Gama, Athyah Danni Surya (2019) Implementasi Analisis Sentimen dalam Klasifikasi Artikel Berita Daring Menggunakan Algoritma Hybrid Convolutional Neural Network (Hybrid CNN-HMM). Other thesis, Institut Teknologi Sepuluh Nopember.
Preview |
Text
06111540000061_Undergraduate_Theses.pdf Download (3MB) | Preview |
Abstract
Era Big Data dan Internet of Things telah merambah hampir pada semua bidang, di antaranya informasi komunikasi, pendidikan, industri bahkan dalam bidang keuangan. Seperti yang telah diketahui, bidang informasi dan komunikasi merupakan bidang yang menjadi sumber paling berpengaruh pada bidang keuangan. Bidang ini akan sangat dipengaruhi oleh informasi yang muncul dan berkembang di masyarakat. Informasi negatif maupun positif yang muncul melalui berita akan mempengaruhi saham. Untuk mengetahui hal tersebut, diperlukan sebuah metode yang mampu melakukan hal tersebut. Salah satu metode yang dapat digunakan adalah analisis sentimen. Pada penelitian ini akan dikembangkan algoritma untuk mencari sentimen/opini yang berkembang melalui pemberitaan di media berita online / daring (dalam jaringan). Untuk mendapatkan berita dari potal berita tersebut, terlebih dahulu akan dilakukan pengambilan data menggunakan metode web crawling. Kemudian untuk melakukan pengenalana sentimen/opini akan digunakan algoritma Hybrid Convolutional Neural Network-Hidden Markov Models (Hybrid CNN-HMM). Hasil dari metode tersebut menunjukkan performansi pembelajaran 99.92% dan performansi pengujian sekitar 70,45 %.
=================================================================================================================================
The Big Data era and the Internet of Things have penetrated almost all fields, including information on communication, education, industry and even in the financial sector. As is well known, the field of information and communication is a field that has become the most influential source of finance. This field will be greatly influenced by information that arises and develops in the community. Negative and positive information that appears through the news will affect shares. To find out this, a method is needed to do this. One method that can be used is sentiment analysis. In this study, algorithms will be developed to find sentiments/opinions that develop through reporting in online news media. To get news from the total news, first, it will take data retrieval using the web crawling method. Then to identify sentiments/opinions will be used Hybrid Convolutional Neural Network algorithm - Hidden Markov Models (Hybrid CNN-HMM). The results of the method show a learning performance of around 99.92% and a testing performance of around 70.45%.
| Item Type: | Thesis (Other) |
|---|---|
| Additional Information: | RSMa 006 Gam i-1 2019 |
| Uncontrolled Keywords: | web crawling, sentiment analysis, Hybrid CNN-HMM, data mining, ekonomi dan keuangan. |
| Subjects: | Q Science > QA Mathematics > QA274.7 Markov processes--Mathematical models. Q Science > QA Mathematics > QA75 Electronic computers. Computer science. EDP Q Science > QA Mathematics > QA76.87 Neural networks (Computer Science) Q Science > QA Mathematics > QA76.9.D343 Data mining. Querying (Computer science) Q Science > QA Mathematics > QA76.9.U83 Graphical user interfaces. User interfaces (Computer systems)--Design. T Technology > T Technology (General) > T57.83 Dynamic programming |
| Divisions: | Faculty of Mathematics and Science > Mathematics > 44201-(S1) Undergraduate Thesis |
| Depositing User: | Athyah Danni Surya Gama |
| Date Deposited: | 26 Oct 2023 07:58 |
| Last Modified: | 26 Oct 2023 07:58 |
| URI: | http://repository.its.ac.id/id/eprint/66406 |
Actions (login required)
 |
View Item |