In complex networks finding influential nodes is a challenging issue as networks are growing fastly. Centrality metrics are the most popular tools that are used for finding the important seed nodes. Few of the existing measures like degree centrality, clustering coefficient, k-core are considered as local centrality which has low accuracy but are not effective. Few measures like betweenness, eigen vector, closeness centrality are considered as global centrality which has high accuracy and high computational time that are not suitable for larger networks. We proposed a new hybrid centrality measure called Hybrid Isolation and Path Centrality (HIPC) that uses neighborhood isolating and localized variation in average shortest path in order to balance the accuracy and effectiveness especially for larger networks. This method uses isolating centrality along with its L-level neighbors isolating centrality and also localized relative variation in average shortest path(LRASP) considering L-Level Neighborhood. An important strength of this approach is its ability to consider the influence of neighboring nodes in detecting influential nodes by integrating local and global information of a network. Diffusion model like SIR model is used to check the reliability of the proposed method to simulate how information spreads through out the network. Experiments are conducted on two real-world datasets demonstrate that HIPC performs well when compares with other six existing measures.