Security is a prime issue in today's world due to the increase in the number of devices and dependencies on the internet. This paper discusses and investigates the irregularities found in different networks which are one of the causes of loopholes in the security of networks. Any number that can be uniquely determined by a graph is called a graph invariant. During the last twenty years’ countless mathematical graph invariants have been characterized and utilized for correlation analysis. The study investigates cloud networks, hierarchical hypercube networks, and block shift networks with the help of the Irregularity sombor index (ISO) which has good potential of prediction to discover irregularities of networks or structure in the field of computer science, mathematics, chemistry, pharmacy, informatics and biology in context with physical and chemical structures. After finding irregularities in the networks, further, the network is solved by K banhatti redefined zagreb indices for the modeling purpose of enhanced networks with best characteristics. These deduced results can be used for the modeling of computer networks like Local area networks (LAN), Metropolitan area networks (MAN), and Wide area networks (WAN), the backbone of the internet and other networks/structures of computers, power generation, bio-informatics and chemical compounds synthesis.