Sentiment analysis of Indonesian hotel reviews: from classical machine learning to deep learning

Abstract

Currently, there are a large number of hotel reviews on the Internet that need to be evaluated to turn the data into practicable information. Deep learning has excellent capabilities for recognizing this type of data. With the advances in deep learning paradigms, many algorithms have been developed that can be used in sentiment analysis tasks. In this study, we aim to compare the performance of classical machine learning algorithms—logistic regression (LR), naïve Bayes (NB), and support vector machine (SVM) using the Word2Vec model in conjunction with deep learning algorithms such as a convolutional neural network (CNN) to classify hotel reviews on the Traveloka website into positive or negative classes. Both learning methods apply hyperparameter tuning to determine the parameters that produce the best model. Furthermore, the Word2Vec model parameters use the skip-gram model, hierarchical softmax evaluation, and the value of 100 vector dimensions. The highest average accuracy obtained was 98.08% by using the CNN with a dropout of 0.2, Tanh as convolution activation, softmax as output activation, and Adam as the optimizer. The findings from the study demonstrate that the integration of the Word2Vec model and the CNN model obtains significantly better accuracy than other classical machine learning methods.

Sentiment Analysis Of Hotel Reviews On Tripadvisor With LSTM And ELECTRA

“…In the text processing process, text data will be processed to remove irrelevant information and improve data quality. In the process of text processing, several actions are performed such as: (Kusumaningrum et al, 2021b) 1) Remove punctuation and make text lowercase 2) Remove stopwords like "the", "and", "a", "to", etc 3) Doing lemmatization by modifying words into basic forms (lemma) such as "relaxed" becomes "relaxed" The results of the text after going through the text processing process are cleaner and have more relevant and concentrated information as shown in the following figure. After the text data cleaning process, the number of sentiments in the resulting data is often unbalanced, where the number of positive or negative sentiments tends to be more than the number of neutral sentiments.…”

“…Previously, sentiment analysis of hotel reviews was generally carried out on datasets originating from various sites such as Agoda (Sambas et al, 2022), Traveloka (Cendani et al, 2023, Google Map (Sambas et al, 2022), andTripadvisor (Baskoro et al, 2021). Various algorithms that have been researched include Random Forest (Utami, 2021a), Convolutional Neural Network (Kusumaningrum et al, 2021a), Long Short-Term Memory (LSTM) (Priyantina & Sarno, 2019), dan Reccurent Neural Network (Utami, 2021a). This research itself will use the LSTM algorithm because it has been widely used for processing text data.…”

Analisis Perbandingan Metode Tf-Idf dan Word2vec pada Klasifikasi Teks Sentimen Masyarakat Terhadap Produk Lokal di Indonesia

“…Proses selanjutnya adalah pembersihan data, dimana angka, tanda baca, dan tanda unik dihilangkan. Langkah selanjutnya adalah lower case, di mana semua kata distandarisasi dengan huruf kecil [18]. Setelah semua kata direduksi menjadi huruf kecil, masuk ke proses stopword removal.…”

“…Dalam tahap ini akan menggunakan confusion matrix, confusion matrix adalah tabulasi dari [17] perhitungan yang didasari pada evaluasi kinerja model klasifikasi berdasarkan jumlah objek penelitian yang diprediksi dengan benar dan salah. Secara singkat confusion matrix memberikan perincian terkait kesalahan klasifikasi [18]. Dalam tahap ini akan dibandingan model mana yang memiliki nilai F1-Score yang paling tinggi.…”

Attention-based Sentence Extraction for Aspect-based Sentiment Analysis with Implicit Aspect Cases in Hotel Review Using Machine Learning Algorithm, Semantic Similarity, and BERT

“…In the condition of determining aspect term keywords, it will certainly cause errors in the text extraction process in determining aspects and opinion terms which will have an impact on inaccurate determination of aspect categories and sentiment polarity. Therefore, a text extraction method is needed to get the right and accurate aspects and opinion terms in hotel reviews [12,13].…”