Maintaining nutrition for glycemic (diabetic) patients in order to retain the blood glucose level is one of the important activity to be followed. Stimulating the amount of carbohydrates, protein, vitamins, and minerals will result in a healthy diet. So, there is a necessity for recommendation of nutrition to those diabetic patients nowadays. Recommender Systems (RS) play a vital role in urging relevant suggestions to the users. To promote improvised and optimized results, Optimization technique plays a significant role in refining the parameters of chosen algorithm. To optimize and to upgrade the accuracy of recommendations, the system has been developed by implementing improved Krill-Herd algorithm. The system which clusters the profiles of diabetic patients using improved k-means clustering algorithm and results has been optimized using Improved Krill-Herd optimization algorithm. The performance will be analysed using different measures like Precision, Recall, F-measure, Accuracy, Matthews correlation, Fallout rate and Miss rate. The evaluation results show that the proposed system performs better and produces optimized results to the diabetic patients with minimum error rate.
K Renuka Devi, J Bhavithra, A Saradha
Dr. Mahalingam College of Engineering and Technology, India
Data Mining, Diabetes Patients, Recommender Systems, Clustering Algorithm, Improved K-Means, Krill Herd Optimization | Published By :
ICTACT
Published In :
ICTACT Journal on Soft Computing
( Volume: 10 , Issue: 3 )
Date of Publication :
April 2020
|