KLASIFIKASI SHORT MESSAGE SERVICE (SMS) SPAM MENGGUNAKAN SUPPORT VECTOR MACHINE (SVM) DAN ARTIFICIAL NEURAL NETWORK (ANN)

PRATAMA, SHATIA EARLANGGA and Rini, Dian Palupi and Rodiah, Desty (2025) KLASIFIKASI SHORT MESSAGE SERVICE (SMS) SPAM MENGGUNAKAN SUPPORT VECTOR MACHINE (SVM) DAN ARTIFICIAL NEURAL NETWORK (ANN). Undergraduate thesis, Sriwjaya University.

[thumbnail of RAMA_55201_09021182126017_cover.pdf] Text
RAMA_55201_09021182126017_cover.pdf - Accepted Version
Available under License Creative Commons Public Domain Dedication.
Download (217kB)
[thumbnail of RAMA_55201_09021182126017.pdf] Text
RAMA_55201_09021182126017.pdf - Accepted Version
Restricted to Repository staff only
Available under License Creative Commons Public Domain Dedication.
Download (5MB)
[thumbnail of RAMA_55201_09021182126017_TURNITIN.pdf] Text
RAMA_55201_09021182126017_TURNITIN.pdf - Accepted Version
Restricted to Repository staff only
Available under License Creative Commons Public Domain Dedication.
Download (8MB)
[thumbnail of RAMA_55201_09021182126017_0023027804_0021128905_01_front_ref.pdf] Text
RAMA_55201_09021182126017_0023027804_0021128905_01_front_ref.pdf - Accepted Version
Available under License Creative Commons Public Domain Dedication.
Download (1MB)
[thumbnail of RAMA_55201_09021182126017_0023027804_0021128905_02.pdf] Text
RAMA_55201_09021182126017_0023027804_0021128905_02.pdf - Accepted Version
Restricted to Repository staff only
Available under License Creative Commons Public Domain Dedication.
Download (941kB)
[thumbnail of RAMA_55201_09021182126017_0023027804_0021128905_03.pdf] Text
RAMA_55201_09021182126017_0023027804_0021128905_03.pdf - Accepted Version
Restricted to Repository staff only
Available under License Creative Commons Public Domain Dedication.
Download (930kB)
[thumbnail of RAMA_55201_09021182126017_0023027804_0021128905_04.pdf] Text
RAMA_55201_09021182126017_0023027804_0021128905_04.pdf - Accepted Version
Restricted to Repository staff only
Available under License Creative Commons Public Domain Dedication.
Download (1MB)
[thumbnail of RAMA_55201_09021182126017_0023027804_0021128905_05.pdf] Text
RAMA_55201_09021182126017_0023027804_0021128905_05.pdf - Accepted Version
Restricted to Repository staff only
Available under License Creative Commons Public Domain Dedication.
Download (671kB)
[thumbnail of RAMA_55201_09021182126017_0023027804_0021128905_06.pdf] Text
RAMA_55201_09021182126017_0023027804_0021128905_06.pdf - Accepted Version
Restricted to Repository staff only
Available under License Creative Commons Public Domain Dedication.
Download (250kB)
[thumbnail of RAMA_55201_09021182126017_0023027804_0021128905_07_ref.pdf] Text
RAMA_55201_09021182126017_0023027804_0021128905_07_ref.pdf - Bibliography
Restricted to Repository staff only
Available under License Creative Commons Public Domain Dedication.
Download (250kB)
[thumbnail of RAMA_55201_09021182126017_0023027804_0021128905_08_lamp.pdf] Text
RAMA_55201_09021182126017_0023027804_0021128905_08_lamp.pdf - Accepted Version
Restricted to Repository staff only
Available under License Creative Commons Public Domain Dedication.
Download (127kB)

Abstract

Short Message Service (SMS) is still used as a communication medium for promotions, notifications, and official information. However, SMS is also prone to misuse in the form of spam that can be disruptive and potentially deceive users. To address this issue, an accurate SMS spam and ham classification system is needed. This study developed an SMS spam classification system using SVM and ANN with a MLP architecture. The dataset used is a secondary dataset from the Kaggle platform, consisting of 1,143 SMS messages with a balanced class distribution of 574 spam and 569 ham messages. Feature extraction was carried out using TF-IDF and n-gram, while feature selection used Pearson Correlation. Model parameters were determined using Grid Search. The study was conducted through six testing scenarios. The results showed that the SVM model with a combination of TF-IDF and n-gram without feature selection achieved the best performance, with an accuracy of 98.25%, precision of 97.50%, recall of 99.15%, and F1-score of 98.32%. The best model used a linear kernel with a C value of 1. The findings indicate that SVM outperformed MLP-based ANN in SMS spam classification.

Item Type: Thesis (Undergraduate)
Uncontrolled Keywords: SMS Spam, Classification, SVM, ANN, MLP, TF-IDF, N-gram, Pearson Correlation, Kaggle, Grid Search
Subjects: Q Science > Q Science (General) > Q334-342 Computer science. Artificial intelligence. Algorithms. Robotics. Automation.
Divisions: 09-Faculty of Computer Science > 55201-Informatics (S1)
Depositing User: Shatia Earlangga Pratama
Date Deposited: 21 Jul 2025 07:34
Last Modified: 21 Jul 2025 07:34
URI: http://repository.unsri.ac.id/id/eprint/179421

Actions (login required)

View Item View Item