Please use this identifier to cite or link to this item: http://hdl.handle.net/10603/306978
Title: Adaptive Data Mining Approaches for Efficient Predictive Analytics
Researcher: Razeef Mohd
Guide(s): Butt,Muheet Ahmed
Keywords: Computer Science
Data Mining
Engineering and Technology
University: University of Kashmir
Completed Date: 2018
Abstract: newline eather prediction plays an important role in the field of agriculture and industries. newlineAccurate rainfall prediction is the only phenomenon for detecting the massive newlinerainfall and to provide the information of warnings regarding the disasters. Modelling newlineweather is considered as an important topic around the world as it can predict the threats newlinebeforehand to prevent the damages caused by extreme rainfall conditions. Prediction of newlineweather plays a vital role as it protects the social infrastructures and is favourable to newlinefarmers for taking precautions before the damage. Various data mining techniques were newlinedeveloped to predict the rainfall data, but effective modeling of rainfall in case of newlinechanging weather conditions are unpredictable. These issues can be handled if occurrence newlineof the weather conditions are predicted in advance and warnings are provided on time. newlineThus, by considering the importance of the rainfall prediction system, this research newlineprovides three contributions for addressing these issues. The first contribution of the newlineresearch is to predict the rainfall in spite of changing weather conditions by adapting newlineNonlinear Autoregressive Network with Exogenous Inputs (NARX), which is trained newlineusing the proposed Self Adaptive Levenberg Marquardt (SALM) algorithm for newlinedetermining the optimal weights. The algorithm is responsible for enhancing the rate of newlinelearning to make it more adaptive for yielding prediction with more accuracy. The second newlinecontribution is to develop a technique; named Grey Wolf- based Linear Regression newline(GWLR), using Grey Wolf Optimizer (GWO) and linear regression, for rainfall prediction. newlineThe linear regression is used for predicting the dependent variable value from an newlineindependent variable using regression coefficient, which is generated using GWLR newlinealgorithm. The third contribution is to design an effective rainfall prediction model by newlineintegrating GWO and Levenberg Marquardt (LM) algorithm, called Grey Wolf newlineLevenberg-Marquardt algorithm (GWLM) algorithm, along with NARX model. GWLM newline......
Pagination: 
URI: http://hdl.handle.net/10603/306978
Appears in Departments:Department of Computer Science

Files in This Item:
File Description SizeFormat 
01_title.pdfAttached File271.01 kBAdobe PDFView/Open
02_certificate.pdf249.59 kBAdobe PDFView/Open
03_declaration.pdf264.15 kBAdobe PDFView/Open
04_acknowledgement.pdf121.04 kBAdobe PDFView/Open
05_dedication.pdf349.2 kBAdobe PDFView/Open
06_contents.pdf151.6 kBAdobe PDFView/Open
07_list_of_tables_figures.pdf101.45 kBAdobe PDFView/Open
08_abbreviations.pdf96.17 kBAdobe PDFView/Open
09_abstract.pdf166.4 kBAdobe PDFView/Open
10_chapter_01.pdf450.58 kBAdobe PDFView/Open
11_chapter_02.pdf469.09 kBAdobe PDFView/Open
12_chapter_03.pdf661.94 kBAdobe PDFView/Open
13_chapter_04.pdf605.85 kBAdobe PDFView/Open
14_chapter_05.pdf1.43 MBAdobe PDFView/Open
15_chapter_06.pdf572.19 kBAdobe PDFView/Open
16_chapter_07.pdf317.87 kBAdobe PDFView/Open
17_publications.pdf300.5 kBAdobe PDFView/Open
18_references.pdf381.49 kBAdobe PDFView/Open
80_recommendation.pdf317.87 kBAdobe PDFView/Open


Items in Shodhganga are protected by copyright, with all rights reserved, unless otherwise indicated.

Altmetric Badge: