the least volatile component is withdrawn from the bottom of one distillation column while a vapor stream enriched in the most volatile component is withdrawn from the top of another distillation column. Of these two distillation columns, the pressure of the distillation column with the bottom liquid enriched in the least volatile component is higher; and this higher pressure distillation column transfers at least two vapor streams from different locations to either one or more other distillation columns within the distillation system. For a ternary mixture, both the vapor streams are transferred to the distillation column with the top vapor enriched in the most volatile component. In the preferred mode, at least one of the vapor transfer is part of a two-way communication between the two columns, i.e., in addition to the transfer of the vapor stream, a return liquid stream is implemented between the same locations of the two columns. This invention allows the flow of all vapor streams from a higher pressure column to a lower pressure column and thereby does not suffer with the operating pressure constraints of the prior art efficient processes.

    摘要翻译： 本发明教导了一种有效且易于操作的蒸馏系统，以将含有三种或更多种组分的混合物分离成富含一种组分的物流。 在本发明中，从一个蒸馏塔的底部排出富含最少挥发性组分的液体物流，同时富含最易挥发组分的蒸气流从另一个蒸馏塔的顶部排出。 在这两个蒸馏塔中，富含最不易挥发组分的底部液体的蒸馏塔的压力较高; 并且该高压蒸馏塔将至少两个蒸汽流从不同位置转移到蒸馏系统内的一个或多个其它蒸馏塔。 对于三元混合物，蒸气流被转移到蒸馏塔中，顶部蒸气富含最易挥发的组分。 在优选的模式中，至少一个蒸汽传递是两列之间的双向连通的一部分，即，除了蒸汽流的转移之外，在两个相同的位置之间实现返回液体流 列。 本发明允许从更高压力塔到低压塔的所有蒸汽流的流动，从而不受现有技术有效过程的操作压力限制的影响。

    Production of argon from a cryogenic air separation process
    74.
    发明授权
    Production of argon from a cryogenic air separation process 失效
    从低温空气分离过程生产氩气

    公开(公告)号：US5970743A

    公开(公告)日：1999-10-26

    申请号：US96009

    申请日：1998-06-10

    申请人： Donn Michael Herron ,  Stephen John Cook ,  Rakesh Agrawal

    发明人： Donn Michael Herron ,  Stephen John Cook ,  Rakesh Agrawal

    IPC分类号： C01B23/00 ,  F25J3/04

    CPC分类号： F25J3/04721 ,  F25J3/04412 ,  F25J3/04624 ,  F25J3/04678 ,  F25J3/04727 ,  F25J3/04872 ,  F25J2200/32 ,  F25J2200/34 ,  F25J2200/90 ,  F25J2205/02 ,  F25J2205/30 ,  F25J2245/58 ,  F25J2250/10 ,  Y10S62/924

    摘要： The present invention relates to a process for the cryogenic separation of air to recover at least a nitrogen-depleted crude argon product, wherein the process is carried out in a primary distillation system comprising at least a first distillation column, which separates a feed mixture comprising nitrogen, oxygen and argon into a nitrogen-enriched overhead and an oxygen-rich bottoms, and a side-arm column which rectifies an argon-containing feed stream fed from the primary distillation column to produce an essentially-oxygen-depleted argon overhead. The improvement of the present invention is characterized in that: (a) a nitrogen-containing, argon-rich side stream is withdrawn from a location of the side-arm column which is above the location of entry of the argon-containing feed stream; (b) the withdrawn, nitrogen-containing, argon-rich side stream of step (a) is fed to a nitrogen rejection column to remove the contained nitrogen, wherein the nitrogen rejection column contains at least a stripping section which is located below the location of the feed of the nitrogen-lean, argon-rich side stream, and wherein the stripping section of the nitrogen rejection column is provided with vapor boilup; (c) the nitrogen-depleted, crude argon product is recovered and removed from the bottom of the nitrogen rejection column; and (d) at least a portion of upward flowing vapor in the nitrogen rejection column is removed and the removed portion is returned to a suitable location of the side-arm column.

    摘要翻译： 本发明涉及一种用于低温分离空气以回收至少一种贫氮粗氩产物的方法，其中该方法在包含至少第一蒸馏塔的一级蒸馏系统中进行，所述第一蒸馏塔分离包含 氮气，氧气和氩气进入富氮塔顶和富氧底物，以及侧臂塔，其对从一级蒸馏塔进料的含氩进料流进行整流，以产生基本上不含氧的氩塔顶馏出物。 本发明的改进的特征在于：（a）含氮富氩侧流从位于含氩进料流进入位置的侧臂塔的位置排出; （b）将步骤（a）的取出的含氮富氩侧流进料至氮排除塔以除去所含的氮，其中氮排除塔至少包含位于位置下方的汽提部分 的贫氮富氩侧流的进料，并且其中氮排除塔的汽提段设置有蒸汽沸腾; （c）从氮排除塔的底部回收并除去贫氮粗氩产物; 和（d）氮排除塔中的向上流动的蒸气的至少一部分被去除，并且将去除的部分返回到侧臂塔的适当位置。

    System and method for parallel mining of association rules in databases
    77.
    发明授权
    System and method for parallel mining of association rules in databases 失效
    数据库中关联规则并行挖掘的系统和方法

    公开(公告)号：US5842200A

    公开(公告)日：1998-11-24

    申请号：US500717

    申请日：1995-07-11

    申请人： Rakesh Agrawal ,  John Christopher Shafer

    发明人： Rakesh Agrawal ,  John Christopher Shafer

    IPC分类号： G06Q30/02 ,  G06F17/30

    CPC分类号： G06Q30/02 ,  G06F2216/03 ,  Y10S707/99931 ,  Y10S707/99933

    摘要： A multiprocessor including a plurality of processing systems is disclosed for discovering consumer purchasing tendencies. Each processing system of the multiprocessor identifies consumer transaction itemsets that are stored in a database that is distributed among the processing systems and which appear in the database a user-defined minimum number of times, referred to as minimum support. Then, the system discovers association rules in the itemsets by comparing the ratio of the number of times each of the large itemsets appears in the database to the number of times particular subsets of the itemset appear in the database. When the ratio exceeds a predetermined minimum confidence value, the system outputs an association rule which is representative of purchasing tendencies of consumers.

