“…The same approach can be used in two class problems where training examples are available only for one class and observations which do not pass a predefined confidence level are assigned to the unknown class (e.g. in detection of motor failure conditions [12]). In that sense the confidence does not refer to a single value limit but to a certain allowed region in the feature space.…”
Section: Interpretation Of Confidencementioning
confidence: 98%
“…[11,12]), the assumption of single component leads to strict requirements for characteristics of the phenomenon: a single basic class which smoothly varies around the class mean. The most significant problem is not typically the smooth behavior but the assumption of unimodality.…”
“…The same approach can be used in two class problems where training examples are available only for one class and observations which do not pass a predefined confidence level are assigned to the unknown class (e.g. in detection of motor failure conditions [12]). In that sense the confidence does not refer to a single value limit but to a certain allowed region in the feature space.…”
Section: Interpretation Of Confidencementioning
confidence: 98%
“…[11,12]), the assumption of single component leads to strict requirements for characteristics of the phenomenon: a single basic class which smoothly varies around the class mean. The most significant problem is not typically the smooth behavior but the assumption of unimodality.…”
“…GMMs have been widely used in all kinds of pattern recognition problems, including speech processing [21], human skin detection [22], realtime tracking [23], hazardous chemical agents detection [24], and bearing damage detection for induction motors [25]. The advantage of GMM is its modeling ability.…”
“…The first-order statistics approach is not sufficient since it simply selects the frequency band where the distance between the expectations is largest, but neglects the variance information, and thus, a significant overlap of the class probabilities may exist [5,4]. In the second-order statistics typically an assumption must be made about forms of probability distributions p x (n) and p y (n) which describe the spread of features of the both classes.…”
Section: Statistical Measures For Discriminative Powermentioning
Abstract. In this study a method for automatic motor condition diagnosis is proposed. The method is based on a statistical discriminance measure which can be used to select the most discriminative features. New signals are classified to either a normal condition class or a failure class. The classification can be done traditionally using training examples from the both classes or using only probability distribution of the normal condition samples. The latter corresponds to typical situations in practice where the amount of failure data is insufficient. The results are verified using real measurements from induction motors in normal condition and with bearing faults.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.