Text Comments Research Articles

Evaluating ChatGPT as a patient resource for frequently asked questions about lung cancer surgery—a pilot study

ObjectiveChat-based artificial intelligence programs like ChatGPT are reimagining how patients seek information. This study aims to evaluate the quality and accuracy of ChatGPT-generated answers to common patient questions about lung cancer surgery. MethodsA 30-question survey of patient questions about lung cancer surgery was posed to ChatGPT in July 2023. The ChatGPT-generated responses were presented to 9 thoracic surgeons at 4 academic institutions who rated the quality of the answer on a 5-point Likert scale. They also evaluated if the response contained any inaccuracies and were prompted to submit free text comments. Responses were analyzed in aggregate. ResultsFor ChatGPT-generated answers, the average quality ranged from 3.1 to 4.2 of 5.0, indicating they were generally “good” or “very good.” No answer received a unanimous 1-star (poor quality) or 5-star (excellent quality) score. Minor inaccuracies were found by at least 1 surgeon in 100% of the answers, and major inaccuracies were found in 36.6%. Regarding ChatGPT, 66.7% of surgeons thought it was an accurate source of information for patients. However, only 55.6% thought they were comparable with answers given by experienced thoracic surgeons, and only 44.4% would recommend it to their patients. Common criticisms of ChatGPT-generated answers included lengthiness, lack of specificity regarding surgical care, and lack of references. ConclusionsChat-based artificial intelligence programs have potential to become a useful information tool for patients with lung cancer surgery. However, the quality and accuracy of ChatGPT-generated answers need improvement before thoracic surgeons would consider this method as a primary education source for patients.

Exploring the Impact of Emojis on Paralanguage in Social Media Communication among University Students

This research investigated the evolving role of emojis in shaping paralanguage within social media communication, with a particular focus on how these visual symbols enhance linguistic and emotional expression. The study examined the nuanced ways in which emojis contribute to the overall tone, context, and meaning in written discourse, while also considering cultural and contextual variations in emoji usage among individuals from diverse linguistic backgrounds. Conducted with a sample of 55 Year-3 Bachelor of Education students at a university in Fiji, where English is a second language, the research employed a mixed methods approach to analyze data on the usage and perception of emojis in social media communication. This paper argues that emojis play a crucial role in conveying emotions, fostering engagement, and adding personality to messages, thereby augmenting textual communication with emotional nuance and clarity, leading to more empathetic and meaningful online interactions. The findings indicate that emojis are widely used across various communication channels, including text messages, comments, and direct messages, reflecting their versatility and widespread adoption. Respondents overwhelmingly perceive emojis as positive tools that enhance the effectiveness of digital communication by conveying emotions, fostering engagement, and adding personality to messages. The study also highlights the significant role emojis play in nonverbal communication, particularly in expressing emotions that may be difficult to articulate through text alone. By providing emotional nuance, clarity, and depth, emojis contribute to more empathetic and meaningful interactions in online spaces. Overall, this research provides valuable insights into the dynamic interplay between emojis and paralanguage in contemporary digital communication, signifying the importance of these symbols in enriching text-based exchanges and fostering more engaging and emotionally resonant social media interactions.

