Implementation of Deep Learning Based Sentiment Classification and Product Aspect Analysis

Abstract

With the increase in E-Commerce businesses in the last decade,the sentiment analysis of product reviews has gained a lot of attention in linguistic research. In literature, the survey depicts the majority of the research done emphasizes on mere polarity identification of the reviews. The proposed system emphasized on classifying the sentiment polarity and the product aspect identification from the reviews. Proposed work experimented with traditional machine learning techniques as well as deep neural networks such as Convolutional Neural Network (CNN), Recurrent Neural Network (RNN) and Long Short Term Memory(LSTM) Networks. The proposed system gives a better understanding of these algorithms by comparing the outcomes. The Deep Learning approach in the proposed work successfully provides a mechanism which identifies the review polarity and intensity of the reviews and also analyses the short form words used by people in the reviews. The experimental results in this work, applied on amazon product dataset, shows that the LSTM model works the best for sentiment analysis and intensity of reviews with 93% accuracy. This research work also predicts polarity for short-form word reviews which is the common trend these days while writing the reviews.

Aspect-Based Sentiment Analysis Pada Aplikasi Pelacakan Kasus Covid-19 (Studi Kasus: Pedulilindungi)

“…Sebuah penelitian menggunakan perbandingan tiga arsitektur deep learning yaitu CNN, RNN, dan LSTM akurasi terbaik menunjukan model LSTM sebesar 93% [7]. ABSA pada domain ulasan data restoran berbahasa Indonesia dengan kombinasi metode BERT-CNN, ELMo-CNN, dan Word2vec-CNN hasil terbaik ditunjukan pada model ELMo-CNN dengan nilai micro-average precision sebesar 0.88%, micro average recall sebesar 84%, dan fl-score sebesar 0.86% untuk klasifikasi sentimen memberikan hasil terbaik pada BERT-CNN dengan nilai precision sebesar 0.89%, recall sebesar 0.86 dan f1-score sebesar 0.91 [8].…”

“…CNN sendiri sangat umum digunakan pada data image. Meskipun CNN telah banyak diaplikasikan pada image classification, CNN juga mampu mengatasi pada permasalahan data teks [7]. CNN terdiri dari beberapa layer, layer pertama menyimpan kata dalam sebuah low-dimensional vector, layer kedua menjalankan convolutions menggunakan multiple filter size, selanjutnya max-pool hasil dari layer convolutional ke dalam long feature vector, menambahkan fully connected dropout regularization, dan mengklasifikasi hasil menggunakan ReLu layer [21].…”

Aspect-Based Sentiment Analysis using Machine Learning and Deep Learning Approaches

“…The results obtained demonstrated that the suggested model boosted the accuracy of ABSA systems by around 5 to 20%. The researchers in [76], suggested a system in which they classified the polarity of sentiments and also identified the aspects of products from the reviews given by customers. In their work, they analyzed the performance of standard ML techniques along with the DL methods including CNN, RNN and LSTM.…”