SPIT callers detection with unsupervised Random Forests classifier

Kentaro Toyoda, Iwao Sasase

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

13 Citations (Scopus)

Abstract

As VoIP (Voice over IP) grows rapidly, it is expected to prevail tremendous unsolicited advertisement calls, which type of calls is referred to SPIT (SPam over Internet Telephony). SPIT detection is more difficult to execute than email SPAM detection since the callee or SPIT detection system does not tell whether it is SPIT or legitimate call until he/she actually takes a call. Recently, many SPIT detection techniques are proposed by finding outliers of call patterns. However, most of these techniques suffer from setting a threshold to distinguish that the caller is legitimate or not and this could cause to high false negative rate or low true positive rate. This is because these techniques analyse call pattern by a single feature e.g. call frequency or average call duration. In this paper, we propose a multi-feature call pattern analysis with unsupervised Random Forests classifier, which is one of the excellent classification algorithms. We also propose two simple but helpful features for better classification. We show the effectiveness of Random Forests based classification without supervised training data and which features contribute to classification.

Original languageEnglish
Title of host publicationIEEE International Conference on Communications
Pages2068-2072
Number of pages5
DOIs
Publication statusPublished - 2013
Event2013 IEEE International Conference on Communications, ICC 2013 - Budapest, Hungary
Duration: 2013 Jun 92013 Jun 13

Other

Other2013 IEEE International Conference on Communications, ICC 2013
CountryHungary
CityBudapest
Period13/6/913/6/13

Fingerprint

Internet telephony
Classifiers
Electronic mail

ASJC Scopus subject areas

  • Electrical and Electronic Engineering
  • Computer Networks and Communications

Cite this

Toyoda, K., & Sasase, I. (2013). SPIT callers detection with unsupervised Random Forests classifier. In IEEE International Conference on Communications (pp. 2068-2072). [6654830] https://doi.org/10.1109/ICC.2013.6654830

SPIT callers detection with unsupervised Random Forests classifier. / Toyoda, Kentaro; Sasase, Iwao.

IEEE International Conference on Communications. 2013. p. 2068-2072 6654830.

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

Toyoda, K & Sasase, I 2013, SPIT callers detection with unsupervised Random Forests classifier. in IEEE International Conference on Communications., 6654830, pp. 2068-2072, 2013 IEEE International Conference on Communications, ICC 2013, Budapest, Hungary, 13/6/9. https://doi.org/10.1109/ICC.2013.6654830
Toyoda K, Sasase I. SPIT callers detection with unsupervised Random Forests classifier. In IEEE International Conference on Communications. 2013. p. 2068-2072. 6654830 https://doi.org/10.1109/ICC.2013.6654830
Toyoda, Kentaro ; Sasase, Iwao. / SPIT callers detection with unsupervised Random Forests classifier. IEEE International Conference on Communications. 2013. pp. 2068-2072
@inproceedings{bab6b5f9ac1f45b1a77e262f8b8be5cf,
title = "SPIT callers detection with unsupervised Random Forests classifier",
abstract = "As VoIP (Voice over IP) grows rapidly, it is expected to prevail tremendous unsolicited advertisement calls, which type of calls is referred to SPIT (SPam over Internet Telephony). SPIT detection is more difficult to execute than email SPAM detection since the callee or SPIT detection system does not tell whether it is SPIT or legitimate call until he/she actually takes a call. Recently, many SPIT detection techniques are proposed by finding outliers of call patterns. However, most of these techniques suffer from setting a threshold to distinguish that the caller is legitimate or not and this could cause to high false negative rate or low true positive rate. This is because these techniques analyse call pattern by a single feature e.g. call frequency or average call duration. In this paper, we propose a multi-feature call pattern analysis with unsupervised Random Forests classifier, which is one of the excellent classification algorithms. We also propose two simple but helpful features for better classification. We show the effectiveness of Random Forests based classification without supervised training data and which features contribute to classification.",
author = "Kentaro Toyoda and Iwao Sasase",
year = "2013",
doi = "10.1109/ICC.2013.6654830",
language = "English",
isbn = "9781467331227",
pages = "2068--2072",
booktitle = "IEEE International Conference on Communications",

}

TY - GEN

T1 - SPIT callers detection with unsupervised Random Forests classifier

AU - Toyoda, Kentaro

AU - Sasase, Iwao

PY - 2013

Y1 - 2013

N2 - As VoIP (Voice over IP) grows rapidly, it is expected to prevail tremendous unsolicited advertisement calls, which type of calls is referred to SPIT (SPam over Internet Telephony). SPIT detection is more difficult to execute than email SPAM detection since the callee or SPIT detection system does not tell whether it is SPIT or legitimate call until he/she actually takes a call. Recently, many SPIT detection techniques are proposed by finding outliers of call patterns. However, most of these techniques suffer from setting a threshold to distinguish that the caller is legitimate or not and this could cause to high false negative rate or low true positive rate. This is because these techniques analyse call pattern by a single feature e.g. call frequency or average call duration. In this paper, we propose a multi-feature call pattern analysis with unsupervised Random Forests classifier, which is one of the excellent classification algorithms. We also propose two simple but helpful features for better classification. We show the effectiveness of Random Forests based classification without supervised training data and which features contribute to classification.

AB - As VoIP (Voice over IP) grows rapidly, it is expected to prevail tremendous unsolicited advertisement calls, which type of calls is referred to SPIT (SPam over Internet Telephony). SPIT detection is more difficult to execute than email SPAM detection since the callee or SPIT detection system does not tell whether it is SPIT or legitimate call until he/she actually takes a call. Recently, many SPIT detection techniques are proposed by finding outliers of call patterns. However, most of these techniques suffer from setting a threshold to distinguish that the caller is legitimate or not and this could cause to high false negative rate or low true positive rate. This is because these techniques analyse call pattern by a single feature e.g. call frequency or average call duration. In this paper, we propose a multi-feature call pattern analysis with unsupervised Random Forests classifier, which is one of the excellent classification algorithms. We also propose two simple but helpful features for better classification. We show the effectiveness of Random Forests based classification without supervised training data and which features contribute to classification.

UR - http://www.scopus.com/inward/record.url?scp=84891355747&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84891355747&partnerID=8YFLogxK

U2 - 10.1109/ICC.2013.6654830

DO - 10.1109/ICC.2013.6654830

M3 - Conference contribution

SN - 9781467331227

SP - 2068

EP - 2072

BT - IEEE International Conference on Communications

ER -