Attitude towards artificial intelligence among medical students in Kerala

Objectives: Surveys conducted on a global scale have revealed a significant level of importance in artificial intelligence (AI) among medical pupils’. There have been no inquiries overseen in Kerala to explore the opinions of medical students regarding the consumption of AI in health care or their level of understanding of AI. We seek to evaluate the depths of Kerala’s medical learners concerning AI, both in the field of health care and its potential integration into the medical program. Methods: A digital scrutiny apparatus was created through a thorough examination of existing text and collaboration with doyens in the field. The inquiry was tested with a trio of medical learners and improved based on their valuable input. We distributed a digital probe to all medical pupils in Kerala, which amounted to around 20,000 students. The survey was available for responses from April 1st, 2024 to June 1st, 2024. The students’ responses were carefully analysed, both in terms of categories and the content of their free text comments. The analysis was conducted using open coding techniques to ensure a thorough qualitative analysis. Results: In total, 1000 students provided comprehensive answers to all the questions. A significant proportion of students (82.0%) fell within the age range of 20-29 years, pursuing medicine as their undergraduate degree (85%). Many students showed a keen interest in AI, with a significant majority (79.9%) expressing their curiosity. A large percentage (83.1%) asserted to have a basic realizing of AI, but when it came to the fundamental computational principles of AI, only a minority (39.8%) agreed that they grasped them. Similarly, more than half of the students (51.6%) acknowledged their lack of knowledge regarding the limitations of AI. Most students (81.1%) had not obtained any instruction in AI. A significant portion of students (57.3%) expressed their support for incorporating AI into medical training, while a majority (76.2%) expressed a desire for increased emphasis on teaching AI in medicine. A significant portion (61.3%) of medical students expressed a lack of worry about the potential effect of AI on their job safety as doctors. According to the survey, the majority of students believe that radiology, pathology, and medical administration are the specialties that will be most affected by AI. On the other hand, psychiatry, palliative care, and obstetrics and gynaecology are considered to be the least likely to be stuck by AI. Through qualitative study of the free text notes, it was discovered that AI is viewed as a valuable tool rather than a replacement for doctors. This finding emerged as a common theme. Conclusion: It seems that medical students in Kerala have shown a keen interest in AI. However, their knowledge about AI is limited and they feel they lack a deep understanding of its fundamental computational principles and limitations. It seems that AI is currently lacking in the medical curriculum in Kerala and India, and a majority of surveyed students expressed their support for its implementation. These findings align with previous surveys conducted on a global scale.

Facilitators and barriers to implementing successful exclusion among children with shiga toxin-producing Escherichia coli: a qualitative analysis of public health case management records

BackgroundShiga toxin-producing Escherichia coli (STEC) infections are a significant public health concern as they can cause serious illness and outbreaks. In England, STEC incidence is highest among children and guidance recommends that children under six diagnosed with STEC are excluded from childcare until two consecutive stool cultures are negative. We aimed to describe the barriers and facilitators to implementing exclusion and the impact of exclusion policies on young children and their families.MethodsIndividual level data was obtained from a wider study focusing on shedding duration among STEC cases aged < 6 years between March 2018 – March 2022. Data was extracted from England’s public health case management system. The case management system includes notes on telephone conversations, email correspondence and meeting minutes relating to the case. Collected data consisted of free text in three forms: (1) quotes from parents, either direct or indirect, (2) direct quotes from the case record by health protection practitioners or environmental health officers, and (3) summaries by the data collector after reviewing the entire case record. We analysed free text comments linked to 136 cases using thematic analysis with a framework approach.ResultsThe median age of included cases was 3 years (IQR 1.5-5), with males accounting for 49%. Nine key themes were identified. Five themes focused on barriers to managing exclusion, including (i) financial losses, (ii) challenges with communication, engagement and collaboration, (iii) issues with sampling, processing, and results, (iv) adverse impact on children and their families and (v) conflicting exclusion advice. Four themes related to facilitators to exclusion, including (i) good communication with parents and childcare settings, (ii) support with childcare, (iii) improvements to sampling, testing, and reporting of results, and (iv) provision of supervised control measures.ConclusionsQualitative analysis of public health case records can provide evidence-based insights around complex health protection issues to inform public health guidelines. Our analysis highlights the importance of considering wider social and economic consequences of exclusion when developing policies and practices for the management of STEC in children.

Joint coordinate attention mechanism and instance normalization for COVID online comments text classification.

The majority of extant methodologies for text classification prioritize the extraction of feature representations from texts with high degrees of distinction, a process that may result in computational inefficiencies. To address this limitation, the current study proposes a novel approach by directly leveraging label information to construct text representations. This integration aims to optimize the use of label data alongside textual content. The methodology initiated with separate pre-processing of texts and labels, followed by encoding through a projection layer. This research then utilized a conventional self-attention model enhanced by instance normalization (IN) and Gaussian Error Linear Unit (GELU) functions to assess emotional valences in review texts. An advanced self-attention mechanism was further developed to enable the efficient integration of text and label information. In the final stage, an adaptive label encoder was employed to extract relevant label information from the combined text-label data efficiently. Empirical evaluations demonstrate that the proposed model achieves a significant improvement in classification performance, outperforming existing methodologies. This enhancement is quantitatively evidenced by its superior micro-F1 score, indicating the efficacy of integrating label information into text classification processes. This suggests that the model not only addresses computational inefficiencies but also enhances the accuracy of text classification.

