I++Mapreduce: Incremental Mapreduce for Mining the Big Data
Journal Title: IOSR Journals (IOSR Journal of Computer Engineering) - Year 2016, Vol 18, Issue 3
Abstract
Abstract: Data mining is an interdisciplinary area of computer engineering. Incremental processing is a challenging approach to refreshing mining results and it uses saved states to avoid the cost of re computation from damage. Here, I propose i2MapReduce is a novel incremental processing expansion to MapReduce in data mining. As compared with the state-of-the-art work on Incoop, instead of using task level re-computation the i 2MapReduce executes the key-value pair level incremental processing . It helps not only one-step computationbut also more sophisticated iterative computation. Then it incorporates a set of novel techniques to decrease input output level for accessing dedicated fine-grain computation states in the data mining. The final results on Amazon EC2 describe the significant performance development of i2MapReduce as compared to iterative and plain MapReduce performing re computation
Authors and Affiliations
Prof. Kanchan Dhomse B
Keywords
Related Articles
- EP ID EP138702
- DOI -
- Views 95
- Downloads 0
A Survey and Research Proposal of Plant Leaf Disease Diagnosis and Classification on Mobile Devices
Abstract: A plant disease is defined as anything that prevents a plant from performing to its maximal potential. A plant disease has impacted society and world olden times. Plant diseases have caused enormous economic lo...
Communication via LED
Abstract: This paper describes the use of light as a medium for communication between two systems. It takes out the “optics from the optical fiber”. It demonstrates transmission and reception of live data between t...
Sentiment analysis for improving healthcare system for women
Abstract: The system proposes a feedback mechanism wherein, sentiment analysis is performed from surveys and tweets based on prevailing health issues among adult women in India and the social opinion on prevalenthealth i...
Search Accelerator
Abstract: Optimization problem consists of maximizing or minimizing a real function by systematically Choosing input values from within an allowed set and the value of the function can be solved. Whenever the use...
A Compound Metric for Identification of Fault Prone Modules
Abstract: Software Quality is a significant non-functional requirement that is not fulfilled by many software products. Faulty modules tend to degrade the software quality which may cause customer’s dissatisfaction. Faul...