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 142
- Downloads 0
Combining Ability Analysis for Qualitative Characters in Indian Hexaploid Wheat
Combining ability effects were estimated in line x tester crossing programme for five characters involving four lines and three testers in Triticum durum. It revealed preponderance of non-additive gene action for protein...
A Review Paper on Cost Effective Cultivation Medium for Biocalcification of Bacillus Pasteurii and its Effect on Concrete
The theme of this study is to isolate and identify the bacillus bacteria are suitable for use in concrete to improve its strength or not. Bacillus is to be incorporated with concrete should be presence of alkali resistan...
Enhanced Heart Disease Analysis and Prediction System [EHDAPS] Using Data Mining
Data mining is the process of analyzing data from different perspectives and summarizing it into useful information. Data mining techniques are used for variety of applications. Data mining techniques have been very effe...
Image Morphing with Data Security
Along with Computer and Network Security.Data Security plays a vital role in today world. Data is of great importance. In this paper, we have discussed how the data can be secured. This paper deals with hiding your encry...
Receipt-Free Multi-Authority Voting with (2,2) Secret Sharing Based Authentication
Election and voting are the important activities in the democratic countries. The traditional voting system becomes inconvenient for the people who cannot visit the polling booth for the vote casting. To overcome the dra...