Deep-learning-based stock market prediction incorporating ESG sentiment and technical indicators.

As sustainability emerges as a crucial factor in the development of modern enterprises, integrating environmental, social, and governance (ESG) information into financial assessments has become essential. ESG indicators serve as important metrics in evaluating a company's sustainable practices and governance effectiveness, influencing investor trust and future growth potential, ultimately affecting stock prices. This study proposes an innovative approach that combines ESG sentiment index extracted from news with technical indicators to predict the S&P 500 index. By utilizing a deep learning model and exploring optimal window sizes, the study explores the best model through mean absolute percentage error (MAPE) as an evaluation metric. Additionally, an ablation test clarifies the influence of ESG and its causality with the S&P 500 index. The experimental results demonstrate improved predictive accuracy when considering ESG sentiment compared to relying solely on technical indicators or historical data. This comprehensive methodology enhances the advantage of stock price prediction by integrating technical indicators, which consider short-term fluctuations, with ESG information, providing long-term effects. Furthermore, it offers valuable insights for investors and financial market experts, validating the necessity to consider ESG for financial assets and introducing a new perspective to develop investment strategies and decision-making processes.

Expansive data, extensive model: Investigating discussion topics around LLM through unsupervised machine learning in academic papers and news.

This study presents a comprehensive exploration of topic modeling methods tailored for large language model (LLM) using data obtained from Web of Science and LexisNexis from June 1, 2020, to December 31, 2023. The data collection process involved queries focusing on LLMs, including "Large language model," "LLM," and "ChatGPT." Various topic modeling approaches were evaluated based on performance metrics, including diversity and coherence. latent Dirichlet allocation (LDA), nonnegative matrix factorization (NMF), combined topic models (CTM), and bidirectional encoder representations from Transformers topic (BERTopic) were employed for performance evaluation. Evaluation metrics were computed across platforms, with BERTopic demonstrating superior performance in diversity and coherence across both LexisNexis and Web of Science. The experiment result reveals that news articles maintain a balanced coverage across various topics and mainly focus on efforts to utilize LLM in specialized domains. Conversely, research papers are more concise and concentrated on the technology itself, emphasizing technical aspects. Through the insights gained in this study, it becomes possible to investigate the future path and the challenges that LLMs should tackle. Additionally, they could offer considerable value to enterprises that utilize LLMs to deliver services.

ESG2PreEM: Automated ESG grade assessment framework using pre-trained ensemble models.

Incorporating environmental, social, and governance (ESG) criteria is essential for promoting sustainability in business and is considered a set of principles that can increase a firm's value. This research proposes a strategy using text-based automated techniques to rate ESG. For autonomous classification, data were collected from the news archive LexisNexis and classified as E, S, or G based on the ESG materials provided by the Refinitiv-Sustainable Leadership Monitor, which has over 450 metrics. In addition, Bidirectional Encoder Representations from Transformers (BERT), Robustly optimized BERT approach (RoBERTa), and A Lite BERT (ALBERT) models were trained to accurately categorize preprocessed ESG documents using a voting ensemble model, and their performances were measured. The accuracy of the ensemble model utilizing BERT and ALBERT was found to be 80.79% with batch size 20. Additionally, this research validated the performance of the framework for companies included in the Dow Jones Industrial Average (DJIA) and compared it with the grade provided by Morgan Stanley Capital International (MSCI), a globally renowned ESG rating agency known for having the highest creditworthiness. This study supports the use of sophisticated natural language processing (NLP) techniques to attain important knowledge from large amounts of text-based data to improve ESG assessment criteria established by different rating agencies.

Integrative transcriptomic and genomic analyses unveil the IFI16 variants and expression as MASLD progression markers.

