Search Results (1,526)

YouTube is a popular social media platform in the contemporary digital landscape. The primary focus of this study is to explore the underlying sentiment in user comments about food-related videos on YouTube, specifically within two pivotal food categories: plant-based and hedonic product. We labeled comments using sentiment lexicons such as TextBlob, VADER, and Google’s Sentiment Analysis (GSA) engine. Comment sentiment was classified using advanced Machine-Learning (ML) algorithms, namely Support Vector Machines (SVM), Multinomial Naive Bayes, Random Forest, Logistic Regression, and XGBoost. The evaluation of these models encompassed key macro average metrics, including accuracy, precision, recall, and F1-score. The results from GSA showed a high accuracy level, with SVM achieving 93% accuracy in the plant-based dataset and 96% in the hedonic dataset. In addition to sentiment analysis, we delved into user interactions within the two datasets, measuring crucial metrics, such as views, likes, comments, and engagement rate. The findings illuminate significantly higher levels of views, likes, and comments in the hedonic food dataset, but the plant-based dataset maintains a superior overall engagement rate. Full article

The healthcare sector constantly investigates ways to improve patient outcomes and provide more patient-centered care. Delivering quality medical care involves ensuring that patients have a positive experience. Most healthcare organizations use patient survey feedback to measure patients’ experiences. However, the power of social media can be harnessed using artificial intelligence and machine learning techniques to provide researchers with valuable insights into understanding patient experience and care. Our primary research objective is to develop a social media analytics model to evaluate the maternal patient experience during the COVID-19 pandemic. We used the “COVID-19 Tweets” Dataset, which has over 28 million tweets, and extracted tweets from the US with words relevant to maternal patients. The maternal patient cohort was selected because the United States has the highest percentage of maternal mortality and morbidity rate among the developed countries in the world. We evaluated patient experience using natural language processing (NLP) techniques such as word clouds, word clustering, frequency analysis, and network analysis of words that relate to “pains” and “gains” regarding the maternal patient experience, which are expressed through social media. The pandemic showcased the worries of mothers and providers on the risks of COVID-19. However, many people also shared how they survived the pandemic. Both providers and maternal patients had concerns regarding the pregnancy risks due to COVID-19. This model will help process improvement experts without domain expertise to understand the various domain challenges efficiently. Such insights can help decision-makers improve the patient care system. Full article

The excess levels of investor participation coupled with irrational behaviour in the South African bond market causes excess volatility, which in turn exposes investors to losses. Consequently, the study aims to examine the effect of market-wide investor sentiment on government bond index returns of varying maturities under changing market conditions. This study constructs a new market-wide investor sentiment index for South Africa and uses the two-state Markov regime-switching model for the sample period 2007/03 to 2024/01. The findings illustrate that the effect investor sentiment has on government bond indices returns of varying maturities is regime-specific and time-varying. For instance, the 1–3-year government index return and the over-12-year government bond index were negatively affected by investor sentiment in a bull market condition and not in a bear market condition. Moreover, the bullish market condition prevailed among the returns of selected government bond indices of varying maturities. The findings suggest that the government bond market is adaptive, as proposed by AMH, and contains alternating efficiencies. The study contributes to the emerging market literature, which is limited. That being said, it uses market-wide investor sentiment as a tool to make pronunciations on asset selection, portfolio formulation, and portfolio diversification, which assists in limiting investor losses. Moreover, the findings of the study contribute to settling the debate surrounding the efficiency of bond markets and the effect between market-wide sentiment and bond index returns in South Africa. That being said, it is nonlinear, which is a better model that uses nonlinear models and alternates with market conditions, making the government bond market adaptive. Full article

