Due to the uncertainty of human subjective judgments, it is sometimes very difficult to obtain accurate evaluation information. The technique for order preference by similarity to ideal solution (TOPSIS) is one of the functional methods used to solve the multicriteria decision‐making (MCDM) problems. However, the traditional TOPSIS is based only on a distance measure and takes into account neither a similarity measure nor a likelihood measure. Therefore, in this article, we define two new information measures which are called as the projection and divergence and more functional than some other existing ones in the neutrosophic sets. Then we develop an innovative TOPSIS (DPL‐TOPSIS) based on the hybrid closeness coefficient which is an optimization of the divergence (distance), projection (similarity), and likelihood (magnitude) closeness coefficients that can also use separately to make the decision in neutrosophic environment. To develop the DPL‐TOPSIS, we define three kinds of analogical factors: the distance‐like positive and negative divergence decision‐making factors, the similarity‐like positive and negative projection decision‐making factors, and the magnitude‐like positive and negative likelihood decision‐making factors. In addition, we provide an objective weight determination model to determine expert judgment with the proposed divergence measure. Finally, to demonstrate the functionality of the developed DPL‐TOPSIS, an application has been made on selection of the masks used as one of the protection methods in the COVID‐19 epidemic, which has recently affected our world, and the results are compared with other existing methods.