Outlier Detection and Analysis using Hybrid Approach for Mixed datasets
Journal Title: International journal of Emerging Trends in Science and Technology - Year 2014, Vol 1, Issue 4
Abstract
There are several approaches of outlier detection employed in many study areas amongst which distance based and density based outlier detection techniques have gathered most attention of researchers.So we are using hybrid of these two methods.The existing system uses distance based method for outlier detection and K-means as clustering method.But distance based method has limitation that it fails for non-uniform datasets.The k-means method requires number of clusters to form as input which is difficult for real life datasets which contains millions of attributes and rows.So we move to proposed model.The proposed model uses hybrid of distance and density outlier detection methods and weighted squuzer method for clustering.Most of the models deals with only single datasets.Here the project deals with mixed datasets.Future scope will be to handle dyanamic data
Authors and Affiliations
Anjali Barmade
Keywords
Related Articles
- EP ID EP220499
- DOI -
- Views 149
- Downloads 0
Evidence Based Psychosocial Intervention
In recent years, there has been significant progress and expansion on the development of evidence-based psychosocial treatments for substance abuse and dependence. A literature review was undertaken using the several ele...
Synthesis and biocidal screening of 4-biphenyl acetamides derivatives.
The present paper deals with the synthesis of 4-biphenyl acetic acid amides by condensation of corresponding acid chlorides with suitable amines. the structure of newly synthesized compounds were elucidated on the basis...
A Study of E-Commerce: Challenges and Oppurtunities In An Indian Economy
Even today; some considerable time after the so called „dot com/Internet revolution‟, electronic commerce (e-commerce) remains a relatively new, emerging and constantly changing area of business management and informatio...
A Study on Stemming Algorithms
Stemming is a process in in which different morphological variants of a word is transformed into its root form. A stemmer involves removing prefixes and suffixes of a word to reduce a word to its root form. This process...
Removal of Impulse Noise in Image Using Reduced Simple Edge Preserving Denoising Technique
Images and videos are mostly depraved by salt and pepper noise during the process of signal accession and transmission. Removal of noise is one of the most important issue in image analysis and processing. Even though wh...