Outlier Regression Analysis

Chapter - 1

INTRODUCTION

1.1 Introduction

Today in Dataset there exist data objects that do not comply with the general behavior or model of the data. Such data objects which are heavy different from or inconsistent with the remaining set of data, are called outliers. An outlier is a data set which is different from the remaining data. Outlier is also denoted to as deformity, deviants or anomalies in the data mining and statistics literature. In most applications the data is produced by one or more generating processes, which could either reflect activity in the system or observations collected about entities. When the developing process behaves in a cursory way it results in the creation of outliers. Therefore, an outlier often contains useful

In many instances such as an intrusion or fraud detection, the outliers can only be seen as a succession of multiple data points, rather than as an individual data level. For example, a fraud event may often find fault the actions of an individual in a particular sequence. The specificity of the sequence is relevant to identifying the anomalous effect. Such anomalies are as well mentioned to as collective anomalies, because they can simply be inferred collectively from a set or sequence of data points. Such collective anomalies typically represent unusual issues, which need to be discovered from the

In this work by showing the experimental work and procedure step for the distributed solving set (DSS), lazy DSS (lazy distributed solving set) and Proposed approach which leads to early stop based on the condition that is Hybrid DSS which is an approach provide the best outlier in less computation time. Objective is to design the hybrid approach as well as performing the simulation of the available data set to find required outlier from it and to use in the real-time dataset to classify available data of categorical dataset.

1.6 SCOPE OF THE WORK

The work defined by us is having the various motivations obtain in order to progress further with the research topic chosen by us, on working with the current given title and scenario can be work further with the various security concerns with the help of outlier detection technique in the larger dimensional dataset.

The research is motivated to find:

• Outlier of user required data from large dataset.

• Working with various pattern mining approaches based on the outlier detection technique in the