In recent years, e-commerce platforms have become popular and transformed the way people buy and sell goods. People are rapidly adopting Internet shopping due to the convenience of purchasing from the comfort of their homes. Online review sites allow customers to share their thoughts on products and services. Customers and businesses increasingly rely on online reviews to assess and improve the quality of products. Existing literature uses natural language processing (NLP) to analyze customer reviews for different applications. Due to the growing importance of NLP for online customer reviews, this study attempts to provide a taxonomy of NLP applications based on existing literature. This study also examined emerging methods, data sources, and research challenges by reviewing 154 publications from 2013 to 2023 that explore state-of-the-art approaches for diverse applications. Based on existing research, the taxonomy of applications divides literature into five categories: sentiment analysis and opinion mining, review analysis and management, customer experience and satisfaction, user profiling, and marketing and reputation management. It is interesting to note that the majority of existing research relies on Amazon user reviews. Additionally, recent research has encouraged the use of advanced techniques like bidirectional encoder representations from transformers (BERT), long short-term memory (LSTM), and ensemble classifiers. The rising number of articles published each year indicates increasing interest of researchers and continued growth. This survey also addresses open issues, providing future directions in analyzing online customer reviews.

Social media reviews are a valuable data source, reflecting consumer experiences and interactions with businesses. This study leverages such data to develop a passive surveillance framework for food safety in urban India. By employing a Bidirectional Encoder Representations from Transformers (BERT)-powered Aspect-Based Sentiment Analysis tool, branded as Eat At Right Place (ERP), the study analyses over 100,000 reviews from 93 restaurants to identify and assess food safety signals. The Causality Assessment Index (CAI) and Severity Assessment Score (SAS) are introduced to systematically evaluate potential risks. The CAI uses pattern recognition and temporal relationships to establish causality while the SAS quantifies severity based on sub-aspects such as cleanliness, food handling, and unintended health outcomes. Results indicate that 40% of the restaurants had a CAI above 1, highlighting significant food safety concerns. The framework successfully prioritizes corrective actions by grading the severity of issues, demonstrating its potential for real-time food safety management. This study underscores the importance of integrating innovative data-driven approaches into public health monitoring systems and suggests future improvements in natural language processing algorithms and data source expansion. The findings pave the way for enhanced food safety surveillance and timely regulatory interventions.

Beyond its immediate health consequences, the COVID-19 pandemic led to an exacerbation in the mental health of the global population. Regular exercise and its lack thereof are also known to affect mental health. Tweets and their content analysis can provide information about aspects of users\' lives including their health habits and mental health. The purpose of this study was to examine individuals\' exercise habits and mental health during the pandemic by means of sentiment and correlational analyses. These results indicate that, while exercise and mental health tweets were more COVID-focused in the first 12 months of the pandemic, exercise tweets became more exercise-focused, and mental health tweets became more mental-health-focused eventually during the pandemic. Efforts to increase exercise participation in individuals may prove beneficial. Further research needs to examine the effects of exercise on mental health in the aftermath of COVID-19.

Infectious diseases have posed a global threat recently, progressing from endemic to pandemic. Early detection and finding a better cure are methods for curbing the disease and its transmission. Machine learning (ML) has demonstrated to be an ideal approach for early disease diagnosis. This review highlights the use of ML algorithms for monkeypox (MP). Various models, such as CNN, DL, NLP, Naïve Bayes, GRA-TLA, HMD, ARIMA, SEL, Regression analysis, and Twitter posts were built to extract useful information from the dataset. These findings show that detection, classification, forecasting, and sentiment analysis are primarily analyzed. Furthermore, this review will assist researchers in understanding the latest implementations of ML in MP and further progress in the field to discover potent therapeutics.

We examined the impact of perpetrator and victim gender on bystander helping choices and assault perceptions. Participants (32 females, 37 males) read about two simultaneously occurring sexual assaults, indicated which victim they would help, and gave their perceptions of the assaults. We used a within-participants design that fully manipulated the perpetrator and victim gender for both assaults. Results showed female victims of male perpetrators and male victims of female perpetrators were most and least likely to be chosen for help, respectively. Cognitive networks derived from open-ended responses provided insight into the rationale used by participants to make helping decisions in ways that differed by perpetrator and victim gender.

This research examines the perceptions of Twitter users regarding the prevalent topics within Work-Life Balance communication before and after the COVID-19 pandemic. The pressing questions surrounding current labour market drivers are addressed, particularly regarding the ongoing Fourth Industrial Revolution and the COVID-19 pandemic\'s impact on communicated themes, particularly in the Human Resource Management field, where Work-Life Balance has emerged as a key concept. Social media platforms like Twitter are pivotal in fostering discussions on Work-Life Balance in society. Over the past decade, Twitter has evolved into a significant research platform researchers utilise in more than ten thousand research articles. The online discourse on Twitter raises awareness of the importance of balancing work and personal life. The COVID-19 pandemic has unveiled new facets of Work-Life Balance, with social media as a key platform for discussing these issues. This research uses Social Media Analysis based on the Hashtag Research framework. A total of 1,768,628 tweets from 499,574 users were examined, and frequency, topic, and sentiment analysis were conducted. Pre-pandemic, the most communicated Work-Life Balance topics were performance and time management, while recruitment and employee development were identified post-pandemic. Pre-pandemic, the highest proportion of negative sentiment was time management and mental health prevention, shifting to time, employee development, and mental health prevention post-pandemic. Despite the limitations of our research, a proposed redefinition of the concept is also presented, including a design for an integrated Work-Life Balance model based on topics communicated by Twitter users. Given the need for a more robust approach to redefining the concept and developing an integrative Work-Life Balance model, the article provides fresh insights for future research.

