Publication Search

This study aims to analyze user sentiment toward the integration of Artificial Intelligence (AI) in online learning platforms, which are increasingly expanding in the digital era. With the growing use of AI technologies in education—such as learning chatbots, material recommendation systems, and automated assessments—it is essential to understand users’ perceptions and reactions to these implementations. The research employs sentiment analysis based on text mining using user review data collected from various online learning platforms. The analysis process includes data preprocessing, sentiment classification using machine learning algorithms, and interpretation of results based on the proportion of positive, negative, and neutral sentiments. The findings indicate that most users express positive sentiments toward AI integration, as it enhances learning efficiency and personalization. However, some users raise concerns regarding data privacy and the lack of human interaction. This study is expected to serve as a reference for educational platform developers to design AI systems that are more adaptive, transparent, and user-centered

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 aims to explore the relevance of Pupuh Kasmaran Mocoan in Lontar Yusup Banyuwangi to the culture of the Banyuwangi community through a literary anthropology approach. Pupuh Kasmaran Mocoan is part of the literary heritage of East Java and is preserved in Lontar Yusup Banyuwangi, reflecting the cultural values, traditions, and worldview of the local society. The research employed a qualitative descriptive method using reading and data classification techniques based on cultural elements. Data were analyzed by identifying cultural aspects embedded in the text and examining their relevance to the social and cultural life of the Banyuwangi community. The findings reveal that Pupuh Kasmaran Mocoan is not merely a literary work but also a representation of the cultural complexity of Banyuwangi society. It contains various cultural values, including social norms, customs, beliefs, and moral teachings that continue to be relevant in contemporary community life. These findings demonstrate that traditional literary works can serve as important cultural documents for understanding local identity and preserving regional wisdom. The study contributes to a deeper understanding of the cultural elements reflected in Lontar Yusup Banyuwangi and emphasizes the significance of literary anthropology in interpreting traditional literature. Furthermore, the findings provide valuable insights for the preservation and development of regional literature while enriching the discourse on literary anthropology.

The rapid growth of social media as a primary channel for information dissemination has triggered a significant surge in the distribution of hoaxes, potentially damaging social order, instigating mass disinformation, and threatening national security. This research aims to design an intelligent algorithm for hoax detection by integrating a critical thinking approach into Natural Language Processing (NLP)-based text processing. The algorithmic model is built using a combination of linguistic features, argument logic, and cognitive indicators such as the detection of unsubstantiated claims, identification of source bias, and evidence testing. To ensure accountability and transparency of the system, an Explainable AI (XAI) approach is applied so that classification results can be understood by non-technical users. The research results show that integrating critical thinking significantly improves detection accuracy to 93.1%, with an increase in precision and recall for detecting hoaxes based on emotional narratives. Beyond technical aspects, this model aligns with the mandate of Law of the Republic of Indonesia Number 11 of 2008 concerning Information and Electronic Transactions (ITE Law), particularly Article 28 paragraph (1), which prohibits the dissemination of false and misleading news that harms the public. Therefore, this system is not only scientifically relevant but also supports law enforcement and strengthens digital literacy in the post-truth era. These findings are expected to be a strategic contribution to the development of an ethical, critical, and responsible digital ecosystem.

The advances in the movie industry have led to the rapid rise of streaming platforms, becoming increasingly popular among people. With countries around the world constantly producing movies, subtitles play an essential role in making films more accessible to a broader audience. Subtitling strategies can be defined as a translation practice involving the display of written text, usually at the bottom of the screen, to convey the original dialogues of the speakers and the information contained in the soundtrack. This study uses data from the lk21 website and analyzes it using Gottlieb’s subtitling strategies theory, which consists of ten specific methods for audiovisual translation: Expansion, Paraphrase, Transfer, Imitation, Transcription, Deletion, Dislocation, Condensation, Decimation, and Resignation. A descriptive qualitative approach was adopted in this study with observation, classification, and note-taking of each subtitle, suitable for the research, and describing the subtitling strategies applied in the movie. The findings show that seven strategies were applied in the Indonesian subtitles, namely paraphrase, transfer, condensation, imitation, deletion, dislocation, and expansion. Among them, paraphrase emerged as the most frequently used strategy, likely because figurative or culture-specific expressions in the source text often required adaptation to preserve their meaning for Indonesian viewers. On the other hand, expansion was identified as the least applied strategy, suggesting that the original dialogues rarely demanded additional explanation. Overall, the strategic use of these methods contributed to producing subtitles that were coherent, culturally adapted, and highly accessible to the target audience.

