Deep learning-based application for multilevel sentiment analysis of Indonesian hotel reviews

Kusumaningrum, Retno; Nisa, Iffa Zainan; Jayanto, Rahmat; Nawangsari, Rizka Putri; Wibowo, Adi

doi:10.1016/j.heliyon.2023.e17147

Cited by 10 publications

(2 citation statements)

References 44 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The acquired dataset will undergo a process of labeling or categorization. This labeling will be applied to 1000 reviews, dividing them into two classes the positive class and the negative class [17]. The positive class will be assigned a value of 1, while the negative class will be assigned a value of 0.…”

Section: Data Labelingmentioning

confidence: 99%

Sentiment Analysis Motorku X Using Applications Naive Bayes Classifier Method

Mustolih,

Arsi,

Subarkah

2023

IJAIDM

View full text Add to dashboard Cite

The rapid development of technology has brought convenience to humans in their daily lives. The continuously evolving technology generates large amounts of data. Data can provide valuable information if processed effectively. The Motorku X application is one of the innovations created by Astra Motor to facilitate consumers or potential customers in servicing and purchasing motorcycles. The Motorku X application generates review data every day. These review data can be utilized for future application development. To make the most of the reviews, sentiment analysis is one of the techniques used to process the review data. Sentiment analysis is a method to measure consumer sentiments in terms of positive or negative reviews. The algorithm used in this research is the Naïve Bayes classifier. One of the advantages of Naïve Bayes is its ability to work quickly and efficiently in terms of computational time. The research consists of several stages: data collection, data labeling, pre-processing, data splitting, tf-idf weighting, implementation of Naïve Bayes classifier, and evaluation of the results. The data comprises 1000 reviews divided into two classes: positive class (number) and negative class (number). The research was conducted with three scenarios of training and testing data sharing: 90%:10%, 80%:20%, and 70%:30%. The best results were achieved with the 90%:10% ratio, with an accuracy of 76%, precision of 76%, and recall of 97%.

show abstract

Section: Data Labelingmentioning

confidence: 99%

Sentiment Analysis Motorku X Using Applications Naive Bayes Classifier Method

Mustolih,

Arsi,

Subarkah

2023

IJAIDM

View full text Add to dashboard Cite

show abstract

“…Online travel agencies (OTAs) are prominent e-commerce platforms that provide hotel bookings. Customers may help each other pick a hotel by submitting reviews [11]. English has received more attention in named entity recognition (NER) studies than any other language in part because of the abundance of highquality annotation datasets in the language.…”

Section: Introductionmentioning

confidence: 99%

Cloud-based machine learning algorithms for anomalies detection

Amarnath,

Gurulakshmanan

2024

IJEECS

View full text Add to dashboard Cite

Gradient boosting machines harnesses the inherent capabilities of decision trees and meticulously corrects their errors in a sequential fashion, culminating in remarkably precise predictions. Word2Vec, a prominent word embedding technique, occupies a pivotal role in natural language processing (NLP) tasks. Its proficiency lies in capturing intricate semantic relationships among words, thereby facilitating applications such as sentiment analysis, document classification, and machine translation to discern subtle nuances present in textual data. Bayesian networks introduce probabilistic modeling capabilities, predominantly in contexts marked by uncertainty. Their versatile applications encompass risk assessment, fault diagnosis, and recommendation systems. Gated recurrent units (GRU), a variant of recurrent neural networks, emerges as a formidable asset in modeling sequential data. Both training and testing are crucial to the success of an intrusion detection system (IDS). During the training phase, several models are created, each of which can recognize typical from anomalous patterns within a given dataset. To acquire passwords and credit card details, "phishing" usually entails impersonating a trusted company. Predictions of student performance on academic tasks are improved by hyper parameter optimization of the gradient boosting regression tree using the grid search approach.

show abstract

Comparative analysis of multi-platform e-commerce online reviews based LDA and sentiment dictionary

Xie,

Zhao,

2024

IJICC

View full text Add to dashboard Cite

PurposeThis study aims to investigate the differences in consumer reviews across multiple e-commerce platforms to better assist consumers in making informed decisions. By examining the specific content of these differentiated reviews, the study seeks to provide insights that can enhance e-commerce services and improve consumer satisfaction.Design/methodology/approachThe research utilizes the latent Dirichlet allocation (LDA) method for text analysis to identify the varying concerns of consumers across different e-commerce platforms for the same product. Additionally, the study expands the sentiment dictionary to address polysemy issues, allowing for a more precise capture of sentiment differences among consumers. A non-parametric test is employed to compare reviews across multiple platforms, providing a comprehensive analysis of review disparities.FindingsThe findings reveal that consumer concerns and sentiments vary significantly across different e-commerce platforms, even for the same product. The combination of text analysis and non-parametric testing highlights the objectivity of the research, offering valuable evidence and recommendations for improving e-commerce services and enhancing the shopping experience.Originality/valueThis study is original in its approach to combining text analysis with non-parametric testing to examine multi-platform review differences. The research not only contributes to the understanding of consumer behavior in the context of e-commerce but also provides practical suggestions for platforms and consumers, aiming to optimize service quality and consumer satisfaction.

show abstract

Deep learning-based application for multilevel sentiment analysis of Indonesian hotel reviews

Cited by 10 publications

References 44 publications

Sentiment Analysis Motorku X Using Applications Naive Bayes Classifier Method

Sentiment Analysis Motorku X Using Applications Naive Bayes Classifier Method

Cloud-based machine learning algorithms for anomalies detection

Comparative analysis of multi-platform e-commerce online reviews based LDA and sentiment dictionary

Contact Info

Product

Resources

About