Abstract

Recently, the health sector is widely adopting artificial intelligence models such as machine learning (ML), deep learning for data analysis, disease prediction, and disease classification. However, the conventional models failed to analyze the data. Therefore, this work is focused on analysis of diabetes prediction using extreme learning machine (DP-ELM) model. Initially, Pima Indian diabetes is considered, which is pre-processed for missing data symbols identification. Then, the statistical features from pre-processed dataset are extracted using principal component analysis (PCA). Then, ELM model is trained with the PCA features and forms the trained feature dataset. Then, a random test combination is applied for ELM testing, which classifies the positive and negative status of diabetes. The simulations proved that; the proposed DP-ELM outperformed in terms of accuracy as compared to existing methods.

Description