This research aims to analyze GoPay user sentiments on the X social media platform (formerly known as Twitter) using the Naive Bayes Classifier algorithm. Sentiment analysis was conducted to understand user perceptions and satisfaction levels towards GoPay digital payment services based on their shared comments and reviews. Data was collected through a tweet crawling process containing the keyword "GoPay" within a specific period. The research stages included data preprocessing (case folding, tokenizing, filtering, and stemming), sentiment labeling (positive, negative), word weighting using TF-IDF, and classification using the Naive Bayes algorithm. The results showed that from a total of 1,431 analyzed tweets, 797 data contained positive sentiments, and 643 data contained negative sentiments. With a classification accuracy rate reaching 82.94%. The most frequently positively commented factors included ease of use and offered promotions, while the main complaints were related to technical issues and customer service. This research provides insights for GoPay developers to improve services according to user feedback.  

The rapid growth of e-commerce platforms has significantly transformed the way consumers share and access product feedback. One of the widely used platforms in Indonesia is Shopee, where customers actively provide reviews of various products, including local skincare brands such as Kahf facial wash. Customer reviews on e-commerce platforms contain valuable information that can be analyzed to understand consumer opinions and preferences. Sentiment analysis, as a branch of natural language processing, enables the classification of textual data into categories such as positive, negative, or neutral. This study aims to classify Shopee user sentiments regarding Kahf facial wash products by implementing the Multinomial Naïve Bayes algorithm, a well-known probabilistic classifier suitable for text categorization. The research methodology consisted of several preprocessing stages, including data cleansing, case folding, tokenizing, stopword removal, and stemming, to prepare raw review texts for further analysis. For feature representation, the Term Frequency–Inverse Document Frequency (TF-IDF) method was applied to capture the importance of words across documents. To evaluate the classification performance, K-Fold cross-validation was employed with K values of 4, 5, 6, and 10 to ensure model reliability and robustness. Considering the issue of imbalanced datasets in user-generated reviews, the Synthetic Minority Over-sampling Technique (SMOTE) was utilized to balance the distribution of sentiment classes. Based on the confusion matrix, the Multinomial Naïve Bayes algorithm demonstrated effective performance in classifying sentiments, achieving satisfactory levels of accuracy, precision, and recall across different folds. These results indicate that the algorithm is capable of handling sentiment analysis tasks for local product reviews effectively. The findings of this study are expected to provide meaningful insights for businesses in understanding consumer perceptions, thereby supporting decision-making processes in product development, marketing strategies, and customer engagement for local brands.

This study aims to analyze the sentiment of customer reviews of the Grand Jatra Hotel Pekanbaru on the Google Review platform using the Naïve Bayes algorithm. Social media and online review platforms are increasingly becoming the primary source of information for potential customers in making purchasing decisions, particularly in the hospitality sector. Therefore, sentiment analysis of customer reviews is crucial for understanding consumer perceptions and providing strategic input for hotels in improving service quality. The research data was collected using web scraping techniques to obtain publicly available customer reviews. The obtained data was then processed through text preprocessing stages including case folding, tokenizing, normalization, stopword removal, and stemming. The Term Frequency-Inverse Document Frequency (TF-IDF) method was then used to weight each word, so that more relevant words have a greater influence in the classification process. The sentiment classification process was carried out into two main categories, namely positive and negative. The Naïve Bayes model was trained using training data and then tested with test data to measure the algorithm's performance in classifying sentiment. The evaluation results show that the model built is able to achieve an accuracy level of 98%, with a precision value of 97% and a recall of 100% in the positive class, and 92% in the negative class. These findings confirm that the Naïve Bayes algorithm can be effectively used in analyzing customer sentiment towards hotel services and facilities. Practically, the results of this study are expected to provide insight for the management of Grand Jatra Hotel Pekanbaru in understanding customer perceptions, identifying service strengths and weaknesses, and formulating more targeted marketing strategies. In addition, this study can also be a reference for the development of similar studies in the hotel industry and other service sectors.

