Klasifikasi Ulasan Buku Menggunakan Algoritma Convolutional Neural Network - Long Short Term Memory

Arham, Anshar Zamrudillah (2018) Klasifikasi Ulasan Buku Menggunakan Algoritma Convolutional Neural Network - Long Short Term Memory. Undergraduate thesis, Institut Teknologi Sepuluh Nopember.

[img] Text
06111340000118-Undergraduate_Theses.pdf - Published Version
Restricted to Repository staff only

Download (3MB) | Request a copy

Abstract

Salah satu parameter untuk mengukur kualitas suatu produk adalah ulasan konsumen terhadap produk tersebut, apakah negatif (mengecewakan) atau positif (memuaskan). Ulasan dari konsumen berbeda dengan ulasan dari pedagang atau distributor yang mengandung unsur promosi. Ulasan konsumen cendrung bersifat jujur berdasarkan opini masing-masing. Pada zaman meluasnya perdagangan hingga ke dunia maya ini, diperlukan proses ekstraksi opini dari ulasan konsumen yang biasa disebut opinion mining. Proses ini dilakukan untuk mengetahui kecendrungan reviewer terhadap objek yang diulasnya. Salah satu model yang digunakan untuk melakukan opinion mining yang sedang berkembang saat ini adalah deep learning. Pada Tugas Akhir ini dilakukan klasifikasi opini konsumen menggunakan kombinasi algoritma Convolutional Neural Network (CNN) dan Long Short Term Memory (LSTM). Metode ini diimplementasikan untuk data ulasan buku yang diperoleh dari Amazon.com. Hasil dari metode tersebut menunjukkan performansi yang lebih baik daripada kombinasi algoritma CNN - L2-SVM. ============= One of parameter to measure the quality of a product is the consumer reviews of the product, whether negative (disappointing) or positive (satisfactory). The customer reviews are different with merchants or distributors reviews that contain promotional value. Consumer reviews tend to be honest based on their respective opinions. In this era of widespread trade to the virtual world, it takes the process of opinion extraction from consumer reviews which commonly called opinion mining. The goal of this process is to determine the reviewer tendency towards the reviewed object. One of the evolve models of opinion mining is deep learning. This Final Project is classifying book using a combination of Convolutional Neural Network (CNN) and Long Short Term Memory (LSTM) algorithms. This method is implemented for book review data obtained from Amazon.com. The results of this method show a better performance than the combination CNN-L2-SVM algorithm.

Item Type: Thesis (Undergraduate)
Uncontrolled Keywords: Reviews; Opinion Mining; Natural Language Processing; Deep Learning
Subjects: Q Science > QA Mathematics
Q Science > QA Mathematics > QA76.6 Computer programming.
Q Science > QA Mathematics > QA76.87 Neural networks (Computer Science)
Divisions: Faculty of Mathematics and Science > Mathematics > (S1) Undergraduate Theses
Depositing User: Anshar Zamrudillah Arham
Date Deposited: 26 Apr 2018 05:50
Last Modified: 26 Apr 2018 05:50
URI: http://repository.its.ac.id/id/eprint/51134

Actions (login required)

View Item View Item