Open AccessFramework for Computing Potential Node with Higher Influence in Complex Social Network
Author(s) -
Selva Kuamr S*,
N Kayarvizhy
Publication year - 2020
Publication title -
international journal of recent technology and engineering
Language(s) - English
Resource type - Journals
ISSN - 2277-3878
DOI - 10.35940/ijrte.e5934.018520
Subject(s) - computer science , node (physics) , identification (biology) , social network (sociolinguistics) , set (abstract data type) , construct (python library) , analytics , dynamic network analysis , distributed computing , data science , data mining , engineering , computer network , botany , structural engineering , biology , world wide web , social media , programming language
The increasing data complexity in social network imposes challenges towards carrying out investigation towards analytics. Similar scenario has started evolving in social network most recently owing to the usage of ubiquitous devices which gives rise to accumulation of complex set of data that offers potential challenge to understand the potential node. Identification of potential node is an essential demand to understand the role player in social network. Therefore, this paper presents a novel analytical methodology to construct a discrete community with higher precision along with a novel progressive algorithm for identifying the potential node that offers a higher degree of influence in social network. Dynamic optimization, as well as probability theory, has been used in order to perform modeling of the proposed system. Along with an effective computational performance, the comparative analysis shows that proposed system offers better performance in contrast to existing techniques with respect to influence degree and information processing time. The inferencing of the quantified outcome status shows that the formulated approach attains ~70% performance improvement in the case of minimizing the processing time as compared to the Chen approach. The identification of potential node in four different types of networks have also significantly improved with negligible computational effort.