Ferdynand, Ferdynand (2024) Aspect based sentiment analysis on Amazon book review using distilbert base = Analisis sentimen berbasis aspek pada ulasan produk buku di Amazon menggunakan distilbert. Bachelor thesis, Universitas Pelita Harapan.
![Title [thumbnail of Title]](http://repository.uph.edu/style/images/fileicons/text.png)
Title.pdf
Restricted to Registered users only
Available under License Creative Commons Attribution Non-commercial Share Alike.
Download (34kB)
![Abstract [thumbnail of Abstract]](http://repository.uph.edu/style/images/fileicons/text.png)
Abstract.pdf
Restricted to Registered users only
Available under License Creative Commons Attribution Non-commercial Share Alike.
Download (265kB)
![ToC [thumbnail of ToC]](http://repository.uph.edu/style/images/fileicons/text.png)
ToC.pdf
Restricted to Registered users only
Available under License Creative Commons Attribution Non-commercial Share Alike.
Download (442kB)
![Chapter1 [thumbnail of Chapter1]](http://repository.uph.edu/style/images/fileicons/text.png)
Chapter1.pdf
Restricted to Registered users only
Available under License Creative Commons Attribution Non-commercial Share Alike.
Download (635kB)
![Chapter2 [thumbnail of Chapter2]](http://repository.uph.edu/style/images/fileicons/text.png)
Chapter2.pdf
Restricted to Registered users only
Available under License Creative Commons Attribution Non-commercial Share Alike.
Download (1MB)
![Chapter3 [thumbnail of Chapter3]](http://repository.uph.edu/style/images/fileicons/text.png)
Chapter3.pdf
Restricted to Registered users only
Available under License Creative Commons Attribution Non-commercial Share Alike.
Download (643kB)
![Chapter4 [thumbnail of Chapter4]](http://repository.uph.edu/style/images/fileicons/text.png)
Chapter4.pdf
Restricted to Registered users only
Available under License Creative Commons Attribution Non-commercial Share Alike.
Download (1MB)
![Chapter5 [thumbnail of Chapter5]](http://repository.uph.edu/style/images/fileicons/text.png)
Chapter5.pdf
Restricted to Registered users only
Available under License Creative Commons Attribution Non-commercial Share Alike.
Download (496kB)
![Bibliography [thumbnail of Bibliography]](http://repository.uph.edu/style/images/fileicons/text.png)
Bibliography.pdf
Restricted to Registered users only
Available under License Creative Commons Attribution Non-commercial Share Alike.
Download (415kB)
![Appendices [thumbnail of Appendices]](http://repository.uph.edu/style/images/fileicons/text.png)
Appendices.pdf
Restricted to Repository staff only
Available under License Creative Commons Attribution Non-commercial Share Alike.
Download (1MB)
Abstract
Product reviews are important sources of information and feedback for both consumers and businesses in e-commerce. However, existing sentiment analysis models are limited in their ability to capture fine-grained opinions on specific aspects of products, especially for books. Therefore, there is a need to create a more comprehensive and accurate Aspect-based Sentiment Analysis (ABSA) model for Amazon book reviews. The proposed solution is to use natural language processing (NLP) techniques, such as the Distillated Bidirectional Encoder Representations from Transformers (DistilBERT), to build an ABSA model tailored for Amazon book reviews, focusing on book-specific aspects. The research will use a dataset of Amazon book reviews, annotated with aspects and sentiment labels, and evaluate the model’s performance using various metrics. In this work, we propose a DistilBERT-based Aspect-based Sentiment Analysis (ABSA) model for Amazon book reviews. Existing sentiment analysis often struggles to capture fine-grained opinions on product aspects. This model addresses this limitation by focusing on book-specific aspects like plot, character development, and writing style. Evaluation using various metrics will ensure the model's accuracy and identify areas for improvement.
/
Ulasan produk adalah sumber informasi dan umpan balik penting bagi konsumen dan bisnis di e-commerce. Namun, model analisis sentimen yang ada terbatas dalam kemampuan mereka untuk menangkap pendapat tentang aspek-aspek tertentu dari produk, terutama untuk buku. Oleh karena itu, ada kebutuhan untuk mengembangkan model Aspect-based Sentiment Analysis (ABSA) yang lebih komprehensif dan akurat untuk ulasan buku Amazon. Solusi yang diusulkan adalah menggunakan teknik Natural Language Processing (NLP), seperti Distillated Bidirectional Encoder Representations from Transformers (DistilBERT), untuk membangun model ABSA yang disesuaikan untuk ulasan buku Amazon, dengan fokus pada aspek khusus buku. Penelitian ini akan menggunakan kumpulan data ulasan buku Amazon, dianotasi dengan aspek dan label sentimen, dan mengevaluasi kinerja model menggunakan berbagai metrik. Dalam penelitian ini, kami mengusulkan model Analisis Sentimen Berbasis Aspek (ABSA) berbasis DistilBERT untuk ulasan buku Amazon. Analisis sentimen yang ada seringkali kesulitan menangkap opini terperinci tentang aspek produk. Model ini mengatasi keterbatasan ini dengan berfokus pada aspek khusus buku seperti plot, pengembangan karakter, dan gaya penulisan. Evaluasi menggunakan berbagai metrik akan memastikan keakuratan model dan mengidentifikasi area untuk perbaikan.
Item Type: | Thesis (Bachelor) |
---|---|
Creators: | Creators NIM Email ORCID Ferdynand, Ferdynand NIM01082200014 ferdynandc@icloud.com UNSPECIFIED |
Contributors: | Contribution Contributors NIDN/NIDK Email Thesis advisor Samosir, Feliks Victor Parningotan NIDN1319049302 feliks.parningotan@uph.edu |
Uncontrolled Keywords: | aspect based sentiment analysis; distilBERT; NLP. |
Subjects: | Q Science > QA Mathematics > QA75 Electronic computers. Computer science |
Divisions: | University Subject > Current > Faculty/School - UPH Karawaci > School of Information Science and Technology > Informatics Current > Faculty/School - UPH Karawaci > School of Information Science and Technology > Informatics |
Depositing User: | FERDYNAND CHANDRA |
Date Deposited: | 23 Jul 2024 01:32 |
Last Modified: | 23 Jul 2024 01:32 |
URI: | http://repository.uph.edu/id/eprint/64222 |