Text

PERBANDINGAN ANALISIS SENTIMEN MENGGUNAKAN METODE K-NEAREST NEIGHBOR DAN MODIFIED K-NEAREST NEIGHBOR DENGAN SELEKSI FITUR INFORMATION GAIN

---

Penilaian

---

Netizens can express opinions on social media in written form to express their feelings or emotions. The number can be very large and the language is not always standard. Therefore, sentiment analysis is needed as a system that can analyze these opinions. In sentiment analysis, the problem with high dimensions of data attributes can have an effect on accuracy so that system performance is not satisfactory. This requires feature selection. In this study, sentiment analysis was carried out using the K-Nearest Neighbor (KNN) and Modified K-Nearest Neighbor (MK-NN) classification methods without feature selection and with Information Gain feature selection. In the PILKADA 1 dataset, the best accuracy results were obtained in the MK-NN test with the Information Gain feature selection by threshold 60% which is 59.8%. In the PILKADA 2 dataset, the best accuracy results were obtained in the MK-NN test with the Information Gain feature selection by threshold 40%, which is 78.2%. From the test results, it is found that MK-NN is not always better than KNN and vice versa. Information Gain can increase accuracy but depends on the selected threshold. MK-NN and Information Gain can produce better accuracy but depends on the choice of threshold. Decrease and increase in accuracy of less than 7%.

---

Availability

Inventory Code	Barcode	Call Number	Location	Status
2207000446	T64958	T649582022	Central Library (Referens)	Available but not for loan - Not for Loan

Detail Information

Series Title

-

Call Number

T649582022

Publisher

Inderalaya : Jurusan Teknik Informatika, Fakultas Ilmu Komputer Universitas Sriwijaya., 2022

Collation

xix, 320 hlm.; ilus.; 29 cm

Language

Indonesia

ISBN/ISSN

-

Classification

006.754 07

Content Type

-

Media Type

-

Carrier Type

-

Edition

-

Subject(s)

Jurusan Teknik Informatika
Situs Jejaring Sosial

Specific Detail Info

-

Statement of Responsibility

SEPTA

Other version/related

File Attachment

Comments

---

You must be logged in to post a comment