Random forests with weighted voting for anomalous query access detection in relational databases

Charissa Ann Ronao, Sung Bae Cho

Research output: Chapter in Book/Report/Conference proceedingConference contribution

8 Citations (Scopus)

Abstract

Data has become more and more important to individuals, organizations, and companies, and therefore, safeguarding these sensitive data in relational databases has become a critical issue. However, despite traditional security mechanisms, attacks directed to databases still occur. Thus, an intrusion detection system (IDS) specifically for the database that can provide protection from all possible malicious users is necessary. In this paper, we present a random forests (RF) method with weighted voting for the task of anomaly detection. RF is a graph-based technique suitable for modeling SQL queries, and weighted voting enhances its capabilities by balancing the voting impact of each tree. Experiments show that RF with weighted voting exhibits a more superior performance consistency, as well as better error rates with increasing number of trees, compared to conventional RF. Moreover, it outperforms all other stateof- The-art data mining algorithms in terms of false positive rate (0.076) and false negative rate (0.0028).

Original languageEnglish
Title of host publicationArtificial Intelligence and Soft Computing - 14th International Conference, ICAISC 2015, Proceedings
EditorsRyszard Tadeusiewicz, Lotfi A. Zadeh, Leszek Rutkowski, Marcin Korytkowski, Rafal Scherer, Jacek M. Zurada
PublisherSpringer Verlag
Pages36-48
Number of pages13
ISBN (Electronic)9783319193687
DOIs
Publication statusPublished - 2015 Jan 1
Event14th International Conference on Artificial Intelligence and Soft Computing, ICAISC 2015 - Zakopane, Poland
Duration: 2015 Jun 142015 Jun 18

Publication series

NameLecture Notes in Artificial Intelligence (Subseries of Lecture Notes in Computer Science)
Volume9120
ISSN (Print)0302-9743

Other

Other14th International Conference on Artificial Intelligence and Soft Computing, ICAISC 2015
CountryPoland
CityZakopane
Period15/6/1415/6/18

Fingerprint

Random Forest
Relational Database
Voting
Anomalous
Query
Intrusion detection
Data mining
Anomaly Detection
Intrusion Detection
False Positive
Balancing
Error Rate
Data Mining
Attack
Necessary
Industry
Graph in graph theory
Experiments
Modeling
Experiment

All Science Journal Classification (ASJC) codes

  • Theoretical Computer Science
  • Computer Science(all)

Cite this

Ronao, C. A., & Cho, S. B. (2015). Random forests with weighted voting for anomalous query access detection in relational databases. In R. Tadeusiewicz, L. A. Zadeh, L. Rutkowski, M. Korytkowski, R. Scherer, & J. M. Zurada (Eds.), Artificial Intelligence and Soft Computing - 14th International Conference, ICAISC 2015, Proceedings (pp. 36-48). (Lecture Notes in Artificial Intelligence (Subseries of Lecture Notes in Computer Science); Vol. 9120). Springer Verlag. https://doi.org/10.1007/978-3-319-19369-4_4
Ronao, Charissa Ann ; Cho, Sung Bae. / Random forests with weighted voting for anomalous query access detection in relational databases. Artificial Intelligence and Soft Computing - 14th International Conference, ICAISC 2015, Proceedings. editor / Ryszard Tadeusiewicz ; Lotfi A. Zadeh ; Leszek Rutkowski ; Marcin Korytkowski ; Rafal Scherer ; Jacek M. Zurada. Springer Verlag, 2015. pp. 36-48 (Lecture Notes in Artificial Intelligence (Subseries of Lecture Notes in Computer Science)).
@inproceedings{a4172d86243948c09e683785a88110ed,
title = "Random forests with weighted voting for anomalous query access detection in relational databases",
abstract = "Data has become more and more important to individuals, organizations, and companies, and therefore, safeguarding these sensitive data in relational databases has become a critical issue. However, despite traditional security mechanisms, attacks directed to databases still occur. Thus, an intrusion detection system (IDS) specifically for the database that can provide protection from all possible malicious users is necessary. In this paper, we present a random forests (RF) method with weighted voting for the task of anomaly detection. RF is a graph-based technique suitable for modeling SQL queries, and weighted voting enhances its capabilities by balancing the voting impact of each tree. Experiments show that RF with weighted voting exhibits a more superior performance consistency, as well as better error rates with increasing number of trees, compared to conventional RF. Moreover, it outperforms all other stateof- The-art data mining algorithms in terms of false positive rate (0.076) and false negative rate (0.0028).",
author = "Ronao, {Charissa Ann} and Cho, {Sung Bae}",
year = "2015",
month = "1",
day = "1",
doi = "10.1007/978-3-319-19369-4_4",
language = "English",
series = "Lecture Notes in Artificial Intelligence (Subseries of Lecture Notes in Computer Science)",
publisher = "Springer Verlag",
pages = "36--48",
editor = "Ryszard Tadeusiewicz and Zadeh, {Lotfi A.} and Leszek Rutkowski and Marcin Korytkowski and Rafal Scherer and Zurada, {Jacek M.}",
booktitle = "Artificial Intelligence and Soft Computing - 14th International Conference, ICAISC 2015, Proceedings",
address = "Germany",

