Data mining method based on computer forensics-based ID3 algorithm

Author

Qin, Iu

Author_Institution

Dept. of Inf. Sci. & Technol., East China Univ. of Political Sci. & Law, Shanghai, China

fYear

2010

fDate

16-18 April 2010

Firstpage

340

Lastpage

343

Abstract

Data mining method based on computer forensics-based ID3 algorithm is presented in the study. Forensics data are unconstant, noisy and dispersive. Based on these characteristic of forensics data, the improved ID3 algorithm from adopting weight and two times information is gained. The examples can be used as the experiment data, and 100 test samples which is independent of training samples are applied to judge error rate of decision tree rules. The experimental results show that the error rate of ID3 is 8.9% and the error rate of improved algorithm is 5.4%, which indicates the accuracy of the proposed method is higher than ID3 algorithm. It can be seen that the improved method used in the computer forensics process is entirely feasible.

Keywords

computer forensics; data mining; decision trees; ID3 algorithm; computer forensics; data mining; decision tree rule; error rate; Algorithm design and analysis; Classification tree analysis; Data mining; Decision trees; Entropy; Error analysis; Forensics; Machine learning algorithms; Mathematical model; Testing; ID3; computer forensics; data mining; decision tree; high precision;

fLanguage

English

Publisher

ieee

Conference_Titel

Information Management and Engineering (ICIME), 2010 The 2nd IEEE International Conference on

Conference_Location

Chengdu

Print_ISBN

978-1-4244-5263-7

Electronic_ISBN

978-1-4244-5265-1

Type

conf

DOI

10.1109/ICIME.2010.5477817

Filename

5477817