Method and apparatus for deriving association rules from data and for
segmenting rectilinear regions
    1.
    发明授权
    Method and apparatus for deriving association rules from data and for segmenting rectilinear regions 失效
    从数据和分割直线区域导出关联规则的方法和装置

    公开(公告)号:US5991752A

    公开(公告)日:1999-11-23

    申请号:US025536

    申请日:1998-02-18

    摘要: The present application discloses a method and apparatus for extracting association rules from data having two or more numeric attributes and a true-false attribute, and for presenting the rules in an easily understandable form. The method comprises the steps of: (i) storing numbers u(i,j) and v(i,j) of data in each pixel whose true-false attribute is true, so as to correspond to each pixel in a plane; (ii) inputting a condition .theta.; (iii) segmenting from the plane a rectilinear region S of the pixels to maximize the equation ##EQU1## ;and (iv) outputting data included in the segmented rectilinear region S. The invention also allows regions to be derived which satisfy a desired support maximization rule, confidence maximization rule, optimized entropy rule, and optimized interclass variance rule.

    摘要翻译: 本申请公开了一种用于从具有两个或多个数字属性和真假属性的数据提取关联规则并且以容易理解的形式呈现规则的方法和装置。 该方法包括以下步骤:(i)在真假属性为真的每个像素中存储数据u(i,j)和v(i,j),以便对应于平面中的每个像素; (ii)输入条件θ; (iii)从平面分割像素的直线区域S以最大化方程;以及(iv)输出包括在分割直线区域S中的数据。本发明还允许导出满足期望的支持最大化规则的区域,置信度 最大化规则,优化熵规则和优化的类间方差规则。

    Method and apparatus for deriving an association rule between data
    2.
    发明授权
    Method and apparatus for deriving an association rule between data 失效
    用于导出数据之间的关联规则的方法和装置

    公开(公告)号:US5812997A

    公开(公告)日:1998-09-22

    申请号:US735911

    申请日:1996-10-25

    摘要: A method is described for finding correlation between a plurality of data having two kinds of numerical attributes and a true-false attribute. The method comprises the steps of: constituting a plane with two numerical attributes, dividing the plane into meshes, and counting the number of data in each mesh (also called a "bucket") and the number of data whose true-false attribute represents true. If each mesh is assumed to be a pixel, such plane can be considered as a plane image in which the number of data corresponds to brilliance, and the number of data whose true-false attribute represents true corresponds to saturation. The method further includes the step of segmenting an admissible image which is convex along an axis of the plane according to a predetermined condition .theta. to find an area with strong correlation. If the segmented area as the admissible image satisfies the above-described condition such as the maximized support rule, the method also presents the area to the user. In addition, necessary attributes for data included in the area are also extracted from a database, as required.

    摘要翻译: 描述了一种用于发现具有两种数值属性的多个数据与真假属性之间的相关性的方法。 该方法包括以下步骤:构成具有两个数值属性的平面,将平面划分为网格,并计算每个网格(也称为“桶”)中的数据数,以及true-false属性表示真实数据的数量 。 如果每个网格被假设为一个像素,则这样的平面可以被认为是其中数据数量对应于辉度的平面图像,并且真假属性表示为真的数据的数量对应于饱和度。 该方法还包括按照预定条件θ分割沿平面轴线凸出的容许图像以找到具有强相关性的区域的步骤。 如果作为容许图像的分割区域满足上述条件,例如最大化的支持规则,则该方法还向用户呈现该区域。 另外,根据需要也从数据库中提取包括在该区域中的数据的必要属性。

    Method and apparatus for computing association rules for data mining in
large database
    3.
    发明授权
    Method and apparatus for computing association rules for data mining in large database 失效
    用于计算大数据库中数据挖掘的关联规则的方法和装置

    公开(公告)号:US5983222A

    公开(公告)日:1999-11-09

    申请号:US738666

    申请日:1996-10-25

    IPC分类号: G06F17/30

    摘要: A method is disclosed for computing an association rule in a database having numerical attributes and 0-1 attributes. First, a numerical attribute is divided into a plurality of intervals (or buckets), and each data set is placed into a single bucket according to the value of the numerical attribute. The number of data sets in each bucket and the number of data sets with a 0-1 attribute being 1 are counted. Second, the starting bucket of an interval to be detected is to be detected. Third, the terminating bucket corresponding to the starting bucket is detected. That is, the largest interval with a confidence equal to or larger than a predetermined value. Fourth, one of the detected pairs of starting and terminating buckets which includes the largest number of customers is the answer to this question. Finally, the required data attributes of data included in this interval is subsequently retrieved.