}

Ronao, CA & Cho, SB 2015, Random forests with weighted voting for anomalous query access detection in relational databases. in R Tadeusiewicz, LA Zadeh, L Rutkowski, M Korytkowski, R Scherer & JM Zurada (eds), Artificial Intelligence and Soft Computing - 14th International Conference, ICAISC 2015, Proceedings. Lecture Notes in Artificial Intelligence (Subseries of Lecture Notes in Computer Science), vol. 9120, Springer Verlag, pp. 36-48, 14th International Conference on Artificial Intelligence and Soft Computing, ICAISC 2015, Zakopane, Poland, 15/6/14. https://doi.org/10.1007/978-3-319-19369-4_4

Random forests with weighted voting for anomalous query access detection in relational databases. / Ronao, Charissa Ann; Cho, Sung Bae.

Artificial Intelligence and Soft Computing - 14th International Conference, ICAISC 2015, Proceedings. ed. / Ryszard Tadeusiewicz; Lotfi A. Zadeh; Leszek Rutkowski; Marcin Korytkowski; Rafal Scherer; Jacek M. Zurada. Springer Verlag, 2015. p. 36-48 (Lecture Notes in Artificial Intelligence (Subseries of Lecture Notes in Computer Science); Vol. 9120).

Research output: Chapter in Book/Report/Conference proceedingConference contribution

TY - GEN

T1 - Random forests with weighted voting for anomalous query access detection in relational databases

AU - Ronao, Charissa Ann

AU - Cho, Sung Bae

PY - 2015/1/1

Y1 - 2015/1/1

N2 - Data has become more and more important to individuals, organizations, and companies, and therefore, safeguarding these sensitive data in relational databases has become a critical issue. However, despite traditional security mechanisms, attacks directed to databases still occur. Thus, an intrusion detection system (IDS) specifically for the database that can provide protection from all possible malicious users is necessary. In this paper, we present a random forests (RF) method with weighted voting for the task of anomaly detection. RF is a graph-based technique suitable for modeling SQL queries, and weighted voting enhances its capabilities by balancing the voting impact of each tree. Experiments show that RF with weighted voting exhibits a more superior performance consistency, as well as better error rates with increasing number of trees, compared to conventional RF. Moreover, it outperforms all other stateof- The-art data mining algorithms in terms of false positive rate (0.076) and false negative rate (0.0028).

AB - Data has become more and more important to individuals, organizations, and companies, and therefore, safeguarding these sensitive data in relational databases has become a critical issue. However, despite traditional security mechanisms, attacks directed to databases still occur. Thus, an intrusion detection system (IDS) specifically for the database that can provide protection from all possible malicious users is necessary. In this paper, we present a random forests (RF) method with weighted voting for the task of anomaly detection. RF is a graph-based technique suitable for modeling SQL queries, and weighted voting enhances its capabilities by balancing the voting impact of each tree. Experiments show that RF with weighted voting exhibits a more superior performance consistency, as well as better error rates with increasing number of trees, compared to conventional RF. Moreover, it outperforms all other stateof- The-art data mining algorithms in terms of false positive rate (0.076) and false negative rate (0.0028).

UR - http://www.scopus.com/inward/record.url?scp=84958537957&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84958537957&partnerID=8YFLogxK

U2 - 10.1007/978-3-319-19369-4_4

DO - 10.1007/978-3-319-19369-4_4

M3 - Conference contribution

AN - SCOPUS:84958537957

T3 - Lecture Notes in Artificial Intelligence (Subseries of Lecture Notes in Computer Science)

SP - 36

EP - 48

BT - Artificial Intelligence and Soft Computing - 14th International Conference, ICAISC 2015, Proceedings

A2 - Tadeusiewicz, Ryszard

A2 - Zadeh, Lotfi A.

A2 - Rutkowski, Leszek

A2 - Korytkowski, Marcin

A2 - Scherer, Rafal

A2 - Zurada, Jacek M.

PB - Springer Verlag

ER -

Ronao CA, Cho SB. Random forests with weighted voting for anomalous query access detection in relational databases. In Tadeusiewicz R, Zadeh LA, Rutkowski L, Korytkowski M, Scherer R, Zurada JM, editors, Artificial Intelligence and Soft Computing - 14th International Conference, ICAISC 2015, Proceedings. Springer Verlag. 2015. p. 36-48. (Lecture Notes in Artificial Intelligence (Subseries of Lecture Notes in Computer Science)). https://doi.org/10.1007/978-3-319-19369-4_4