BACKGROUND AND AIMS: Metabolic dysfunction-associated steatotic liver disease (MASLD) encompasses a broad and continuous spectrum of liver diseases ranging from fatty liver to steatohepatitis. The intricate interactions of genetic, epigenetic, and environmental factors in the development and progression of MASLD remain elusive. Here, we aimed to achieve an integrative understanding of the genomic and transcriptomic alterations throughout the progression of MASLD. APPROACH AND RESULTS: RNA-Seq profiling (n = 146) and whole-exome sequencing (n = 132) of MASLD liver tissue samples identified 3 transcriptomic subtypes (G1-G3) of MASLD, which were characterized by stepwise pathological and molecular progression of the disease. Macrophage-driven inflammatory activities were identified as a key feature for differentiating these subtypes. This subtype-discriminating macrophage interplay was significantly associated with both the expression and genetic variation of the dsDNA sensor IFI16 (rs6940, A>T, T779S), establishing it as a fundamental molecular factor in MASLD progression. The in vitro dsDNA-IFI16 binding experiments and structural modeling revealed that the IFI16 variant exhibited increased stability and stronger dsDNA binding affinity compared to the wild-type. Further downstream investigation suggested that the IFI16 variant exacerbated DNA sensing-mediated inflammatory signals through mitochondrial dysfunction-related signaling of the IFI16-PYCARD-CASP1 pathway. CONCLUSIONS: This study unveils a comprehensive understanding of MASLD progression through transcriptomic classification, highlighting the crucial roles of IFI16 variants. Targeting the IFI16-PYCARD-CASP1 pathway may pave the way for the development of novel diagnostics and therapeutics for MASLD.

Sentiment Impact of Public Health Agency communication Strategies on TikTok under COVID-19 Normalization: Deep Learning Exploration.

Aim: The accessibility of social media data has allowed researchers to measure official-public interactions during COVID-19. However, previous work analyzing official posts or public comments has failed to explore the link between the two. Therefore, this study investigates the relationship between the communication strategies of public health agencies (PHAs) on TikTok and public emotional/sentiment tendencies in COVID-19 normalization. Subject and methods: This study uses the 2022 Shanghai city closure event as a public health communication case study in the context of COVID-19 normalization, using TikTok as a data source. We first analyze the communication strategies adopted by the PHA based on the Crisis and Emergency Risk Communication (CERC) model. Then, we classify the sentiment of public comments using the Large-Scale Knowledge Enhanced Pre-Training for Language Understanding and Generation (ERNIE) pre-training model. Finally, we explore the connection between PHA communication strategies and public sentiment tendencies. Results: First, the public's sentiment tendencies differ at different stages. Therefore, appropriate communication strategies should be developed stage-by-stage. Second, the public's emotional disposition to different communication strategies varies: government statements, vaccines, and prevention and control programs are more likely to produce a friendly comment environment, while policy and new cases per day are more likely to produce unfavorable comment content. However, this does not mean that policy and new cases per day should be avoided; the judicious use of these two strategies can help PHAs understand the current issues causing public dissatisfaction. Third, videos with celebrity appearances can significantly increase positive public sentiment and, thereby, public participation. Conclusion: We propose an improved CERC guideline for China based on the Shanghai lockdown case.

I know my teammates: the role of Group Member Familiarity in Computer-Supported and face-to-face collaborative learning.

Considering the importance of group member familiarity in collaborative learning in classroom learning environments, this study examined the impact of group member familiarity on CSCL (computer-supported collaborative learning) in a networked setting. Also, the differences between CSCL in the online environments and FtF (face-to-face) collaborative learning were also compared. An analysis using structural equation modeling revealed that group member familiarity increased teamwork satisfaction, which in turn increased student engagement and perceived knowledge construction. A multi-group analysis revealed that while FtF collaborative learning exhibits higher levels of group member familiarity, teamwork satisfaction, student engagement, and perceived knowledge construction, the mediating effect of teamwork satisfaction was more pronounced in online settings. The study findings provided insights for teachers to improve collaborative learning experiences and adapt different teaching strategies.

Machine Learning Algorithm for Efficient Design of Separated Buffer Super-Junction IGBT.

