Comparative analysis of machine learning techniques in prognosis of type II diabetes

Sarwar, Abid; Sharma, Vinod

doi:10.1007/s00146-013-0456-0

Cited by 32 publications

(24 citation statements)

References 13 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…An application has been developed in [7] where NNs have been trained using data related to potential diabetic patients. On the basis of certain parameters like DoB, Sex, Smoking, Drinking, Thirst, Urination, Height, Weight, Fatigue etc, the application predicts the potential diabetic patient.…”

Section: Resultsmentioning

confidence: 99%

“…Neural networks are used in case where developing solution is either very complex or the possiblility of developing an algorithmic solution is very low. This makes Neural Networks useful in various applications like pattern recognition, prediction, classification, data compression, decision making, and optimization [7,10].…”

Section: Artificial Neural Networkmentioning

confidence: 99%

“…This architecture followed here is often used for ANN in medical research [7]. It's a directed graph of multiple layers of artificial neurons where each layer is connected to next layer.…”

Section: Eq 1: S-shaped Fuzzy Membership Functionmentioning

confidence: 99%

“…The hidden layer allows ANNs to develop its representation of input-output mapping. In case of back propagation, the error data at the output layer is propagated back to earlier layers, thereby allowing incoming weights to these layers be updated and adjusted such that the error between the input and desired output is as least as possible [7]. The topology employed is 10 inputs, one hidden layer and one output in case of Normal and Fuzzified data and is represented as:…”

Section: Eq 1: S-shaped Fuzzy Membership Functionmentioning

confidence: 99%

See 3 more Smart Citations

Preserving Privacy and Optimizing Neural Network Classification by using a Mix of Soft Computing Techniques

Malik¹,

Asger²,

Ali³

et al. 2016

IJCA

View full text Add to dashboard Cite

An Artificial Neural Network (ANN) commonly called as Neural Network (NN) is a mathematical tool that is used in data mining for the analysis of data. Inherently human brain has its own limitations, the inability to handle large volumes of data being one of them. NNs with its power to handle large volumes of data along with imitation of human brain provides good solutions where the size of data is too large, beyond human interpretation capacity or beyond the capacity of conventional computational methods. NNs find applicability in healthcare among other fields like business, genetics, bioinformatics, pharmaceuticals, etc. NNs are mostly not used in data mining due to the reason that it takes lots of time for training of the networks. But as far as the utility of NNs in data mining is concerned, it is a valuable technique. Data mining in general and NNs in particular are being used in various areas where privacy of individuals is at stake. This study attempts to optimize the results of NN classification on one hand and address the problem of training time consumption on the other along with preserving privacy of the stakeholders.

show abstract

Section: Resultsmentioning

confidence: 99%

Section: Artificial Neural Networkmentioning

confidence: 99%

“…This architecture followed here is often used for ANN in medical research [7]. It's a directed graph of multiple layers of artificial neurons where each layer is connected to next layer.…”

Section: Eq 1: S-shaped Fuzzy Membership Functionmentioning

confidence: 99%

Section: Eq 1: S-shaped Fuzzy Membership Functionmentioning

confidence: 99%

See 2 more Smart Citations