Heusler alloys are an incredible class of inter-metallic materials with different compositions and over 1500 members. Though discovered a century back, they are an active area of physics and material science research. Novel properties and potential fields of applications materialize constantly. Even the alloy system is extensively investigated owing to its shape memory behavior and prospective relevance in the development of actuator devices, where strains are controlled by applying an external magnetic field. Heusler alloys are currently the material of interest due to their properties leading to their use as shape memory alloys and topological insulators. Hence, predicting and determining their composition and structure is imperative before synthesis. Utilizing the conventional method in determining the possible changes in the properties and the structure of the proposed compositions is tedious and time-consuming. In the current consumerism-driven environment, we require a faster method to predict the structure of the proposed alloy or compound or other parameters for the desired application. Once the prediction is made, it must be tested experimentally by synthesizing the material and characterizing its behavior. This analysis is focusing on network analysis with a supervised machine learning approach to study the properties of Heusler alloys with their application as shape memory alloys.