Considerations Around the Inclusion of Children and Young People's Time in Economic Evaluation: Findings from an International Delphi Study.

People's time is a finite resource and a valuable input that ought to be considered in economic evaluations taking a broad, societal perspective. Yet, evaluations of interventions focusing on children and young people (CYP) rarely account for the opportunity cost of time in this population. As a key reason for this, health economists have pointed to uncertainty around when it is appropriate to include CYP time-related costs in an economic evaluation and highlighted the lack of clear guidance on the topic. With this in mind, we carried out a Delphi study to establish a list of relevant considerations for researchers to utilise whilst making decisions about whether and when to include CYP time in their economic evaluations. Delphi panellists were asked to propose and rate a set of possible considerations and provide additional thoughts on their ratings. Ratings were summarised using descriptive statistics, and text comments were interrogated through thematic analysis. A total of 73 panellists across 16 countries completed both rounds of a two-round Delphi study. Panellists' ratings showed that, when thinking about whether to include displaced CYP time in an economic evaluation, it is very important to consider whether: (1) inclusion would be in line with specified perspective(s) (median score: 9), (2) CYP's time may already be accounted for in other parts of the evaluation (median score: 8), (3) the amount of forgone time is substantial, either in absolute or relative terms (median score: 7) and (4) inclusion of CYP's time costs would be of interest to decision-makers (median score: 7). Respondents thought that considerations such as (1) whether inclusion would be of interest to the research community (median score: 6), (2) whether CYP's time displaced by receiving treatment is 'school' or 'play' time (median score: 5), and (3) whether CYP's are old enough for their time to be considered valuable (median score: 5) are moderately important. A range of views was offered to support beliefs and ratings, many of which were underpinned by compelling normative questions.

Awareness, support, and opinions of healthy food and drink policies: a survey of staff and visitors in New Zealand healthcare organisations

BackgroundIn 2016, a voluntary National Healthy Food and Drink Policy (hereafter, “the Policy”) was released to encourage public hospitals in New Zealand to provide food and drink options in line with national dietary guidelines. Five years later, eight (of 20) organisations had adopted it, with several preferring to retain or update their own institutional-level version. This study assessed staff and visitors’ awareness and support for and against the Policy, and collected feedback on perceived food environment changes since implementation of the Policy.MethodsCross-sectional electronic and paper-based survey conducted from June 2021 to August 2022. Descriptive statistics were used to present quantitative findings. Free-text responses were analysed following a general inductive approach. Qualitative and quantitative findings were compared by level of implementation of the Policy, and by ethnicity and financial security of participants.ResultsData were collected from 2,526 staff and 261 visitors in 19 healthcare organisations. 80% of staff and 56% of visitors were aware of the Policy. Both staff and visitors generally supported the Policy, irrespective of whether they were aware of it or not, with most agreeing that “Hospitals should be good role models.” Among staff who opposed the Policy, the most common reason for doing so was freedom of choice. The Policy had a greater impact, positive and negative, on Māori and Pacific staff, due to more frequent purchasing onsite. Most staff noticed differences in the food and drinks available since Policy implementation. There was positive feedback about the variety of options available in some hospitals, but overall 40% of free text comments mentioned limited choice. 74% of staff reported that food and drinks were more expensive. Low-income staff/visitors and shift workers were particularly impacted by reduced choice and higher prices for healthy options.ConclusionsThe Policy led to notable changes in the healthiness of foods and drinks available in NZ hospitals but this was accompanied by a perception of reduced value and choice. While generally well supported, the findings indicate opportunities to improve implementation of food and drink policies (e.g. providing more healthy food choices, better engagement with staff, and keeping prices of healthy options low) and confirm that the Policy could be expanded to other public workplaces.

How well do faculty do in providing general surgery EPA feedback?

