PERBAIKAN FEATURE SELECTION PADA SUPPORT VECTOR MACHINE MENGGUNAKAN PARTICLE SWARM OPTIMIZATION UNTUK SENTIMENT ANALYSIS TWITTER PARIWISATA INDONESIA

Wahyu Setiawan, Kusrini
UNIVERSITAS AMIKOM YOGYAKARTA.2017

A B S T R A C T

With the rapid technological developments at this time, there is a lot of information presented through social media. The social media is allow us to send a news to friends, relatives, media with ease of use. Examples such as the use of Twitter. Through text mining, we can collect statements contained in Twitter as the main material for sentiment analysis. Therefore, Support vector machine is one of many approaches in machine learning to predict sentiment classification task.

Particle Swarm Optimization is an approach to classify data based on properties of each particle which can support selection task in the Feature Selection Process. Feature Selection is an activity that generally can be done by preprocessing. Main task of this process is to choose the features that effect and override a feature where the features does not affect in any activity modeling or analyzing data. So that the opinion can be identified can be classified in the negative, positive and neutral based on the properties of each particle.

Keywords - Support Vector Machine, Particle Swarm Optimization, Data Mining, Kecerdasan Buatan, Sistem, Optimalisasi, Sentiment Analysis.

CategoryUndergraduate Thesis
Posted Date( undocumented )
Modified Date14 September 2017
Download File Publikasi_15.21.0845.pdf
Google Scholar