Page 47 - Conference Book The 3rd IConVET "Future of TVET Graduates: Developing Talent for Industry 4.0 and The New Normal"

The 3 International Conference
rd
on Vocational Education and Technology (IConVET)
Universitas Pendidikan Ganesha - Bali | 7 November, 2020
th
DETECTION OF DOS ATTACKS USING NAIVE BAYES
METHOD BASED ON INTERNET OF THINGS (IOT)

F F Setiadi , M W A Kesiman and K Y E Aryanto 1,c
1,a
1,b

1 Department of Computer Science, Graduate Program, Universitas
Pendidikan Ganesha, Singaraja, Bali, Indonesia

b
a
Email : fransferysetiadi@gmail.com, antara.kesiman@undiksha.ac.id,
c yota.ernanda@undiksha.ac.id

Abstract. Internet of Things (IoT) is one form of technology that is trending
today. Interconnecting networks on IoT, that useful in the automation
process, have vulnerabilities to network-based disruptions and attacks such
as Denial of Service (DoS). This study aims to implement the Naive Bayes
algorithm to predict attribute classes using training datasets from NSLKDD
with the KDD99 format and testing data obtained from the log process of
DoS attacks on IoT-based devices. The advantage of using Naive Bayes is
that this method only requires a small amount of training data to determine
the estimated parameters needed in the classification process. The results of
research conducted have been able to detect attacks on IoT devices by using
the help of snort tools to capture traffic logs. The results from the log are
then converted into KDD99 format and processed by the Naive Bayes
method. This research uses a training dataset from NSLKDD with KDD99
format which is widely used in various studies and testing data obtained from
the IDS log process on the Raspberry Pi 3. The attributes used are 9 attributes
namely service, flag, src_bytes, dst_bytes, srv_serror_rate, same_srv_rate,
diff_srv_rate, dst_host_srv_diff_host_rate and dst_host_srv_serror_rate.
The results of the research analysis showed an accuracy of 64.02%. These
results are good, but some are still different from the actual results because
the testing data and training data are taken from two different datasets, so
they have different characteristics.

| IConVET 2020 | 47

https://conference.undiksha.ac.id/iconvet/ | 47 |