Cost Effective Disaster Finding Using Supervised Learning Methods On Hadoop
Journal Title: Elysium Journal of Engineering Research and Management - Year 2017, Vol 4, Issue 4
Abstract
Streams of data is collected and the required information is separated from these data is known as data mining. Various data sources are combined for processing to provide further consistent, precise and important information is known as data fusion. The feature and stages of processing is utilized to evaluate the data fusion. The fused data is then classified by data fusion classifier. The accuracy is increased by eliminating undesirable features from the data fused. Extract, Transform and Load (ETL) on Hadoop is the proposed method which is executed to enhance the performance metrics such as scalability, reliability, maintenance cost, CPU utilization and throughput. Support Vector Machine is the learning method that analyzes the data for classification analysis. K-means algorithm is suitable method for classification in which the data set is organized into clusters. In Hadoop, Clustering is performed and Map/Reduce function is used to map the key/value pair and ignore the irrelevant features or redundant data. Data efficiency is improved by applying Map/reduce function. The classification accuracy is increased by implementing the proposed system. Map/reduce approach is in Apache Hadoop to reduce the cost barriers for analyzing and processing a large data. The evaluation outcomes shows that the proposed method is reliable and scalable and also throughput and CPU utilization is improved.
Authors and Affiliations
Sathya M, Sangeetha K
Keywords
Related Articles
- EP ID EP363052
- DOI -
- Views 113
- Downloads 0
COMMUNITY DISCOVERY: COMPARITIVE SURVEY BETWEEN PAGE RANK, HITS, SALSA AND HEIGEN
his paper presents the comparative study of ranking algorithm such as PageRank, HITS, SALSA and Heigen. These algorithms share a common factor that is it finds the principal eigen vector. We experimented the...
AN EFFICIENT AND SECURE POLICY UPDATE OUTSOURCING SCHEME FOR BIG DATA STORAGE IN CLOUD
Store the big data in cloud is a best and an effective option due to its high volume and velocity. Generally, the cloud has the capability to store the big data and to process the high volume of user access requests. In...
A Statistical Pattern Recognition Approach for Tracking Left Ventricle Endocardium
The Left Ventricle endocardium has a difficult problem can be determined by a new statistical pattern recognition approach. In this pattern, resampling algorithm along with the current time step is estimated and appearan...
SELECTION OF OPTIMAL MINING ALGORITHM FOR OUTLIER DETECTION - AN EFFICIENT METHOD TO PREDICT/DETECT MONEY LAUNDERING CRIME IN FINANCE INDUSTRY
Today, Money Laundering (ML) poses a serious threat and unique challenge to financial institutions. Most of the financial institutions internationally have been implementing Anti-Money Laundering solutions (AML) to fight...
NFC SIGNALS BASED ON AN EFFECTIVE MOBILE ROBOT LOCALIZATION
A Complete Localization system for an indoor self - ruled vehicle equipped with Near Field Communication (NFC) reader to Interrogate tags located on the environment. Phase Measurement for real t...