Publication Search

User reviews on the Shopee e-commerce platform represent an important source of information for understanding consumer perceptions of products and services. Sentiment analysis is commonly applied to classify user opinions into positive, neutral, and negative sentiment categories based on textual data. This study aims to analyze the performance of the Long Short-Term Memory (LSTM) method in sentiment classification of Shopee user reviews. The dataset used in this study consists of Indonesian-language user reviews that have undergone preprocessing stages, including case folding, text cleaning, tokenization, and stopword removal. The LSTM model was trained using preprocessed text represented as word sequences. Model performance was evaluated using overall accuracy and class-wise classification results. The experimental results indicate that the LSTM method achieved an overall accuracy of 87.62%. In addition, the classification performance for the positive sentiment class reached 95.27%, the neutral class achieved 4.96%, and the negative class reached 74.26%. These results demonstrate that the LSTM method performs well in classifying sentiment in Shopee user reviews, particularly for positive sentiment. This study is expected to provide insights and references for the application of deep learning methods in sentiment analysis of Indonesian e-commerce review data.

This study presents a Systematic Literature Review on Artificial Intelligence (AI) and Natural Language Processing (NLP) applications for customer support automation and digital service optimization. The review follows the PRISMA framework to ensure methodological rigor and transparency, focusing on literature published between 2020 and 2025 from the Scopus database. The findings reveal that AI-driven technologies, including Machine Learning, Deep Learning, and Large Language Models, have significantly improved efficiency, response time, and customer satisfaction in customer support and digital service. Common NLP applications include sentiment analysis, ticket classification, and automated response generation. Among these, hybrid and transformer-based models demonstrate superior accuracy and contextual understanding compared to traditional algorithms. However, several challenges persist, including data quality limitations, privacy and security concerns, algorithmic bias, and linguistic ambiguities such as sarcasm and negation. Moreover, issues related to trust and ethical adoption continue to influence user acceptance of AI systems. This review provides a comprehensive synthesis of current methodologies, trends, and research gaps, offering insights for future studies to develop explainable, secure, and human-centered AI systems that enhance the sustainability and transparency of digital customer support services.

This study enhances Jiang et al.'s compression-based classification algorithm by addressing its limitations in detecting semantic similarities between text documents. The proposed improvements focus on unigram extraction and optimized concatenation, eliminating reliance on entire document compression. By compressing extracted unigrams, the algorithm mitigates sliding window limitations inherent to gzip, improving compression efficiency and similarity detection. The optimized concatenation strategy replaces direct concatenation with the union of unigrams, reducing redundancy and enhancing the accuracy of Normalized Compression Distance (NCD) calculations. Experimental results across datasets of varying sizes and complexities demonstrate an average accuracy improvement of 5.73%, with gains of up to 11% on datasets containing longer documents. Notably, these improvements are more pronounced in datasets with high-label diversity and complex text structures. The methodology achieves these results while maintaining computational efficiency, making it suitable for resource-constrained environments. This study provides a robust, scalable solution for text classification, emphasizing lightweight preprocessing techniques to achieve efficient compression, which in turn enables more accurate classification.

This study aims to analyze the sentiment of iPhone product reviews fromAmazon using the BERT (Bidirectional Encoder Representations from Transformers) model to classify reviews as either positive or negative. The dataset, sourced from Kaggle, includes text reviews and star ratings, where high ratings indicate positive sentiment and low ratings indicate negative sentiment. After text preprocessing steps, including data cleaning, tokenization, and sentiment labeling, the BERT model was fine-tuned for sentiment classification, with the data split into training, validation, and test sets. Evaluation results demonstrate that the BERT model achieves a high classification accuracy, with an accuracy rate of 93.9% and a balanced F1 score between precision and recall. Confusion matrix evaluation also indicates that the model consistently identifies both positive and negative sentiments. This study shows that Transformer-based models like BERT are highly effective in understanding customer opinions in e-commerce, with broad application potential for data-driven decision-making in marketing strategies and product development.