An improved structure for an Insulated Gate Bipolar Transistor (IGBT) with a separated buffer layer is presented in order to improve the trade-off between the turn-off loss (Eoff) and on-state voltage (Von). However, it is difficult to set efficient parameters due to the increase in the new buffer doping concentration variable. Therefore, a machine learning (ML) algorithm is proposed as a solution. Compared to the conventional Technology Computer-Aided Design (TCAD) simulation tool, it is demonstrated that incorporating the ML algorithm into the device analysis could make it possible to achieve high accuracy and significantly shorten the simulation time. Specifically, utilizing the ML algorithm could achieve coefficients of determination (R2) of Von and Eoff of 0.995 and 0.968, respectively. In addition, it enables the optimized design to fit the target characteristics. In this study, the structure proposed for the trade-off improvement was targeted to obtain the minimum Eoff at the same Von, especially by adjusting the concentration of the separated buffer. We could improve Eoff by 36.2% by optimizing the structure, which was expected to be improved by 24.7% using the ML approach. In another way, it is possible to inversely design four types of structures with characteristics close to the target characteristics (Eoff = 1.64 µJ, Von = 1.38 V). The proposed method of incorporating machine learning into device analysis is expected to be very strategic, especially for power electronics analysis (where the transistor size is comparatively large and requires significant computation). In summary, we improved the trade-off using a separated buffer, and ML enabled optimization and a more precise design, as well as reverse engineering.

Effect of daily new cases of COVID-19 on public sentiment and concern: Deep learning-based sentiment classification and semantic network analysis.

Aim: This study explored the influence of daily new case videos posted by public health agencies (PHAs) on TikTok in the context of COVID-19 normalization, as well as public sentiment and concerns. Five different stages were used, based on the Crisis and Emergency Risk Communication model, amidst the 2022 Shanghai lockdown. Subject and Methods: After dividing the duration of the 2022 Shanghai lockdown into stages, we crawled all the user comments of videos posted by Healthy China on TikTok with the theme of daily new cases based on these five stages. Third, we constructed the pre-training model, ERNIE, to classify the sentiment of user comments. Finally, we performed semantic network analyses based on the sentiment classification results. Results: First, the high cost of fighting the epidemic during the 2022 Shanghai lockdown was why ordinary people were reluctant to cooperate with the anti-epidemic policy in the pre-crisis stage. Second, Shanghai unilaterally revised the definition of asymptomatic patients led to an escalation of risk levels and control conditions in other regions, ultimately affecting the lives and work of ordinary people in the area during the initial event stage. Third, the public reported specific details that affected their lives due to the long-term resistance to the epidemic in the maintenance stage. Fourth, the public became bored with videos regarding daily new cases in the resolution stage. Finally, the main reason for the negative public sentiment was that the local government did not follow the central government's anti-epidemic policy. Conclusion: Our results suggest that the methodology used in this study is feasible. Furthermore, our findings will help the Chinese government or PHAs improve the possible behaviors that displease the public in the anti-epidemic process.

How does online social interaction promote students' continuous learning intentions?

Learning from online videos using synchronized Danmu comments provides a co-learning experience. To explore the factors influencing learning with or without Danmu videos, an initial set of reasons and impediments was compiled based on a pilot study of 24 Chinese university students with learning experience using Danmu videos. Three hundred such students were surveyed to determine the factors that influence their motivations and hindrances with regard to using Danmu videos. The potential predictors of users' continuance intentions were also examined. The results showed that the frequency of using Danmu videos is related to the continuous learning intention. Seeking information, social presence, and perceived entertainment motivate learners and positively impact their intention to continue learning using Danmu videos. Hindrances such as information pollution, attention failure, and visual obstacle were found to be negatively associated with learners' continuance intention in the long run. Our findings provided constructive suggestions for addressing the issue of dropout rates, and novel ideas for future studies were proposed.

Correction to: Effect of daily new cases of COVID-19 on public sentiment and concern: Deep learning-based sentiment classification and semantic network analysis.

[This corrects the article DOI: 10.1007/s10389-023-01833-4.].

Treatment outcomes of oral doxycycline versus intravenous azithromycin in adults hospitalized with scrub typhus: A retrospective study using inverse probability treatment weighting (IPTW) propensity analysis.