    摘要翻译： 公开了一种包括多个处理系统的多处理器，用于发现消费者购买倾向。 多处理器的每个处理系统识别存储在数据库中的消费者事务项目集，该数据库分布在处理系统之间，并且在数据库中出现用户定义的最小次数，称为最小支持。 然后，通过比较数据库中出现的每个大项目集的次数与数据库中出现的项目集的特定子集的次数之间的比例，系统发现项目集中的关联规则。 当比率超过预定的最小置信度值时，系统输出代表消费者购买倾向的关联规则。

    Database system and method employing data cube operator for group-by operations
    78.
    发明授权
    Database system and method employing data cube operator for group-by operations 失效
    数据库系统和采用数据多维数据集算子进行分组操作的方法

    公开(公告)号：US5832475A

    公开(公告)日：1998-11-03

    申请号：US624283

    申请日：1996-03-29

    申请人： Rakesh Agrawal ,  Ashish Gupta ,  Sunita Sarawagi

    发明人： Rakesh Agrawal ,  Ashish Gupta ,  Sunita Sarawagi

    IPC分类号： G06F17/30

    CPC分类号： G06F17/30489 ,  Y10S707/99931 ,  Y10S707/99932 ,  Y10S707/99933 ,  Y10S707/99934 ,  Y10S707/99937

    摘要： Disclosed is a system and method for performing database queries including GROUP-BY operations, in which aggregate values for attributes are desired for distinct, partitioned subsets of tuples satisfying a query. A special case of the aggregation problem is addressed, employing a structure, called the data cube operator, which provides information useful for expediting execution of GROUP-BY operations in queries. Algorithms are provided for constructing the data cube by efficiently computing a collection of GROUP-BYs on the attributes of the relation. Decision support systems often require computation of multiple GROUP-BY operations on a given set of attributes, the GROUP-BYs being related in the sense that their attributes are subsets or supersets of each other. The invention extends hash-based and sort-based grouping methods with optimizations, including combining common operations across multiple GROUP-BYs and using pre-computed GROUP-BYs for computing other GROUP-BYs. An extension of the cube algorithms handles any given collection of aggregates.

    摘要翻译： 公开了一种用于执行包括GROUP-BY操作的数据库查询的系统和方法，其中期望满足查询的元组的不同分配子集的属性的聚合值。 解决聚合问题的一个特殊情况，采用称为数据多维数据集运算符的结构，该结构提供了有助于在查询中加速执行GROUP-BY操作的信息。 提供了通过有效地计算关系的属性的GROUP-BY集合来构造数据立方体的算法。 决策支持系统通常需要对给定的一组属性计算多个GROUP-BY操作，GROUP-BY在其属性是彼此的子集或超集的意义上相关。 本发明通过优化来扩展基于散列和分类的分组方法，包括组合跨多个GROUP-BY的常规操作，并使用预先计算的GROUP-BY来计算其他GROUP-BY。 多维数据集算法的扩展可处理任何给定的聚合集合。

    Method and system for mining quantitative association rules in large relational tables
    79.
    发明授权
    Method and system for mining quantitative association rules in large relational tables 失效
    在大型关系表中挖掘定量关联规则的方法和系统

    公开(公告)号：US5724573A

    公开(公告)日：1998-03-03

    申请号：US577945

    申请日：1995-12-22

    申请人： Rakesh Agrawal ,  Ramakrishnan Srikant

    发明人： Rakesh Agrawal ,  Ramakrishnan Srikant

    IPC分类号： G06F17/30

    CPC分类号： G06F17/3061 ,  Y10S707/99931 ,  Y10S707/99933 ,  Y10S707/99936

    摘要： A method and apparatus are disclosed for mining quantitative association rules from a relational table of records. The method comprises the steps of: partitioning the values of selected quantitative attributes into intervals, combining adjacent attribute values and intervals into ranges, generating candidate itemsets, determining frequent itemsets, and outputting an association rule when the support for a frequent itemset bears a predetermined relationship to the support for a subset of the frequent itemset. Preferably, the partitioning step includes determining whether to partition and the number of partitions based on a partial incompleteness measure. The candidate generation includes discarding those itemsets not meeting a user-specified interest level and those having a subset which is not a frequent itemset. The frequent itemsets are determined using super-candidates that include information of the candidate itemsets. Preferably, each super-candidate has a data structure, such as a multi-dimensional tree or array, representing quantitative attributes common to the replaced candidate itemsets.

    摘要翻译： 公开了一种从记录的关系表中挖掘定量关联规则的方法和装置。 该方法包括以下步骤：将所选择的定量属性的值分为间隔，将相邻属性值和间隔组合成范围，生成候选项集，确定频繁项集，以及当对频繁项集的支持具有预定关系时输出关联规则 支持频繁项目集的一个子集。 优选地，分割步骤包括基于部分不完全性测量确定是否划分分区和分割数。 候选生成包括丢弃不符合用户指定的兴趣级别的那些项目集，以及具有不是频繁项目集的子集的项目集。 使用包括候选项集的信息的超级候选来确定频繁项集。 优选地，每个超级候选具有诸如多维树或阵列的数据结构，其表示替换的候选项集合共同的定量属性。