This study aims to utilize data mining techniques, specifically the K-Nearest Neighbors (KNN) algorithm, to classify leaf diseases in sugarcane (Saccharum officinarum). Early and accurate detection of leaf disease types is a crucial step in prevention and control strategies, thereby reducing potential crop losses caused by pathogen attacks. Leaf diseases in sugarcane, such as leaf scald, rust, and mosaic virus, are known to affect photosynthesis, inhibit growth, and reduce the quality and quantity of sugarcane produced. The classification process in this study was carried out through image analysis of infected sugarcane leaves, where features such as color, texture, and shape were extracted using digital image processing techniques. The KNN algorithm was chosen because of its non-parametric nature, ease of implementation, and its ability to provide accurate classification results even with limited data size. The working principle of KNN is to determine the class of a new sample based on the majority class of its k nearest neighbors in the feature space, making it very suitable for the case of leaf disease image classification. In addition to building a classification model, this study also examines disease prevention strategies based on the identification results. These strategies include the use of disease-resistant sugarcane varieties, the implementation of appropriate planting patterns, land moisture management, regular plantation sanitation, and the measured and environmentally friendly use of pesticides or fungicides. Model performance evaluation was conducted using accuracy, precision, recall, and F1-score metrics to assess model effectiveness across various data scenarios. The results of this study are expected to not only contribute to the development of decision support systems for farmers and related parties but also support the application of artificial intelligence-based technology in the agricultural sector.

This research focuses on the development of an Android-based vegetable detection application by utilizing digital image processing technology and data communication through Application Programming Interface (API). This application is designed to make it easier for users to visually recognize different types of vegetables using the device's camera. The detection process is carried out by sending the image to a cloud server, where the image analysis process is carried out to identify the type of vegetable, displaying its name, characteristics, and benefits. The app's implementation includes an intuitive and user-friendly user interface, with key features such as login, registration, and an interactive dashboard. The dashboard displays user information, location, ambient temperature, vegetable detection history, and direct access to the camera for real-time detection processes. The utilization of cloud computing technology not only keeps application performance lightweight and responsive, but also enables high processing efficiency and data scalability. This allows the application to continue to evolve according to the increasing number of users and incoming data. Image processing is done with machine learning algorithms that are trained to recognize the shape, color, and texture of different types of local vegetables. In addition, this system is also equipped with a periodic data update feature to be able to adjust to the development of new vegetable classifications. The test results show that the app is able to recognize different types of vegetables with a high level of accuracy, as well as provide additional relevant information quickly and accurately. Tests are carried out on a variety of lighting and background conditions to ensure the reliability of the system. The success of the development of this application reflects the integration of modern technology in supporting the digital agriculture sector.

The formation of new lexical items through the combination of two or more lexemes is referred to as compounding, a significant process within the field of morphology. As noted by O’Grady (1996), compound words are typically categorized into two types: endocentric and exocentric compounds. Katamba (1994:320) emphasizes that endocentric compounds exhibit greater productivity compared to their exocentric counterparts. This study aims to examine the morphological structures of endocentric compound words found in a specific object—namely, the speech text School Strike for Climate. Employing a qualitative-descriptive method, the research analyzes and interprets the semantic classification of the identified compounds. The speech transcript reveals 26 occurrences of endocentric compounds, all of which function as compound nouns and consist of nominal components.

Mining activities in the Raja Ampat area have sparked various public reactions, both supportive and critical, particularly on social media platforms such as Twitter. This study aims to analyze public sentiment regarding the mining operations by employing two classification algorithms. A total of 500 tweets related to Raja Ampat were collected from the X platform, and after data cleaning, 168 were identified as positive sentiments and 303 as negative. Sentiment analysis was conducted using text mining techniques by comparing two algorithms: Support Vector Machine (SVM) and Naïve Bayes. To address the issue of data imbalance, the Synthetic Minority Over-sampling Technique (SMOTE) was applied. The analysis results showed that SVM achieved an accuracy of 80%, outperforming Naïve Bayes, which reached only 68%. This indicates that SVM performed better in classifying sentiment. Additionally, the application of SMOTE effectively enhanced both algorithms’ abilities to detect positive sentiment, as reflected in the precision, recall, and F1-score metrics. For SVM, precision reached 85%, recall 80%, and F1-score 80%, while Naïve Bayes recorded a precision and recall of 69%, and an F1-score of 68%.

