Improving the Prediction Rate of Diabetes Diagnosis Using Fuzzy, Neural Network, Case Based (FNC) Approach

Thirugnanam, Mythili; Kumar, Praveen; Srivatsan, Shruti; Nerlesh, C.R.

doi:10.1016/j.proeng.2012.06.208

Cited by 39 publications

(17 citation statements)

References 8 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…Second stage applied rule based algorithms on the outcomes from first stage. The study concluded that the proposed method gives the high classification rate to predict diabetes 58 . Jefri Junifer Pangaribuan et al, (2014) conducted the study to diagnose diabetes in women aged 21 years.…”

Section: Fig 8: Rate Of Accuracies Achieved Using Data Mining Technimentioning

confidence: 87%

Untitled

2018

IJPSR

View full text Add to dashboard Cite

Diabetes is one of the deadliest and non-contagious diseases that can adversely affect several parts of human body. Early prognosis of diabetes can inkling the grievous complications and help to save human life. Several researchers have used different data mining (Iterative Dichotomiser 3, Random Forest, Support Vector Machine, k-Nearest Neighbour, C4.5) and soft computing (Genetic Algorithm, Ant Colony Optimization, Particle Swarm optimization, Artificial Bee Colony) techniques to prospect diabetes in human beings. In last 10 years, C4.5 was the most preferred choice for mining diabetic patients. Likewise, in soft computing, maximum number of researchers have used genetic algorithm. Furthermore, the usage of pre-processing techniques is significantly increasing in diabetes diagnosis. It is also observed that rate of accuracy achieved in diagnosing diabetes using traditional data mining lies in 68.5%-95.3%. Likewise, the range for soft computing and their hybridized use lies in 74%-100%. In addition, rate of accuracy achieved using GA based hybridized approach is better than the accuracy obtained using PSO as well as ABC. Most of the researchers have used textual and numeric data for diabetes diagnosis. Few researchers have used images for the same. However, no significant research is found where diabetes has been diagnosed using audio or sound. Moreover, the diagnostic results obtained using image based data are not as good as obtained using textual or discrete data. Therefore, an attention is still obligatory to develop smart diabetes diagnostic system that can effectively work on different types of data like text, images as well as sound. INTRODUCTION: Diabetes is a persistent metabolic and pancreas affected human disorder. The devastating rise in obesity and sedentary life style has made it as a universal epidemic 1, 2. Numbers of individuals are getting affected by different types of diabetes. Human body changes most of the food into glucose and it is the primary source of energy in the body. Pancreas assists in digestion and releases a hormone called insulin that transforms glucose into body cells.

show abstract

Section: Fig 8: Rate Of Accuracies Achieved Using Data Mining Technimentioning

confidence: 87%

Untitled

2018

IJPSR

View full text Add to dashboard Cite

show abstract

“…However, there are no real studies in the literature for fuzzy-CBR systems for diabetes diagnosis. Thirugnanam et al [39] built a hybrid CDSS system for diabetes diagnosis using a neural network, fuzzy, and CBR. This study used the fuzzy and CBR reasoning mechanisms separately, and no fuzziness has been added to enhance the CBR functionality.…”

Section: Regarding the Fuzzification Of Medical Datamentioning

confidence: 99%

A fuzzy-ontology-oriented case-based reasoning framework for semantic diabetes diagnosis

El–Sappagh

Elmogy

Riad

2015

Artificial Intelligence in Medicine

View full text Add to dashboard Cite

“…In this paper [2] of "Improving prediction rate of diabetes Diagnosis using Fuzzy, neural, Case based approach "author MythilTirugnanam,Dr Praveen kumar proposed new approach would called FNC approach using newly design influenced parameter. Technique gave better resultIn this The above survey indicate different technique used for prediction of diabetes but all system had nuber of problem and prediction rate not appreciable.so new system is proposed that improve prediction rate.…”

Section: Related Workmentioning

confidence: 99%

Prognosis of Diabetes using Neural Network, Fuzzy Logic, Gaussian Kernel Method

Giri¹,

Todmal²

2015

IJCA

View full text Add to dashboard Cite

In Today's world there is an increase in the prevalence of diabetes mellitus and therefore the disease is recognising as a major global public health problem.medical data mining extracts hidden patterns from medical data. This is to design system for diabetes prediction.. The soft computing technique is most useful and powerful technique used for diagnosis purpose. The proposed system a novel approach for diagnosis of diabetes which has two stages to predict the diabetes status. Initial Stage we are using Gaussian kernel function which help to distribution of data and second stage adopt two computational intelligence and knowledge engineering technique such as fuzzy logic and neural network. The benefit applying these is that accuracy of prediction rate will be higher than most of the suggested system for predicting the occurrence of diabetes mellitus. The dataset used for the Experimental is based on Pima Indian Dataset from University of California.

show abstract

Improving the Prediction Rate of Diabetes Diagnosis Using Fuzzy, Neural Network, Case Based (FNC) Approach

Cited by 39 publications

References 8 publications

Untitled

Untitled

A fuzzy-ontology-oriented case-based reasoning framework for semantic diabetes diagnosis

Prognosis of Diabetes using Neural Network, Fuzzy Logic, Gaussian Kernel Method

Contact Info

Product

Resources

About