Fuzzy and Probabilistic Clustering with Shape and Size Constraints
نویسندگان
چکیده
More sophisticated fuzzy clustering algorithms, like the Gustafson–Kessel algorithm [11] and the fuzzy maximum likelihood estimation (FMLE) algorithm [10] offer the possibility of inducing clusters of ellipsoidal shape and different sizes. The same holds for the expectation maximization (EM) algorithm for a mixture of Gaussians. However, these additional degrees of freedom can reduce the robustness of the algorithm, thus sometimes rendering their application problematic. In this paper we suggest methods to introduce shape and size contraints that handle this problem effectively.
منابع مشابه
Support vector regression with random output variable and probabilistic constraints
Support Vector Regression (SVR) solves regression problems based on the concept of Support Vector Machine (SVM). In this paper, a new model of SVR with probabilistic constraints is proposed that any of output data and bias are considered the random variables with uniform probability functions. Using the new proposed method, the optimal hyperplane regression can be obtained by solving a quadrati...
متن کاملGenerating Optimal Timetabling for Lecturers using Hybrid Fuzzy and Clustering Algorithms
UCTTP is a NP-hard problem, which must be performed for each semester frequently. The major technique in the presented approach would be analyzing data to resolve uncertainties of lecturers’ preferences and constraints within a department in order to obtain a ranking for each lecturer based on their requirements within a department where it is attempted to increase their satisfaction and develo...
متن کاملMulti-granulation fuzzy probabilistic rough sets and their corresponding three-way decisions over two universes
This article introduces a general framework of multi-granulation fuzzy probabilistic roughsets (MG-FPRSs) models in multi-granulation fuzzy probabilistic approximation space over twouniverses. Four types of MG-FPRSs are established, by the four different conditional probabilitiesof fuzzy event. For different constraints on parameters, we obtain four kinds of each type MG-FPRSs...
متن کاملPublished in the Proceedings of “Information Processing and Managing of Uncertainty” (IPMU’2002) pages 225-232. A new fuzzy clustering technique based on pdf estimation
Among the set of clustering methods that do not make assumptions concerning the size and shape of clusters, we consider a method based on the estimation of the probability density function. We show that within this framework, crisp clustering can also be extended to fuzzy clustering. We illustrate the usefulness of this extension using the example of singleand multi-component image segmentation...
متن کاملRefining membership degrees obtained from fuzzy C-means by re-fuzzification
Fuzzy C-mean (FCM) is the most well-known and widely-used fuzzy clustering algorithm. However, one of the weaknesses of the FCM is the way it assigns membership degrees to data which is based on the distance to the cluster centers. Unfortunately, the membership degrees are determined without considering the shape and density of the clusters. In this paper, we propose an algorithm which takes th...
متن کامل