Most previous stock investing methods were unable to predict newly listed stocks because they did not have historical data on newly listed stocks. In this paper, we use the Q-learning algorithm based on a convolutional neural network and add sentiment analysis to establish a prediction method for Chinese stock investment tasks. There are 118 companies that are ranked in the Chinese top 150 list for two consecutive years in both 2022 and 2023. We collected all comments under the stock bar of these 118 stocks for each day from 1 January 2022 to 1 July 2024, totaling nearly 10 million comments. There are 90 stocks left after the preprocessing of 118 stocks. We use these 90 stocks as the dataset. The stock’s closing price, volume, and comment text data are fed together to the agent, and the trained agent outputs investment behaviors that maximize future returns. We apply the trained model to two test sets that are completely different from the training set and compare it to several other methods. Our proposed method called SADQN-S obtains results of 1.1229 and 1.1054 on the two test sets. SADQN-S obtained higher final total assets than the other methods on both test sets. This shows that our model can help stock investors earn high returns on newly listed stocks. Full article

Text augmentation plays an important role in enhancing the generalizability of language models. However, traditional methods often overlook the unique roles that individual words play in conveying meaning in text and imbalance class distribution, thereby risking suboptimal performance and compromising the model’s generalizability. This limitation motivated us to develop a novel technique called Text Augmentation with Word Contributions (TAWC). Our approach tackles this problem in two core steps: Firstly, it employs analytical correlation and semantic similarity metrics to discern the relationships between words and their associated aspect polarities. Secondly, it tailors distinct augmentation strategies to individual words based on their identified functional contributions in the text. Extensive experiments on two aspect-based sentiment analysis datasets demonstrate that the proposed TAWC model significantly improves the classification performances of popular language models, achieving gains of up to 4% compared with the case of data without augmentation, thereby setting a new standard in the field of text augmentation. Full article

Generative models have shown excellent results in aspect-based sentiment analysis tasks by predicting quadruples by setting specific template formats. The existing research predicts sentiment elements and enhances the dependency between elements using the multi-template prompting method, but it does not realize the information interaction in the generation process, and it ignores the dependency between the prompt template and the aspect terms and opinion terms in the input sequence. In this paper, we propose a Dual-template Prompted Mutual Learning (DPML) generative model to enhance the information interaction between generation modules. Specifically, this paper designs a dual template based on prompt learning and, at the same time, develops a mutual learning information enhancement module to guide each generated training process to interact with iterative information. Secondly, in the decoding stage, a label marking the interactive learning module is added to share the explicit emotional expression in the sequence, which can enhance the ability of the model to capture implicit emotion. On two public datasets, our model achieves an average improvement of 5.3% and 3.4% in F1 score compared with the previous state-of-the-art model. In the implicit sentiment analysis experiment, the F1 score of the proposed model in the data subset containing implicit words is increased by 2.75% and 3.42%, respectively. Full article

This study analyzed 2421 microblogs posted between the year 2012 to March 2022 reflecting the microbloggers’ attitudes toward different cultures. Results indicated that (1) the number of microblog posts expressing the users’ intercultural attitudes increased distinctly from 2019 to March 2022, with females users in general posting more microblogs than males; (2) females posted more microblogs encompassing positive emotions to show their interest and motivation to learn about foreign cultures, and the tendency to value and appreciate cultural differences, whereas males created more sentimentally neutral posts that revealed their recognition of the existence of cultural differences, and females and males posted a similar number of microblogs containing negative emotions; and (3) more posts involved “small c” culture were posted than those containing themes belonging to the “Big C” culture. Gender gap was further observed regarding the cultural themes concerned by the microbloggers. Implications were discussed in the context of intercultural education. Full article

This purpose of this research is to understand the role of networked narratives in social media in modulating viewer prejudice toward ethnic neighborhoods. We designed experimental videos on YouTube based on intergroup contact theory and narrative frameworks aimed at (1) gaining knowledge, (2) reducing anxiety, and (3) fostering empathy. Despite consistent storytelling across the videos, we observed significant variations in viewer emotions, especially in replies to comments. We hypothesized that these discrepancies could be explained by the influence of the surrounding digital network on the narrative’s reception. Two-stage research was conducted to understand this phenomenon. First, automated emotion analysis on user comments was conducted to identify the varying emotions. Then, we explored contextual factors surrounding each video on YouTube, focusing on algorithmic curation inferred from traffic sources, region, and search keywords. Findings revealed that negative algorithmic curation and user interactivity result in overall negative viewer emotion, largely driven by video placement and recommendations. However, videos with higher traffic originating from viewers who had watched the storyteller’s other videos result in more positive sentiments and longer visits. This suggests that consistent exposure within the channel can foster more positive acceptance of cultural outgroups by building trust and reducing anxiety. There is the need, then, for storytellers to curate discussions to mitigate prejudice in digital contexts. Full article

