High Impact Factor : 4.396 icon | Submit Manuscript Online icon | UGC Approved icon

Decision Trees for Analyzing Different Versions of KDD Cup Datasets

Author(s):

Vikas Kaushik ; Sachin Gavhane; Amruta Pokhare; Snigdha Bangal; Komal Mahajan

Keywords:

Classification C4.5, Confusion Matrix, KDD Cup Datasets, Training Algorithm

Abstract

Many Organizations release standard datasets for researchers to work. One amongst them is KDD cup dataset and its versions. Different datasets gives different theory, however we are here to talk about KDD Cup 1999 and KDD cup 2015. In KDD Cup 2015 we get a dataset of users behavior while watching online videos provided by MOOC (Massive Open Online Courses), we get information about user’s behavior – whether he is interested in a particular video or not. This information can be analyzed so as to help an organization to suggest videos to user in which he may be more interested. This ultimately in a long run can save time of both user and organization and keep user with Organization. In KDD cup 1999, we can study similar dataset of network traffic and can train algorithm for various attacks. In proposed system correct classes are predicted by applying C4.5 algorithm and are compared to other algorithms using parameters of confusion matrix.

Other Details

Paper ID: NCTAAP165
Published in: Conference 4 : NCTAA 2016
Publication Date: 00/00/0000
Page(s): 715-720

Article Preview




Download Article