Please use this identifier to cite or link to this item:
http://hdl.handle.net/10603/565923
Title: | Performance analysis on machinelearning techniques for sparse and densely distributed big data analytics |
Researcher: | Kalyana Saravanan A |
Guide(s): | Tamilarasi A |
Keywords: | Big Data Analytics Data Mining Machine Learning |
University: | Anna University |
Completed Date: | 2022 |
Abstract: | Big data analytics is a strategy for estimate the enormous data. The newlineprocess of Big data analytics is determined to perceive the important data. In newlineorder to owing the big data analytics is difficult task in improved volume of newlinedata by using data mining. The huge data is managed by several machine newlinelearning techniques for identifying the valuable data from large dataset. newlineHowever, the dimensionality reduction was a difficult task. From data mining, newlineclustering method is obtained to reduce the dimensionality through grouping newlinesimilar type of data in dataset. In addition, classification is one of the newlineimportant data mining methods to categorize the data into relevant classes for newlinebig data analytics. Now a lot of clustering and classification techniques were newlinedetermined to utilize the big data in an important manner. However, it failed newlineto extend the accuracy and quality of data with minimum dimensionality. To newlineovercome the above such issues, three different techniques are developed in newlinethis research for improving accuracy of big data analytics with minimum time newlineand space complexity. newlineInitially, a novel technique is called as Proximity Fuzzy Likelihood newlineMaximization Data Clustering (PFLMDC) technique. The designed technique newlineis to improve the performance of clustering by using big data analytics. The newlineproposed technique is achieving both sparse and dense data clustering. From newlinePFLMDC technique, sparse data clustering is performed by computing newlineProximity Manhattan distance. This aids to set the similar sparse data into newlinecluster with better accuracy. After that, Fuzzy Expected Maximum Likelihood newlineEstimation is applied to set the dense data into separate cluster to minimize newlinethe dimensionality. newline |
Pagination: | xvi,187p. |
URI: | http://hdl.handle.net/10603/565923 |
Appears in Departments: | Faculty of Information and Communication Engineering |
Files in This Item:
File | Description | Size | Format | |
---|---|---|---|---|
01_title.pdf | Attached File | 82.08 kB | Adobe PDF | View/Open |
02_prelimpages.pdf | 3.06 MB | Adobe PDF | View/Open | |
03_contents.pdf | 581.36 kB | Adobe PDF | View/Open | |
04_abstracts.pdf | 126.83 kB | Adobe PDF | View/Open | |
05_chapter1.pdf | 210.77 kB | Adobe PDF | View/Open | |
06_chapter2.pdf | 372.71 kB | Adobe PDF | View/Open | |
07_chapter3.pdf | 934.05 kB | Adobe PDF | View/Open | |
08_chapter4.pdf | 998.4 kB | Adobe PDF | View/Open | |
09_chapter5.pdf | 664.31 kB | Adobe PDF | View/Open | |
10_chapter6.pdf | 638.72 kB | Adobe PDF | View/Open | |
11_annexures.pdf | 109.92 kB | Adobe PDF | View/Open | |
80_recommendation.pdf | 93.98 kB | Adobe PDF | View/Open |
Items in Shodhganga are licensed under Creative Commons Licence Attribution-NonCommercial-ShareAlike 4.0 International (CC BY-NC-SA 4.0).
Altmetric Badge: