Searcharxiv⌕ Search

arXiv subjects

K. Thanushkodi

Publications and source records attributed to K. Thanushkodi.

2 recordsLinked to original sources
arXiv2010

A Novel Rough Set Reduct Algorithm for Medical Domain Based on Bee Colony Optimization

Feature selection refers to the problem of selecting relevant features which produce the most predictive outcome. In particular, feature selection task is involved in datasets containing huge number of features. Rough set theory has been one of the most successful methods used for feature selection. However, this method is still not able to find optimal subsets. This paper proposes a new feature selection method based on Rough set theory hybrid with Bee Colony Optimization (BCO) in an attempt to combat this. This proposed work is applied in the medical domain to find the minimal reducts and experimentally compared with the Quick Reduct, Entropy Based Reduct, and other hybrid Rough Set methods such as Genetic Algorithm (GA), Ant Colony Optimization (ACO) and Particle Swarm Optimization (PSO).

cs.LG↗
arXiv2009

Log Management support for recovery in mobile computing Environment

Rapid and innovative improvement in wireless communication technologies has led to an increase in the demand for mobile internet transactions. However, internet access from mobile devices is very expensive due to limited bandwidth available on wireless links and high mobility rate of mobile hosts. When a user executes a transaction with a web portal from a mobile device, the disconnection necessitates failure of the transaction or redoing all the steps after reconnection, to get back into consistent application state. Thus considering challenges in wireless mobile networks, a new log management scheme is proposed for recovery of mobile transactions. In this proposed approach, the model parameters that affect application state recovery are analyzed. The proposed scheme is compared with the existing Lazy and Pessimistic scheme and a trade off analysis between the cost invested to manage log and the return of investment in terms of improved failure recoverability is made. From the analysis, the best checkpoint interval period that yields the best return of investment is identified.

cs.NI↗