Read the abstract, see who wrote it and who cites it, and get the citation in APA, MLA, Chicago or Harvard.
About 10,000 papers for “topic distribution”
Huzeyfe Ayaz (20114109), Muhammed Hasan Celik (13210148), Huseyin Zeyd Koytak (13210145) et al. · Figshare · 2024
<p>The topic distribution of Twitter posts of users who posted multiple tweets. Each diagram represents a group of users interested in a specific topic and illustrates how their other tweets are distributed.</p>
Cited by 0Open access
Nor Ilyani Mohamed Nazar (22008724), Norny Syafinaz Ab Rahman (22008727), Nor Elina Alias (22008730) et al. · Figshare · 2025
<div> The escalating global burden of chronic diseases necessitates innovative approaches to enhance medication adherence and quality use of medicines (QUM), particularly in underserved rural populations. This study developed and evaluated <i>Impak Sihat</i>, a telehealth system tailored to address systemic healthcare barriers in rural Malaysia through a three-phase mixed-methods design. Phase 1 involved qualitative interviews with 15 villagers, revealing smartphone ownership, inconsistent internet connectivity, high social media engagement, and limited critical appraisal of online health information. Phase 2 utilised these insights to create a dual-component system: a public portal with Malay-language educational materials, appointment booking, and a practitioner platform featuring secured patient data management. Phase 3 assessed feasibility via community demonstrations with 77 participants (mean age 53.4 ± 11.8 years), showing high acceptance scores (73–87%) across six domains. Key findings included strong usability (87.0 ± 16.3) and interface design (74.8 ± 23.9), though older adults scored significantly lower on interface design for learnability (ρ=−0.29, <i>p</i> < 0.01). The system’s offline functionality and WhatsApp integration mitigated rural connectivity constraints, yet challenges persisted in data confidentiality (lowest score: 73.1 ± 26.7). Healthy participants consistently rated the system significantly higher across multiple domains (Interface Design: p = 0.003, User Experience: p = 0.018, Healthcare Delivery: p = 0.002, and Overall Satisfaction: p = 0.003). These results underscore the potential of context-specific telehealth systems to bridge urban-rural health disparities while highlighting critical implementation barriers. This work highlights the importance of engaging key stakeholders, such as healthcare providers and community leaders, to ensure system sustainability and scalability. Overall, the study demonstrates that digital health interventions, when appropriately tailored to the specific needs of rural populations, can significantly contribute to reducing healthcare disparities and promoting patient empowerment. </div>
Cited by 0Open access
Heini Utunen (19146725), Giselle Balaciano (6801320), Elham Arabi (19146728) et al. · Figshare · 2024
<div><p>Background</p><p>Keeping the health workforce and the public informed about the latest evolving health information during a health emergency is critical to preventing, detecting and responding to infectious disease outbreaks or other health emergencies. Having a well-informed, ready, willing, and skilled workforce and an informed public can help save lives, reduce diseases and suffering, and minimize socio-economic loss in affected communities and countries. Providing “just in time” support and opportunities for learning in health emergencies is much needed for capacity building. In this paper, ‘learning intervention’ refers to the provision of ad-hoc, focused, or personalized training sessions with the goal of preparing the health workers for emergencies or filling specific knowledge or skill gaps. We refer to ‘training methods’ as instructional design strategies used to teach someone the necessary knowledge and skills to perform a task.</p><p>Methods</p><p>We conducted a scoping review to map and better understand what learning interventions and training methods have been used in different types of health emergencies and by whom. Studies were identified using six databases (Pubmed/Medline, Embase, Hinari, WorldCat, CABI and Web of Science) and by consulting with experts. Characteristics of studies were mapped and displayed and major topic areas were identified.</p><p>Results</p><p>Of the 319 records that were included, contexts most frequently covered were COVID-19, disasters in general, Ebola and wars. Four prominent topic areas were identified: 1) Knowledge acquisition, 2) Emergency plans, 3) Impact of the learning intervention, and 4) Training methods. Much of the evidence was based on observational methods with few trials, which likely reflects the unique context of each health emergency. Evolution of methods was apparent, particularly in virtual learning. Learning during health emergencies appeared to improve knowledge, general management of the situation, quality of life of both trainers and affected population, satisfaction and clinical outcomes.</p><p>Conclusion</p><p>This is the first scoping review to map the evidence, which serves as a first step in developing urgently needed global guidance to further improve the quality and reach of learning interventions and training methods in this context.</p></div>
Cited by 0Open access
Zhuoyuan Tang (17972847), Xuan Pan (187031), Zhouyi Gu (17972850) · Figshare · 2024
