Analyzing the Deep Learning Techniques Based on Three Way Decision Under Double Hierarchy Linguistic Information and Application

In the real life world, the multi-criteria group decision-making (MCGDM) is very important in the decision process. The intelligent decision process for deep learning is playing a very vital role in the selection of deep learning technique for health care. The main aim of the present work is develop an intelligent based three way decision models for analyzing and ranking the deep learning techniques for health care under the double hierarchy linguistic information. The double hierarchy linguistic term set (DHLTS), which enables more adaptable representation of uncertainty and fuzziness. The Yager operational laws for DHLTSs has considered and defined to develop the Yager aggregation operators for DHLTSs. We defined double hierarchy linguistic Yager weighted averaging (DHLYWA) operators, double hierarchy linguistic Yager hybrid aggregation (DHLYHA) operators, double hierarchy linguistic Yager geometric aggregation (DHLYWG) operator, double hierarchy linguistic Yager geometric aggregation (DHLYHG) operator. Next, three way intelligent decision model developed by using the proposed aggregation operators with the help of EDAS decision making. The three way decision making is based on the loose function. The proposed three way intelligent decision making model apply to deep learning techniques to ranking and analyze the K-nearest neighbor, support vector machines, decision tree, A 4: naive Bayes, linear regression, and k-means clustering. The entropy method is used to determine the unknown weights of the decision expert and criterion. Then, we find the conditional probability using the EDAS method. The expected losses are then calculated by combining the loss functions using the double hierarchy linguistic Yager aggregation operators. The proposed models compare with already developed models and also provide the sensitivity analysis of the proposed models.