A statistical approach to class separability
نویسندگان
چکیده
We propose a new statistical approach for characterizing the class separability degree in . This approach is based on a nonparametric statistic called “the Cut Edge Weight”. We show in this paper the principle and the experimental applications of this statistic. First, we build a geometrical connected graph like Toussaint’s Relative Neighbourhood Graph on all examples of the learning set. Second, we cut all edges between two examples of a different class. Third, we compute the relative weight of these cut edges. If the relative weight of the cut edges is in the expected range of a random distribution of the labels on all the neighbourhood of the graph’s vertices, then no neighbourhood-based method provides a reliable prediction model. We will say then that the classes to predict are non-separable.
منابع مشابه
Evaluation of Tests for Separability and Symmetry of Spatio-temporal Covariance Function
In recent years, some investigations have been carried out to examine the assumptions like stationarity, symmetry and separability of spatio-temporal covariance function which would considerably simplify fitting a valid covariance model to the data by parametric and nonparametric methods. In this article, assuming a Gaussian random field, we consider the likelihood ratio separability test, a va...
متن کاملانجام یک مرحله پیش پردازش قبل از مرحله استخراج ویژگی در طبقه بندی داده های تصاویر ابر طیفی
Hyperspectral data potentially contain more information than multispectral data because of their higher spectral resolution. However, the stochastic data analysis approaches that have been successfully applied to multispectral data are not as effective for hyperspectral data as well. Various investigations indicate that the key problem that causes poor performance in the stochastic approaches t...
متن کاملA heuristic multi-criteria classification approach incorporating data quality information for choropleth mapping.
Despite conceptual and technology advancements in cartography over the decades, choropleth map design and classification fail to address a fundamental issue: estimates that are statistically indifferent may be assigned to different classes on maps or vice versa. Recently, the class separability concept was introduced as a map classification criterion to evaluate the likelihood that estimates in...
متن کاملA comparison of three class separability measures
Measures of class separability can provide valuable insights into data, and suggest promising classification algorithms and approaches in data mining. We compare three simple class separability measures used in supervised machine learning. Their relative effectiveness is evaluated through their functional relationships and their random projections of data onto R for visualization. We conclude t...
متن کاملSeath - a New Tool for Automated Feature Extraction in the Context of Object-based Image Analysis
In order to avoid the time-consuming trial-and-error practice for seeking significant features for optimal class separation in object-based classification, an automatic feature extraction methodology, called SEaTH has been developed. SEaTH calculates the SEperability and the corresponding THresholds of object classes for any number of given features on the basis of a statistical approach. The s...
متن کامل