Note: Current work in the field of Activity Recognition.
Neighborhood Rough Set Model for Knowledge Acquisition Using Map Reduce
International Journal of Communication Networks & Distributed Systems | Paper ID: 10.1504/IJCNDS.2015.070975, Volume 15,No. 2/3,2015
- Advisor : Prof. B.K. Tripathy, VIT
- A neighborhood based rough set model to propose a method to determine reduced neighborhood subsets derived from samples of the universal set to efficiently handle heterogeneous real datasets.
- Proposed new algorithms to enhance knowledge acquired from large datasets by applying the concepts of Neighborhood Systems to Rough Sets.
- We compute the accuracy and coverage of the computations obtained by using parallel rough set-based methods using the conventional MapReduce technique
- Inspired by the work in “Parallel Rough Set Based Knowledge Acquisition using Map Reduce”.
Rainfall Prediction Using Artificial Neural Network
International Journal of Applied Engineering Research (IJAER) | Paper ID: 29348 – IJAER, Volume 9, No. 23, Dec 2014
- Advisor : Prof. Govinda K, VIT
- The Artificial Neural Network (ANN) model that was used is based on 'prediction' by smartly 'analysing' the trend from an already existing voluminous historical set of data.
- The simulation to predict rainfall using ANN and a few inputs that were not used previously such as cloud cover and evapo-transpiration resulted in an accuracy of 85.2%.
Efficient Approach for Query Optimization in Rough Data
International Conference on Computing, Cybernetics and Intelligent Information System - CCIIS 2013, At VIT University, Vellore
- Advisor : Prof. Sharmila Banu K, VIT
- Representation of an efficient algorithm that tries to partition the data and assign it to the respective virtual data centers.
- Practical utility includes efficient storage of the huge amounts of data thereby reducing the space and time complexity for storage and retrieval of data respectively
Efficient Clustering Algorithm for Storage Optimization in the Cloud
International Journal of Scientific and Engineering Research (IJSER), France | Paper ID: I033713, Volume 4, Issue 11, Nov 2013
- Advisor : Prof. Shalini L, VIT
- Representation of an efficient query optimization for the multi-valued rough relational database that follows the indiscernibility relation in its domain
- Practical utility includes reduction in response time for obtaining results from Rough Relational Databases (RRDs)