Papers by Dr. Rika Sharma
Lecture notes in networks and systems, 2023
Journal of Intelligent & Fuzzy Systems, 2018
Rainfall prediction is one of the complex nonlinear dynamic phenomena. This is due to uncertainti... more Rainfall prediction is one of the complex nonlinear dynamic phenomena. This is due to uncertainties associated with the climatic parameters used for rainfall prediction. Fuzzy system has the capability to deal with the uncertainties and is efficient when the conventional linear statistical models are not able to perform well due to the nonlinear nature of the climatic parameters. In the present study, a data driven Fuzzy Inference System for high-dimensional data is developed to predict rainfall of the Indian subcontinent. Indian monsoon is an important climatic phenomenon due to its direct impact on socioeconomic growth. The parameters Sea Surface Temperature, Sea Level Pressure, El Niño-Southern Oscillation, Indian Ocean Dipole Mode and the Equatorial Indian Ocean Oscillation have been used for analyses and prediction. The variability of Indian rainfall is considered for the period of 25 years from 1990-2014 and the possibility of prediction is explored using Fuzzy Inference System. In fuzzy inference system the membership functions are the building blocks and computing its range is a crucial task. We have used triangular membership function and in order to define the range of membership function, this study proposes two methods, divisive method for input parameters and clustering based method for output parameter. The experimental results obtained using the proposed fuzzy inference system is compared with Multiple Linear Regression and Multiple Adaptive Regression Splines. The proposed Fuzzy based predictive model shows better results in terms of the accuracy with 84% and correlation 0.78 between actual and predicted rainfall.
International Journal of Fuzzy Systems, 2019
Shared nearest neighbor (SNN) clustering algorithm is a robust graph-based, efficient clustering ... more Shared nearest neighbor (SNN) clustering algorithm is a robust graph-based, efficient clustering method that could handle high-dimensional data. The SNN clustering works well when the data consist of clusters that are of diverse in shapes, densities, and sizes but assignment of the data points lying in the boundary regions of overlapping clusters is not accurate. In order to overcome this problem, we have presented an extension of shared nearest neighbor algorithm that have better capability of handling the data points lying in the boundary regions specifically for overlapping cluster by means of fuzzy concept. Extensive experiments were carried out to compare the proposed approach fuzzy shared nearest neighbor clustering (FSNN) with existing clustering methods K-means, Fuzzy C-means, Density_clust, and Shared Nearest Neighbor. The effectiveness of FSNN is evaluated in benchmark datasets. Experimental results using FSNN method show that it can accurately cluster the data points lying in the overlapping partition and generate compact and well-separated clusters as compared to state-of-the-art clustering algorithm. The results obtained using different clustering methods are validated by standard cluster validation measures.
Soft Computing, 2017
Massive amount of Earth science data open an unprecedented opportunity to discover potentially va... more Massive amount of Earth science data open an unprecedented opportunity to discover potentially valuable information. Earth science data are complex, nonlinear, high-dimensional data, and the sparsity of data in highdimensional space poses major challenge in clustering of the data. Shared nearest neighbor clustering (SNN) algorithm is one of the well-known and efficient methods to handle high-dimensional spatiotemporal data. The SNN clustering method does not cluster all the data forming rigid boundary selection. This paper reports fuzzy shared nearest neighbor (FSNN) algorithm which is an enhancement of the SNN clustering method that has the capability of handling the data lying in the boundary regions by means of a fuzzy concept. The clusters obtained can be characterized by the cluster centroid, which summarizes the behavior of the ocean points in the cluster. The statistical measure is used to find the significant relation between the cluster centroids and the existing climate indices. In this study, correlation measure is used to find the significant pattern, such as teleconnection or dipole. The experimentation is performed on Indian continent latitude range 7.5 • −37.5 • N and longitude range 67.5 • −97.5 • E. Extensive experiments are carried out to compare the proposed approach with existing clustering methods such as K-means, fuzzy C-means and SNN. The proposed method, FSNN algorithm, not only handles the data lying in the Communicated by V. Loia.
Natural Hazards, 2014
ABSTRACT
Natural Hazards, 2014
ABSTRACT
2014 Fourth International Conference on Communication Systems and Network Technologies, 2014
ABSTRACT
This paper attempts to explore regression approach for the prediction of tropical cyclone genesis... more This paper attempts to explore regression approach for the prediction of tropical cyclone genesis in North Indian Ocean. The cyclonic storms from the Bay of Bengal are the usual phenomenon in east coast of India. The present study mainly focuses on two such very severe cyclonic storms, PHAILIN and HUDHUD in 2013 and 2014, respectively. Various meteorological parameters that clearly explain the favorable situation for the occurrence of cyclonic storm are investigated and analyzed. The study emphasizes on selection of the potential predictor variables using stepwise regression method to analyze the efficiency of model. Regression analysis yields a predicted value resulting from a linear combination of the predictors. For the regression model, the predictor set consists of sea level pressure (SLP) and wind speed parameters. Different types of the regression methods have been used in the study and compared to find the efficient regression model for cyclone genesis prediction. The thresh...
Uploads
Papers by Dr. Rika Sharma