Political dynasty is a political power exercised by a group of people who are related by family, with the aim of obtaining power and ensuring that this power remains within the group by passing it on to other family members. This study conducts a sentiment analysis on comments related to the Supreme Court decision which is believed to pave the way for Kaesang Pangarep in support of Jokowi's political dynasty. Sentiment analysis is carried out using the Naive Bayes method, a commonly used algorithm for text classification based on probability. The data used consists of comments from videos taken from social media platforms. These comments are then categorized into positive, negative, and neutral sentiments. The results of the study show the distribution of public sentiment towards this issue, providing an overview of how the public responds to the decision. The Naive Bayes method is chosen for its simplicity and its ability to provide reasonably accurate results in text analysis.

In the rapidly developing digital era, social media such as Twitter has become part of everyday life and facilitates the rapid dissemination of information, including information about criminals. This research aims to analyze public sentiment towards information about criminals spread on Twitter using the Naive Bayes algorithm. This algorithm was chosen because of its simplicity and effectiveness in text classification. Data was collected through a crawling process from Twitter, followed by a preprocessing stage to remove noise. The research results show that public sentiment towards information about criminals on Twitter is divided into three categories: positive, neutral and negative. After classification, it was found that neutral sentiment increased significantly to 63.4%, while positive and negative sentiment decreased to 10.5% and 26.1%. These findings indicate that people tend to be more careful in reacting to sensitive information. This research provides important insights for related parties in managing information about criminals on social media and can be a reference for developing further policies and strategies.

The Halal Product Assurance Agency (BPJPH) is an agency under the auspices of the Ministry of Religion with the task of ensuring the halalness of products in Indonesia. BPJPH has become a public concern after establishing the new halal logo. On February 10, 2022 the new halal logo was ratified by the Head of BPJPH, Muhammad Aqil Irham. This has become a topic of public discussion either directly or through social media, one of which is social media twitter. The number of opinion tweets about the change of the halal logo can be used as a data source to obtain information about public opinion on the change of the halal logo through sentiment analysis. Sentiment analysis can be done by machine learning approach, one of these is the SVM algorithm . In this research, oversampling and undersampling are applied to handle data that has an unbalanced sentiment class. The results showed that the Support Vector Machine (SVM) model using oversampling training data got the highest accuracy, recall, precision, and f1-score, namely 71% accuracy, 67% precision, 61% recall, and 61% f1-score while training using undersampling training data has the lowest performance, namely getting 56% accuracy, 51% precision, 57% recall, and 52% f1-score.

The rapid growth of e-commerce in Indonesia has generated a massive and continuous volume of product reviews. This user-generated content is vital for business intelligence, yet its sheer scale makes manual analysis inefficient, subjective, and practically impossible. Automated sentiment analysis is therefore crucial for businesses to efficiently understand customer feedback and market perception. This research addresses this gap by implementing the Naïve Bayes Classifier (NBC) algorithm to automatically classify the sentiment of Indonesian-language e-commerce product reviews. This study utilized a dataset of 2,000 reviews collected from a major e-commerce platform's "Electronics" category. The data underwent critical text preprocessing stages (case folding, tokenizing, stopword removal, and stemming using the Sastrawi library) to handle the complexities of informal Indonesian text. The dataset was split using an 80/20 ratio, resulting in 1,600 training reviews and 400 testing reviews. Model performance was then evaluated using a Confusion Matrix, focusing on the key metrics of Accuracy, Precision, and Recall. The test results showed excellent performance, achieving an Accuracy of 90.00%, Precision of 91.93%, and Recall of 95.00%. These results demonstrate that the Naïve Bayes algorithm, when supported by robust preprocessing, is a highly effective, reliable, and computationally efficient method for this task, providing a valuable tool for e-commerce stakeholders.

Classification is one of the most widely used techniques in machine learning. Text classification is the process of classifying data according to pre-determined groups or classes. Where in most cases, text classification uses labeled training data to obtain the rules used to classify test data into predefined groups. In this study, it is proposed to use CountVectorizer for Indonesian text classification which will be compared with TF-IDF Term Weighting and its three feature levels, namely Character Level, Word Level and N-gram Level as feature extraction which is implemented together with Naive Bayes classification and the BPPPTIndToEngCorpusHalfM dataset. To compare the classification performance, this study uses 10-Fold Cross Validation and Split Data using a ratio of 90:10, while to evaluate the accuracy of the authors using the F1-Score and AUC with the hope that this study will get good accuracy results so that it can be used as a reference to be developed using another method. The F1-Score accuracy obtained in this study was 0.93 and the AUC score was 0.95.