Every day, enormous amounts of intricate information are created in numerous fields. Data sets that are so vast that traditional database administration and data analysis technologies are unable to handle them are referred to as complex information. Medical big data management and analysis involve numeral experiments through the organization, storage, and analysis of the data. To avoid the problem, in this paper adaptive honey badger (AHB) algorithm with the selected features is presented for dimensionality reduction. The proposed approach consists of three phases such as pre-processing, feature selection and classification. At first, the medical data are fed to the pre-processing phase to eliminate the missing values and redundant data. Then, to minimize computational complexity, time consumption and storage, the optimal features are selected using the AHB algorithm. Then, the selected features are fed to the adaptive deep neural network (ADNN) classifier to classify data as normal or diseased data. The experimental results show that our presented model is attained an average accuracy of 96.8% for the cervical cancer dataset and 98% for the Wisconsin cancer dataset.

Cervical cancer, adaptive honey bee, ADNN, LSTM, CNN, attention mechanism, dimension reduction and big data.