OBJECTIVE: Only a few well-designed studies that have investigated the effectiveness of azithromycin in treating adult patients hospitalized with scrub typhus are currently available. The purpose of our study was to compare the effects of intravenous azithromycin administration with those of oral doxycycline, and to evaluate cardiovascular death associated with intravenous azithromycin in adult patients hospitalized with scrub typhus. METHODS: This retrospective study investigated Korean National Infectious Disease Cohort Collaborative-registered scrub typhus-infected patients who were hospitalized between January 1, 2013, and December 31, 2021, and who were ≥18 years old. The primary outcome was time to fever clearance and the secondary outcomes were treatment failure, relapse, scrub typhus-related death, or azithromycin-related cardiovascular death. To address any indication bias, inverse probability of treatment weighting (IPTW) analysis was performed. Times to fever clearance between the doxycycline and azithromycin groups were compared using log-rank tests and Kaplan-Meier curves. RESULTS: A total of 326 consecutive patients with laboratory-confirmed scrub typhus were included in this study of whom 109 were treated with azithromycin and 217 with doxycycline. Using IPTW, there were no statistically significant differences in the following end points between the azithromycin and doxycycline groups: median time to fever clearance (3 days vs. 3 days, P = 0.649), treatment failure (0.71% vs. 0.42%, P = 0.702), relapse (0.0% vs. 0.0%), and scrub typhus-related death (5.12% vs. 0.0%, P = 0.155). No azithromycin-related cardiovascular deaths occurred. In the sensitivity analyses, there were no significant changes in effect size. CONCLUSIONS: Our study showed that the therapeutic effects and safety of intravenous azithromycin are comparable to those of oral doxycycline administration in patients hospitalized with scrub typhus. A well-designed randomized controlled trial may help further evaluate the most adequate route of administration, dose and duration of treatment with azithromycin.

How public health agencies communicate with the public on TikTok under the normalization of COVID-19: A case of 2022 Shanghai's outbreak.

Objective: As life with COVID-19 became a norm in 2022, the public's demand for and perception of COVID-19-related information has changed. This study analyzed the performance and responses of Healthy China and the public at various stages of COVID-19 normalization using the crisis and emergency risk communication (CERC) theory. Methods: This study was based on the 2022 Shanghai COVID-19 outbreak and data from "Healthy China," the official TikTok account of the National Health Commission of the People's Republic of China (NHCC). First, we divided the Shanghai lockdown into five stages in accordance with the CERC. Second, the videos released by Healthy China were open-coded. Third, to understand the distribution of strategies across the stages, we used counts and percentages to summarize the categorical variables. Fourth, we investigated the distribution of public participation indicators using descriptive statistical analysis. Finally, the relationship between stage and communication strategy was examined using the chi-square test and negative binomial regression. Results: (1) Healthy China adopted a more flexible approach to communication strategies; (2) new cases per day was the commonly used substrategy for uncertainty reduction; (3) there was a significant difference in the strategies used by Healthy China at different stages; (4) public participation was highest in the pre-crisis period; and (5) the stage had a significant positive impact on the number of views, favorites, likes, and shares. Conclusions: This research provides insight into effective communication strategies for the government or public health agencies to employ during COVID-19 normalization.

Changes in Sleep Problems in Patients Who Underwent Surgical Treatment for Degenerative Spinal Disease with a Concurrent Sleep Disorder: A Nationwide Cohort Study in 3183 Patients during a Two-Year Perioperative Period.

Sleep disturbance is prevalent in patients with degenerative spinal disease, and recent studies have reported that surgical treatment is more effective for improving sleep quality than conservative treatment. We aimed to investigate the perioperative changes of sleep problems in patients who underwent surgical treatment for degenerative spinal disease with a concurrent sleep disorder, and presented them according to various clinical profiles possibly associated with sleep disturbance. In addition, we identified factors associated with poor sleep improvement after surgery. This study used data from the Korea Health Insurance Review and Assessment Service database from 2016 to 2018. We included 3183 patients aged ≥19 years who underwent surgery for degenerative spinal disease and had a concurrent sleep disorder. Perioperative changes in the two target outcomes, including the use of sleep medication and hospital visits owing to sleep disorders, were precisely investigated according to factors known to be associated with sleep disturbance, including demographics, comorbidities, and spinal regions. Logistic regression analysis was performed to identify factors associated with poor improvement in terms of sleep medication after surgery. All estimates were validated using bootstrap sampling. During the 1-year preoperative period, the use of sleep medications and hospital visits owing to sleep disorder increased continuously. However, they abruptly decreased shortly after surgical treatment, and throughout the 1-year postoperative period, they remained lower than those in the late preoperative period. At the 1-year follow-up, 75.6% (2407 of 3183) of our cohort showed improvement in sleep medication after surgery. Multivariable analysis identified only two variables as significant factors associated with non-improvement in sleep medication after surgery: depressive disorder (odds ratio (OR) = 1.25 [1.06-1.48]; p = 0.008), and migraine (OR = 1.42 [1.04-1.94]; p = 0.028). We could not investigate the actual sleep quality and resultant quality of life; however, our results justify the necessity for further high-quality studies that include such information and would arouse clinicians' attention to the importance of sleep disturbance in patients with degenerative spinal disease.