This study explores the integration of the Grounded Theory (GT) approach, developed by Anselm Strauss and Barney Glaser, into thematic hadith studies (maudhū‘ī) as a methodological alternative for extracting meaning and constructing theory from hadith texts. This approach is considered capable of producing more contextual, relevant, and applicable understandings of contemporary social issues such as justice, leadership, and environmental ethics. By employing GT’s systematic stages open coding, axial coding, and selective coding hadith research goes beyond mere thematic classification and evolves into a theorization process directly rooted in hadith data. While GT offers advantages in flexibility and contextual depth, its application also faces epistemological and technical challenges, particularly in terms of data validity, interpretive bias, and the integration of modern qualitative methods with the principles of ulūm al-ḥadīth. Therefore, an interdisciplinary approach is necessary, involving collaboration between hadith scholars and social researchers, alongside strong mastery of qualitative methodology. In conclusion, GT is not a replacement for classical methods, but a complementary tool that broadens the horizons of hadith studies. By combining the strengths of the Islamic scholarly tradition and contemporary methodologies, this approach has the potential to enrich Islamic epistemology and to present hadith as a living, dynamic, and relevant source of values for modern society.

This study presents an advanced AI-powered framework to detect and prevent cyberbullying across diverse social media platforms using a multiclass classification approach. Addressing the growing complexity and linguistic diversity of online abuse, the research integrates various machine learning (RF, LR, SVM) and deep learning (Bi-LSTM, BERT) models trained on a balanced dataset covering bullying categories based on religion, age, ethnicity, gender, and neutral content. Data preprocessing, tokenization, feature extraction via TF-IDF and CountVectorizer, and class balancing using SMOTE were applied to enhance model accuracy. The proposed system further supports real-time detection through social media APIs, offering dynamic monitoring and intervention capabilities. Among the tested models, Random Forest and BERT achieved the highest classification performance with 94% accuracy. Despite its robust architecture, limitations include dependence on English-language datasets, exclusion of multimodal data (e.g., memes, audio), and API restrictions that challenge scalability. Future development will focus on incorporating vision-language models and optimizing the system for real-time, multilingual, and multimodal environments. This study contributes to digital safety efforts by proposing a scalable and adaptive detection system suitable for safeguarding users from evolving forms of online harassment.

This study is a literature study that aims to analyze the practice of withholding and collecting Income Tax (PPh) Article 23 in the context of tax regulations in Indonesia. Data were obtained through a literature review of laws and regulations, tax textbooks, and relevant scientific journals. The results of the analysis indicate that although the provisions of PPh Article 23 have been regulated in detail, there is still the potential for differences in interpretation in practice, especially regarding the classification of tax objects and withholding rates. This study recommends increasing the socialization of regulations and simplifying tax administration in order to improve taxpayer compliance.

This study aims to analyze the derivational affixes found in the Regina Caeli speech delivered by Pope Leo XIV. The main focus of the research is to identify the types of derivational affixes and determine their frequency, including both prefixes and suffixes. The research applies a mixed methods approach, combining both quantitative and qualitative descriptive methods. The quantitative aspect involves identifying and counting the types and frequency of derivational affixes found in Pope Leo XIV’s speech at Regina Caeli. The qualitative aspect focuses on analyzing the contextual meaning and function of those affixes within the speech. Data collection techniques include textual analysis.. Based on the analysis, a total of 13 types of derivational affixes were identified, consisting of one prefix and 12 suffixes. Furthermore, derivational affixes in this study are classified into four types: noun, verb, adjective, and adverb derivation. This study is expected to encourage English learners to understand and apply derivational affixes effectively, as doing so can significantly enrich vocabulary and enhance linguistic competence by recognizing root words and how new forms are created. This classification helps English learners expand vocabulary and understand how word forms are systematically built, thus enhancing their language skills.

