专利内容由知识产权出版社提供
专利名称：Method for estimating similarity function
coefficients from object classification data
发明人：Schwanke, Robert W.,Hanson, Stephen J.申请号：EP92107661.8申请日：19920506公开号：EP0513653A3公开日：19940302
专利附图：
摘要：Given a set of objects {A, B, C, ...}, each described by a set of attribute values, andgiven a classification of these objects into categories, a similarity function accounts wellfor this classification only a small number of objects are not correctly classified. This is
obtained when coefficients are found for the similarity function which result in an errorrate that is considered to be an acceptable level. A method for estimating a coefficientfor a similarity function comprises the steps of: (a) selecting an initial value w = w as theinitial coefficient of a similarity function SIM(a,b); (b) selecting an initial value k = k definingthe number of most-similar objects to use when testing the classification of an object; (c)computing similarity measures by using the current estimate of SIM for an object S toneighboring objects G in the given category and for the object S to neighboring objects Boutside the given category, and forming a training set of respective data 〈S, SIM(S,G),SIM(S,B)〉 for members in the category and those outside the category; (d) reducing thetraining set by eliminating all data except for the k-nearest (most similar) objects of Gand B; (e) estimating new coefficients for the similarity function by adjusting w to minimizean error rate measured in terms of the extent to which neighboring objects B outside thecategory have higher similarity measures than neighboring objects G within the category;and (f) repeating steps (c), (d), and (e) until the error rate is reduced below apredetermined low level.
申请人：SIEMENS AKTIENGESELLSCHAFT
地址：Wittelsbacherplatz 2 D-80333 München DE
国籍：DE
更多信息请下载全文后查看
