Image of ANALISIS SENTIMEN ULASAN RESTORAN PADA TRIPADVISOR MENGGUNAKAN MERTODE SELEKSI FITUR INFORMATION GAIN DAN ALGORITMA SUPPORT VECTOR MACHINE (SVM)
Bookmark Share

Text

ANALISIS SENTIMEN ULASAN RESTORAN PADA TRIPADVISOR MENGGUNAKAN MERTODE SELEKSI FITUR INFORMATION GAIN DAN ALGORITMA SUPPORT VECTOR MACHINE (SVM)

Tripadvisor is a website that provides information about restaurants, user-generated restaurant reviews.data in the form of comment text. However, there are many features in review makes textual data ambiguous, thus causing difficulties sentiment analysis. To overcome this challenge, this research uses Information Gain feature selection method to reduce high features dimensions in sentiment analysis of Tripadvisor restaurant reviews. Exam The research results show that implementing the Information Gain feature selection method in the linear SVM kernel algorithm with a parameter C value 1 produces the highest performance. Results accuracy, precision, recall, and f-measure are 0.89, 0.89, 0.77, and 0.8, every. Next use this feature selection approach reduced the number of features and computing time from 12,198 to 2885 features and computing time of just 0.8 seconds. Besides that, The use of the SVM algorithm without feature selection produces inferior performance with accuracy 0.87, precision 0.89, recall of 0.74, f-measure of 0.79, and computing time of 0.98 seconds, considering a total of 12,198 features. These results show that accurate parameter selection and application of Information Obtaining feature selection methods can improve efficiency, effectiveness, and accuracy of sentiment analysis. This research seeks to improving sentiment analysis methods on large text data a number of features.

Availability
#
Central Library (References) T1554202024
T155420
Available but not for loan - Not for Loan
Detail Information
Series Title
-
Call Number
T1554202024
Publisher
Indralaya : Prodi Teknik Informatika, Fakultas Ilmu Komputer Universitas Sriwijaya.,
Collation
xv, VI-2 hlm.; ilus.; tab, 29 cm
Language
Indonesia
ISBN/ISSN
-
Classification
006.307
Content Type
Text
Media Type
unmediated
Carrier Type
other (computer)
Edition
-
Subject(s)
Kecerdasan Buatan
Prodi Teknik Informatika
Specific Detail Info
-
Statement of Responsibility
Other version/related
File Attachment
Comments
You must be logged in to post a comment