<p>The figure displays the distribution of secondary topics within a two-dimensional scaling space.</p>
Cited by 0Open access
Zhuoyuan Tang (17972847), Xuan Pan (187031), Zhouyi Gu (17972850) · Figshare · 2024
<p>The figure displays the distribution of primary topics within a two-dimensional scaling space.</p>
Cited by 0Open access
Podong Song (20390102), Wiseong Jin (18271694), Shaowei Chen (1391854) et al. · Figshare · 2024
<div><p>Cyberspace is emerging as a critical living environment, significantly influencing sustainable human development. Internet public opinion is a crucial aspect of cyberspace governance, serving as the most important form of expressing popular will. However, perceiving public opinion can be challenging due to its complex and elusive nature. In this paper, we propose a novel framework for perceiving popular will, managing public opinion, and influencing people’s behavior, based on machine learning and game theory approaches. Our framework leverages deep learning techniques to analyze public opinion, active learning methods to reduce costs, and game theory to make optimal management decisions. We verify the effectiveness of our framework using empirical data collected from Chinese provinces Y and G, and provide theoretical support by analyzing the interrelationship between public opinion, online public opinion, and people’s behavior. Our framework can be applied inexpensively to studies in other regions, thereby offering valuable insights into cyberspace governance and public opinion management.</p></div>
Cited by 0Open access
Wei Yang (109917), Jiayin Pei (21799598), Zheyu Lin (10904295) · Figshare · 2025
<div> In recent years, the frequent occurrence of food safety incidents has posed a serious threat to people’s lives and properties, and triggered heated discussions in the whole society. If the relevant departments shirk each other’s responsibilities or deceive the public in handling the incident, it will further lead to the generation of network public opinion, which will have an impact on the stability of the society and the credibility of the government. In order to study the generation of food safety network public opinion, this article takes the sudden “mouse head and duck neck” incident in a university in Jiangxi Province, China as an example, and combines text mining with grounded theory. Firstly, the LDA topic clustering model is used to identify six main concerns (topics) of the public during the public opinion period, and the topic words and words with high TF-IDF values under each topic is counted. Based on this, the grounded theory method is used for three-level coding, and then a network public opinion generation model is constructed. It was found that the four main categories of national subjects, public’s responses, value orientations, and news media play an important role in the generation process of network public opinion. These findings not only provide a reference for the governance of food safety public opinion in China, but also shed light on public opinion management in other countries, especially when responding to food safety incidents of general significance. </div>
Cited by 0Open access
Juan Zhao (109151), QiPing Feng (5692676), Patrick Wu (6331307) et al. · Figshare · 2019
<p>To visualize the prevalence of each topic in the cohort, we assigned an individual to the topic with the maximum score.</p>
Cited by 0Open access
Wiseong Jin (18271694), Kwisik Min (18271697), Xufang Hu (13105013) et al. · Figshare · 2024
<div><p>The rural B&B industry is a key component of rural tourism, local economic development, and the wider rural revitalization strategy. Despite the abundance of tourism resources in Yunnan, the B&B sector faces significant challenges. It is therefore imperative to accurately identify the most pressing issues within the current B&B industry and formulate appropriate solutions to advance Yunnan’s rural revitalization efforts. This study uses recent reviews of rural B&Bs on Ctrip.com and employs machine learning techniques, including Bert, CNN, LSTM, and GRU, to identify the key management challenges currently facing Yunnan’s rural B&B industry. An analysis is then conducted to identify the key stakeholders involved in the process of improving the management of Yunnan’s B&Bs. To assess the willingness of each stakeholder to support the improvement of the rural B&B industry, this paper establishes a three-party evolutionary game model and examines the dynamic evolutionary process of management improvement within Yunnan’s rural B&B industry. Two scenarios of evolutionarily stable strategies are analyzed, and parameters impacting stakeholders’ strategy choices are simulated and evaluated. The results show that: i) Improving the "human factor" is the top priority for the current management improvement because tourists are most concerned about the emotional experience. Operators need to focus on improving service attitude and emotional experience; ii) The main stakeholders in the current management optimization process of Yunnan B&Bs are the local government, B&B operators, and tourists. Under appropriate conditions, the evolutionarily stable strategy of (1, 1, 1) is reachable. iii) variables such as additional costs, tourists’ choice preferences, and government penalties significantly affect the strategy choices of stakeholders, especially B&B operators. This paper offers effective strategies for improving B&B management that can benefit the government, B&B operators, and tourists, and ultimately contribute to the promotion of quality rural revitalization. The paper not only identifies focal areas for improving B&B management in rural Yunnan, but also provides an in-depth understanding of stakeholder dynamics. As a result, it provides valuable insights to further the cause of quality rural revitalization.</p></div>
