Enhancing topological index of calcium chloride network through feature selection methods exploration.

With the chemical formula CaCl, calcium chloride is a salt as well as an inorganic material. At room temperature, it has the consistency of a white, crystalline solid and is very water-soluble. It can be created by neutralizing calcium hydroxide with hydrochloric acid. Calcium chloride is a solution with a large enthalpy change. It is extensively utilized in research facilities, manufacturing facilities, and pharmaceuticals, including all types of food-graded applications, the treatment of acute illnesses, packaging for drying tubes, dust controllers, and de-icing, among other uses. In this paper, firstly we compute the topological indices, coindices, and reverse indices of CaCl. Further, we employ machine learning strategies to capture the best suitable set of indices for the proximity of the prediction of distinct physio-chemical properties of CaCl. To strengthen the results, different regression techniques are implemented to predict HOF of CaCl based on our features, and the most influential features were detected to verify our results.