Graph-Based Semi-Supervised Deep Learning for Indonesian Aspect-Based Sentiment Analysis

Chamid, Ahmad Abdul; Widowati, Widowati; Kusumaningrum, Retno

doi:10.3390/bdcc7010005

Cited by 11 publications

(5 citation statements)

References 45 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Looking at the overall kappa value, it is known that this study obtained better kappa results than previous studies [32], In addition, the kappa value obtained from this study is superior to that [33]. So the labeling results in the form of a dataset can be used for classification tasks, especially to develop the ABSA model, and we have used these results to develop the Indonesian ABSA model with good performance results [27].…”

Section: Class Aspectmentioning

confidence: 59%

“…Some classification tasks whose results are not good can be influenced by the data labeling process and other possibilities [26]. However, based on a good data labeling process, it has been proven to produce good classification [27,28]. So in this research, a data labeling process and consistency test were carried out using the Cohen Kappa method, to obtain good labeled data that can be used for classification tasks, especially aspect-based sentiment analysis.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Labeling Consistency Test of Multi-Label Data for Aspect and Sentiment Classification Using the Cohen Kappa Method

Chamid,

Widowati,

Kusumaningrum

2024

ISI

View full text Add to dashboard Cite

Classification is part of machine learning, and developing it requires labeled data. Most data is available in an unlabeled form. Data labeling is a step that researchers must take. Good labeled data will produce a good classification model. The data labeling process cannot be ignored and needs to be done carefully and consistently. Because the classification process requires well-labeled data that can be accounted for. In addition, good labeled data will produce a good classification model. The role of an expert (rater) is needed to label the data and ideally at least two experts. However, involving two raters will become a new problem because it is likely that the results of the inter-rater labeling will be different. We propose the Cohen Kappa method to overcome this problem. We used data from scraping user reviews of the Indonesian marketplace, there were 4.307. Based on the calculation results, Kappa=0.909 for aspect detection, Kappa=0.893 for sentiment classification, and Kappa=0.971 for class aspect. Based on the kappa value, the labeling results for aspect detection, sentiment classification and aspect class were declared "almost perfect agreement", so that the results of this research obtained labeled data that can be used for classification tasks, especially for developing aspect-based sentiment analysis models.

show abstract

Section: Class Aspectmentioning

confidence: 59%

Section: Introductionmentioning

confidence: 99%

Labeling Consistency Test of Multi-Label Data for Aspect and Sentiment Classification Using the Cohen Kappa Method

Chamid,

Widowati,

Kusumaningrum

2024

ISI

View full text Add to dashboard Cite

show abstract

“…Topic Dataset Used Classification Method [33] Negation in Sentence Kaggle Naïve Bayes, SVM, ANN, RNN [34] Market Value of AI and ML Google API Services Naïve Bayes, Decision Tree, Random Forest [35] Indonesian Hotel Review Traveloka Website LSTM [36] Subsistence Marketplace Respondents' Data in Bangladesh Common Method Variance (CMV) [37] Mature Destination TripAdvisor, St. Mark Square, and the Doge's Palace LSTM [38] Indonesian-Based Aspect Sentiment Indonesian Marketplace…”

Section: Refsmentioning

confidence: 99%

Sentiment Analysis on Marketplace in Indonesia using Support Vector Machine and Naïve Bayes Method

Dakwah,

Firdaus,

Furizal

et al. 2024

J. Ilm. Tek. Elektro Komput. Dan Inform

View full text Add to dashboard Cite

This research addresses the challenges of marketplace customer feedback, which is an important aspect in today's era of online transactions. Marketplaces often receive many unsatisfactory comments from their customers through social media platforms. One approach that can be used to address this is sentiment analysis. This research contributes new insights as recommendations for marketplaces based on customer opinions on available services and delivery. The sentiment analysis methods used are Naive Bayes and Support Vector Machine because they are considered the best methods in training text-based classification models. Before being classified, the data goes through preprocessing stages such as cleaning, case folding, filtering, stemming, and tokenizing, as well as feature extraction stages using Term Frequency -Inverse Document Frequency (TF-IDF). The objects analyzed are divided into several well-known marketplaces in Indonesia such as Tokopedia, Lazada, and Shopee in discussing services and delivery of goods. The data used in this study comes from Twitter (X) social media accessed on August 27, 2023, using crawling techniques and successfully obtained as much as 2057 Tweet data. The best accuracy is obtained in the SVM method when compared to the Naive Bayes method. Words obtained based on service talks include price, service, application service, feedback, independence, and others. As for the delivery of goods, common words such as COD, delivery, package, courier, cheap, price, and others appear. Both methods used have good accuracy and can be recommended for use in similar research.

show abstract

“…Furthermore, a domain-independent dynamic ABSA model [12] was introduced in 2021, automating the aspect extraction and sentiment analysis process using Efficient Named Entity Recognition (E-NER) guided dependency parsing and Neural Networks (NN). In 2022 [13], efforts were made to enhance aspect-based sentiment analysis for reviews in the Indonesian language through a deep learning approach in semi-supervised graph-based, such as GCN and GRN for aspect and opinion relationships detection, while polarity classification in CNN and RNN demonstrating superior performance over existing models.…”

Section: Related Workmentioning

confidence: 99%

Fine-Tuning BERT for Aspect Extraction in Multi-domain ABSA

Akram,

Sabir

2023

IJCAI

View full text Add to dashboard Cite

Aspect extraction plays a crucial role in understanding the fine-grained nuances of text data, allowing businesses and researchers to gain deeper insights into customer opinions, sentiment distributions, and preferences. This study presents a BERT-based framework for aspect extraction in ABSA and evaluates its performance. Our research focuses on the comprehensive analysis of aspect extraction as we test our method using the SEMEVAL dataset of various consumer evaluations across diverse domains, including laptops, restaurants, and Twitter. By fine-tuning BERT on a large dataset, we aim to overcome the limitations of traditional approaches and improve the accuracy and efficiency of aspect extraction in ABSA. The experimental findings provide evidence of the efficacy of our methodology with a noteworthy aspect extraction accuracy of 0.98, highlighting its capacity to properly and consistently extract features. The article also explores the applicability of our approach to new domains and its possible applications in real-world scenarios.Povzetek: Ta študija predstavlja okvir za ugotavljanje aspekta v ABSA, ki temelji na sistemu BERT.

show abstract

Graph-Based Semi-Supervised Deep Learning for Indonesian Aspect-Based Sentiment Analysis

Cited by 11 publications

References 45 publications

Labeling Consistency Test of Multi-Label Data for Aspect and Sentiment Classification Using the Cohen Kappa Method

Labeling Consistency Test of Multi-Label Data for Aspect and Sentiment Classification Using the Cohen Kappa Method

Sentiment Analysis on Marketplace in Indonesia using Support Vector Machine and Naïve Bayes Method

Fine-Tuning BERT for Aspect Extraction in Multi-domain ABSA

Contact Info

Product

Resources

About