Abstract
The proposed relational fuzzy clustering method, called FRFP (fuzzy relational fixed point), is based on determining a fixed point of a function of the desired membership matrix. The ethod is compared to other relational clustering methods. Simulations show the method to be very effective and less computationally expensive than other fuzzy relational data clustering methods. The membership matrices that are produced by the proposed method are less crisp than those produced by NERFCM and more representative of the proximity matrix that is used as input to the clustering process.
Highlights
The first stage of knowledge acquisition concerning a group of objects is to partition or divide the objects into groups based either on individual object (IO) data such as feature vectors or on inter-object (OR) data incorporated in proximity matrices
As the ultimate test of clustering quality, the membership matrices that are produced through automatic clustering based on the proximity matrix, are examined to determine how well they correspond to the visual representation of the proximity matrices produced by multidimensional scaling (MDS)
A method (FRFP) of relational clustering that is based on solving a function of the membership matrix for a fixed point is in several cases superior to non-euclidian relational fuzzy c-means (NERFCM)
Summary
Simulations are done to compare the following 4 methods consisting of the proposed method and three existing methods. 1. Proposed - FRFP with average distance 2. For fuzzy relational fixed point (FRFP) clustering all 3 formulations expressed by (15), (16) and (17) were tested. The results reported here are with the use of (17) which is solved iteratively starting with a random matrix for M. The methods are compared on the basis of visual clustering and the proposed quantitative measure. As the ultimate test of clustering quality, the membership matrices that are produced through automatic clustering based on the proximity matrix, are examined to determine how well they correspond to the visual representation of the proximity matrices produced by MDS. The quantitative clustering quality measure is in turn validated by comparing its value to the result of the qualitative visual comparison
Sign-in/Register to view highlights & summary Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a callMore From: International Journal of Computational Intelligence Systems
Disclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.