Dimensional aspect-based sentiment analysis (dimABSA) aims to recognize aspect-level quadruples from reviews, offering a fine-grained sentiment description for user opinions. A quadruple consists of aspect, category, opinion, and sentiment intensity, which is represented using continuous real-valued scores in the valence–arousal dimensions. To address this task, we propose a hybrid approach that integrates the BERT model with a large language model (LLM). Firstly, we develop both the BERT-based and LLM-based methods for dimABSA. The BERT-based method employs a pipeline approach, while the LLM-based method transforms the dimABSA task into a text generation task. Secondly, we evaluate their performance in entity extraction, relation classification, and intensity prediction to determine their advantages. Finally, we devise a hybrid approach to fully utilize their advantages across different scenarios. Experiments demonstrate that the hybrid approach outperforms BERT-based and LLM-based methods, achieving state-of-the-art performance with an F1-score of 41.7% on the quadruple extraction. Full article

The COVID-19 pandemic has had a profound impact on global health, inspiring the widespread use of machine learning in combating the disease, particularly in prediction models. This study aimed to assess academic publications utilizing machine learning prediction models to combat COVID-19. We analyzed 2422 original articles published between 2020 and 2023 with bibliometric tools such as Histcite Pro 2.1, Bibliometrix, CiteSpace, and VOSviewer. The United States, China, and India emerged as the most prolific countries, with Stanford University producing the most publications and Huazhong University of Science and Technology receiving the most citations. The National Natural Science Foundation of China and the National Institutes of Health have made significant contributions to this field. Scientific Reports is the most frequent journal for publishing these articles. Current research focuses on deep learning, federated learning, image classification, air pollution, mental health, sentiment analysis, and drug repurposing. In conclusion, this study provides detailed insights into the key authors, countries, institutions, funding agencies, and journals in the field, as well as the most frequently used keywords. Full article

In the competitive landscape of Airbnb hosting, optimizing pricing strategies for properties is a complex challenge that requires revenue maximization with high occupancy rates. This research aimed to introduce a solution that leverages big data and machine learning techniques to help hosts improve their property’s market performance. Our primary goal was to introduce a solution that can augment property owners’ understanding of their property’s market value within their urban context, thereby optimizing both the utilization and profitability of their listings. We employed a multi-faceted approach with diverse models, including support vector regression, XGBoost, and neural networks, to analyze the influence of factors such as location, host attributes, and guest reviews on a listing’s financial performance. To further refine our predictive models, we integrated natural language processing techniques for in-depth listing review analysis, focusing on term frequency-inverse document frequency (TF-IDF), bag-of-words, and aspect-based sentiment analysis. Integrating such techniques allowed for in-depth listing review analysis, providing nuanced insights into guest preferences and satisfaction. Our findings demonstrated that AirBnB hosts can effectively utilize both state-of-the-art and traditional machine learning algorithms to better understand customer needs and preferences, more accurately assess their listings’ market value, and focus on the importance of dynamic pricing strategies. By adopting this data-driven approach, hosts can achieve a balance between maintaining competitive pricing and ensuring high occupancy rates. This method not only enhances revenue potential but also contributes to improved guest satisfaction and the growing field of data-driven decisions in the sharing economy, specially tailored to the challenges of short-term rentals. Full article