IntroductionEntrustable Professional Activities (EPAs) provide a framework for competency-based assessment in surgery. EPA descriptions include observable behaviors by trainees at progressive levels of autonomy. The American Board of Surgery (ABS) required all General Surgery (GS) residency programs to implement assessment of 18 EPAs at the beginning of academic year 2023–2024. Microassessments provide formative self-reflection by the resident and feedback by faculty upon completion of the EPA. These frequent assessments culminate in a resident performance profile utilized by the trainee for formative growth and the clinical competency committee for summative feedback. Assessor free text comments are an opportunity to provide meaningful, constructive feedback to residents. Our aim was to analyze comments provided by faculty to residents in terms of their alignment with EPA descriptors and provision of actionable feedback. MethodsA total of 540 ​GS EPA assessments for inguinal hernia, gallbladder disease, appendicitis, trauma, and surgical consultation were evaluated from 6/2021-12/2022. We assessed free text EPA comments from faculty compared to EPA behavior descriptions for alignment with the selected EPA level of entrustment. The comments were judged on a binary scale of “Align” vs “Not Align” by two independent evaluators, with a third evaluator to address discordance. Comments were then evaluated for resident behavioral descriptions, suggestions for improvement, and positive or negative feedback. ResultsApproximately 77 ​% of EPA microassessments had alignment between level of autonomy and free text feedback. A common example of feedback discordant with level of autonomy was rating a trainee at an intraoperative level 4 (independent practice) with comments such as “required some guidance with retrocecal case and upsizing port.” Based on behavior descriptions this would be a level 3 (indirect supervision). Approximately 88 ​% of feedback contained positive comments with minimal negative feedback (e.g., “this did not go well.”). Actionable feedback including “work on optimization of retracting hand” or “continue to work clamp/tie technique and square off each knot” was present in 28.3 ​% of feedback. ConclusionsThe majority of faculty provide feedback that is aligned with the behavioral anchors of the EPAs assessed, but frequently did not provide actionable feedback to the resident regarding how to advance to the next level of entrustment. EPA entrustment behaviors provide a framework for the development of practice-ready behaviors, and if assessors anchor their feedback in the behaviors for a given entrustment level and project how a resident could proceed to the next level, they can provide a clear trajectory for skill development. Faculty development should focus on improving the frequency of actionable free text feedback, outlining how residents can advance in the future.

The Postmodern Narrative Techniques in Tim O'Brien's "How to Tell a True War Story"

The Vietnam War had an irreversible negative impact on the United States, Vietnam, and the world at large. Its effects extended beyond geopolitics and seeped into the domain of literature, where numerous works have profoundly depicted and contemplated the conflict. Tim O'Brien's "How to Tell a True War Story" stands as a prime example of such literature, set against the backdrop of the Vietnam War. Drawing on Jean-Franois Lyotard's theory of postmodern metanarratives, this paper examines how O'Brien employs postmodern narrative techniques that diverge from traditional grand narratives to articulate the essence of his war fiction. Upon analysis, the paper contends that Tim O'Brien adopts the narrative mode of metafiction, blending narrative and commentary texts alongside real and fictive elements. By doing so, he seeks to alert readers to the constructed nature of both the textual and real worlds, encouraging them to perceive war through the lens of individuals' traumatic war experiences.

CommentClass: A Robust Ensemble Machine Learning Model for Comment Classification

Enormous amounts of data are generated in the form of feedback or comments from online platforms such as social media, e-commerce, education, and programming. This feedback and comments hold significant value for making important strategic decisions; therefore, effectively analyzing them poses a major challenge. This research addresses the imperative need for an efficient comment classification model. To fill this research gap, we propose a robust ensemble machine learning (ML) model called CommentClass (RF+AdaBoost+SVM+Soft-Voting), specifically designed for the comment classification task. First, we developed eight (08) pipelines using various combinations of ML algorithms. Next, the fundamental ensemble techniques such as stacking, blending, hard-voting, soft-voting, and averaging are incorporated into these pipelines to improve comment classification performance. These ensemble models are able to discern the latent characteristics of diverse text comments, classifying them to achieve superior accuracy. The proposed CommentClass ensemble model achieved an impressive accuracy and F1-score of approximately 98% for comment classification on the YouTube dataset. This result represents an improvement in accuracy by approximately +3% compared to prior research on the same dataset. Moreover, the proposed CommentClass model obtained higher F1-scores of 90.26%, 87.04%, and 75.74%, on the Spambase, IMDB, and Twitter datasets, respectively, compared to other sophisticated models. Furthermore, the proposed CommentClass model exhibited significant accuracy on the SMS dataset and two distinct synthetic datasets.

