Qualification:
MSc, BSc
jayakumarp@am.amrita.edu

Jayakumar P. currently serves as an Assistant Professor at the Department of Computer Science Applications at Amrita School of Engineering, Amritapuri. He has completed M. Sc. in Computer Science and also qualified UGC NET (June 2012).

## Publications

### Publication Type: Conference Paper

Year of Publication Title

2018

P. Tatikonda, Kuppili, H., Paila, A., Pail, J., and Jayakumar P., “Construction of Multiplicative Graph Spanners using Minimum Connected Dominating Set with Bounded Diameter”, in 2018 International Conference on Advances in Computing, Communications and Informatics (ICACCI), Bangalore, India, 2018.[Abstract]

A Multiplicative Spanner is a spanning sub-graph H(V, E') of a graph G(V, E) such that where distance(u, v, G) is the shortest distance between the vertices u and v in G. The parameter t is called the multiplicative stretch of the spanner. When the size of the graph is reduced to construct a spanner, the shortest distance between the vertices increases, consequently the stretch factor also increases. It is known that the construction of spanners with optimum size-stretch is hard. Many researchers proposed efficient algorithms that yield proven near optimal results. In this paper we propose a quadratic time algorithm to construct multiplicative t-spanners with a bound on the stretch factor.

More »»

2018

Y. Veluri, Jayakumar P., and Nair, J. J., “Computing Multiplicative Spanners Efficiently for a Class of Simple Graphs”, in 2018 International Conference on Data Science and Engineering (ICDSE), Kochi, India, 2018.[Abstract]

The notion of a multiplicative t - spanner was defined by David Peleg and Alejandro Scha̅fer in 1989. A vertexspanning sub graph S, of a simple graph H(V, E) in which the pairwise distance between any two vertices is less than or equal to t times that in H is called a puerly multiplicative t - spanner. The real valued parameter t (&gt; 1) denotes the maximum possible multiplicative stretch of u-v distances in the spanner S. Although the combinatorial construction of a t-spanner with m or fewer edges for a given m and t (MTS problem henceforth) is known to be difficult, it is possible to construct MTS efficiently for graphs which satisfies certain properties. In this paper we characterize an infinite family of graph for which the MTS problem can be solved efficiently. Based on the characterization we posit a (n) algorithm for constructing an MTS.

More »»

### Publication Type: Journal Article

Year of Publication Title

2018

A. Prabhendran, P, P., and Jayakumar P., “On Identifying Related Objects from a Set”, Journal of Advanced Research in Dynamical and Control Systems, vol. 6, pp. 1245-1251, 2018.

### Publication Type: Conference Proceedings

Year of Publication Title

2018

Krishna Prasad T. R., Jayakumar P., and Dr. Sajeev G. P., “A K-Clique Based Clustering Protocol for Resource Discovery in P2P Network”, 2018 International Conference on Advances in Computing, Communications and Informatics (ICACCI). pp. pp. 2475-2480, 2018.[Abstract]

Data transfer amongst peers in a network without a central authority to regulate traffic, is on the ascendancy in the recent years. In this paper, we propose a k-clique based overlay network formation using multi key-single value pair mapping mechanism within a peer to peer network. This clique based model aims to discover resources of the same metadata type within a cluster of subnet with a minimum of hops as possible subject to the nodes having certain properties. The discovery of a subset of resources sharing similar characteristics is essential in the context of requirements being dynamic, in fields related to Internet of Things and Cloud systems. The simulated experiment validates our approach as it discovers resources in very less number of hops.

More »»

2016

P. R. Sharma, Yathu, E. M., and Jayakumar P., “Optimization Strategies for Product Rating Using User Review Analysis”, International conference on computing paradigms(ICCP 2016), International Journal of Control Theory and Applications . pp. 4553-4562, 2016.[Abstract]

E-commerce web has evolved as a key source of unstructured data rich in customer opinion. Understanding user reviews and mining opinion is an important research area in computer science which mostly relies on statistical inference methods. A deterministic and efficient algorithm to mine opinion remains far in the future. In this paper, domain specific strategies for optimizing existing product rating methods are discussed. The proposed strategies include a simple method to filter out irrelevant words from extracted nouns, an application of improved apriori algorithm and a new mathematical estimation of the overall rating. These methods can improve the efficiency in terms of running time of algorithms and accuracy of results. An empirical analysis of accuracy and efficiency of our methods is also presented.

More »»
Faculty Research Interest: