Explore the vast range of titles available.

This study investigates the application of large language models (LLMs) in sentiment analysis of financial news and their use in developing effective investment strategies. We conducted sentiment analysis on news articles related to the top 30 companies listed on Nasdaq using both discriminative models such as BERT and FinBERT, and generative models including Llama 3.1, Mistral, and Gemma 2. To enhance the robustness of the analysis, advanced prompting techniques—such as Chain of Thought (CoT), Super In-Context Learning (SuperICL), and Bootstrapping—were applied to generative LLMs. The results demonstrate that long strategies generally yield superior portfolio performance compared to short and long–short strategies. Notably, generative LLMs outperformed discriminative models in this context. We also found that the application of SuperICL to generative LLMs led to significant performance improvements, with further enhancements noted when both SuperICL and Bootstrapping were applied together. These findings highlight the profitability and stability of the proposed approach. Additionally, this study examines the explainability of LLMs by identifying critical data considerations and potential risks associated with their use. The research highlights the potential of integrating LLMs into financial strategy development to provide a data-driven foundation for informed decision-making in financial markets.

As an extension of the transformer architecture, the BERT model has introduced a new paradigm for natural language processing, achieving impressive results in various downstream tasks. However, high-performance BERT-based models—such as ELECTRA, ALBERT, and RoBERTa—suffer from limitations such as poor continuous learning capability and insufficient understanding of domain-specific documents. To address these issues, we propose the use of an attention mechanism to combine BERT-based models with neural topic models. Unlike traditional stochastic topic modeling, neural topic modeling employs artificial neural networks to learn topic representations. Furthermore, neural topic models can be integrated with other neural models and trained to identify latent variables in documents, thereby enabling BERT-based models to sufficiently comprehend the contexts of specific fields. We conducted experiments on three datasets—Movie Review Dataset (MRD), 20Newsgroups, and YELP—to evaluate our model’s performance. Compared to the vanilla model, the proposed model achieved an accuracy improvement of 1–2% for the ALBERT model in multiclassification tasks across all three datasets, while the ELECTRA model showed an accuracy improvement of less than 1%.

In today’s world, artificial intelligence (AI) and human intelligence coexist, and no field is free from the impact of AI. At present, education cannot be discussed without mentioning AI, which has an omnidirectional impact on all its areas, including the purpose, content, method, and evaluation system. This study aimed to explore the future direction of education by examining the current impact and predicting future impacts of AI. It also examined research trends and collaboration status by country through network analysis, topic modeling and global research trends in AI in education (AIED), by applying the Latent Dirichlet Allocation algorithm. Over the past 20 years, the number of papers on AIED has steadily increased, with a dramatic rise since 2015. The research can be broadly classified into eight topics, including “changes in the content of teaching and learning.” Using a linear regression model, three hot topics, two cold topics and trend changes for each research topic were identified. The study found that AIED research should be more thematically diversified and in-depth; this directly applies AI algorithms and technologies to education, which should be further promoted. This study provides a reference for exploring the direction of future AIED research.

Recently, there has been growing educational interest in competency. Global organizations, such as the United Nations (UN) and Organization for Economic Co-operation and Development (OECD), which are leading the discourse on education reform, are undertaking the lead in spreading awareness regarding competency education. Since 2015, the number of published articles on competency education has been rapidly increasing. This paper aims to provide significant implications for creating a sustainable future of competency education. A topic modeling method was used to empirically analyze latent topics and international research trends in 26,532 articles published on competency-based education (CBE). As a result of the analysis, 15 topics were derived, including “approach to competency development.” In addition, five topics including “learning skills” and “teacher training” were found to be hot topics with the increasing article publication. The rapidly changing modern society is calling for a transformation in education. We hope that the results of this study paves the way for further research exploring new directions for education, such as competency education.