Buyers’ Negative Ratings and Textual Comments on eBay: Reasons for Posting Ratings and Factors in Denouncing Sellers

Buyers’ Negative Ratings and Textual Comments on eBay: Reasons for Posting Ratings and Factors in Denouncing Sellers

Perceptions on Penicillin Allergy Labels among Nurses and Prescribers in Three Pediatric Urgent Care Sites

Background: National guidelines recommend penicillins (PCN) as first-line treatment for many common pediatric infections in the outpatient setting. Although less than 1% of the United States population has a true, IgE-mediated PCN-allergy, approximately 10% of patients are labeled with a PCN-allergy. Accurate adverse drug reaction (ADR) documentation plays an important role in this over-labeling. We have previously shown that nurses feel assessment and documentation of PCN-allergies are critical to their role. However, additional evidence purports nurse hesitancy to interrogate allergy accuracy or reclassify parent’s response to side effect. Our objective was to explore frontline clinicians’ confidence in assessing, documenting, and responding to PCN-allergy labels. Methods: To expose barriers and prioritize improvement ideas for a multidisciplinary quality improvement (QI) project aimed to improve PCN-allergy labeling in our pediatric urgent care clinics, we deployed this investigator-developed survey to prescribers and nurses. It’s comprised of 14-questions scored on a 5-point Likert scale (4 demographic, 4 PCN/safety, 3 allergy types, 4 allergy documentations, 3 treatment options), and 1 optional free-text. We used descriptive statistics to compare survey responses between prescribers and nurses and evaluated free text comments for themes. Results: Eighty-seven clinicians across 3 sites participated, with a response rate of 35%, with variation by sites (25.3% to 41.4%). Forty-one percent of (n=36) responders have been in practice &gt;15 years and 40.2% (n=35) have worked at our hospital &gt; 15 years (Table 1). Overall, perceived knowledge of PCN-allergies and safety was favorable (Table 2). Prescribers reported higher confidence with: 1) perceiving many patients who believe they are allergic to PCN can safety take PCN (prescribers median=5 [IQR: 4, 5] vs. nurses median=4 [4,4], p = 0.003); and 2) perceiving that time pressures influenced their ability to reconcile allergies and side effects (prescribers median=4 [4, 5] vs. nurses median=3 [2, 4], p = 0.001). Both prescribers and nurses reported lower confidence in continuing to administer or prescribe an antibiotic in the setting of a reported ADR. Thirteen respondents (15%) provided comments with specific requests for additional family education and practice guidance, including the referral process to subspecialty clinics for PCN-allergy testing. Conclusions: Our survey results identified barriers to accurate PCN-allergy labels, including knowledge on documentation, time pressures, hesitancy to challenge parent report, and uncertainty on referral process for PCN-allergy testing. This survey will inform future drivers for our QI. Opportunities include electronic medical record refinement, improving referrals to PCN-allergy de-labeling clinics, and the development of scripted education to guide family discussions.Disclosure: Rana El Feghaly: Merck- grant funding. Amanda Nedved: Contracted Research – Merck

Service quality dimensions in Austrian food retailing – a text mining approach for physical retail stores

