Performance Analysis of Classifier Models to Predict Diabetes Mellitus

Kandhasamy, J. Pradeep; Balamurali, S.

doi:10.1016/j.procs.2015.03.182

Cited by 210 publications

(92 citation statements)

References 5 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…A good research work by [18] has proposed a PPG-based research work to assist the clinicians in diabetes screening and for the adopting of a suitable treatment plan towards the preventing of end-organ damage. In seeking the development of good and reliable classifiers for the prediction of diabetes, several classifiers were developed [19][20][21]. A study by [22] claimed that they can use a deep learning algorithm for the detection of a prevalent diabetes by the analysis of PPG signal with a plausible difference.…”

Section: Literature Reviewsmentioning

confidence: 99%

Neural Network-based Diabetic Type II High-Risk Prediction using Photoplethysmogram Waveform Analysis

Qawqzeh¹

2019

IJACSA

View full text Add to dashboard Cite

This work aims to predict and classify patients into diabetic and nondiabetic subjects based on age and four independent variables extracted from the analysis of photoplethysmogram (PPG) morphology in time domain. The study has two main stages, the first one was the analysis of PPG waveform to extract b/a, RI, DiP, and SPt indices. These parameters contribute by some means to the prediction of diabetes. They were statistically significant and correlated with the HbA1C test. The second stage was building a neural network based classifier to predict diabetes. The model showed an accuracy of 90.2% in training phase and an accuracy of 85.5% in testing phase. The findings of this research work may contribute towards the prediction of diabetes in early stages. Also, the proposed classifier showed a high accuracy in predicting the existence of diabetes in Saudi people population.

show abstract

Section: Literature Reviewsmentioning

confidence: 99%

Neural Network-based Diabetic Type II High-Risk Prediction using Photoplethysmogram Waveform Analysis

Qawqzeh¹

2019

IJACSA

View full text Add to dashboard Cite

show abstract

“…Pradeep et al [5]in this study the researchers concentrate on different datasets including Diabetes Dataset(DD). The researchers have investigated and constructed the universally good models and capability for varies/different medical datasets (MDs).…”

Section: Related Workmentioning

confidence: 99%

Minimal Rule-Based Classifiers using PCA on Pima-Indians-Diabetes-Dataset

Reddy*¹,

Nagendra²

2019

IJITEE

View full text Add to dashboard Cite

Diabetes mellitus has been a very complex and chronic lifelong disease. Hence, it has been identified as a clinically highly significant disease which further attracted the healthcare industry to define most relevant clinical directories and also to apply efficient automated prediagnoses and further care. Rule-based classifier technique has proven its strength in diabetes diagnosis when used the computational methods. There has been a considerable development in the classifier's performance, to classify the disease, in the past decades by which they were highly recommended. In this paper, a classifier is proposed based on minimum rules which further uses Principal Component Analysis (PCA). To apply these techniques Pima Indians diabetes dataset is used from the UCI Machine LearningRepository. Set of experiments were conducted on the data set with PCA to evaluate the performance among the decision tree, Naïve Bayes, and Support Vector Machine.

show abstract

“…Data baru akan diklasifikaikan ke dalam kelas dari mayoritas data training (neighbors) yang terdekat [5]. Jarak data baru dengan neighbors dapat diukur dengan euclidean distance disajikan pada persamaan (1) [11].…”

Section: K-nearest Neighbors (Knn)unclassified

“…Pengujian dengan menghitung accuracy, sensitivity dan specificity. Nilai accuracy 100 % diperoleh dari algoritma KNN dengan k=1 dan random forest [5]. Kinerja KNN juga terbukti unggul dibandingkan dengan binary logistic regression, multilayer perceptron dimana dengan observasi sebanyak 100 data dengan 7 atribut yaitu PG concentration, Diastolic BP, Tri Fold Thick, Serum Ins, BMI, DP function, age dan disease.…”

unclassified

Implementasi Web Service Dan Analisis Kinerja Algoritma Klasifikasi Data Mining Untuk Memprediksi Diabetes Mellitus

Setyawan

Suradi

2017

Simet

View full text Add to dashboard Cite

Salah satu penyakit yang ditimbulkan akibat kesalahan pola gaya hidup adalah Diabetes Mellitus (DM). Gejala penyakit diabetes sering dilalaikan oleh kebanyakan orang, sehingga mereka cenderung untuk mengabaikannya dan tidak mau melakukan medical check up. Di Indonesia jumlah penderita DM terus mengalami peningkatan dari tahun ke tahun. World Health Organization (WHO) memperkirakan jumlah penderita DM tipe 2 di Indonesia akan mengalami peningkatan secara signifikan hingga 21,3 juta jiwa pada tahun 2030 mendatang. Ternyata dengan bantuan ilmu data mining, data pasien diabetes dapat digunakan untuk memprediksi apakah sesorang positif diabetes atau tidak. Tahapan awal dilakukan preprocessing data untuk menangani missing dan non numeric values. Kemudian traning dan testing menggunakan k-fold cross validation dengan algoritma K-Nearest Neighbors (KNN), random forest dan naive bayesian. Pengujian dilakukan dengan menghitung accuracy, sensitivity dan specificity. Dari hasil uji 10-fold cross validation diperoleh rata-rata akurasi tertinggi ketika menggunakan naive bayesian yaitu 75,65%, sedangkan KNN 75,53% dan random forest 73,69%. Perhitungan sensitivity dan specificity dengan membagi 786 data menjadi 594 data training dan 192 data testing. Untuk KNN diperoleh sensitivity 56,72% dan specificity 78,68%, random forest diperoleh sensitivity 53,73% dan specificity 86,4%, sedangkan naive bayesian diperoleh sensitivity 62,69% dan specificity 84%. Implementasi restful web service diterapkan pada model dengan akurasi tertinggi yaitu naive bayesian dengan format json sebagai return value.

show abstract

Performance Analysis of Classifier Models to Predict Diabetes Mellitus

Cited by 210 publications

References 5 publications

Neural Network-based Diabetic Type II High-Risk Prediction using Photoplethysmogram Waveform Analysis

Neural Network-based Diabetic Type II High-Risk Prediction using Photoplethysmogram Waveform Analysis

Minimal Rule-Based Classifiers using PCA on Pima-Indians-Diabetes-Dataset

Implementasi Web Service Dan Analisis Kinerja Algoritma Klasifikasi Data Mining Untuk Memprediksi Diabetes Mellitus

Contact Info

Product

Resources

About