Indonesian language has an important role in education, especially in shaping students’ literacy skills through text-based learning. One type of text taught in the Merdeka Curriculum is anecdote text. This study aims to identify and describe the forms of syntactic errors in anecdote texts in the book Cerdas Cergas Berbahasa dan Bersastra Indonesia for SMA / SMK class X. The method used is descriptive qualitative with error analysis approach according to Ellis’ model. The data were collected through listening and note-taking techniques, and analyzed through the stages of identification, classification, and evaluation of errors. The results of the analysis show the existence of various errors, including the use of non-standard words, incorrect word writing, inappropriate diction, excessive word usage, conjunction errors, and punctuation errors. The findings emphasize the importance of improving the quality of language in textbooks, as errors in books can affect students' language skills. Theoretically, this study expands the study of syntactic errors in the narrative text genre. Practically, the results of this study can be a reference for teachers, book writers, and curriculum developers in improving and enhancing the quality of Indonesian teaching materials.

This study aims to analyze market sentiment towards Gold Financing Products (PKE) in Islamic banking before and after the Trump Effect phenomenon using the text mining method. This technique involves extracting information from unstructured text data to then be visualized and analyzed using the Natural Language Processing (NLP) approach and a RoBERTa-based classification model. Data was collected through web scraping from the X application with the help of API and processed using Google Colab. From a total of 4,074 tweets analyzed, it was found that the majority of public sentiment was neutral (59%), followed by negative (24%) and positive (17%). This reflects the public's tendency to discuss informatively rather than emotionally, although there was a spike in negative sentiment in certain periods indicating sensitivity to global dynamics, especially the impact of the Trump Effect on gold prices. The resulting wordcloud reveals key topics such as gold prices, buying and selling activities, and institutions such as Pegadaian Syariah and BSI. Terms such as "sharia", "riba", and "principles" emphasize the importance of Islamic financial values ​​in public perception. The results of this study indicate that text mining-based sentiment analysis is effective in capturing the dynamics of public opinion in real-time and can be a strategic tool for Islamic financial institutions in responding to market changes.

Digital learning applications like HeyJapan are increasingly popular. User reviews on platforms such as Google Play Store contain valuable information on user perceptions and experiences. To process this information systematically, this study employs a Natural Language Processing (NLP) approach to analyze sentiment toward the HeyJapan application. Data was collected using web scraping techniques with Python and the google play scraper library, resulting in 1,000 latest user reviews. The analysis included data collection, preprocessing, sentiment labeling using TextBlob, visualization, modeling with Logistic Regression, and evaluation. After preprocessing, 923 valid reviews were classified into three sentiment categories based on polarity which are positive, neutral, and negative. Results showed 71.4% of reviews positive, 26.1% neutral, and 2.5% negative. Visualizations in pie charts and word clouds provided an overview of user perceptions. Modeling with TF-IDF and Logistic Regression achieved 88% accuracy with the highest f1-score in the positive sentiment category. Evaluation indicates the model is fairly reliable in classifying sentiments, especially for positive and neutral categories, though negative sentiment classification needs improvement. This study shows the NLP approach can evaluate user perceptions of educational applications based on reviews and serve as a basis for improving foreign language learning app quality.

The rapid development of financial technology (fintech), particularly digital wallet applications like OVO, has significantly transformed transaction patterns in society. However, issues such as server instability and unsatisfactory user experiences frequently emerge on social media platforms. This study aims to analyze user sentiments toward OVO on platform X (formerly Twitter) by comparing the performance of two machine learning algorithms: Naïve Bayes and Support Vector Machine (SVM). Data were collected through web scraping from 1,000 Indonesian-language tweets containing the keyword "OVO." The research methodology included text preprocessing (data cleaning, tokenization, stopword removal), feature extraction using TF-IDF, and sentiment classification (positive, negative, neutral). Evaluation results demonstrated that SVM achieved the highest accuracy of 85.2%, while Naïve Bayes reached 78.5%. SVM also outperformed in precision (87%) and recall (83%) due to its ability to handle non-linear data. These findings provide actionable recommendations for OVO developers to enhance server stability and features based on user feedback. Additionally, this study serves as a reference for future sentiment analysis research employing algorithmic comparisons.