Entropy-Reduced Transformation Approach to Pattern Recognition of Complex Data Set
نویسندگان
چکیده
The term of complex data set indicates that the target da ta set t o be recognized consists of multiple categories of pattern samples. Recognition of complex da ta set is a challenging research topic in computer vision and pattern recognition. The entropy reduction approach has been widely used t o solve the problem of recognition of single category data set. In this paper we generalize this concept in terms of Entropy-Reduced Transformation (ERT) which contains several important properties which enable us t o produce the concrete solution for the practical applications. Validation of the generalized approach is demonstrated by an example. Index Terms : Pattern recognition, complex da ta set, large set, entropy-reduced, t,ransformation, multiple categories text.
منابع مشابه
Prediction of dispersed mineralization zone in depth using frequency domain of surface geochemical data
Discrimination of the blind and dispersed mineralization deposits is a challenging problem in geochemical exploration. The frequency domain (FD) of the surface geochemical data can solve this important issue. This new exploratory information can be achieved using the interpretation of FD of geochemical data, which is impossible in spatial domain. In this research work, FD of the surface geochem...
متن کاملPrediction of mineral deposit model and identification of mineralization trend in depth using frequency domain of surface geochemical data in Dalli Cu-Au porphyry deposit
In this research work, the frequency domain (FD) of surface geochemical data was analyzed to decompose the complex geochemical patterns related to different depths of the mineral deposit. In order to predict the variation in mineralization in the depth and identify the deep geochemical anomalies and blind mineralization using the surface geochemical data for the Dalli Cu-Au porphyry deposit, a ...
متن کاملINFORMATION MEASURES BASED TOPSIS METHOD FOR MULTICRITERIA DECISION MAKING PROBLEM IN INTUITIONISTIC FUZZY ENVIRONMENT
In the fuzzy set theory, information measures play a paramount role in several areas such as decision making, pattern recognition etc. In this paper, similarity measure based on cosine function and entropy measures based on logarithmic function for IFSs are proposed. Comparisons of proposed similarity and entropy measures with the existing ones are listed. Numerical results limpidly betoken th...
متن کاملIdentifying Useful Variables for Vehicle Braking Using the Adjoint Matrix Approach to the Mahalanobis-Taguchi System
The Mahalanobis Taguchi System (MTS) is a diagnosis and forecasting method for multivariate data. Mahalanobis distance (MD) is a measure based on correlations between the variables and different patterns that can be identified and analyzed with respect to a base or reference group. MTS is of interest because of its reported accuracy in forecasting small, correlated data sets. This is the type o...
متن کاملWeighted similarity measure on interval-valued fuzzy sets and its application to pattern recognition
A new approach to define the similarity measure betweeninterval-valued fuzzy sets is presented. The proposed approach isbased on a weighted measure in which the normalized similaritiesbetween lower functions and also between upper functions arecombined by a weight parameter. The properties of this similaritymeasure are investigated. It is shown that, the proposed measurehas some advantages in c...
متن کامل