Urban forest parks play a vital role in promoting physical activities (PAs) and providing cultural ecosystem services (CESs) that enhance citizens’ well-being. This study aims to reevaluate CESs by focusing on the physical activity experiences of park visitors to optimize park management and enhance citizen satisfaction. This study utilized social media data and employed natural language processing techniques and text analysis tools to examine experiences related to physical activities in Beijing Olympic Forest Park, Xishan Forest Park, and Beigong Forest Park. A specialized sports activity dictionary was developed to filter and analyze comments related to PA, emphasizing the impact of natural environments on enjoyment and participation in PA. The importance–performance analysis (IPA) method was used to assess the service characteristics of each park. The findings reveal that urban forest parks are highly valued by citizens, particularly for their natural landscapes, leisure activities, and the emotional fulfillment derived from PA, with 82.58% of comments expressing positive sentiments. Notably, appreciation for natural landscapes was exceptionally high, as evidenced by the frequent mentions of key terms such as ‘scenery’ (mentioned 2871 times), ‘autumn’ (mentioned 2314 times), and ‘forest’ (mentioned 1439 times), which significantly influence park usage. However, 17.11% of the reviews highlighted dissatisfaction, primarily with the management of facilities and services during sports and cultural activities. These insights underscore the need for performance improvements in ecological environments and sports facilities. This study provides a novel perspective on assessing and optimizing urban forest parks’ functions, particularly in supporting active physical engagement. The rich CESs offered by these parks enhance physical activity experiences and overall satisfaction. The findings offer strategic insights for park managers to better meet citizens’ needs and improve park functionality. Full article

Studies have found that competitive products’ online review ratings (ORRs) have a spillover effect on the focal product’s sales. However, the spillover effect of online review sentiment (ORS) as an essential component of online review analysis has yet to be studied. In this study, we analyze online review content from JD.com using the latent Dirichlet allocation to identify the product attribute topics that consumers are most concerned about. We then construct a baseline regression model of ORS and ORRs to explore the effects of online competitive product reviews on focal product sales. Moreover, we examine how the interaction between ORS and critical factors of online reviews affect sales. Our results indicate that the ORS of competitive products has a negative effect on focal product sales, and the effect is greater than the ORS and ORRs of focal products, respectively. In addition, the ORS of competitive products inhibits the sale of focal products as evaluations of product attributes become more positive or online review usefulness increases. We also find that the effect of ORRs of competitive products is not significant, which may be because clothing, as an experiential product, requires consumers to gain more information about specific usage scenarios before making a decision. This study provides a more accurate basis for consumer decision-making and offers retailers a novel approach to developing marketing strategies. Full article

Emerging research indicates that sentiment analyses of Dubrovnik focus mainly on hotel accommodations and restaurants. However, little attention has been paid to attractions, even though they are an important aspect of destinations and require more care and investment than amenities. This study examines how visitors experience Dubrovnik based on the reviews published on the Tripadvisor platform. Data were collected by implementing a web-scraping script to retrieve reviews of the tourist attraction “Old Town” from Tripadvisor, while data augmentation and random oversampling techniques were applied to address class imbalances. A sentiment analysis model, based on the pre-trained RoBERTa, was also developed and evaluated. In particular, a sentiment analysis was performed to compare reviews from 2022 and 2023. Overall, the results of this study are promising and demonstrate the effectiveness of this model and its potential applicability to other attractions. These findings provide valuable insights for decision makers to improve services and to increase visitor engagement. Full article

Designing securities applications for mobile devices is challenging due to their inherent complexity, necessitating improvement through the analysis of online reviews. However, research applying deep learning techniques to the sentiment analysis of Korean text remains limited. This study explores the use of Aspect-Based Sentiment Analysis (ABSA) as an effective alternative to traditional user research methods for securities application design. By analyzing large volumes of text-based user review data of Korean securities applications, the study identifies critical elements like “update”, “screen”, “chart”, “login”, “access”, “authentication”, “account”, and “transaction”, revealing nuanced user sentiments through techniques such as PMI, SVD, and Word2Vec. ABSA offers deeper insights compared to overall ratings, uncovering hidden areas of dissatisfaction despite positive biases in reviews. This research demonstrates the scalability and cost-effectiveness of ABSA in mobile-application design research. Full article