Cited by 0Open access
Podong Song (20390102), Wiseong Jin (18271694), Shaowei Chen (1391854) et al. · Figshare · 2024
<div><p>Cyberspace is emerging as a critical living environment, significantly influencing sustainable human development. Internet public opinion is a crucial aspect of cyberspace governance, serving as the most important form of expressing popular will. However, perceiving public opinion can be challenging due to its complex and elusive nature. In this paper, we propose a novel framework for perceiving popular will, managing public opinion, and influencing people’s behavior, based on machine learning and game theory approaches. Our framework leverages deep learning techniques to analyze public opinion, active learning methods to reduce costs, and game theory to make optimal management decisions. We verify the effectiveness of our framework using empirical data collected from Chinese provinces Y and G, and provide theoretical support by analyzing the interrelationship between public opinion, online public opinion, and people’s behavior. Our framework can be applied inexpensively to studies in other regions, thereby offering valuable insights into cyberspace governance and public opinion management.</p></div>
Cited by 0Open access
John Dagdelen (14552112), Amalie Trewartha (11710181), Haoyan Huo (6883787) et al. · Figshare · 2023
<p>Fraction of papers published each month belonging to each of the 10 LDA topics over the course of the pandemic. Lines are smoothed using the 2-month moving average.</p>
Cited by 0Open access
Ning Chen (127610), Yun Liu (84346), Yijie Cheng (796160) et al. · Figshare · 2015
Cited by 0Open access
Jaeah You, Youngjoong Ko · 2023
Dealing with multiple topics should be considered an important issue in dialogue summarization, because dialogues, unlike documents, are prone to topic drift. Thus, we propose a new dialogue summarization model that reflects dialogue topic distribution to consider all topics present in the dialogue. First, the distribution of dialogue topics is estimated by an effective topic discovery model. Then topic-informed prompt transfers estimated topic distribution information to the output of encoder and decoder vectors. Finally, the topic extractor estimates the summary topic distribution from the output context vector of decoder to distinguish its difference from the dialogue topic distribution. To consider the proportion of each topic distribution appeared in the dialogue, the extractor is trained to reduce the difference between the distributions of the dialogue and the summary. The experimental results on SAMSum and DialogSum show that our model outperforms state-of-the-art methods on ROUGE scores. The human evaluation results also show that our framework well generates comprehensive summaries.
Cited by 3Open access
Association for Computational Linguistics 2023, Ko, Youngjoong, You, Jaeah · Underline Science Inc. · 2023
Dealing with multiple topics should be considered an important issue in dialogue summarization, because dialogues, unlike documents, are prone to topic drift. Thus, we propose a new dialogue summarization model that reflects dialogue topic distribution to consider all topics present in the dialogue. First, the distribution of dialogue topics is estimated by an effective topic discovery model. Then topic-informed prompt transfers estimated topic distribution information to the output of encoder and decoder vectors. Finally, the topic extractor estimates the summary topic distribution from the output context vector of decoder to distinguish its difference from the dialogue topic distribution. To consider the proportion of each topic distribution appeared in the dialogue, the extractor is trained to reduce the difference between the distributions of the dialogue and the summary. The experimental results on SAMSum and DialogSum show that our model outperforms state-of-the-art methods on ROUGE scores. The human evaluation results also show that our framework well generates comprehensive summaries.
Cited by 0Open access
J. A. Sim, Namgyu Kim · Journal of Intelligence and Information Systems · 2024
최근 ERP 도입 기업의 증가와 함께 ERP 유지보수 업무도 활발히 진행되고 있으며, 이에 따라 ERP 사용 고객의 의견을 분석해 고객 대응 전략을 수립하고자 하는 연구에 기업의 관심이 높아지고 있다. 고객 의견은 대부분 텍스트 데이터의 형태로 존재하며, 이에 대한 효과적인 분석을 위해 텍스트 마이닝 기법 중 하나인 토픽 모델링이 널리 활용되고 있다. 다양한 토픽 모델링 기법 중 작성자 정보를 활용하여 토픽을 생성함으로써 작성자의 관심분야 식별에 유용한 장점을 가진 Author-Topic Model(ATM)에 대한 관심이 높지만, 해당 방법은 지나치게 많은 수의 작성자로 인해 작성자간 차이를 해석하기가 어렵다는 한계를 갖는다. 따라서 본 연구는 ATM을 기반으로 도출된 작성자별 토픽 분포에 데이터의 계층 정보를 반영하는 계층적 토픽 분포 집계를 수행하고, 이를 통해 산출된 집단의 토픽 분포를 활용하여 집단간 토픽을 비교하는 새로운 방법론을 제안한다. 제안 방법론의 유용성을 평가하기 위해 SAP ERP 전문 컨설팅 기업이 유지보수 서비스를 제공하는 4개 회사의 고객 서비스 요청 데이터 1,903건을 적용해 실험을 진행하였다. 구체적으로는 요청자별 ATM 분석을 수행하고, 그 결과에 상위 수준 집단과 하위 수준 작성자의 계층 구조를 반영하여 고객사별로 토픽 분포를 집계한 후 고객사별 토픽의 차이를 비교하였다. 실험 결과, 작성자 수준에서는 확인이 어려웠던 집단간 차이를 본 방법론을 통해 명확히 파악할 수 있음을 확인하였다.
Cited by 0Open access
Xiaobao Wu, Chunping Li, Yan Zhu et al. · 2020
Topic models have been prevailing for many years on discovering latent semantics while modeling long documents. However, for short texts they generally suffer from data sparsity because of extremely limited word cooccurrences; thus tend to yield repetitive or trivial topics with low quality. In this paper, to address this issue, we propose a novel neural topic model in the framework of autoencoding with a new topic distribution quantization approach generating peakier distributions that are more appropriate for modeling short texts. Besides the encoding, to tackle this issue in terms of decoding, we further propose a novel negative sampling decoder learning from negative samples to avoid yielding repetitive topics. We observe that our model can highly improve short text topic modeling performance. Through extensive experiments on real-world datasets, we demonstrate our model can outperform both strong traditional and neural baselines under extreme data sparsity scenes, producing high-quality topics.
Cited by 60Open access
Anna Wexler (7422443), Anahita Davoudi (8707998), Davy Weissenbacher (6158144) et al. · Figshare · 2020
Cited by 0Open access
Alan Cooke, Juliet Homer, Lisa Schwartz · OSTI OAI (U.S. Department of Energy Office of Scientific and Technical Information) · 2018
The U.S. Department of Energy (DOE) recently completed a study on how regulators in 16 states have dealt with questions related to distribution system planning (DSP) and distributed energy resource (DER) utilization. During the study, DOE and three of its national laboratories developed a significant amount of information on state-level activities including grid modernization planning; distribution planning; tools for integrating distributed energy resources (DERs) (e.g., hosting capacity analysis and tools for maximizing the value of DERs to the distribution system such as locational value analyses); and linkages between distribution planning and related processes such as cost recovery through general rate cases or rate surcharges. The Oregon Public Utility Commission (Commission) requested assistance from DOE on a list of specific questions related to DSP. The Commission was specifically interested in identifying guidance on the creation of comprehensive and transparent utility distribution system plans; on the linkages between such plans and other utility planning processes such as integrated resource planning; improved forecasting and representation of DERs in distribution system plans and other plans such as integrated resource planning; valuation of DERs; and a roadmap for optimized locational planning for DERs. In response to their interests, we have identified 1) specific topics from the original study that would be of assistance to the Commission and 2) states that could provide potential examples or templates to draw from. In this report, we have summarized and reorganized the pertinent original material by subject. We also developed a matrix that points to specific examples for each topic and provide one or two paragraph summaries of each possible template for each topic area with web links directly to the source documents. For some cases in which states have taken significant steps since the publication of the original DOE study, the original source information has been supplemented to ensure the information provided to the Commission remains current. Finally, we called out a small number of especially pertinent or immediately useful documents and have highlighted these in the introductory section. This report documents the material developed for the Oregon Commission by DOE.
Cited by 15Open access
Sibele Fausto (113102), Fabio A. Machado (113105), Luiz Fernando J. Bento (113109) et al. · Figshare · 2015
Cited by 0Open access
Jinpeng Wen (13137660), Hongxing Han (22605462) · Figshare · 2025
<div> As construction of the Guangdong-Hong Kong-Macao Greater Bay Area advances, talent has become the core driving force for its development. This study quantitatively analyzes 94 talent policies from Guangdong, Hong Kong, and Macao, revealing the strategic focuses and thematic evolution of these policies across different stages from a time-series perspective. It also distills relevant experiences to propose optimization strategies and a development blueprint for talent policies in the region. By combining historical review and thematic modeling, this study categorizes talent policies from the three regions over the past five years (2019–2024) into three phases: initial phase, developmental phase, and deep plowing phase. Theme heat analysis and similarity analysis are then applied to clarify the strategic focus and evolutionary trajectory of each phase, while the implementation effects of the talent policies are evaluated based on four dimensions: talent inflow volume, talent satisfaction, economic impact, and social influence. The Latent Dirichlet Allocation (LDA) model uncovers seven strategic talent themes and their heat distribution across the three phases. The study reveals that “Hong Kong and Macao tax incentives” have consistently served as the cornerstone of talent policy throughout the development process, while also fostering the rapid emergence of related themes. Looking ahead, efforts will be dedicated to building a comprehensive talent ecosystem integrating education, science and technology, and social security, thereby positioning the Guangdong-Hong Kong-Macao Greater Bay Area as an international first-class bay area. Finally, this paper draws on the successful experiences of the Yangtze River Delta, Beijing-Tianjin-Hebei region, and other international bay areas to formulate targeted strategies for the development of the Guangdong-Hong Kong-Macao Greater Bay Area. </div>
Cited by 0Open access