Adamic–Adar similarity indexed Wald boost data classification for diabetic disease diagnosis with big data

Big data contains vast amounts of information that can be processed in the future, leading to improved prediction outcomes. In the realm of big healthcare data, analyzing and preventing hospital errors, along with disease prevention and cost savings, are crucial. Healthcare data analysts play a pivotal role in enhancing healthcare outcomes by gathering, integrating, and analyzing data from diverse sources. Disease diagnosis involves identifying the disease or condition underlying a person's symptoms and signs. Ensemble learning methods, utilized in science and technology, have introduced various classification and clustering techniques to achieve more accurate and time-efficient disease diagnosis. To address these challenges, we introduce the Adamic–Adar Similarity Indexed Wald Boost Data Classification (AASIWBDC) Technique. This technique conducts feature selection and data classification to enhance the efficiency of diabetic disease diagnosis. It utilizes the Adamic–Adar Similarity Index to gauge the similarity between features and their objectives. Leveraging selected features, the AASIWBDC Technique implements the Wald Boost Classification Process to categorize patient data into normal and abnormal groups with greater accuracy. As a result, it achieves effective diabetic disease diagnosis in an efficient manner. We evaluate the performance of the AASIWBDC Technique in terms of diabetic disease diagnosis accuracy, diagnosis time, and error rate using the Diabetes 130-US hospitals dataset from 1999–2008. Experimental results illustrate that the AASIWBDC Technique enhances diabetic disease diagnosis accuracy, reduces diagnosis time, and lowers error rates compared to conventional methods.