Effects of postpolymerization conditions on the physical properties, cytotoxicity, and dimensional accuracy of a 3D-printed dental restorative material.

STATEMENT OF PROBLEM: Although the introduction of high-speed 3-dimensional (3D) printing technology has significantly reduced printing time, the time required for postpolymerization is a speed-determining step because of the long wait time. How postpolymerization conditions affect material properties is unclear. PURPOSE: The purpose of this in vitro study was to assess the physical properties, accuracy, and biosafety of a 3D-printed dental restorative material according to postpolymerization conditions. MATERIAL AND METHODS: Specimens were prepared by 3D printing with a digital light processing 3D printer with 1 interim dental material (C&B MFH). All printed specimens underwent a postpolymerization process with 5 different postpolymerization devices and were designated as groups D1 (D102H), FO (Form Cure), LC (LC-3DPrintBox), ME (Medusa), and MP (MP100). The light intensity and temperature of each device were measured, and the Vickers hardness, flexural strength and modulus, degree of conversion (DC), cytotoxicity, and polymerization shrinkage were analyzed. Statistical analyses were conducted with 1-way analysis of variance, the Tukey post hoc test, and regression testing (α=.05). Scanning electron microscopy was used to assess the fracture surface characteristics of the specimens. RESULTS: Light intensity was strongest with the ME device, and the temperature inside the device during postpolymerization showed the highest increase with the LC device and the lowest increase with the D1 device. The LC group specimens showed the highest mean Vickers hardness, and the MP group showed the lowest. The flexural strength was ≥100 MPa in all groups, with a flexural modulus ranging from 1.17 to 1.5 GPa. The DC results were similar to the physical properties test results. The D1, FO, LC, and ME groups all showed ≥70% cell viability, indicating no toxicity. The FO group showed the highest shrinkage rate of 0.52%. CONCLUSIONS: When the light intensity was strong, the surface was sufficiently hard, and toxic substances were not eluted even after a short postpolymerization time, suggesting that light intensity modulation and time management can be used to improve the postpolymerization process.

Factors Affecting Intention of Consumers in Using Face Recognition Payment in Offline Markets: An Acceptance Model for Future Payment Service.

Face recognition payment (FRP), an innovative financial technology service, is a recently developed mode of payment service that has garnered attention in the offline market, particularly in China. However, studies examining the adoption of FRP by consumers are scarce. Therefore, this study proposed a causal model built on the Unified Theory of Acceptance and Use of Technology, and key predictors related to the intention of using FRP were identified. The structural equation model-based results obtained from 305 Chinese participants demonstrated that the intention was most affected by relative advantage. In addition, performance expectancy, effort expectancy, social influence, and perceived risk also had a significant impact. However, trust was found to not significantly affect consumers' intentions, despite it negatively influencing perceived risk. Thus, the results of this study are expected to provide a set of guidelines for companies regarding the implementation of FRP.

A Visualization Analysis of Crisis and Risk Communication Research Using CiteSpace.

This study aims to understand the research status and development trend of crisis and risk communication research (CRCR) through a visual analysis in CiteSpace, thereby providing a more comprehensive perspective for future research agenda. First, we retrieved published papers from Web of Science (1986-2020) and Scopus (1979-2020) with a title search. Subsequently, we analyzed the main research strengths and main topics of CRCR from two dimensions: co-authorship network and co-citation network. We conducted an in-depth co-citation network analysis from four perspectives: cluster analysis, high co-cited literature analysis, burst analysis, and turning points analysis. These results revealed the main research topics in the CRCR field, the most eye-catching research literature, the emerging research hotspots in each period, and the turning points of the overall development. Finally, we suggested further research directions for future avenues.