ABSTRACT Investigating Service Quality (SQ) and its implications for customer satisfaction has become an increasingly popular research area. Especially in the context of physical grocery retailing, ensuring customer satisfaction has become a key success factor. Most previous research studies apply traditional methods (e.g. surveys) for physical shops and utilize text mining-based approaches mainly for e-commerce. In our paper, we propose a novel approach based on LDA. By combining expert-based word counting analysis with the LDA approach, we confirm that unsupervised text mining based on LDA can be used as a reliable approach to cluster textual comments to SQ dimensions in physical retail settings. We have analyzed over 163,000 publicly available textual customer reviews related to the Austrian market, which is special in terms of its extremely high density of retail outlets, high price levels and a tendency towards traditional form of shopping. Our results show that personal interaction, policies, and product-related aspects seem to be positive drivers of customer satisfaction, while reliability is a clear driver of customer dissatisfaction. Results also show a significantly higher relevance of personal interaction in smaller stores and cities with fewer than 5,000 inhabitants than in other store types and bigger cities. Regarding practical implications, hypermarkets should focus on physical aspects to reduce negative reviews and increase efforts in personal interaction to increase positive reviews. On the other hand, smaller stores should continue to rely on personal interaction to avoid negative reviews and might focus on higher reliability to generate more positive reviews. The applied text-mining approach enables future research with a starting base to analyze SQ dimensions and their relevance in additional countries or area, as e.g. fashion or hardware retail.

Sentiment Analysis and Ratings of Professors: A Comparison of Rate-My-Professor and Department Results

Previous studies employing numeric scores have observed that anonymous postings on the Rate-My-Professor (RMP) site tend to be more negative than ratings within a school. In this study, rather than using numeric evaluations, we compared the official UNIV evaluations with the RMP evaluations at the individual faculty course level, employing sentiment analysis on the text comments in evaluations. We compared positive sentiments in ratings on RMP to those in a specific business school department at a university. Our results show a statistically significant difference, with higher positive ratings at the university. We also analyzed emotions using NRC, finding a significant difference, with RMP having higher levels of negative emotions.

Sentiment analysis of the Hamas-Israel war on YouTube comments using deep learning

Sentiment analysis aims to classify text based on the opinion or mentality expressed in a situation, which can be positive, negative, or neutral. Therefore, in the world, a lot of opinions are available on various social media sites, which must be gathered and analyzed to assess the general public’s opinion. Finding and monitoring comments, as well as manually extracting the information contained in them, is a difficult task due to the vast diversity of ideas on YouTube. Identifying public opinion on war topics is crucial for offering insights to opposing sides based on popular opinion and emotions about the ongoing war. To address the gap, we build a model on YouTube comment sentiment analysis of the Hamas-Israel war to determine public opinion. In this study, we address the gaps by developing a deep learning-based approach for sentiment analysis. We have collected 24,360 comments from popular YouTube News Channels including BBC, WION, Aljazeera, and others about the Hamas-Israel War using YouTube API and Google spreadsheet and labeled them by linguistic experts into three classes: positive, negative, and neutral. Then, textual comments were preprocessed using natural language processing (NLP) techniques, and features were extracted using Word2vec, FastText, and GloVe. Moreover, we have used the SMOTE data balancing technique and used different data splits, but the 80/20 train-test split ratio has the highest accuracy. For classification model building, commonly used classification algorithms LSTM, Bi-LSTM, GRU, and Hybrid of CNN and Bi-LSTM were applied, and their performance is compared. As a result, the Hybrid of CNN and Bi-LSTM with Word2vec achieved the highest performance with 95.73% accuracy for comments classifications.

What drives employers’ favorability ratings on employer review platforms? The role of symbolic, personal, and emotional content

AbstractEmployer review platforms have changed the recruitment landscape by allowing current and former employees to post messages about an employer outside of direct company control. Therefore, they have emerged as an important form of third‐party employer branding. However, we know little about how such open‐ended comments relate to the key variable in employer reviews: employers’ favorability rating. Therefore, we start by situating this variable among other constructs in the employer branding space. Next, we build theory on how content in the open‐ended comments of an employer review relates to the positivity or negativity of the reviews’ favorability rating. We test our hypotheses via a text‐mining analysis of approximately half a million employer reviews. The results reveal an intriguing discrepancy. Although instrumental, impersonal, and cognitive content is more prevalent in employer reviews, symbolic, personal, and emotional content dominates employer reviews’ favorability rating. In terms of practical implications, this result shows that merely inspecting the frequency of attributes mentioned in employer review text comments as a basis for changing company policies of employer branding efforts might be misguided. We discuss implications for theory and future research, and provide our dictionary for further scholarly and practical use.

