Support Vector Machine VS Information Gain: Analisis Sentimen Cyberbullying di Twitter Indonesia

Destitus, Christevan; Wella, Wella; Suryasari, Suryasari

doi:10.31937/si.v11i2.1740

Cited by 5 publications

(3 citation statements)

References 3 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…Cyberbullying dapat diartikan sebagai bentuk penyalahgunaan yang dimaksudkan untuk melecehkan, mengancam, mempermalukan dan mengejek orang lain (Destitus, Wella, & Suryasari, 2020). Pada dasarnya kasus cyberbullying melalui platform media sosial seperti Instagram merupakan salah satu upaya untuk mengintimidasi dan menggretak orang dengan mengirim pesan mengintimidasi atau mengancam, Cyberbulliying juga menjadikan ujaran kebencian sebagai salah satu bentuk intimidasi terhadap seseorang atau kelompok tertentu (Rachmayanti & Candrasari, 2022).…”

Section: Pendahuluanunclassified

Komparasi Optimasi Analisis Sentimen Cyberbullying Pada Instagram Berbasis Particle Swarm Optimization

Herliana,

Muawiyah

2024

j. responsif ris. sains inform.

View full text Add to dashboard Cite

Sejak masa pandemi covid-19 melanda dunia, sekitar 78,19% manusia di Indonesia mengandalkan media internet sebagai penunjang utama kegiatan sehari-hari. Hal ini membuat aktivitas manusia mayoritas dilakukan melalui dunia maya, salah satunya adalah sebagai bentuk eksistensi. Media sosial seperti Instagram, menjadi pilihan dari banyak manusia di dunia utamanya Indonesia untuk menyalurkan segala macam bentuk aspirasinya. Efek dari meningkatnya postingan di media sosial ini juga berimbas kepada tingginya tingkat perundungan melalui dunia maya yang sering dikenal dengan istilah cyberbullying. Salah satu bentuk cyberbullying yang marak terjadi adalah melalui ujaran kebencian dan kata-kata yang tidak baik terhadap postingan yang diunggah. Pada penelitian kali ini akan dilakukan optimasi untuk mengetahui analisis sentimen terhadap berbagai tindak cyberbullying yang ada pada media sosial Instagram agar dapat ditindak lebih lanjut dengan menggunakan metode Particle Swarm Optimization (PSO). Sedangkan untuk metode klasifikasi dari analisis sentiment pada penelitian kali ini dilakukan komparasi dengan menggunakan algoritma support vector machine (SVM) dan naïve bayes. Dari hasil penelitian diketahui bahwa performa metode PSO memberikan hasil yang lebih baik jika dikombinasikan dengan metode SVM yang mencapai nilai akurasi 78,60% dengan dukungan 100% class precission. Sedangkan hasil naïve bayes hanya mencapai nilai akurasi 78,00% dengan dukungan class precission sebesar 99,74%.

show abstract

Section: Pendahuluanunclassified

Komparasi Optimasi Analisis Sentimen Cyberbullying Pada Instagram Berbasis Particle Swarm Optimization

Herliana,

Muawiyah

2024

j. responsif ris. sains inform.

View full text Add to dashboard Cite

show abstract

“…This also includes sharing personal information that can lead to embarrassment or humiliation. [2] Figure 1. Number of Internet Users in Indonesia in 2018 -2022 [3] The statistics regarding internet usage in Indonesia are quite staggering, with 212.9 million internet users at the beginning of this year, which is around 77% of the country's total population of 276.4 million.…”

Section: Introductionmentioning

confidence: 99%

Analysis of Cyberbullying on Social Media Using A Comparison of Naïve Bayes, Random Forest, and SVM Algorithms

Ahmad,

Insani,

Salim

2024

J. teknol. inf. pendidik.

View full text Add to dashboard Cite

Social media allows the public, especially the younger generation, to access information and knowledge or communicate with others online. Unfortunately, the phenomenon of bullying has evolved into cyberbullying, encompassing various forms of violence such as taunting, insults, intimidation, or harassment carried out by young individuals through digital technology or social media platforms. Therefore, considering the available data, there is a need for a method to classify text comments on social media, whether they fall into the category of cyberbullying or not. One of the methods used is the creation of a cyberbullying classification model using a Support Vector Machine (SVM), Random Forest (RF), and Naive Bayes algorithms. This research aims to analyze cyberbullying in social media by comparing three different algorithms, namely Naïve Bayes, Random Forest, and SVM. The research results show that in the classification analysis, the Support Vector Machine (SVM) model performed the best, with an accuracy of 85%, precision of 79.93%, and recall of 94.29%. The Naive Bayes model also provided satisfactory results, with an accuracy of around 82.19%, precision of 81.29%, and recall of 85.10%. Meanwhile, the Random Forest (RF) model had a lower accuracy of approximately 73.15%, with a precision of 74.05% and a recall of 77.79%.

show abstract

“…Currently, there is much research on detecting hoax news on Twitter. Several methods are often used in hoax news detection research on Twitter, such as the Support Vector Machine method [1]. Another study is Naïve Bayes was researched by Rahutomo et al [2], Backpropagation was researched by Lhaksmana et al [3], Naïve Forecast Method [12] and Prasetyo, Rino [4] studied Modified K-Nearest Neighbor.…”

Section: Introductionmentioning

confidence: 99%

The Detection of Indonesian Hoax Content about COVID-19 Vaccine using Naive Bayes Multinomial Method

Kirana

Prasetyo

Lestari

2023

Indones.J.electronic.electromed.med.inf

View full text Add to dashboard Cite

One media currently famously used in all worlds is twitter. The ease of dissemination and the exchange of information is accelerating. Every day, millions of tweets exist using various information, such as politics, technology, sports, academics, and others. The information that is widely found is about COVID-19-19 nowadays. The information on Twitter is not entirely accurate or according to facts and needs to be proven true. Therefore, this study aims to try to detect the information contained in Indonesia using methods of Naive Bayes Multinomial by using the Information Gain feature selection. The classification process is carried out by crawling tweets, preprocessing, then using feature selection, namely Information Gain, and classification using the Multinomial Naive Bayes method. Meanwhile, the validation needs in this study use k-fold cross-validation where the existing dataset is divided into training and testing data that will be tested with a confusion matrix. Researchers have carried out the confusion matrix testing process using 720 datasets divided as train data & the test data received an average accuracy value of 81.39%, precision of 80.36%, and recall of 79.73%. The highest accuracy is using k-fold two. The accuracy value reaches 88.8%, the precision value is 79.1%, and the recall value is 86.3%. The lowest accuracy was obtained on the 8th k-fold with an accuracy value of 73.6%, precision 75.4%, and recall 86.9%.

show abstract

Support Vector Machine VS Information Gain: Analisis Sentimen Cyberbullying di Twitter Indonesia

Cited by 5 publications

References 3 publications

Komparasi Optimasi Analisis Sentimen Cyberbullying Pada Instagram Berbasis Particle Swarm Optimization

Komparasi Optimasi Analisis Sentimen Cyberbullying Pada Instagram Berbasis Particle Swarm Optimization

Analysis of Cyberbullying on Social Media Using A Comparison of Naïve Bayes, Random Forest, and SVM Algorithms

The Detection of Indonesian Hoax Content about COVID-19 Vaccine using Naive Bayes Multinomial Method

Contact Info

Product

Resources

About