Please use this identifier to cite or link to this item:
http://localhost:8081/jspui/handle/123456789/17602
Title: | OUTLIER DETECTION BASED ON FREQUENT PATTERNS |
Authors: | Rajput, Anuj |
Keywords: | Unfortunately;Identification;Detection;Calculating |
Issue Date: | Jun-2013 |
Publisher: | I I T ROORKEE |
Abstract: | An outlier in a dataset is an observation or a data point that is very much dissimilar to or inconsistent with the rest of the data, these are also called deviant objects or exceptions. Detection of such outliers is important for many applications and has recently attracted much attention in the data mining research area. Identification of outlier patterns is very important in modern-day engineering problems such as credit card fraud detection and network intrusion detection. Most previous studies focused on finding outliers that are hidden in numerical datasets but cannot be directly apply to categorical sets where there is little difficulty in calculating distances among data points. Unfortunately, those outlier detection methods were not directly applicable to real life transaction databases. In this dissertation work, we focus on the outlier detection method based on frequent patterns. We describe frequent pattern based core algorithm and LFP method and then we analyze the drawback of LFP method. Targeting at the drawback of LFP method, we propose an improved method. Finally we evaluate the two methods by using real datasets. Experimental results show that proposed method outperforms the LFP method. |
URI: | http://localhost:8081/jspui/handle/123456789/17602 |
metadata.dc.type: | Other |
Appears in Collections: | MASTERS' THESES (E & C) |
Files in This Item:
File | Description | Size | Format | |
---|---|---|---|---|
G22938.pdf | 6.99 MB | Adobe PDF | View/Open |
Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.