A deep feature interaction and fusion model for fake review detection: Advocating heterogeneous graph convolutional network

In contemporary real-world scenarios, opinion spammers are hired to fabricate reviews that unfairly promote or demote particular products or services for personal gain. Although considerable attention has been devoted to addressing the problem, existing approaches often overlook the heterogeneous nature of reviewer–product interactions. Specifically, the correlation between review text (comments) and overall ratings, which provides various latent rich information to expose fake reviews, remains inadequately explored. Current methodologies focus on limited interactions, such as reviewer–review, product–review, or reviewer–product interactions, while neglecting significant aspects like reviewer–review–product and reviewer–rating–product interactions, leading to inadequate classifier performance. Motivated by this observation, this study proposes a novel Deep Feature Interaction and Fusion Model (DFIFM) whose ideas are five-folds: (a) constructing a reviewer–product interaction bipartite graph that represents heterogeneous feature node interactions through review text and overall rating values; (b) recognizing the existing mutual interactive relationship between review text and overall rating features, we construct a unified GCN to gain additional insights into feature relationships and capture mutual heterogeneous interactions between nodes; (c) to handle the encoding of unstructured review text features as edge attributes, we adopt a convolutional neural network (CNN); (d) attention mechanisms and fusion techniques are employed to capture interdependencies among reviewer–product latent features; and (e) a Multilayer Perceptron (MLP) utilizes the resulting latent feature representation for review classification. Experimental results on three publicly available datasets demonstrate its superiority over state-of-the-art baselines.

Exploring factors influencing quality of life variability among individuals with coeliac disease: an online survey

ObjectivePatients with coeliac disease (CD) need to follow a strict gluten-free diet to manage symptoms and prevent complications. Restrictions imposed by the diet can be challenging and affect quality of...

Deep Pattern Analysis for Cyberbullying Detection using Probabilistic Analytical Learning

Abstract: Cyberbullying has emerged as a significant issue in today's social media landscape, posing various detrimental effects. The combination of photo sharing and text comments has exacerbated the severity of cyberbullying incidents. Automated detection tools are crucial for ensuring the health and security of these platforms. However, traditional approaches that analyze text and images separately may fail toidentify all instances of cyberbullying, especially whenseemingly innocent content conveys bullying messages when posted together. This research proposes a novel system that extracts combined features from text and images to identify diverse cases of cyberbullying. The system can extract profiles based on behavior and uncover latent ties between users and groups with similar behaviors. Our approach utilizes methods log mining, business analysis, complex networks, and graph theory to achieve this. This paper outlines the entire process, from log file analysis to the construction of the user graph, witha particular focus on the step known as The finding of user behavioral patterns.

Healthcare professionals’ knowledge, understanding and confidence to manage chronic pain after cancer treatment: A UK survey

IntroductionHealthcare professionals are vital in preparing people living with and beyond cancer about the risks of chronic pain after cancer treatment. To do so, healthcare professionals need to be knowledgeable and confident about chronic pain after cancer treatment, yet little is known about their understanding or confidence of this common long-term and late side effect of cancer treatment. AimTo identify healthcare professionals’ knowledge and understanding of chronic pain after cancer treatment and consider how confident they are to inform, listen and signpost people living with and beyond cancer to appropriate information and support. MethodA cross sectional online survey was distributed to healthcare professionals in the UK via cancer and primary care networks, cancer alliances and social media. The survey consisted of four domains: 1) knowledge and understanding, 2) information and support, 3) confidence and 4) barriers. Quantitative data were analysed with descriptive statistics and free text comments were analysed using qualitative content analysis. ResultsHealthcare professionals reported limited knowledge and understanding of chronic pain after cancer treatment. Healthcare professionals lacked confidence to talk to people about chronic pain after cancer treatment and viewed their lack of knowledge as a barrier. Additional barriers included ‘Limited service provision’, ‘Conflict between services’, ‘Not my role’ and ‘Challenges in diagnosing chronic pain in cancer survivors’. ConclusionChronic pain after cancer can be a significant issue for those living with and beyond cancer, yet healthcare professionals report limited knowledge of it or understanding of the impact. More education is needed to increase healthcare professionals’ knowledge and confidence in chronic pain after cancer treatment.