    摘要翻译: 公开了一种用于计算具有数值属性和0-1属性的数据库中的关联规则的方法。 首先,数值属性被划分为多个间隔(或桶),并且根据数值属性的值将每个数据集放置在单个存储桶中。 对每个桶中的数据集数和0-1属性为1的数据集的数量进行计数。 第二,要检测待检测间隔的起始桶。 第三,检测与开始桶对应的终止桶。 也就是说,具有等于或大于预定值的置信度的最大间隔。 第四,检测到的起始和终止桶之一包括最大数量的客户是这个问题的答案。 最后,随后检索包含在该间隔中的数据所需的数据属性。

    Method and apparatus for deriving an optimized confidence rule
    4.
    发明授权
    Method and apparatus for deriving an optimized confidence rule 失效
    用于推导优化置信度规则的方法和装置

    公开(公告)号:US5870748A

    公开(公告)日:1999-02-09

    申请号:US738151

    申请日:1996-10-25

    IPC分类号: G06F12/00 G06F17/17 G06F17/30

    摘要: A method is disclosed for determining the correlation among data sets having a numerical attribute and a 0-1 attribute. First, a numerical attribute is divided into a plurality of buckets, and each data set is placed into a single bucket according to the value of the numerical attribute. The number of data sets in each bucket and the number of data sets with a 0-1 attribute of 1 are counted. Second, an axis corresponding to the total number of data sets in a first through a particular buckets (X axis) and an axis corresponding to the total number of data sets with a 0-1 attribute of 1 in a first through a particular buckets (Y axis) are virtually established, and points corresponding to the respective values of the first through the particular buckets are virtually plotted. Third, after a plane is constructed in this manner, one of the pairs of points separated at an interval of T.times.N or T or larger which has the largest slope is found. This step is most important to fast processing, and this invention employs the nature of convex hulls to reduce the number of points to be considered. Finally, once this pair of points has been determined, the corresponding pair of buckets can be determined, resulting in the output of the corresponding segment. Also, once this process has been finished, the user can retrieve the required part of the data included in this section.

    摘要翻译: 公开了一种用于确定具有数值属性和0-1属性的数据集之间的相关性的方法。 首先,数值属性被划分为多个桶,并且根据数值属性的值将每个数据集放置在单个桶中。 每个桶中的数据集数量和0-1属性值为1的数据集的数量被计数。 其次,对应于在第一至特定桶(X轴)中的数据集合的总数和与第一至特定桶中具有0-1属性1的数据集合的总数对应的轴( Y轴),并且虚拟地绘制对应于第一至特定桶的各个值的点。 第三,在以这种方式构造平面之后,找到以一个具有最大斜率的TxN或T或更大的间隔分开的一对点中的一个。 该步骤对于快速加工是最重要的,本发明使用凸包的性质来减少要考虑的点的数量。 最后,一旦确定了这对点,就可以确定相应的一对桶,从而产生相应段的输出。 此外,一旦该过程完成,用户可以检索本节中包含的数据的必需部分。

    Image segmenting method and apparatus
    9.
    发明授权
    Image segmenting method and apparatus 失效
    图像分割方法和装置

    公开(公告)号:US5729628A

    公开(公告)日:1998-03-17

    申请号:US640304

    申请日:1996-04-30

    申请人: Takeshi Tokuyama

    发明人: Takeshi Tokuyama

    CPC分类号: G06T7/0081

    摘要: A clear partial image of predetermined size can be segmented at high speed. The method for segmenting a partial image from a whole image comprising a plurality of pixels with gray levels comprises the steps of: inputting a number k of pixels of the partial image to be segmented; detecting a partial image maximizing: ##EQU1## where n is the number of pixels of the whole image, g(i, j) is the gray level of a pixel (i, j), .alpha. is a parameter, and .vertline.S.vertline. is the number of pixels of the partial image S; comparing the number .vertline.S.vertline. of pixels of the partial image S maximizing U.sub..alpha. (S) with the inputted number k; increasing a value of .alpha. if .vertline.S.vertline.>k and decreasing the value of .alpha. if .vertline.S.vertline.

    摘要翻译: 可以高速地分割预定尺寸的清晰的部分图像。 用于从包括具有灰度级的多个像素的整个图像分割部分图像的方法包括以下步骤:输入要分割的部分图像的数目k个像素; 检测部分图像最大化:其中n是整个图像的像素数,g(i,j)是像素(i,j)的灰度级,α是参数,| S | 是部分图像S的像素数; 比较数| S | 的部分图像的像素S使输入的数字k最大化U alpha(S); 如果| S |> k则增加α的值,并且如果| S |