The disease with unknown cause is Rheumatoid Arthritis (RA). The progression of this disease can be prevented by the early diagnosis of it. The treatment and early diagnosis of RA can be explored in medical data mining field. In recent days, Fuzzy Cognitive Maps (FCMs) is a soft computing technique developed as an advanced decision support tool. This problem can be modelled by using Particle Swarm Optimization (PSO) and FCMs in association with the knowledge of medical experts. The severity of RA disease is computed using the same. But still the performance of the PSO is not enhanced because of the local optima issue. Ant Colony Optimization (ACO) is proposed to FCM clustering. The casual relationships among gene samples and classes can be described by FCM. The dynamic behavior of a system is determined by this. The severity of RA disease is computed by ACO Algorithm which is combined with FCMs termed as Optimized Ant Colony Based Fuzzy Cognitive Maps (OACFCM) along with E-GEOD-64707 dataset. Computational problems are solved by a probabilistic technique called Ant Colony Optimization Algorithm (ACO). Improved adjacency matrix is contained by ACO algorithm and produces better performance in searching when compared to FCM. The searching is done by two evaluations in ACO and it has static and dynamic. Proposed OACFCM system and existing methods are evaluated using a parameters like accuracy, f-measure, precision, and recall. Results of proposed method OACFCM are compared with RAF, FCM-PSO. These methods are implemented via the use of the Matrix Laboratory (MATLAB) which can assist gene samples in predicting the patients with RA accurately in early stages.
Volume 11 | 11-Special Issue
Pages: 168-175
DOI: 10.5373/JARDCS/V11SP11/20192944