MapReduce-based big data framework using modified artificial neural network classifier for diabetic chronic disease pred
- PDF / 1,368,602 Bytes
- 11 Pages / 595.276 x 790.866 pts Page_size
- 98 Downloads / 185 Views
(0123456789().,-volV)(0123456789(). ,- volV)
METHODOLOGIES AND APPLICATION
MapReduce-based big data framework using modified artificial neural network classifier for diabetic chronic disease prediction R. Ramani1 • K. Vimala Devi2 • K. Ruba Soundar1
Springer-Verlag GmbH Germany, part of Springer Nature 2020
Abstract Recently, healthcare data consist of an enormous amount of information, which is challenging to maintain by manual methods. Due to the development of big data in the communities of biomedical and health care, accurate study of the medical data helps the recognition of the disease in early stage, patient care and community services. It mainly focuses on predicting and exploring the conditions due to some significant effects on health which are on the increase in multiple cities. The existing system in the medical field cannot extract complete information from the chronic disease database. It is complicated for the healthcare practitioner to analyze and diagnose constant disease since it plays a challenging task. This paper presents a modified artificial neural network (ANN) classifier technique with a MapReduce framework for the prediction of disease. For preprocessing, min–max normalization is carried out to enhance the accuracy of system. This MapReduce is applied for providing a feasible framework in predictive programming algorithms for the map and reduce functions. This is a simple programming interface, which helps in efficiently solving predictive problems. The primary intention of the proposed system is to analyze accurate, fast and optimal results on chronic disease datasets. It increases the throughput and redundancy in cases of retrieving the vast data. Thus, integrating a modified ANN classifier with a reduced framework is useful in providing better outcomes. The experimental results over chronic diabetic dataset prove that the proposed artificial neural network with MapReduce structure is capable of predicting the precision, sensitivity and specificity level modified on comparing with other existing deep neural network approaches. Keywords MapReduce Diabetic chronic disease Feature selection Aggregation function Disease prediction Artificial neural networks and big data
1 Introduction Diabetic chronic disorders are the primary health challenge nowadays and the most significant reason for causing death in the world. For the past few years, the leading diabetic chronic diseases such as cardiovascular disease, cancer, persistent respiratory disease, chronic kidney disease and diabetes cause around 29 million of the deaths worldwide. Chronic illness is a physical, mental or emotional condition
Communicated by V. Loia. & R. Ramani [email protected] 1
Department of Computer Science and Engineering, P. S. R. Engineering College, Sivakasi, India
2
School of Computer Science and Engineering, Vellore Institute of Technology, Vellore 632014, India
that can be fixed for the significant period. Chronic diseases have extensive effects that can be extended beyond the dull physi
Data Loading...