Few studies have assessed outcomes associated with the use of traditional medicine therapies to manage infertility in clinical practice. The aim of this study was to investigate the clinical characteristics of and infertility treatment effects among patients who visited a Korean medicine (KM) clinic to aid in achieving pregnancy. This study consisted of a 10-year analysis of patient records from a KM clinic. A retrospective 10-year (2010-2020) chart analysis was performed using the medical records of infertile patients who visited a KM clinic in South Korea for fertility treatment (ICD-10, infertility symptoms: 59 codes). Of the 6194 patients who visited the clinic during the selected time frame, 1786 were female patients seeking fertility treatment to achieve pregnancy. Among the 1786 infertile women, 586 women succeeded in becoming pregnant (32.8%). Among the 586 patients who became pregnant, 476 women had received KM, 92 had been treated using KM and in vitro fertilization (IVF), and 18 had received KM and undergone intrauterine insemination (IUI). The live birth rates achieved with these treatments were 66.0%, 68.8%, and 66.7%, respectively. The most frequently prescribed medicines were Gamiboher-tang (Jiaweiwuxu-tang), Gamiguibi-tang (Jiaweiguipi-tang), and Gamidanggui-san (Jiaweidanggui-san). Additionally, the most frequent adjunct therapies administered to these patients were acupuncture and moxibustion. Infertility therapies using KM may be a successful option to treat infertility when used alone or in addition to IVF and IUI. However, further pharmacological investigations and clinical trials are required to ensure the objectivity of the efficacy evaluation.

Multiplayer Online Battle Arena (MOBA) games, exemplified by titles such as League of Legends and Dota 2, have attained global popularity and have been formally recognized as an official event in the 2022 Hangzhou Asian Games, thus establishing their significance in the esports industry. In this study, we proposed a machine learning-based model for predicting comeback victories by leveraging the object bounty mechanism, a critical yet underexplored aspect of previous research. By closely examining the game environment following the activation of the bounty system, we identified pivotal variables and constructed novel indicators that contribute to successful comebacks. Furthermore, an individualized case analysis based on SHapley Additive exPlanations (SHAP) provides new insights to support strategic in-game decision-making and enhance the player experience. The experimental results demonstrate that the indicators introduced in this study, such as the weighted team champion mastery and similarity in champion mastery among the team’s main champions, significantly influence the likelihood of a comeback victory. By capturing the intrinsic dynamism of MOBA games, the proposed model is expected to improve player engagement and satisfaction.

This study aimed to develop an automated computer scoring system (ACSS) incorporating a Korean spell checker to assess students’ constructed responses and to check the efficacy of this system. To accomplish this, we examined the performance of automatic spelling correction in reporting and correcting spelling errors, the interaction of gender and grade level in making spelling errors, the relationship between spelling errors and academic achievement, and the scoring efficacy of an ACSS that incorporated a spell checker. The analysis of percentage, two-way ANOVA, t-test, Pearson’s correlation, and human–computer correspondence were conducted. The results revealed that an automatic spelling correction system could report 66.44% and correct 26.78% of all total misspelled words. We also found gender and grade-level differences in misspelling words. Students misspelled fewer words as they advanced in grade level, and male students misspelled more words than females. In terms of the relationship between spelling errors and concepts, we found that the number of concepts included in student’s responses had a significant relationship with the total number of written words and misspelled words. This indicates that students who made more spelling errors had discussed more concepts in their responses. Based on these results, we discuss practical implications for preventing students’ responses being scored lower due to spelling errors caused by being less attentive using an ACSS with a spelling correction system.

Predicting cross-sectional stock returns is challenging due to low signal-to-noise ratios and evolving market regimes. Classical factor models offer interpretability but limited flexibility, while deep learning models achieve strong performance yet often underutilize financial priors. We address this gap with PRISM-VQ (PRior-Informed Stock Model with Vector Quantization), a dynamic factor framework that integrates expert prior factors, vector-quantized discrete latent factors learned from cross-sectional structure, and a structure-conditioned Mixture-of-Experts to generate time-varying factor loadings. Vector quantization acts as an information bottleneck that suppresses noise while capturing robust market structure, with discrete codes serving both as latent factors and as routing signals for temporal expert specialization. Experiments on CSI 300 and S&P 500 show consistent improvements in cross-sectional return prediction and portfolio performance over strong baselines while preserving interpretability. Our code is available at https://github.com/finxlab/PRISM-VQ.

A Markov-chain Monte Carlo sampling algorithm samples a new point around the latest sample due to the Markov property, which prevents it from sampling from multi-modal distributions since the corresponding chain often fails to search entire support of the target distribution. In this paper, to overcome this problem, mode switching scheme is applied to the conventional MCMC algorithms. The algorithm separates the reducible Markov chain into several mutually exclusive classes and use mode switching scheme to increase mixing rate. Simulation results are given to illustrate the algorithm with promising results. KCI Citation Count: 0