BACKGROUND: Commonly offered as supportive care, therapist-led online support groups (OSGs) are a cost-effective way to provide support to individuals affected by cancer. One important indicator of a successful OSG session is group cohesion; however, monitoring group cohesion can be challenging due to the lack of nonverbal cues and in-person interactions in text-based OSGs. The Artificial Intelligence-based Co-Facilitator (AICF) was designed to contextually identify therapeutic outcomes from conversations and produce real-time analytics.
OBJECTIVE: The aim of this study was to develop a method to train and evaluate AICF\'s capacity to monitor group cohesion.
METHODS: AICF used a text classification approach to extract the mentions of group cohesion within conversations. A sample of data was annotated by human scorers, which was used as the training data to build the classification model. The annotations were further supported by finding contextually similar group cohesion expressions using word embedding models as well. AICF performance was also compared against the natural language processing software Linguistic Inquiry Word Count (LIWC).
RESULTS: AICF was trained on 80,000 messages obtained from Cancer Chat Canada. We tested AICF on 34,048 messages. Human experts scored 6797 (20%) of the messages to evaluate the ability of AICF to classify group cohesion. Results showed that machine learning algorithms combined with human input could detect group cohesion, a clinically meaningful indicator of effective OSGs. After retraining with human input, AICF reached an F1-score of 0.82. AICF performed slightly better at identifying group cohesion compared to LIWC.
CONCLUSIONS: AICF has the potential to assist therapists by detecting discord in the group amenable to real-time intervention. Overall, AICF presents a unique opportunity to strengthen patient-centered care in web-based settings by attending to individual needs.
UNASSIGNED: RR2-10.2196/21453.

Sentiment analysis also referred to as opinion mining, plays a significant role in automating the identification of negative, positive, or neutral sentiments expressed in textual data. The proliferation of social networks, review sites, and blogs has rendered these platforms valuable resources for mining opinions. Sentiment analysis finds applications in various domains and languages, including English and Arabic. However, Arabic presents unique challenges due to its complex morphology characterized by inflectional and derivation patterns. To effectively analyze sentiment in Arabic text, sentiment analysis techniques must account for this intricacy. This paper proposes a model designed using the transformer model and deep learning (DL) techniques. The word embedding is represented by Transformer-based Model for Arabic Language Understanding (ArabBert), and then passed to the AraBERT model. The output of AraBERT is subsequently fed into a Long Short-Term Memory (LSTM) model, followed by feedforward neural networks and an output layer. AraBERT is used to capture rich contextual information and LSTM to enhance sequence modeling and retain long-term dependencies within the text data. We compared the proposed model with machine learning (ML) algorithms and DL algorithms, as well as different vectorization techniques: term frequency-inverse document frequency (TF-IDF), ArabBert, Continuous Bag-of-Words (CBOW), and skipGrams using four Arabic benchmark datasets. Through extensive experimentation and evaluation of Arabic sentiment analysis datasets, we showcase the effectiveness of our approach. The results underscore significant improvements in sentiment analysis accuracy, highlighting the potential of leveraging transformer models for Arabic Sentiment Analysis. The outcomes of this research contribute to advancing Arabic sentiment analysis, enabling more accurate and reliable sentiment analysis in Arabic text. The findings reveal that the proposed framework exhibits exceptional performance in sentiment classification, achieving an impressive accuracy rate of over 97%.

Climate security refers to the risks posed by climate change on nations, societies, and individuals, including the possibility of conflicts. As an emerging field of research and public debate, where conceptual definitions are not yet fully agreed upon, gaining insights into global discussions on climate security enables systematizing its various interpretations and framings, mapping thematic priorities, and understanding information gaps that need to be filled. Considering Twitter as an important digital forum for information exchanges and dialogue, the dataset was created through the development of a query strategy based on a snowball scraping technique, which collected tweets containing hashtags related to climate security between January 2014 to May 2023. The dataset comprises 636,379 tweets. Content analysis was performed using text mining and network analysis techniques to generate additional data on sentiment, countries mentioned in the body of tweets, and hashtag co-occurrences. With almost 10 years of data, the utility of this dataset lies in the ability to assess the discursive evolution of a particular topic since its inception.

Aspect-level sentiment analysis within multimodal contexts, focusing on the precise identification and interpretation of sentiment attitudes linked to the target aspect across diverse data modalities, remains a focal research area that perpetuates the advancement of discourse and innovation in artificial intelligence. However, most existing methods tend to focus on extracting visual features from only one facet, such as face expression, which ignores the value of information from other key facets, such as the textual information presented by the image modality, resulting in information loss. To overcome the aforementioned constraint, we put forth a novel approach designated as Multi-faceted Information Extraction and Cross-mixture Fusion (MIECF) for Multimodal Aspect-based Sentiment Analysis. Our approach captures more comprehensive visual information in the image and integrates these local and global key features from multiple facets. Local features, such as facial expressions and textual features, provide direct and rich emotional cues. By contrast, the global feature often reflects the overall emotional atmosphere and context. To enhance the visual representation, we designed a Cross-mixture Fusion method to integrate this local and global multimodal information. In particular, the method establishes semantic relationships between local and global features to eliminate ambiguity brought by single-facet information and achieve more accurate contextual understanding, providing a richer and more precise manner for sentiment analysis. The experimental findings indicate that our proposed approach achieves a leading level of performance, resulting in an Accuracy of 79.65 % on the Twitter-2015 dataset, and Macro-F1 scores of 75.90 % and 73.11 % for the Twitter-2015 and Twitter-2017 datasets, respectively.