Image of ANALISIS SENTIMEN PENGGUNA INSTAGRAM TERHADAP POSTINGAN PROGRAM MAKAN SIANG GRATIS DENGAN KLASIFIKASI SENTIMEN BIDIRECTIONAL ENCODER REPRESENTATIONS FROM TRANSFORMERS
Bookmark Share

Skripsi

ANALISIS SENTIMEN PENGGUNA INSTAGRAM TERHADAP POSTINGAN PROGRAM MAKAN SIANG GRATIS DENGAN KLASIFIKASI SENTIMEN BIDIRECTIONAL ENCODER REPRESENTATIONS FROM TRANSFORMERS

Social media is widely used today for communication and information exchange. One of the most popular social media platforms is Instagram. The comments section on an Instagram post creates discussions among users regarding issues that are currently being widely discussed. One of the issues that is currently being widely discussed is the free lunch program, which is a new program created by the government and has garnered mixed reactions from the public due to its significant impact. Public sentiment needs to be analyzed to determine whether the free lunch program is really needed by the community, especially the community in South Sumatra. The data for sentiment analysis comes from Instagram user comments in South Sumatra, which were obtained using scraping techniques. The data was then pre-processed and divided. Next, fine-tuning was performed to train the BERT model. Finally, an evaluation was conducted to assess the model's performance in classifying user sentiment as positive, neutral, or negative towards the program. The sentiment analysis uses the BERT (Bidirectional encoder representations from transformers) method, which is a language model based on a transformer architecture that understands context in two directions. The results show that positive sentiment received the highest F1 score of 0.92 and the lowest F1 score in the negative class of 0.81. The BERT model can recognize sentiment quite effectively, especially in positive comments. Overall, the BERT model successfully analyzed sentiment with an accuracy rate of 88.26%.

Availability
#
Central Library (Reference) T1946112026
T194611
Available but not for loan - Not for Loan
Detail Information
Series Title
-
Call Number
T1946112026
Publisher
Indralaya : Prodi Sistem Informasi, Fakultas Ilmu Komputer Universitas Sriwijaya.,
Collation
xiv, 46 hlm.; ilus.; 29 cm
Language
Indonesia
ISBN/ISSN
-
Classification
006.350 7
Content Type
Text
Media Type
unmediated
Carrier Type
-
Edition
-
Subject(s)
Prodi Sistem Informasi
Analisis Sentimen (Ilmu Komputer)
Specific Detail Info
-
Statement of Responsibility
Other version/related
File Attachment
Comments
You must be logged in to post a comment