Published Paper Details:

Probabilistic graphical models, often known as PGMs, provide a strong framework for modelling intricate interactions between various components. Through the incorporation of data about the elemental composition and structural characteristics, these models make it possible to deduce the attributes of materials from a probabilistic point of view. This technique bears promise attempts towards expediting the design of materials discovery, as it makes it easier to predict the characteristics of a wide variety of materials, including their electrical and mechanical properties, as well as their thermal and optical behaviour. The use of PGMs in the field of materials science is an example of a sophisticated approach that is used to harness data-driven insights in order to direct the discovery of novel materials that have specific functions. The objective of this study is to conduct a literature review with the intention of examining the possible applications of data science ideas, big data, and machine learning in the context of the production of artificial intelligence. This is a strategy that involves reviewing the existing literature in order to get an understanding of the use and application of computational intelligence in the cutting-edge research and innovation in materials science. Using machine learning to solve complex chemical issues that would otherwise be intractable is shown by the results of this study. Predicting the characteristics of novel materials based on their composition and structure may be accomplished via the use of PGMs, which provides a potential route within this field.