Examining Crisis Communication Using Semantic Network and Sentiment Analysis: A Case Study on NetEase Games.

The mobile game "Immortal Conquest," created by NetEase Games, caused a dramatic user dissatisfaction event after an introduction of a sudden and uninvited "pay-to-win" update. As a result, many players filed grievances against NetEase in a court. The official game website issued three apologies, with mix results, to mitigate the crisis. The goal of the present study is to understand user feedback content from the perspective of Situational Crisis Communication Theory through semantic network analysis and sentiment analysis to explore how an enterprise's crisis communication strategy affects users' attitudes. First, our results demonstrate that the diminishing crisis communication strategies (excuse and justification) do not change players' negative attitudes. It was not a failure because it successfully alleviated the players' legal complaints and refocused their attention on the game itself. Second, the rebuild (apology & compensation) strategy was effective because it significantly increased the percentage of positive emotions and regenerated expectations for the game. The litigation crisis was identified within gamer communications with respect to Chinese gaming companies for the first time. Nevertheless, this does not indicate an increase in overall legal awareness among the larger Chinese population. It may only reflect greater legal awareness among Chinese online gamers. Fourth, gamers emphasized that they and enterprises should be equally involved when communicating with each other. Finally, in-game paid items should be reasonably priced, otherwise, they will drive users to competitors.

My video game console is so cool! A coolness theory-based model for intention to use video game consoles.

With the outbreak of COVID-19, the video game console market is thriving again. In this study, we attempted to explore users' intention to use video game consoles by developing a causal model mainly based on coolness theory and the technology acceptance model. To better illustrate user experience for video game consoles, we added several concepts to the causal model, including hedonic motivation, system and service quality, perceived cost, and game variety. Through examining survey-based data from 360 Koreans, we discovered that the model had a high explanatory power for users' intention to use video game consoles. The key findings were as follows: First, among the components of coolness theory, individuals' attitude toward consoles was significantly related to subcultural appeal and originality, but not to attractiveness. Second, originality positively influenced subcultural appeal significantly. Overall, this study implied that the novel coolness theory is effective for exploring user experience regarding of specific devices and services.

Dental students' perceptions on a simulated practice using patient-based customised typodonts during the transition from preclinical to clinical education.

PURPOSE/OBJECTIVES: Dental students experience difficulties during the transition from preclinical to clinical curriculum. In order to help the students to adapt to the clinical education programme, a simulated practice using patient-based customised models was introduced in this study to prepare for their first clinical practice. METHODS: This study included 45 third-year predoctoral students (D3 students) who were about to perform the preparation of a single crown abutment on their first patient. After practicing abutment preparation using simulated models and providing the actual treatment to their own patient, the students were surveyed to investigate their perceptions on the simulated practice using the 3D-printed customised typodont model. The statistical analysis of the quantitative data and the thematic analysis of the qualitative data were conducted. RESULTS: Regarding this simulation, more than 80% of the students gave positive feedback on their practice of (a) operative positions and postures, (b) finger rest, (c) occlusal reduction, (d) axial reduction and (e) proximal reduction. Student responses on the open-ended questions about how they perceived the usefulness of this simulation were categorised as "First clinical case," "Patient-based model" and "Realistic simulation environment." In addition, a number of improvements of the simulation were also suggested by the students including the typodont and the manikin. CONCLUSIONS: This study gives insights into the significance of simulated practice using patient-based customised typodonts as a transitional education tool and its direction of development in the field of restorative treatments accompanied by irreversible tooth preparations.

Improvement in Turn-Off Loss of the Super Junction IGBT with Separated n-Buffer Layers.

In this study, we propose a super junction insulated-gate bipolar transistor (SJBT) with separated n-buffer layers to solve a relatively long time for carrier annihilation during turn-off. This proposition improves the turn-off characteristic while maintaining similar on-state characteristics and breakdown voltage. The electrical characteristics of the devices were simulated by using the Synopsys Sentaurus technology computer-aided design (TCAD) simulation tool, and we compared the conventional SJBT with SJBT with separated n-buffer layers. The simulation tool result shows that turn-off loss (Eoff) drops by about 7% when on-state voltage (Von) and breakdown voltage (BV) are similar. Von increases by about 0.5% and BV decreases by only about 0.8%.